Knowledge Base

20 CJBL knowledge docs + 1 cross-project docs

CJBL Advanced Stats Extraction Framework

CJBL Advanced Stats Extraction Framework

From Play-by-Play to Actionable Intelligence


Overview

Ratings tell you what a player should do. Advanced stats tell you what they actually do. This framework extracts metrics that predict wins better than raw ratings.

Core Principle: Volume × Efficiency = Actual Impact


Data Structure (PBP Format)

Each possession contains:

[Time] [Team] [Score] [Event description]
[Play sequence] > [Action] > [Zone] > [Result]
Set: [Play call]
Closest Defender: [Name], [contestedness]
Defensive Set: [Man/Zone type]
Coverage: [Guard coverage] Big: [Roll coverage] (assignments)

Key extractable fields: - Shot location (distance + zone) - Play type (ISO, PnR BH, PnR RM, Spot-Up, Transition, Post-Up, Off Screen, Hand Off, Cut) - Defender closeness (wide open, open, guarded, tight, v tight) - Defensive scheme faced (Man, 2-3 Zone, 1-2-2 Zone, etc.) - Make/Miss/Turnover outcome - Foul drawn (yes/no)


Tier 1: Efficiency Metrics (Most Predictive)

Effective Field Goal % (eFG%)

Formula: (FGM + 0.5 × 3PM) / FGA

Why it matters: Weights three-pointers appropriately. A 40% 3PT shooter (1.2 PPP) beats a 50% 2PT shooter (1.0 PPP).

Extraction method:

For each shot attempt:
- Parse "made/missed X-foot [shot type]"
- Track 3PT vs 2PT via "three pointer" in description
- Calculate: (2PT_makes + 1.5 × 3PT_makes) / total_FGA

Morey Ball target: eFG% > 52%


True Shooting % (TS%)

Formula: PTS / (2 × (FGA + 0.44 × FTA))

Why it matters: Incorporates free throws. Critical for evaluating volume scorers who draw fouls.

Extraction method:

Track per player:
- Total points (from "X PTS" in parentheses)
- FGA (shot attempts)
- FTA (from free throw lines)

Benchmark: TS% > 55% = efficient scorer


Points Per Possession (PPP) by Play Type

Why it matters: THE efficiency measure. Identifies what actions work vs. what's wasted motion.

Extraction method:

For each play sequence ending in shot/turnover/foul:
1. Parse play type from action chain (e.g., "P&R Ball Handler > High P&R")
2. Track outcome: Points scored (0, 2, or 3) + FTA × 0.75
3. Calculate: Total_points / Total_possessions_of_type

Benchmarks by play type:

Play Type Good Average Poor
Transition 1.10+ 0.90-1.10 <0.90
PnR Ball Handler 0.90+ 0.75-0.90 <0.75
Isolation 0.85+ 0.70-0.85 <0.70
Spot-Up 0.75+ 0.60-0.75 <0.60
Post-Up 0.90+ 0.75-0.90 <0.75
Off Screen 0.85+ 0.70-0.85 <0.70

Tier 2: Usage & Volume Metrics

3PT Attempt Rate (3PAr)

Formula: 3PA / FGA

Why it matters for Morey Ball: A player with great 3PT rating who takes 2 threes per game is useless. This measures whether they actually shoot.

Extraction method:

Count "three pointer" in shot descriptions
Divide by total shot attempts

Morey Ball target: 3PAr > 40% for perimeter players


Rim Attempt Rate (RimAr)

Formula: Rim_attempts / FGA

Why it matters: Morey Ball wants 3s and rim attacks. This confirms players are attacking paint, not settling for mid-range.

Extraction method:

Count shots ≤ 4 feet ("layup", "dunk", "1-foot", "2-foot", "3-foot", "4-foot")
Divide by total FGA

Target: RimAr > 25% for guards, > 50% for bigs


Mid-Range Attempt Rate (MidAr)

Formula: Mid_range_attempts / FGA

Why it matters: Should approach zero in Morey Ball. Identifies players who can't adapt.

Extraction method:

Count shots 10-22 feet that aren't three-pointers
Includes: "jump shot" without "three pointer" at 10+ feet

Morey Ball target: MidAr < 15%


Turnover Rate (TOV%)

Formula: TOV / (FGA + 0.44 × FTA + TOV)

Why it matters: High ratings mean nothing if the player bleeds possessions.

Extraction method:

Parse "turnover (lost ball/bad pass/offensive foul)" events
Calculate per 100 possessions

Benchmark: TOV% < 15% = reliable, > 20% = liability


Assist-to-Turnover Ratio (AST/TOV)

Formula: AST / TOV

Why it matters: Measures ball security and playmaking efficiency.

Extraction method:

Count "assisted by [Player]" occurrences
Divide by turnover count

Benchmark: > 2.0 = excellent, < 1.0 = concerning


Tier 3: Shot Quality Metrics

Contested Shot Breakdown

Formula: Track makes by defender closeness

Why it matters: Identifies shot selection quality and ability to create separation.

Extraction method:

Parse "Closest Defender: [Name], [level]"
Levels: wide open, open, guarded, tight, v tight
Track FG% at each level

Analysis: - High "wide open" % = good off-ball movement or team creating looks - High "v tight" volume + low % = poor shot selection - High "v tight" volume + high % = elite shot maker (rare)


Shot Zone Efficiency

Why it matters: Identifies where each player is most/least effective.

Extraction method:

Parse shot distance and location from action chains
Zones:
- Rim (0-4 feet)
- Paint non-rim (5-9 feet)
- Short mid-range (10-15 feet)
- Long mid-range (16-22 feet, non-3PT)
- Corner 3 (extracted from "corner" in description)
- Above-break 3 (all other 3PT)

Morey Ball priority: Rim > Corner 3 > Above-break 3 > Everything else


Tier 4: Defensive & Hustle Metrics

Defensive Rebounding Rate (DRB%)

Formula: DRB / (DRB + Opp_ORB)

Why it matters: Per-possession rebounding shows actual impact vs. playing more minutes.

Extraction method:

Track "Defensive rebound by [Player]"
Calculate as % of available defensive rebounds when player on court

Benchmark: > 20% = elite rebounder


Steal + Block Rate

Formula: (STL + BLK) / Minutes × 40

Why it matters: Per-40-minute disruption rate.

Extraction method:

Count "stolen by [Player]" and "blocked by [Player]"
Normalize to per-40 minutes

Benchmark: > 4.0 combined = disruptive defender


Opponent Shot Quality Allowed

Why it matters: Measures defensive impact beyond counting stats.

Extraction method:

When player listed as "Closest Defender":
- Track opponent FG% when guarded
- Track shot distance distribution forced
- Track contestedness level achieved

Analysis: Good defenders force: - Lower FG% than team average - Longer shots (further from rim) - Higher "tight/v tight" contest rate


Tier 5: Scheme-Specific (Morey Ball Critical)

Zone Defense Efficiency Differential

Formula: PPP_vs_zone - PPP_vs_man

Why it matters: Zone efficiency differentials reveal scheme-specific weaknesses. (Note: SIUE's originally reported -32% was a parsing error; v2-corrected value is -0.5%. Specific zone types like 2-1-2 (0.794 PPP) and 2-3 Matchup (0.714 PPP) are the real vulnerabilities.)

Extraction method:

Parse "Defensive Set: [type]" for each possession
Track PPP when facing:
- Man-to-man
- 2-3 Zone
- 1-2-2 Zone
- 3-2 Zone
- Other

Red flag: > 0.15 PPP drop vs zone = spacing problem


Spot-Up Efficiency

Why it matters: Morey Ball depends on spot-up shooters punishing defensive collapses.

Extraction method:

Filter plays containing "Spot-Up" in action chain
Calculate PPP and eFG%

Morey Ball requirement: Spot-up eFG% > 50%


PnR Ball Handler Efficiency

Why it matters: Primary action in Morey Ball offense.

Extraction method:

Filter plays containing "P&R Ball Handler"
Track:
- PPP
- % going to basket vs. pull-up vs. pass
- Efficiency by coverage type (Drop, Hedge, Switch, etc.)

Benchmark: PnR BH PPP > 0.85 = usable


Implementation: Python Parser Template

import re
from collections import defaultdict

def parse_pbp(filename):
    """Extract advanced stats from CJBL play-by-play file."""

    player_stats = defaultdict(lambda: {
        'fga': 0, 'fgm': 0, '3pa': 0, '3pm': 0,
        'fta': 0, 'ftm': 0, 'pts': 0,
        'rim_att': 0, 'rim_make': 0,
        'mid_att': 0, 'mid_make': 0,
        'tov': 0, 'ast': 0,
        'drb': 0, 'orb': 0,
        'stl': 0, 'blk': 0,
        'play_types': defaultdict(lambda: {'pos': 0, 'pts': 0}),
        'vs_zone': {'pos': 0, 'pts': 0},
        'vs_man': {'pos': 0, 'pts': 0},
        'contest_levels': defaultdict(lambda: {'att': 0, 'make': 0})
    })

    current_possession = {}

    with open(filename, 'r') as f:
        for line in f:
            # Shot made
            if match := re.search(r'(\w+ \w+) made (\d+)-foot (.+?) \((\d+) PTS', line):
                player, distance, shot_type, pts = match.groups()
                distance = int(distance)
                pts = int(pts)

                stats = player_stats[player]
                stats['fgm'] += 1
                stats['fga'] += 1
                stats['pts'] = pts  # Update running total

                if 'three pointer' in shot_type:
                    stats['3pm'] += 1
                    stats['3pa'] += 1
                elif distance <= 4:
                    stats['rim_make'] += 1
                    stats['rim_att'] += 1
                elif 10 <= distance <= 22:
                    stats['mid_make'] += 1
                    stats['mid_att'] += 1

                current_possession['shooter'] = player
                current_possession['made'] = True
                current_possession['pts'] = 3 if 'three' in shot_type else 2

            # Shot missed
            elif match := re.search(r'(\w+ \w+) missed (\d+)-foot (.+)', line):
                player, distance, shot_type = match.groups()
                distance = int(distance)

                stats = player_stats[player]
                stats['fga'] += 1

                if 'three pointer' in shot_type:
                    stats['3pa'] += 1
                elif distance <= 4:
                    stats['rim_att'] += 1
                elif 10 <= distance <= 22:
                    stats['mid_att'] += 1

                current_possession['shooter'] = player
                current_possession['made'] = False
                current_possession['pts'] = 0

            # Assist
            elif match := re.search(r'assisted by (\w+ \w+)', line):
                player_stats[match.group(1)]['ast'] += 1

            # Turnover
            elif match := re.search(r'(\w+ \w+) turnover', line):
                player_stats[match.group(1)]['tov'] += 1

            # Defensive set
            elif match := re.search(r'Defensive Set: (.+)', line):
                def_set = match.group(1)
                current_possession['vs_zone'] = 'Zone' in def_set

            # Play type
            elif any(pt in line for pt in ['P&R Ball Handler', 'Spot-Up', 'ISO',
                                            'Transition', 'Post-Up', 'Off Screen']):
                for pt in ['P&R Ball Handler', 'Spot-Up', 'ISO', 'Transition',
                          'Post-Up', 'Off Screen', 'Hand Off', 'Cut']:
                    if pt in line:
                        current_possession['play_type'] = pt
                        break

            # Closest defender
            elif match := re.search(r'Closest Defender: .+?, (\w+ ?\w*)', line):
                current_possession['contest'] = match.group(1).strip()

    return player_stats


def calculate_advanced(stats):
    """Calculate advanced metrics from raw stats."""

    advanced = {}

    # eFG%
    if stats['fga'] > 0:
        advanced['eFG%'] = (stats['fgm'] + 0.5 * stats['3pm']) / stats['fga']

    # TS%
    possessions = stats['fga'] + 0.44 * stats['fta']
    if possessions > 0:
        advanced['TS%'] = stats['pts'] / (2 * possessions)

    # Shot distribution
    if stats['fga'] > 0:
        advanced['3PAr'] = stats['3pa'] / stats['fga']
        advanced['RimAr'] = stats['rim_att'] / stats['fga']
        advanced['MidAr'] = stats['mid_att'] / stats['fga']

    # Turnover rate
    usage_possessions = stats['fga'] + 0.44 * stats['fta'] + stats['tov']
    if usage_possessions > 0:
        advanced['TOV%'] = stats['tov'] / usage_possessions

    # AST/TOV
    if stats['tov'] > 0:
        advanced['AST/TOV'] = stats['ast'] / stats['tov']

    return advanced

Quick Reference: Red Flags

Metric Red Flag Meaning
MidAr > 25% Scheme misfit Player won't adapt to Morey Ball
3PAr < 20% (perimeter) Role mismatch Can't space the floor
TOV% > 22% Ball security Bleeding possessions
Spot-up eFG% < 45% Can't shoot Kills spacing
Zone PPP < 0.60 Spacing disaster Team collapses vs zone
Rim FG% < 55% Finishing problem Can't convert attacks

Quick Reference: Green Flags

Metric Green Flag Meaning
TS% > 58% Elite efficiency Maximizing possessions
3PAr > 45% + eFG% > 52% True shooter Volume + efficiency
AST/TOV > 2.5 Ball security Creates without bleeding
Spot-up PPP > 0.85 Gravity Punishes help defense
RimAr > 30% + Rim FG% > 60% Finisher Collapses defenses

Next Steps

  1. Parse all 36 games into consolidated player stat lines
  2. Build comparative rankings vs. OVC averages
  3. Identify scheme fit scores based on shot distribution + efficiency
  4. Create scouting templates for transfer targets using same metrics
CJBL Assistant AD - Project Instructions

CJBL Assistant AD - Project Instructions

Role

You are my Assistant Athletic Director for the CJBL simulation basketball league. Your job is to provide strategic recommendations. You never make decisions autonomously—all actions require my approval.

Interaction Modes

Chat Mode

When I ask questions, provide: - Direct answer first - Supporting reasoning - Risk assessment (what could go wrong) - Alternative options if applicable

Check-In Mode

When I paste new data (roster changes, portal updates, phase announcements), immediately: 1. Identify what changed 2. Flag urgent items requiring action 3. Flag opportunities worth considering 4. Note any conflicts with current strategy

Trigger phrases: "Check-in:" or "Update:" at the start of my message.

Response Format

For Player Evaluations

[Name] | [Position] | [Year] | [Rating if known]
- Scheme fit: [1-5 score with brief reason]
- Risk level: [Low/Medium/High with reason]
- Recommendation: [Action]

For Recommendations

Action: [What to do]
Rationale: [Why]
Downside: [What could go wrong]
Confidence: [Low/Medium/High]

For Phase Transitions

  1. Priority actions (ranked)
  2. Deadlines
  3. Dependencies (what must happen before what)

Constraints

  • Be specific with numbers (not "several" — say "3")
  • Flag when you need data you don't have
  • Distinguish between confirmed mechanics and community speculation
  • Default to conservative recommendations when uncertain
  • Never recommend breaking promises unless tournament implications are clear

Knowledge Hierarchy

When information conflicts: 1. Commish announcements > Community consensus 2. Recent information > Older information 3. Confirmed mechanics > Speculation

Current Context

Refer to Team_Context document for current team state. I will update this document as the season progresses.

Check-In Template

When I start a session, I may paste something like:

Check-in:

Phase: [Current phase]
Days remaining: [X]

Recent changes:
- [What happened]

Current concerns:
- [What I'm worried about]

Questions:
- [Specific questions]

Respond by: 1. Acknowledging the phase and timeline 2. Analyzing changes and their implications 3. Addressing concerns with specific recommendations 4. Answering questions directly 5. Proactively flagging anything I should know

Philosophy Awareness

Always consider my team's offensive philosophy when making recommendations. A great player who doesn't fit the system is worse than a good player who does.

Prestige Awareness

Scale recommendations to prestige tier: - High+ prestige: Win-now pressure, can under-promise playing time - Average: Balance development and winning - Low: Long-term development focus, over-promise to attract talent

Red Flags to Always Call Out

  • Promises that seem unsustainable given depth chart
  • Transfer targets that don't fit philosophy
  • Scheduling decisions that don't match roster strength
  • Recruiting players at positions where we have depth
  • Any action that risks AD Trust unnecessarily
CJBL Core Rules & Mechanics

CJBL Core Rules & Mechanics

Prestige System

9-tier system (lowest to highest): 1. Extremely Low (EL) 2. Very Low 3. Low 4. Below Average 5. Average 6. Above Average 7. High 8. Very High 9. Extremely High

How prestige increases: - Tournament runs (deeper = bigger boost) - Conference titles - Players drafted to JBL (pro league)

How prestige decreases: - Missing tournament (penalty scaled to current prestige tier) - Losing conference record - Sustained underperformance

Prestige affects: - Recruiting access (higher prestige = access to better recruits) - Hot seat pressure (higher prestige = stricter expectations) - Transfer market leverage (players accept less playing time at higher prestige schools) - Coach hiring (some coaches won't accept low prestige offers)


Hot Seat Mechanics

Contract structure: - 6-season initial contracts - Evaluated on: tournament performance, conference finish, W-L record, prestige maintenance

Risk by prestige tier: - High/Very High/Extremely High = highest hot seat risk - Average = moderate risk - Low/Very Low/EL = minimal risk (safe to rebuild)

Contract extensions: - Championship = 6-year extension - Meeting expectations = 3-year extension - Underperforming = 1-year extension

Job security: - Other ADs can apply for your job if you're on hot seat - Hot seat determined by performance vs expectations (prestige-adjusted)


Philosophy System

Core rule: You MUST play your head coach's offensive philosophy. Cannot change without firing coach (offseason only).

Philosophy Overview

Philosophy Pace Primary Actions Best For
Seven Seconds Very Fast (9-10) Spot Up, Transition, PnR Max development, athletes
Pace & Space Fast (7-8) Spot Up, PnR, Transition Versatile, modern, rebuilds
Perimeter Centric Average (5-7) Spot Up, PnR, Off Screen Shooter-heavy rosters
Morey Ball Average (4-6) Spot Up, PnR, Transition Rim/three only, no mid-range
Balanced Average Flexible Safe choice, adaptable
Post Centric Slow (1-3) Spot Up, PnR, Post Up Traditional bigs, grind
Triangle Slow (3-5) Spot Up, Isolation, Post Up Veterans, high BBIQ
Grit & Grind Slow Defense-first TBD

Philosophy Details

Seven Seconds or Less: - Highest pace = most possessions = best for player development (XP) - Requires: ball handlers, athletes, rim runners, fearless shooters - De-emphasizes: post-up, isolation, hand-off - Risk: depth requirements, fatigue management

Pace & Space: - Modern motion offense, ball movement, drive-and-kick - Requires: shooters, passers, off-ball movement, roll gravity bigs - Recommended for rebuilds (adaptable, multiple attack vectors) - Fast pace still good for development

Morey Ball: - Efficiency-obsessed: rim or three only, NO mid-range - Requires: 3PT shooters, isolation creators, athletic rim runners - High variance (live by three, die by three) - Mid-range shooting is wasted rating points

Perimeter Centric: - Find the open man, movement shooter focus - Requires: shooters, passers - Safe three-point philosophy - Mixed community results (some struggle despite good rosters)

Post Centric: - Slowest pace = fewest possessions = worst for development - Requires: post scorers, face-up forwards, distributing guards - Needs rare personnel (skilled post players hard to find)

Triangle: - Complex read-and-react system - Requires: passing bigs, isolation wings, high BBIQ, veterans - TRAP FOR NEW ADs: needs cohesion, slow pace hurts development - Only system with LOW PnR frequency - Avoid unless inheriting veteran roster with high BBIQ

Philosophy Tier List (for new ADs)

S-Tier: Seven Seconds, Pace & Space A-Tier: Perimeter Centric, Morey Ball, Balanced C-Tier: Post Centric (slow development, rare personnel) D-Tier: Triangle (complex, cohesion-dependent, rebuild trap)


Chemistry vs Cohesion

These are SEPARATE systems that both affect performance.

Chemistry: - Based on: personalities, role satisfaction, coaching fit - Affected by: promise fulfillment, role changes, coach changes - Low chemistry = locker room issues, transfer risk

Cohesion: - Based on: time playing together - Seniors have edge (years in system) - Affected by: roster turnover, scheme proficiency - Coach change reduces scheme proficiency


Captain Mechanic

How it works: The captain gives a performance boost to other players when on the floor with them. This is an active, minutes-dependent boost — not a passive team-wide aura.

Selection priority: 1. Minutes matter most — the boost only applies when the captain is playing, so it should be a starter or heavy-minutes player. A bench player with elite leadership but zero floor time wastes the boost. 2. Ideal: Junior or Senior with great intangibles and leadership 3. Open question: Underclassman with better intangibles vs upperclassman with worse intangibles — no confirmed answer on which is better. Use judgment based on the specific intangible gap vs minutes projection.

Implication: Always give captaincy to a regular-minutes player. Do not assign it to a backup who never touches the floor, even if that backup has perfect leadership ratings.

Source: Community intel (confirmed mechanic)


Positional Flexibility

There is no positional proficiency mechanic. If a player has the right ratings and skills for a position, plug him in immediately — no learning curve, no development time needed. (Source: Morgado - USC)

  • System position labels (PG, SG, SF, PF, C) are just labels, not constraints
  • Players routinely play out of position with no penalty: SFs at PG, PGs at SG, SFs at PF, etc.
  • Example: 2049 POTY (Craig at UNC) was a "SF" playing PG
  • Recruiting implication: Target best available players by ratings, not by listed position. A 6'5 "PG" can play SG day one if the skills fit.
  • Roster construction implication: Position versatility is free — evaluate players by skill profile, not position label

Player Development (XP System)

When ratings change: Training Camp (offseason only). Little to no in-season change.

XP earning sources:

Playing Time: - Minutes played (linear scaling) - Road games > home games - Clutch minutes (<5 min remaining, ≤5 pt margin) = bonus - Overtime = extra XP

Team Success: - Close win (1-3 pts) = max XP - Win 4-9 pts = less XP - Blowout win = slightly less - Large loss = 0 XP (critical: playing young guys in blowouts hurts development)

Veterans: - Young players gain XP playing with veterans - Veteran leadership rating multiplies this effect

Playoffs: - Round 1 = ~4x regular season XP - Conference Finals = ~8x regular season XP - Scales by round

Training Camp focus areas: - Two focus areas per player, weighted distribution - Focus areas are BUNDLES (e.g., "Playmaking" = Playmaking + Passing + Ball Handling + Self-Creation equally) - Cannot micro-target single ratings - Usage matters: player who shot 200 threes more likely to improve 3PT than player who shot none

Pro Potential vs Development (COMMISH CONFIRMED): - Pro potential is the player's ceiling — how much "space" they have to grow, not the rate or magnitude of growth. - High pro potential does NOT mean faster or bigger development strides vs low pro potential players. - Low-potential players can still improve; they just have a lower floor/starting point and a lower cap on how good they can get. - Pro potential has no correlation with current college ability — a 3.5★ pro potential player can be a 2.5★ college ability senior. - For CJBL recruiting: Pro potential is irrelevant. Focus on current ratings and college ability.

Development reality: Development isn't just "play young guys." It's "play young guys in COMPETITIVE minutes on the ROAD in CLOSE games with VETERANS while WINNING." Narrow window, especially for low-prestige teams.


Rating Accuracy (Fog of War)

Scouting accuracy by relationship: - Own team: High accuracy - Conference opponents: Average accuracy - Everyone else: Low accuracy (significant uncertainty range)

Star ratings: Summaries, not gospel. System fit matters more than raw stars.


Fatigue & Rest

Halftime rest: ~2 blocks of recovery (not full recovery) Quarter breaks (JBL pro only): ~1 block of rest

Implications: - Fast pace = high fatigue accumulation - Can't just ride starters, rest at halftime, repeat - Depth required for fast-paced systems - Endurance ratings affect fatigue accumulation


Scheduling

Recruiting budget boost: Road games at higher prestige opponents = more recruiting $ Trade-off: Harder games vs better budget

Best practice for new ADs: - Don't schedule all OOC games before recruiting - Save 2-3 slots until after seeing recruiting class - Balance SOS needs with expected roster strength

Warning: If you've applied for another team, DO NOT schedule games (schedules persist with team, not AD)


Walk-Ons

  • Can be released anytime
  • Can be converted to scholarship
  • Wait until AFTER recruiting to convert (don't burn scholarship early)
  • Junior walk-ons must be offered scholarship by end of Transfer Phase 1 or auto-released

Player Release

Players released go to "Division II" (permanently gone). They don't return.


Coach Mechanics

Hiring: - Each phase = 24 hours - Must have Head Coach before Assistant Coaches - If you fail to hire in time, auto-assigned

Firing: - Firing inherited coach (previous AD's hire) = no penalty - Firing coach YOU hired = recruiting budget reduction - Firing can unsettle team, reduce scheme proficiency - Players may transfer after coach change

Extensions: - One extension per coach maximum - Must have 1-2 years remaining (not 0/free agent) - Your prestige must be high enough for their salary demand - Location: My Team > Front Office

Development: - Coaches develop slowly (25+ seasons) - Experience helps with unruly locker rooms - Rookie coaches are viable but less control


Key Thresholds to Remember

  • 16 MPG: Players below this threshold are generally dissatisfied with playing time
  • 85% of promised minutes: Below this risks breaking promise
  • 0 XP: Players in blowout losses earn nothing
  • 3-5 players retained: Expected carryover after philosophy change
  • 2 pitches: Per offseason (transfers + recruiting combined)
CJBL Meta Knowledge

CJBL Meta Knowledge & Community Insights

Undefined Terms Glossary

Black Sunday

Status: Partially defined Likely meaning: Day when hot seat firings happen and coach carousel opens Need to confirm: Exact timing

Tickover

Status: Defined Meaning: End-of-season processing that includes: - Prestige recalculation - Awards distribution - Season stats finalization - Hot seat evaluations


Community-Discovered Mechanics

Ball Dominance (BD)

Mechanic: Determines who terminates possessions - High BD = terminates plays early (more freelance) - Low BD = keeps passing (runs play to completion)

Hierarchy Interaction (Source: Vince, Rockets1 CM): - BD creates a usage floor — high-BD players placed low in hierarchy still command elevated usage - Example: 33% USG player dropped from #1 to #4 → only fell to 29% - Conversely, low-BD players (e.g., BD 11) placed at #1 will accept more touches than they'd seek naturally - Cannot fully suppress high-BD players through hierarchy alone; plan for their usage even when deprioritized - Not all high-usage players handle reduced usage well — some thrive (Moss: most efficient season of career at reduced role), others may not

Offensive Hierarchy Mechanics (Source: Vince, Rockets1 CM)

Primary function: 95% of hierarchy is a lever to control usage distribution.

Top-of-hierarchy special effects (beyond usage): - Interacts with "Clutch: Prefer Star Iso" setting — #1 hierarchy player gets Iso looks in clutch - Gets more touches near end of games even WITHOUT Prefer Star Iso checked - Gets the ball when play breaks down / shot clock winding down → player at #1 MUST have Iso/Self-Creation skills or these possessions are wasted

Heliocentric setting — three-tiered usage bump: | Hierarchy Position | Heliocentric Bump | |---|---| | Top 6 | Base bump | | Top 3 | Base bump + additional bump | | #1 | Base bump + additional bump + third bump |

Strategic implication: To suppress a ball-dominant player's usage, place them below #3 to avoid the compounding Heliocentric bonuses. Placing them at #4-6 still gives the base bump but removes the extra two tiers.

Real-world example (Vince's Austin Rockets, Prime): - Briggs Greenaway (heliocentric monster, 33% USG at #1) → moved to #4 → dropped to 29% USG - Antoine Rigaux (C, BD 11, not a natural shot-seeker) → placed #1 → pulled up to 24% USG - Cardez Moss (SF, career 27% USG) → placed below #3 → dropped to 21% USG, most efficient season of career - Hierarchy order: Rigaux #1, [Wills #2], [Link #3], Briggs #4, Moss #5, DPJ #6 - Final usage: Rigaux 24%, Briggs 29%, Moss 21%, Wills 19%, DPJ 18%, Link 16%

Key takeaway: You can use hierarchy position AND Heliocentric tier breaks to sculpt your usage distribution. Low-BD players at the top pull touches they wouldn't seek naturally; high-BD players below #3 lose the compounding bonuses but still maintain a usage floor from their BD rating.

Hierarchy Does Not Affect Mood (Source: THS - OKC2)

Hierarchy position does not affect player mood/satisfaction. Only role (offer importance) and minutes matter for mood. A Marquee player getting 30 minutes at the bottom of the hierarchy will be fine — they care about minutes and role promise fulfillment, not where they sit in the usage order.

Implication: Use hierarchy purely as a usage-sculpting tool without worrying about upsetting players by placing them low.

Pro Potential Star Scale (Source: THS - OKC2)

Approximate mapping of scouted pro potential to JBL ability stars:

Pro Potential Stars JBL Ceiling
5★ MVP level
4★ All-Star / All-League
3★ Starter
2★ Rotation player
1★ Deep bench

COMMISH CONFIRMED: Pro potential is the player's ceiling — how much "space" they have to grow. Lower-potential players can still improve; they just likely have a lower floor/starting point. High pro potential does NOT mean faster or bigger development strides compared to low pro potential players. (Source: commish)

Additional context (Source: Brandon - Indiana University): Pro potential has no correlation with current CJBL ability. A 3.5★ pro potential player can easily be a 2.5★ college ability senior who only gained half a star over 3 seasons at 10 mpg. Development depends on playing time, TC emphasis, coaching, and luck. Pro potential is not relevant for CJBL recruiting decisions — focus on current college ability and ratings instead.

Captain Boost Mechanic (Source: Community Intel)

The captain provides a performance boost to teammates when on the floor with them. This means: - Captaincy should go to a starter or high-minutes player — the boost is wasted on bench players - Best case: Junior/Senior with great intangibles and leadership - Underclassman with better intangibles vs upperclassman with worse intangibles = open question, no confirmed answer - Never assign captaincy to a backup who doesn't play, regardless of leadership ratings

Pressing Frequency

"Only like 10-15% of the time and most of that is half court" - "Often" press ≠ majority of possessions

Trade Block Effects

Players know they're on block, affects performance (personality-dependent).

Analysis Sample Size Thresholds (Source: Vince, Rockets1 CM)

Analysis Type Directional Signal Trustworthy Data
Setting/hierarchy tweaks 10 games 20 games
Lineup combinations 100 minutes 200 minutes

Rule: One change at a time. 10 games to sniff-test, 20 games to commit or revert.

Locker Room Transfer Chances

Even happy players show 20%+ transfer chance. System tuned to encourage movement.

Initial List

Definition: A prospect's top 10 schools that he is interested in — regardless of whether those schools have expressed interest in him.

Key Insights (Community Confirmed): - Not being on a prospect's Initial List is not disqualifying - "Easy to overcome" with proper recruiting effort - Only becomes difficult "if multiple top 10 teams are offering him"

Strategic Implications: - Don't eliminate targets just because you're not on their Initial List - Focus on being first to offer - Use coach visits and campus visits strategically - Deploy pitch only when facing multiple competing offers from schools on his top 10

Coach Visit Success Rates

  • HC Visit: Higher success chance, lower failure chance, higher cost
  • AC Visit: Lower success chance, higher failure chance, lower cost

Rule: Use HC visits on must-get Tier 1 targets where failure is unacceptable.

Campus Visit Timing Bonus

Offering a campus visit in an earlier phase gives a bonus compared to schools that offer in a later phase. This stacks with the general "early actions have greater impact" mechanic but is specifically confirmed for campus visits. Implication: Don't wait to offer campus visits on your top targets — being first matters.

Pitch ("Bag") Timing Bonus (Source: THS - OKC2, confirmed by Morgado - USC)

Note: "Dropping a bag" is community slang for using a pitch on a recruit.

Offering a pitch in an earlier phase provides more benefit than offering it in a later phase. This is consistent with the general "early actions have greater impact" mechanic — confirmed to apply to virtually all recruiting actions (calls, visits, campus visits, pitches). The earlier you commit resources, the greater their impact on the recruit's evaluation.

Implication: If you've decided to use a pitch on a recruit, deploy it as early as possible. Waiting until Phase 4+ loses the timing multiplier.

Pitch Timing (Source: James - Jaguars1, UNC/NMSU)

Risk/reward tradeoff: - Early pitch (Phase 3) = bigger impact on recruit's decision. If he's your guy, pitch now. - Wait (Phase 4) = more time to confirm he's the right target and avoid wasting a pitch against too-stiff competition. - Pitches do NOT roll over year to year — use them or lose them. (Source: robinsmd - Evansville)

Implication: Since pitches expire, the only reason to delay is uncertainty about the target or competition — not to bank pitches for later.

Recruiting Call Limit

Max 3 calls per recruit — cannot offer more. (Source: Secret - UNL/UWGB)

Not calling a recruit = knocked down when phase runs. If you have calls available, always use them on active targets or you lose ground passively.

pitch Budget

2 pitches per offseason, shared between transfers and recruiting. Plan allocation carefully — once spent, they're gone for the cycle. (Source: Community Intel, 2026-02-10)

Commitment Mechanics (Source: THS - OKC2)

  • Commits are not phase-locked — once a recruit has sufficient interest, he can commit at any time from Phase 5 onward.
  • Early decider timeline: Deciding phase starts at Phase 5, can commit as early as Phase 6 if they like the offer.
  • Interest level is the driver, not a specific phase window.
  • Early deciders with no offers: They do NOT enter deciding until the next phase flip. If they get a good enough offer, they'd likely become a decider at that phase flip. (Source: Coach Martin - Georgia Tech/Western Illinois)

Upgrading Offer Importance Mid-Recruiting (Source: PDM - St. Francis, THS - OKC2)

Upgrading offer importance during deciding phase (e.g., Starter → Marquee) does get considered in the recruit's final evaluation. It increases chances but not as much as if done with the initial promise. It doesn't move the interest bar much, but if interest is already maxed, the bar doesn't matter — it's the offer comparison that counts.

Implication: Front-load your best offer when possible. Late upgrades help but at diminished returns.

Scholarship Revocation Penalty (Source: Calvin - Michigan State/Campbell)

Revoking a scholarship offer has two consequences: - Same phase as offered: Lose interest in that recruit (no AD trust penalty). - Later phase than offered: Lose interest in that recruit plus a small AD trust hit.

Implication: If you need to revoke, do it the same phase you offered to avoid the trust penalty. Be deliberate with scholarship offers — don't offer speculatively if you might need to pull back.

Always Execute All Actions on Top Targets (Community Consensus)

Even with a full interest bar, no competition, and an active offer — keep doing every available action (visits, calls) on your top targets. The community consensus is unanimous: don't get complacent and redeploy resources elsewhere until a player is actually committed. The only exception is if the player is already in deciding phase, where visits may be less impactful. (Sources: Calvin - Michigan State/Campbell, Trevor - Skyhawks4/UTEP/Bellarmine, THS - OKC2)

Prestige Gap in Recruiting (Source: Morgado - USC, Scottie - Eastern Washington)

If programs 3-4 prestige levels above you are committing major resources to a recruit (Marquee/Important offer, 24-28+ min, pitch), competing against that is a losing play. Focus on realistic targets where prestige gap is manageable.

EL recruiting reality (Source: Scottie - Eastern Washington): At Extremely Low prestige, you have effectively last pick. Focus on recruits with no other offers, and only compete against other EL and VL teams. Fighting above that is a losing play.

Especially relevant for SIUE (Very Low prestige): Don't burn pitches or scholarships fighting VH/H programs for shared targets. Target recruits where the competition is EL/VL/Low at most.

Low-Prestige Recruiting Tactics (Source: THS, Coach Martin, Jackie, Kevin, Josh, Bill, Brad — community thread)

Can EL/VL beat A/AA teams? Yes, but it requires stacking every advantage:

How to win above your weight class (when you must): 1. Offer a significantly better role — Marquee/Important + 28+ min vs their Rotation/12. The offer gap must be large enough to overcome prestige deficit. 2. Scout the competition's roster — Check what they're returning, what they've already signed. Gauge whether their offer to your target is likely strong or an afterthought. 3. Player values are FUNCTIONAL, not flavor — Pay close attention to what the recruit values: - Low "Prestige" value = much more winnable for low-prestige teams. High prestige value = gravitates to bigger programs even with a worse offer. - High "Distance" value = local teams have a major edge - High "Playing Time" value = your 28-min offer beats their 12-min offer - Academics = if your school has good academics (rare at EL) - Even EH programs lose when they mismatch on recruit values. UConn (EH) missed on Leslie (prestige was his #1 value, UConn had M28 + pitch, only EH on him) — went to Wichita State (H), likely because of academic preference (Wichita = EL academics) and slight edge in success. Prestige being #1 doesn't make it the ONLY factor. (Source: Mr Clean - UConn, THS - OKC2) 4. Add a pitch ("bag") when close — Josh (VL) beat an AA team with IP/28 + pitch on an in-state recruit who valued being close to home 5. Pitch if you're close and think you can seal it (Source: Coach Martin)

The safer strategy (what you should do most of the time): 1. Target uncontested recruits — guys with no other offers or only EL/VL competition 2. Prioritize early deciders — lock them up before higher prestige teams can pivot to them late (Source: Bill, Kevin) 3. Find early-deciding guys you can win quick and uncontested — the safest path at EL/VL (Source: Kevin). Bigger schools sometimes save recruits as pivot targets — if you identify an early decider they haven't committed to, you can grab him before they arrive. 4. Be prepared to lose contested battles — "EL prestige life is largely about losing most contested guys" (Source: Kevin). Budget your pitches and emotional energy accordingly. 5. Always a luck element — even doing everything right, you can lose. Brad (L prestige) lost with I/28 + pitch against H prestige offering Future Starter/12. The game has variance. Mr Clean (EH, UConn) has "never landed my favorite recruit entering phase 1, for any school." Multiple valid strategies exist — experienced ADs with years of success fundamentally disagree on individual player evaluations. 6. Don't chase favorites — chase the gettable. You will generally NOT sign your favorite recruits. The chances of them wanting your specific program are small. Excellent recruiting classes are still possible without getting your top targets — the difference between your #1 choice and a very good alternative is usually small, but the difference in gettability can be enormous. (Source: Morgado - USC)

Real examples: - Jackie (Butler, A) lost to an EL team that had a better offer — Butler only had Rotation/4 - Josh (VL) beat AA with IP/28 + pitch on in-state recruit who valued distance - Kevin (EL with Miami OH) lost all contested battles vs higher prestige, won only vs other EL teams and uncontested - Brad (L) lost to H prestige despite better offer — prestige weight can override - PDM (St. Francis, VL) went 3/3 on contested recruits in his first cycle: beat BA in 1v1, beat BA+AA in 1v2, beat VL+A in 1v2. All "punching up." Acknowledges luck was involved, but demonstrates VL can consistently win contested battles when stacking advantages. - Morgado (USC, EH) got NONE of his top 3 bigs (Lefort/Duke, Dozier/VT, Esteller/UVA) but still landed a very good class (#4 Rozhkov, #37 Bowen, #64 Zabelin, #573 Sabb) by pivoting to gettable alternatives - Butter (Dayton, H) gave Marquee + pitch to Ndukwe (#14 SF) instead of swinging at Rozhkov — filled the must-have SF role with less competition. "Worked out." - Mr Clean (UConn, EH) keeps backup options "warmed up" with coach visits every phase — if the primary target falls through, the backup is already at high interest

Pitch ("Bag") Strategy — The Powell Doctrine (Source: Morgado - USC, THS - OKC2, Vince - Rockets1, sabotai - La Salle)

Note: "Dropping a bag" / "bag money" is community slang for using a pitch. You only get 2 pitches per offseason (shared between transfers and recruiting), so deployment strategy is critical.

Core principle: The pitch is a tiebreaker — an additive factor that moves the mean of your random variable. It cannot be a tiebreaker if all resource commits are the same.

When to use a pitch: 1. Overwhelming superiority — You can slam the door shut and make it a near-guaranteed win you absolutely need. USC has never lost on any pitch because they only deploy them when they'll win with overwhelming superiority. (Source: Morgado) 2. Thin competition — Only 2-3 teams are offering, and you're confident you'll make the final 3. Using a pitch in a 3-way race gives you a meaningful edge. (Source: THS) 3. Must-win scholarships — When filling a starting spot or critical roster need, offer to win. "Winning is paramount." (Source: Morgado)

When NOT to use a pitch: 1. Crowded fields (4+ teams also pitching) — If everyone at the same prestige level is throwing pitches, there's no differentiator. It's down to RNG. A pitch in a 6-way race is a crap shoot. (Source: THS, Morgado) 2. Low gettability targets — A 33% chance at an outstanding player is inferior to a 95% chance at a very good player who fills your needs. (Source: Morgado) 3. Home run swings on surplus players — If the recruit is nice-to-have but not must-have, the pitch is better saved for a fight you need to win. (Source: Vince, Butter)

Statistical framing (Morgado): "I'm expecting the pitch to be an additive factor that moves the mean of my random variable. I want the 95% confidence intervals of my package to not overlap with opposing offers. Yes, there is some nonzero tail probability that someone gets a superfreak dice roll, but what can you do?"

Exception — high-upside gambles: sabotai (La Salle) pitched Jackson Hodge (generational talent, appears every 3-4 seasons) knowing it was a risk because La Salle was a frontrunner from Phase 1. Lost, but would do it again for the next Hodge/Deprez-level player. Sometimes the upside justifies the risk if your baseline position is strong.

Adversary Reading & Contingency Planning (Source: Morgado - USC, Mr Clean - UConn, Vince - Rockets1)

How well you plan alternatives and react to other programs' behavior on phase flips dictates recruiting success.

Two questions to always be asking: 1. Who do I have to beat to land this recruit? — Scout the competition's offers, prestige, and resources. (Source: Mr Clean) 2. What's my play if I swing and miss? — Have warmed-up backup options ready at all times. (Source: Mr Clean)

Reading the field: - Watch interest bars and competing school lists on each phase flip — the landscape changes constantly - If competition thins out on a recruit you didn't expect, consider pivoting to them (Morgado pivoted to Rozhkov when competition was thinner than expected on Zabelin) - If competition intensifies on your primary target, be ready to cut bait and redirect to backups (Vince dropped Yusong and Esteller when VH/EH pressure made the wins unlikely, pivoted to Weidenaar where only AA schools competed)

Keeping backups warm: - Do coach visits on 2-3 backup recruits every phase even when your primary target looks good (Source: Mr Clean) - This lets you immediately increase offers on backups if the primary falls through - Missing on a big recruit isn't the end of the world if your backup is already at high interest

The pivot is the skill: Original targets often don't work out — Morgado's final class of 4 included only 1 player from his initial Phase 0 targets. USC pivoted 3 times based on reading competition. The ability to read and react faster than opposing ADs is the core competitive advantage.

Roster Composition & Class Balance (Source: Morgado - USC, Anthony - Warriors2/Bradley)

Class balance matters — target ~3 players per class year (Fr/So/Jr/Sr) with 13 scholarships. The system targets an average of 3 vacancies per recruiting class.

Using transfers to rebalance classes: - If your class distribution is skewed (e.g., 1 SO, 5 SR), deliberately target SOs in the transfer portal to rebalance - Transfers are a great opportunity to reshape class cohorts — the portal pool is large enough to find decent players at any class year - Plan roster composition deliberately across BOTH transfers and recruiting (Source: Morgado)

You don't need 13 scholarship players. An 8-player rotation is easy to set up. Carrying 10-11 scholarship players is fine if your rotation is strong and you'd rather save scholarships for next year's class. (Source: Anthony - Warriors2/Bradley, DeaconBuc16 - UMBC)

Counterpoint — balance isn't mandatory in Year 1. PDM (St. Francis, VL) deliberately skipped class balance in his first cycle (4 FR, 1 SO, 3 JR, 5 SR) and was happy with the results. His philosophy: "just went after the guys that fit the best." Class balance is the most consistently recommended advice, but prioritizing fit over balance can work, especially when rebuilding. (Source: PDM - St. Francis)

Ideal rotation construction: 2 PGs, 3 Wings, 3 Bigs = 8-man rotation. (Source: DeaconBuc16)

Pipeline Recruiting Boost (Source: ChrisSpi - Stetson)

Pipelines provide a small initial bump to player interest. The effect is "probably quite small" — not a major factor in recruiting decisions. Build pipelines organically through commitments rather than chasing pipeline bonuses.

Positional Flexibility (Source: Morgado - USC)

  • No positional proficiency mechanic — there's no learning curve for new positions. If a player has the right ratings/skills, plug him in at any position immediately.
  • System position labels (PG, SG, SF, etc.) are just labels, not constraints. "Positions are a social construct."
  • Example: 2049 POTY (Craig at UNC) was a "SF" playing PG. SFs at PF/PG, PGs at SG all work fine.
  • Recruiting implication: Target best available players by ratings, not by listed position.

Coaching Attribute Priority (COMMISH CONFIRMED)

Priority Order: 1. Tactics — Most critical attribute 2. Coach Offense / Coach Defense — Core competency 3. Adaptability — Prevents "locked state" tactical rigidity 4. Position-specific (Guards, Forwards, Centers) 5. Culture, Patience, etc.

What Adaptability Does: - Low Adaptability = coach gets stuck playing one way - Cannot flex within the engine - Cannot do things "outside the box" - Creates ceiling regardless of talent


Community Philosophy Rankings

S-Tier (fun + effective): - Seven Seconds or Less (high variance) - Pace & Space (adaptable, recommended for rebuilds)

A-Tier: - Morey Ball (fantastic with right personnel) - Balanced (quietly effective)

B-Tier: - Post Centric - Perimeter Centric

F-Tier: - Triangle (terrible for rebuilds)

Morey Ball Community Notes

Mr Clean: "Fantastic with right personnel... Just get a prime Zbro" (tongue-in-cheek about difficulty)

Key insight: Morey Ball is harder in college than pro because: - Harder to recruit elite shooters at low prestige - Less roster flexibility - Development time needed for non-shooters


Phase 0 Recruiting Insights

Class Perception Pattern

Every class looks weak in Phase 0. They turn out ok. Don't panic.

Efficient Scouting Workflow

  1. Regional filter first — Focus on home state + surrounding states
  2. Scout to Level 4 before expressing interest — Character filter eliminates ~77%
  3. OutS rating > 3PT% — Statistics don't match ratings
  4. Budget allocation: ~$950/Level 4 scout (Great Lakes)
  5. Express interest last — Only on clean-character survivors
  6. Target 7-10 clean prospects — Expect to land 1-2 at Very Low prestige

Ratings vs. Statistics Disconnect

High school 3PT% does NOT correlate with OutS rating. Trust ratings, not statistics.

HS Box Score Stats — What They Actually Tell You (Source: 2050 Class Analysis, n=450 with ratings, 21K with stats)

HS stats predict archetype (guard vs big), not talent level. SPG (r=+0.794 vs avg rating), RPG (r=-0.742), BPG (r=-0.726) have the highest correlations with skill ratings — but these reflect position separation. High-SPG recruits have guard skills (Speed, Handling, Perimeter D), high-RPG/BPG recruits have big skills (Rim Protection, Post, Rebounding).

Best HS stat → rating mappings: - BPG → Rim Protection (r=+0.922, R2=84.9%) — extremely reliable - 3PAr → Spacing (r=+0.912, R2=83.1%) - RPG → Speed (r=-0.903, R2=81.5%) — inverse (guards are fast, bigs rebound) - SPG → Off-Ball Defense (r=+0.892, R2=79.5%) - APG → Passing (r=+0.814, R2=66.2%) - PPG → Basketball IQ (r=+0.270, R2=7.3%) — BBIQ is nearly invisible from box scores

BBIQ warning: The #1 college win predictor (coef 0.0186, 1.5x next skill) has only 7.3% variance explained by the best HS stat. You CANNOT identify high-BBIQ recruits from their HS box scores alone — scouting cards are essential.

PPG is the best balanced talent indicator (r=+0.575 vs avg rating) because it doesn't separate as strongly by position. Use PPG for rough talent ranking, but trust ratings for role-specific evaluation.

Star rating predictors: PPG is the best single stat for star rating (r=+0.430, R2=18.5%). But Production Score (r=+0.495) and Value Score (r=+0.463) from the model beat raw stats. Morey Score does NOT predict star rating (r=+0.089, R2=0.8%) — Morey Ball fit and raw talent are orthogonal measurements.

HS PPG by star rating: 1-star avg 13.1 PPG, 3-star 13.3, 4-star 15.1, 5-star 18.1. Overlap is enormous — PPG alone can't reliably distinguish stars from non-stars.

Character Filtering Data (2050 Class -- Phase 2 Updated)

  • Sample: 16 players scouted to Level 5 with full character data
  • Clean character rate: 20% (3 of 16). Worse than 23% Phase 1 estimate.
  • "Maddeningly inconsistent" rate: ~80% of scouting reports contain this flag

CRITICAL UPDATE (Phase 2 L5 results): - "Maddeningly inconsistent" has ZERO correlation with Discipline. Players with Disc 20 (maximum) still get tagged "inconsistent" (Iruafemi, Ratcliff, Crossland). The flag describes on-court performance variance (likely low BBIQ), NOT character discipline. - Positive WE language predicts WE only, NOT Disc/Prof. "Obsessive dedication" = high WE (Abdallah WE 19) but Disc/Prof can still be terrible (Abdallah Disc 7, Prof 7). WE, Disc, and Prof are fully independent ratings. - Personality types provide signal: Workhorse = guaranteed CLEAN. Dedicated = guaranteed CLEAN (same intangible signature as Workhorse). Professional = reliable Disc/Prof. Troublemaker = catastrophic fail. Leader = high Prof but Disc varies (19% CLEAN). Driven = high WE/Prof but Disc varies (12% CLEAN). Introverted = high WE but Disc/Prof vary (5% CLEAN). Personality does NOT predict skill ratings (ANOVA p=0.40). - Prof 20 is NOT a strong positive signal. Many players with poor Disc have Prof 20 (Anderton, Bingaya, Glasford, Stockton, Crossland). Prof appears easier to max than Disc. - Clean character threshold: WE 15+, Disc 12+, Prof 12+ with no Troublemaker personality. Below Disc 10 or Prof 10 is disqualifying regardless of WE.


Useful URLs

  • Recruit Stats: jblfl.com/cjbl/recruit-stats
  • Transfer Portal: jblfl.com/cjbl/transfers?year=[YEAR]
  • Coach Free Agency: jblfl.com/cjbl/coachfreeagency?year=[YEAR]

Season Timeline (Typical)

  1. Season ends
  2. Tickover (prestige updates, awards, hot seat evaluations)
  3. Coach carousel opens
  4. Transfer Portal Phase 1-3
  5. Recruiting Phase 0-10
  6. Walk-ons
  7. Training Camp
  8. Scheduling
  9. Season begins

Note: Tickover happens BEFORE coach carousel. This means: - Hot seat evaluations use updated prestige - ADs apply for jobs knowing their new prestige tier - Prestige at carousel = prestige post-tickover

Offseason Phase Sequence

  • Phase 3: Scholarships can be offered. Pitches available (earlier = bigger impact).
  • Phase 5: EARLY players enter deciding phase. Can commit as early as Phase 6 if interest is sufficient.
  • Phase 7: MID players commit
  • Phase 9: LATE players commit

Note: Commits are not strictly phase-locked. Once a recruit reaches sufficient interest, he can commit at any time from his deciding phase onward — Phase 5/7/9 are when each tier starts deciding, not the only window they can commit.


SIUE-Specific Learnings (2049-2050)

Zone Defense Vulnerability (CORRECTED — v2 Parser Final)

Original analysis claimed: 0.38 PPP vs zone (catastrophic -32% drop)

Corrected v2 analysis found: 0.961 PPP vs zone, 0.966 PPP vs man (-0.5% — essentially no difference)

The original finding was a PBP parsing methodology error (shots with "draws foul" swallowed by foul handler, all 3PT counted as 2PT, inverted turnover attribution). Zone offense is not a team-level weakness, but specific zone types (2-1-2 at 0.794 PPP, 2-3 Matchup at 0.714 PPP) do require preparation.

Zone Performance by Player (v2 Corrected)

Player Man PPP Zone PPP Diff Note
Totten 0.907 1.312 +0.405 Elite zone killer
Topercer 1.096 1.234 +0.138 Zone killer (graduating)
Levett 1.172 1.250 +0.078 Great (graduating)
Jihad 1.172 1.195 +0.023 Neutral
Coddington 1.032 0.970 -0.062 Slight zone weakness
Ross 1.093 0.800 -0.293 Zone liability

Road Performance

  • 1-13 road record vs 15-7 home
  • Root cause: Defensive breakdown on road, not offensive
  • Road offense actually scored MORE than home
  • Mental toughness manifests defensively

Philosophy Transition (2050)

  • Balanced → Morey Ball
  • Josh Carrington fired (0% penalty — inherited)
  • Taavi Allen hired (Off 17, Def 13, Tactics 10, Adapt 8)
  • Noah Davis hired as AC (Def 17, Tactics 10, Adapt 9)
  • Critical gap: Both coaches weak in most important attribute (Tactics)

Coaching Staff Limitation

Problem: Allen (Tactics 10, Adapt 8) + Davis (Tactics 10, Adapt 9) = cannot adapt mid-game

Implication: Staff cannot counter opponent adjustments or flex scheme. Must win with talent execution, not tactical outmaneuvering. Simplify schemes.

Morey Ball Roster Fit Assessment

Player OutS Fit
Coddington 18 ★★★★★ Perfect
Robinzine 18 ★★★★★ Perfect (1-year)
Mimms 18 ★★★★ Excellent
Totten 16 ★★★ Acceptable
Sliwoski 15 ★★★ Acceptable
Kunchik 12 ★★ Below threshold
Sanderson 9 ★ Poor — PROBLEM
Jihad 6 ★★ Rim-runner only

Transfer Portal Acquisition (2050)

  • Loyd Robinzine (SF, Sr) — OutS 18, Gravity 18, On-Ball D 19
  • 1-year rental solving SF shooting gap
  • Character concern: "minimal appetite for extra work"
  • Promise: Starter, 28 MPG, no pitch used

Recruiting Priority (2050 Class) -- POST-HTML EXTRACTION UPDATE

Strategy shift: Guard shooting is #1 predictor (r=+0.688). HTML page extractions revealed character failures in key Group A targets.

Priority Player Pos OutS Character Interest Commit Notes
#1 Emanuel Delaney PG 18 CLEAN (20/15/16) 4th of 10 EARLY Only clean PG. pitch reserved. Green Bay leads.
#2 Jon Haug SG 17 BORDERLINE (20/11/10) 2nd of 3 Mid ELEVATED -- ONLY Morey Ball philosophy match. Ultra-local (13 mi). Lowest competition.
? Clent Seumalo G 19 FAIL (20/8/9) 2nd of ~5 Mid Best OutS but Disc 8/Prof 9 both below 10 floor. Character decision pending.
#3 Sai Iruafemi SG/SF 15 CLEAN (20/20/20) Very High Mid Perfect character trifecta. 7'1 WS. Local (73 mi).
#4 Kenny Danridge PG 18 PENDING (L3) Very High Mid Scout L5 Phase 3. Size risk: 5'9/6'0 WS.
#5 Ikenna Ranson SF 15 CLEAN (20/13/14) High EARLY Iron man 3&D. End 19.
#6 Jules Toone PG/SG 17 CLEAN (20/19/20) -- -- Best character. End 11 limits minutes.
DROP Cannon SF 16 HARD FAIL (Disc 6) 1st of 3 EARLY Disc 6 catastrophic. Morey fit also weak.
DROP Mangrum SF 18 -- Out of running EARLY SIUE out of running (Above Avg interest).
DROP Antwine W 16 FAIL (Disc 5) -- EARLY Character fail.

Scholarship Allocation (2 available): - If enforcing character threshold: Delaney + Haug (best clean/borderline combo with Morey Ball fit) - If accepting Seumalo's character risk: Delaney + Seumalo (best raw talent combo) - If Delaney lost: Haug + Iruafemi (system fit + perfect character)

Critical Finding: PG Barren

Every PG alternative to Delaney failed character screening: - Allegri, McGruder, Rosga, Duckworth, Hockman -- all failed - Danridge PENDING L5 (scout Phase 3) but 5'9/6'0 size is a real concern

If Delaney goes elsewhere, no proven backup PG exists in this class. Contingency: Trust Mimms, address PG in 2051.

Key HTML Extraction Finding: Philosophy Match

Haug is the ONLY target in the entire pool whose preferred philosophy is Morey Ball. All other major targets (Delaney, Seumalo, Cannon) prefer Balanced. Combined with Low prestige weight (47), ultra-local distance (13 mi), and only 3 competing schools, Haug is SIUE's most closeable target.

Scaling from Small Conference to P6 — What Changes (Source: Secret - UNL/UWGB, Kevin - Jailbirds3/Syracuse)

Question: What differentiates impactful players in small conferences vs P6?

Community consensus on what matters MORE at higher prestige/P6: 1. Defense and size — P6 programs are bigger and can punish undersized or weak defenders. You can get away with defensive liabilities in small conferences; in P6, they get exploited. (Source: Secret) 2. Basketball IQ — Small conferences have fewer high-IQ players, so you can win with athletes who make mistakes. At P6, opponents exploit low-IQ players. (Source: Kevin)

Implication for SIUE (Very Low, OVC): - At our level, we can prioritize offensive fit (shooting, spacing) and accept some defensive/size/IQ gaps - As we climb prestige, defense, size, and BBIQ become increasingly non-negotiable - This aligns with the College Ridge model: BBIQ is the #1 win predictor (2x any other skill), and its importance likely scales with competition quality

Open question: Whether it's fundamentally different (new skills needed) or mostly the same philosophy at higher rating thresholds. The thread leans toward "same archetypes, but defense + size + IQ become harder to sacrifice."


Role Player Efficiency Principle

Stars (25%+ USG): TS% 54%+ acceptable — they take difficult shots Role players (<15% USG): TS% 58%+ required — should only take good shots

Inefficient role players = wasted possessions. Bench them regardless of counting stats.


PnR Execution: College vs Pro Paradox (2026-02-11)

Finding: PnR Execution is NEGATIVE in the college Ridge win model (coefficient -0.0077) but is the #2 pro win predictor (+0.0252). This is NOT a contradiction — it's a level-of-analysis artifact.

Why it's negative at the team level: - The college model predicts wins from team-average ratings across all roster players - PnR Execution is a role-specific skill — only the ball handler and roll man need it (2 of 5 players on the floor) - High roster-wide PnR Execution means too many players were recruited for PnR skills, signaling a roster over-indexed on bigs/roll men and lacking shooters - The other 3 players on the floor need spacing skills (OutS, Gravity, BBIQ) — if they have high PnR instead, the team is misbuilt

Why it's positive at the pro level: - Pro model uses individual player ratings mapped to team wins — captures the VALUE of a great PnR executor without penalizing the roster for non-PnR players having low PnR ratings - At the pro level, elite PnR Execution in the right players drives offense

Recruiting implication: - Do NOT recruit broadly for PnR Execution — it's not a roster-wide skill - DO target PnR Execution specifically for PnR Ball Handler (PG) and Roll Man (C) roles - For wings and shooting guards, PnR Execution is irrelevant — prioritize OutS, Gravity, BBIQ, Help Defense - A prospect with low PnR Execution is NOT penalized unless they're being recruited for a PnR-specific role

Source: CJBL College Ridge regression (368 teams, R2=0.411) vs JBL Pro Ridge regression (36 teams, adj-R2=0.642)


Corrected Morey Ball Analysis Reference

Date: February 6, 2026

Cross-league analysis of 31 confirmed Morey Ball teams revealed critical strategic corrections: - Guard shooting (not PnR BH) is #1 predictor at low prestige (r=+0.688) - SF OutS importance was overstated (negligible after correction) - Post-scoring bigs are toxic (r=-0.631) - Too many creators hurts (r=-0.443) - Lighter, faster rosters win (avg weight r=-0.568) - Shooter stacking works (r=+0.576) - Wichita State benchmark was INVALID (wasn't running Morey Ball)

Full document: CJBL_Morey_Ball_Corrected_Analysis.md


Defensive Archetypes — Descriptors, Not Modifiers (Source: THS - OKC2, Donte - MXC2)

Status: Partially confirmed — no known penalties, but some uncertainty remains.

Defensive archetypes (Point of Attack, Wing Stopper, Chaser, Rim Protector, etc.) are descriptors of what a player is suited for based on their ratings, not bonus/penalty modifiers tied to specific defensive actions.

  • No confirmed penalty for guarding actions outside your archetype (e.g., Point of Attack player guarding off-screen or spot-up actions)
  • Use archetypes as a guide for assignment, not a hard constraint

Unresolved question — Wing Stopper on PG: - Historical community discussion about wing stoppers struggling when assigned to guard point guards, even if they're the team's best on-ball defender - Theory: the specific actions a PG runs (PnR ball handling, dribble penetration) may not align with what a wing stopper's ratings are optimized for - THS has done it and reports it "didn't seem disastrous" but has no empirical proof either way - No consensus. Treat as a yellow flag, not a hard rule. If your wing stopper has strong on-ball + help defense ratings, it's probably fine.

Practical implication: Don't overthink defensive archetypes. Assign defenders based on their actual ratings (On-Ball Defense, Help Defense, Steal, Quickness, etc.) rather than archetype labels. But be aware that positional matchup quirks may exist even if unconfirmed.


Hidden Ratings — Impact & Detection (Source: THS - OKC2, Scottie - Eastern Washington, Donte - MXC2, Butter - Barons)

Mechanic: Hidden ratings have roughly equal impact on winning as visible ratings. A player with mediocre visible ratings but strong hidden ratings can post elite advanced metrics (BPM, on-off splits) that don't match the eye test on his scouting card.

Case study: ML Reis (Eastern Washington) — unremarkable visible ratings but one of the best BPMs in the country and 2nd-best on-off splits on a top team. Explanation: great hidden ratings + highly cohesive system full of playmakers and floor spacers amplifying his output. On a bad team with unskilled players, his stats would crater.

Key hidden ratings to watch for clues: - Clutch — late-game performance - Consistency — game-to-game variance - Positioning — defensive/offensive awareness beyond visible ratings - Durability — injury resistance - Turnover Proneness — ball security independent of Ball Handling rating

How to detect hidden ratings when recruiting: - ~30% from high school stats (performance beyond what visible ratings explain) - ~30% from scouting report blurbs (strengths/weaknesses section drops clues) - ~40% luck (Source: Scottie - Eastern Washington)

Implication for recruiting: Don't dismiss prospects whose HS stats outperform their visible ratings — hidden ratings may explain the gap. Conversely, prospects with great visible ratings but poor stats may have bad hidden ratings. Always read the scouting blurb carefully for clues about Clutch, Consistency, Turnover Proneness, etc.

Implication for player evaluation: Advanced metrics (BPM, on-off, PER) capture hidden rating effects that the scouting card doesn't show. A role player posting elite on-off splits likely has strong hidden ratings even if his card looks pedestrian. System context matters too — the same player on a bad team may look average.


Community Intel (Unconfirmed)

Bigs may need rebounding + scoring ability, not just spacing/rolling.

Source: Community member, not Commish-confirmed.

UPDATE (Corrected Analysis): Cross-league data CONTRADICTS this for Morey Ball specifically. Big insideShot (r=-0.631), offensive rebounding (r=-0.347), and finish (r=-0.364) are all negatively correlated with winning at low prestige. For Morey Ball, bigs should stretch or rim-run, not post-score or crash boards. This community intel may apply to other philosophies but is harmful for Morey Ball.

CJBL New User Guide

CJBL New User Guide

Source: Official CJBL documentation


Welcome to the CJBL

As an Athletic Director of a CJBL team, you are responsible for hiring a coach, recruiting players and managing your roster in an attempt to reach (and win) the CJBL Tournament.

The CJBL is part of the JBL universe, running for over 30 real-life years. Players recruited from high schools may go on to the pro league (JBL) or international leagues after graduating. All teams (college and pro) are run by humans.

What Makes CJBL Stand Out

  • Depth: The simulation engine takes into account hundreds of factors from player ratings and team chemistry to coaching styles and game-specific situations.
  • Community: Active Slack channels for strategy, conference chatter, help. Passionate user base.
  • Competition: Hundreds of dedicated users managing programs. Don't expect to win a championship in your first season.
  • Realism: Conference structures, recruiting battles, transfer portal, hot seat pressure. Evolves its own history and rivalries.
  • Commissioner & Development: Developer is the commissioner. New features constantly. Still technically in "beta."

Setting Expectations

The CJBL has a steep learning curve. First season is about learning, experimenting, getting comfortable. Little pressure to win immediately. Be patient, ask questions, enjoy the process. It takes time to build a successful program.


League Basics

Slack and Community

  • All college discussion: #collegejbl channel
  • Join your conference channel (e.g., #ivy)
  • Questions: #help-and-tips
  • Being part of the community is a vital part of being in the CJBL

League Structure and Conferences

  • 32 conferences, 367 total teams, geographically organized
  • 14-17 out-of-conference games + conference regular season (each opponent 1-2 times) = ~35-40 games total
  • Conference Tournament at end of regular season (seeded by conference record, not overall)
  • Conference Tournament winner gets automatic CJBL Tournament bid
  • At-large teams ranked by NET formula (record, SOS, strength of opponents)
  • CJBL Tournament: 64 teams, six-round knockout
  • JIT Tournament: Next 64 teams
  • WMG Tournament: Next 64 teams
  • Total: 192 teams make postseason

Prestige

Nine tiers (highest to lowest): 1. Extremely High 2. Very High 3. High 4. Above Average 5. Average 6. Below Average 7. Low 8. Very Low 9. Extremely Low

Prestige matters for recruiting — higher-ranked players want high-prestige programs for pro draft exposure.

Prestige changes yearly based on: National Championship, Tournament depth, conference regular season title, having a player drafted in JBL. Losing prestige from losing conference record or missing Tournament.

Impact is scaled by current tier: Missing the Tournament as EH = huge drop. Making it as EL = big boost. Expected performance for your tier doesn't move the needle.

The Hot Seat

  • Every AD starts with a 6-season contract (~18 months real time)
  • Evaluated on: Tournament performance, conference finish, win-loss record, prestige expectations
  • Safe: Meeting/exceeding expectations
  • Precarious (Hot Seat): Underperforming, on notice
  • Fired: Consistently underperformed
  • Contract extensions: Championship = 6 years, meeting expectations = 3 years, underperforming = 1 year
  • Tier 1 (High+): Highest hot seat risk. Must compete for titles.
  • Tier 2 (AA to BA): Generally safe, but consistent failure leads to review.
  • Tier 3 (Low to EL): Safe from performance-based hot seat. Focus on development.
  • If fired, other ADs can apply for your job. You can always find a new opening.

Polls and Rankings

  • RPI and Strength of Schedule used for overall rankings
  • Top 25 Poll, Mid-Major and AD rankings use algorithmic formula (SOS, RPI, record, roster strength, etc.)
  • Quadrant wins system exists (see Rules & FAQ)

League Phases

Regular Season / Postseason

If joined mid-season: roster is set, can't add/release players, can't change coach or philosophy.

Regular Season phases: - Out of Conference: 14 games against non-conference opponents - Early Season Tournament: 1-3 games, auto-assigned knockout - Conference Challenge Game: 1 game, power-matched from rival conference - Conference Regular Season: 16-20 games - Conference Tournament: 2-4 games, winner gets CJBL Tournament bid

Postseason: - CJBL Tournament: Top 64 (conference winners + at-large by NET) - JIT Tournament: Next 64 (includes all conference regular season winners not in CJBL Tournament)

Regular Season Responsibilities

  • Watch games, review boxscores and statistics
  • Manage depth chart, strategy, rotations based on performance/injuries
  • Maintain healthy locker room (role and playing time satisfaction)
  • Participate in Slack discussion

Activity Requirements

  • Minimum: Log in at least once every 7 days
  • Recommended: Check in every 2-3 days
  • Counts as activity: Any team action (logging in, setting lineups, recruiting, checking scores)
  • Vacation: Message Commissioner if away
  • Consequence: Teams released after 7+ days inactivity without notice

Scouting Players

  • View roster: My Team > Roster or My Team > Home > Roster tab
  • View ratings: My Team > Home > Ratings tab
  • All ratings are 1-20 (20 = league best in CJBL, scaled relative to college level)
  • Scouting accuracy:
  • Own team: High Accuracy
  • Conference players: Average Accuracy
  • All other players: Low Accuracy
  • Lower accuracy = ratings may be higher or lower than actual ("fog of war")

Star Ratings

  • College Ability: Relative ability against other CJBL players (5-star = best in CJBL)
  • Pro Potential: Ceiling relative to JBL (5★=MVP, 4★=All-Star, 3★=Starter, 2★=Rotation, 1★=Deep bench). Shows growth "space," not development speed. No correlation with current college ability — irrelevant for CJBL recruiting.
  • Star ratings are weighted sum of visible + hidden ratings — rough guide only
  • Chemistry, cohesion and system fit are more important than raw star ratings

The Daily Sim

  • Games simulated daily ~6-7pm ET
  • Four JBL Days per sim session = team plays ~once per real day
  • Games are "live" — watch play-by-play with live boxscore via Simcast
  • Results posted in Slack channel after each Day

Team Strategy

Depth Chart and Rotations

  • Depth chart: 4 players deep per position (each player listed only once per position)
  • Rotations: set at two-minute intervals, assumes no foul trouble
  • If depth chart/rotations invalid, AI sets them — make sure they're correct
  • Strategy page: My Team > Strategy

Offensive Settings

  • Pace: Faster = more transition but more turnovers. Coach philosophy limits range.
  • Focus: Balanced, Outside, Inside, Drive, Bully Ball, Pick & Roll Heavy. Choose based on roster strengths.
  • Ball Distribution: Natural (based on Ball Dominance), Heliocentric (star-focused), Egalitarian (share). Uses Offensive Hierarchy.
  • Offensive Freedom: Balance between set plays and freelance. More freedom = more creation, less structure.
  • Clutch: Star takes over vs. run set plays in late game.
  • Playmakers: Designate primary ball-handlers.
  • Emphasis Points: Up to 2 per game (e.g., Box Out, Look for Early Threes, Get to the Rim).

Defensive Settings

  • Intensity: Aggressive (more turnovers forced, more fouls) vs. Conservative (safer, more open looks given up).
  • Focus: Balanced, Perimeter, Interior.
  • Screen Defense: Guards (Over, Under, Switch) and Bigs (Drop, Hedge, Blitz). Critical for stopping PnR.
  • Press Frequency: Never, Rarely, Situational, Often.
  • Zone Frequency: Never, Rarely, Situational, Often, Always.
  • Emphasis Points: Up to 2 (e.g., No Fouls, Limit Fast Breaks, Force Turnovers).
  • Defensive Assignments: Set archetypes and specific matchups.

Offensive Philosophies (8 Types)

Philosophy Description
Balanced Versatile. PnR, spot-ups, transition. Good for mixed rosters.
Pace & Space Fast, spread floor, 3PT shooting, quick ball movement. (Golden State)
Grit & Grind Slow, physical, half-court. Post-ups, inside scoring, tough defense.
Perimeter Centric Outside shooting emphasis, creating open looks from deep. Needs shooters + ball movement.
Post Centric Traditional inside-out. Strong post players, feeding inside. Slower pace.
Seven Seconds or Less Extremely fast, transition-focused. Score before defense sets. (D'Antoni Suns)
Triangle Structured spacing, passing, cutting, reading defense. (Phil Jackson)
Morey Ball Three-pointers and rim only. Higher ISO plays. High variance. (mid-2010s Rockets)

Key: Choose philosophy via head coach. Don't force a square peg into a round hole. Match roster strengths to philosophy.


Offseason

Offseason Phases (in order)

  1. League Break — Downtime while JBL/other leagues catch up. Plan your offseason.
  2. JBL Draft — Seniors auto-eligible. Underclassmen may declare. Undrafted return. Drafted = freed scholarship.
  3. Out of Conference Scheduling — Schedule 14 OOC games (min 5 home, 5 away). Road games earn budget for NEXT season. Higher prestige opponents = more road money. Use #college_scheduling. Don't schedule too early — wait for recruiting class.
  4. Coach Hiring/Firing — HC and AC. Budget set by prestige. Full budget offered automatically. HC hiring happens before AC (HC needed for recruiting). Fire via My Team > Front Office. Firing mid-contract may cost recruiting budget. Firing impacts transfers, cohesion, scheme proficiency.
  5. Transfer Portal — Phase 1: Discuss with transferring players to convince them to stay. Phase 2: Other schools can call, visit, offer. Phase 3: Final chance. Actions refresh each phase and don't cost budget.
  6. Recruiting — See Recruiting Guide for full detail.
  7. Walk-On Tryouts — Request positions (My Team > Scholarships). Random allocation from local high schools. Walk-ons stay max 2 years (must offer scholarship or cut before junior year).
  8. Training Camp — Set development focuses.

Scheduling Strategy

  • Schedule based on team ability and expectations
  • EH teams need harder schedules; EL teams should play peers but take a few paydays for budget
  • Hard conferences: easier OOC to find lineups. Easy conferences: harder OOC to boost SOS.
  • Balance difficulty — too hard = bubble losses, too easy = weak SOS

Player Roles / Status

Role Limit Expectation
Marquee Player 1 Start, 28-32+ MPG
Important Player 2 Start every game, 24+ MPG
Starter 2 Start every game, 20+ MPG
Sixth Man 1 First off bench, 14-18 MPG
Rotation No limit Play in most games
Backup No limit Injuries and blowouts only
  • Roles lock after third week of season
  • Unmet expectations = unhappy player (check Locker Room)

Training Camp and Development

Experience Points (XP) System

Players earn hidden XP throughout the season via: - Playing time (minutes) - Starting games - Tournament games - Clutch situations - Road games - Playing with veteran teammates - Team wins

More XP = more development opportunities during Training Camp. Give young players meaningful minutes freshman/sophomore years.

Training Camp Focuses

Set two focus areas per player (My Team > Training Camp). Allocate percentage split (50/50, 100/0, etc.):

Focus Skills Improved
Shooting Inside, Mid, Outside, FT, Finishing
Rebounding Off Reb, Def Reb, Strength, Positioning
Playmaking Playmaking, Passing, Ball Handling, Self-Creation
Offensive Execution Awareness, ISO, Post, PnR Execution, BBIQ
Perimeter Defense Perimeter D, Off-Ball D, Help D
Post Defense Post D, Off-Ball D, Help D, Rim Protection
Endurance Strength, Endurance, Gain Weight
Athletics Athletics, Quickness, Speed, Lose Weight
Tendency Work Shoot More/Less 3s, More/Less Mid, Drive More, Catch & Shoot More
  • Player still develops in other areas naturally (XP, coach, work ethic, DNA)
  • Effectiveness depends on work ethic + coaching skills + room to improve
  • Coach impact: 80% HC, 20% AC
  • Freshmen have no XP but still get coaching bonuses
  • Not all attributes are trainable (especially Intangibles, BBIQ)
  • Results shown at TC conclusion: blue = improvement, red = decline

Chemistry vs Cohesion

Team Chemistry

How well players get along off the court. Influenced by: - Player personalities and how they mesh - Coaching style compatibility - Locker room dynamics and happiness - Success on the court (winning helps) - Role satisfaction (playing time, role contentment)

High chemistry = overperform talent. Low chemistry = derail even talented rosters.

Team Cohesion

On-court familiarity — how long players have played together: - Better play execution through familiarity - Better handling of momentum swings - More resilient during tough stretches - Senior-heavy teams generally have better cohesion

Building cohesion takes time. Keep core groups together when possible.


Advanced Topics

Statistics Available

  • Traditional: points, rebounds, assists, steals, blocks
  • Advanced: eFG%, TS%, assist rate, usage rate
  • Team metrics: offensive/defensive efficiency, pace, turnover rate
  • Plus-minus and lineup analysis

Pro Pipeline

  • All seniors eligible for JBL Draft
  • Underclassmen may declare early
  • Undrafted players return to roster
  • Having players drafted boosts prestige
  • Developing players for pros is part of AD legacy

Pricing & Credits

  • CJBL: 600 credits/season (~$6 USD)
  • JBL: 3,000 credits/season (~$30 USD)
  • New users: 600 free credits (one CJBL season)
  • Rate: $1 per 100 credits
  • Bulk buying: Bonus credits at higher tiers
  • Credits never expire
  • Deducted at season lock, not when you join
  • ~4 seasons per real-life year (~3 months each)
  • Payments via Stripe
  • Credits are non-refundable and non-transferable (gift codes available)
CJBL Opponent Scouting Framework

CJBL Opponent Scouting Framework

Source

Methodology developed with experienced AD Morgado. Proven across 6 case studies with consistent results.


A. Defensive Philosophy

Core Principles: 1. Execute YOUR offensive plan. Don't react to opponents offensively. Offense is roster-driven and mostly stable. 2. Defensively, push opponents out of their comfort zone WITHOUT committing fouls. Force teams to do what they don't want to do. 3. Conservative intensity is the default. Always. Reduces gambling, reaching, and foul-prone plays. 4. Don't get too fancy with settings. Identify the 2-3 settings that actually matter for a given matchup, not every lever. 5. Break scoreless streaks. Any score (1, 2, or 3 points) is better than going cold. Prioritize getting the ball in the hoop. 6. High-percentage mechanisms win. Build around what you do efficiently. If inside guys, go inside. If outside, go outside. Or get fouled.

Two Defensive Modes: | Mode | When | Settings | |------|------|----------| | Box Out / Conservative (default) | Default for most games | Conservative intensity, No Fouls emphasis | | High Intensity Steals | Rarely, when personnel supports it or opponent is turnover-prone | Aggressive intensity, Force Turnovers emphasis |

Always choose Conservative to avoid fouling and giving up high-probability shots.


B. Available Defensive Settings

Intensity

Option Use When
Conservative Default. Always. Avoid fouls.
Balanced Almost never. Conservative is better.
Aggressive Only if you have elite steal personnel AND opponent is turnover-prone

Focus (PRIMARY counter-lever)

Option Use When
Interior Opponent is paint-heavy (high Points in Paint share). Pack the paint, force them to shoot threes they don't want.
Perimeter Opponent is perimeter-heavy (high 3PA, high 3PT%). Pack the line, force them to drive/post where they're uncomfortable.
Balanced Opponent is good at BOTH inside and outside. Don't create openings. Also use when threats are split (perimeter alpha + paint alpha).

Screen Defense

Setting Guards Options Bigs Options
Guards Over / Under / Switch -
Bigs - Drop / Hedge / Blitz

Press Frequency

Never / Rarely / Situational / Often / Always

Zone Frequency

Never / Rarely / Situational / Often / Always

Zone Defense Type

2-3 / 2-1-2 / 3-2 / 1-3-1 / 1-2-2 / 1-1-2-1 / 2-3 Matchup

Defensive Emphasis (pick 2)

Checkbox What It Does
No Fouls Reduces fouls, denies free throws. Use vs high FTA teams.
Limit Fast Breaks Slows transition. Use vs high transition teams.
Force Turnovers Increases steals/gambling. Use with elite steal personnel only.
Protect The Rim Paint protection. Use vs extreme paint teams (like Elon).
Box Out (Def Rebounding) Ends possessions. Default conservative choice.
Defensive Communication Handles screens/PnR. Use vs high screen assist / PnR-heavy teams.
Stifle Star Players Targets alpha. Use ONLY when opponent is alpha-dependent (1 player 25%+ scoring).
Disrupt Passing Lanes Breaks ball movement. Use vs high assist / kick-shot teams.
Challenge Every Shot Contests everything. Use vs catch-and-shoot-heavy teams.

Defensive Archetypes (per player)

Archetype Role
Anchor Big (Drop Big) Drop coverage, rim protection
Mobile Big (Versatile Bigs) Defends perimeter and PnR
Helper Rotates to help on drives and at rim
Chaser (Off Ball Wing) Chases shooters and cutters off-ball
Wing Stopper (On Ball Stopper) Main defender vs wings/secondary creators
Point of Attack (On Ball Stopper) Defends primary ball handler/PnR
Low Activity (Hidden) Conserve energy or hide poor defender

C. Step-by-Step Scouting Methodology

Step 1: Team-Level Profile

Normalize everything by pace. Raw counting stats are meaningless without pace context. Use Offensive Rating and Defensive Rating (per 100 possessions), NOT points per game.

Calculate shares, not raw volume: - Paint Share = Points in Paint / Points Scored - Three Share = (3PM x 3) / Points Scored - FT Share = FTM / Points Scored - Transition Share = Transition Points / Points Scored

High share in one area = that's their identity = that's what you target.

Data points to check (in order): 1. Pace (national rank) -- How fast do they play? 2. Transition Points (rank) -- Do they score in transition? 3. 3PA and 3PT% (rank) -- Perimeter threat level? 4. Points in Paint (as share of total scoring) -- Inside identity? 5. FTA (rank) -- Do they draw fouls? 6. Bench Scoring -- Starter-dependent or deep? 7. Assists and Screen Assists -- Ball movement? Screen-based offense? 8. Defensive Rebounds, Blocks, Steals -- Defensive profile 9. Offensive Rating / Defensive Rating (per 100 possessions)

Step 2: Individual Player Analysis

  1. Sort roster by MPG to push starters to top
  2. Identify alphas -- Is scoring concentrated (1 player 25%+ of team scoring) or diversified (4-5 in double figures)?
  3. For each key player, examine:
  4. Shot location distribution (% of FGA from three, mid-range, rim)
  5. Play type volume (where do their possessions come from?)
  6. Play type efficiency (eFG% or PPP by play type -- what ACTUALLY works?)
  7. Separate volume from actual threats:
  8. High volume + high efficiency = REAL threat -- deny this
  9. High volume + low efficiency = NOT a threat -- let them shoot, they're hurting their own team
  10. Low volume + high efficiency = Monitor, don't over-commit

Step 3: Threat Prioritization

Rank all threats as: Primary / Secondary / Acceptable

Key questions: - Which actions are BOTH high-volume AND efficient? - Which players are dangerous vs. which are hurting their own team? - If their alpha is a volume shooter but eFG stinks across all play types, you WANT him to shoot - "If he's truly dangerous at something, lean into defending that"

Scoring distribution matters: | Profile | Indicator | Implication | |---------|-----------|-------------| | Alpha-dependent | 1 player with 25%+ of team scoring | "Stifle Star Players" is high-value | | Diversified | 4-5 in double figures | "Stifle Star Players" is low-value; focus on scheme |

Play type analysis for alphas: - Don't defend what they do MOST -- defend what they do BEST - A player might run lots of PnR but actually score efficiently on spot-ups and transition - Look at efficient actions, not overall role

Step 4: Settings Selection

Intensity: Conservative (locked default)

Focus Decision Tree:

Is opponent clearly paint-heavy (high paint share, low 3PA)?
  YES --> Interior Focus (pack the paint)
Is opponent clearly perimeter-heavy (high 3PA, good 3PT%)?
  YES --> Perimeter Focus (pack the line)
Is opponent good at BOTH inside and outside?
  YES --> Balanced Focus (don't create openings)
Is opponent EXTREMELY one-dimensional (e.g., #1 paint, 336th 3PA)?
  YES --> Full opposite Focus + double-down with Emphasis

CRITICAL: Don't overcommit Focus if it opens a secondary threat. If they have both a perimeter alpha and a paint alpha, stay Balanced and use Emphasis checkboxes to cover both.

Emphasis Checkbox Selection (pick 2): Match opponent's top 2 statistical strengths to counter-checkboxes:

Opponent Statistical Strength Counter Checkbox
High FTA / FT rate No Fouls
High assists Disrupt Passing Lanes
High transition points Limit Fast Breaks
High offensive rebounds Box Out (Defensive Rebounding)
Alpha-dependent scoring (1 player 25%+) Stifle Star Players
High 3PA / perimeter-heavy Challenge Every Shot
High points in paint Protect The Rim
High screen assists / PnR-heavy Defensive Communication

The goal is marginal disruption, not total shutdown. You're shaving percentage points off their best actions and forcing them into Plan B. The checkboxes are small tilts, not silver bullets. Cumulative small edges over 40 minutes.

Step 5: Big Game Loss Pattern Analysis

For important games only (conference rivals, tournament matchups):

  1. Pull opponent's losses
  2. Identify common statistical patterns across those losses
  3. Build your settings to replicate those conditions
  4. "Other teams give you the roadmap"

Example: If every loss has "poor outside shooting + lost rebound battle," your game plan is: - Perimeter Focus (contest threes, lower their 3PT%) - Box Out emphasis (win the rebound battle)

You're not guessing -- you're reverse-engineering what actually beats them.


D. Key Analytical Insights

  1. "Spot-up is bad but 3PT% is good" = pull-up threes in transition. The threes are coming in transition (pull-ups), not half-court (catch-and-shoot). Limit Fast Breaks is more valuable than Disrupt Passing Lanes in this case.

  2. Pace does not equal Transition. Virginia Tech: #37 pace but only #93 transition. They push pace for half-court advantages, not fast breaks. Don't assume high pace = transition team.

  3. Screen assists reveal scheme-based offense. Elite screen assists = motion/PnR offense. Counter with Defensive Communication.

  4. Don't overcommit Focus if it opens secondary threats. When threats are mixed (perimeter + PnR), stay Balanced and use emphasis checkboxes to cover multiple threats without tilting Focus.

  5. Co-alpha teams require Balanced Focus. If the two best players are split inside/outside, you can't tilt either way.

  6. Inefficient volume scorers are a gift. If a player has high usage but low efficiency across all play types, they're hurting their own team. Let them cook.

  7. For big games, study losses, not wins. Opponent losses reveal vulnerabilities more clearly than wins reveal strengths.


E. Quick Reference Decision Matrix

Focus Selection (one decision)

Paint-heavy opponent  --> INTERIOR
Perimeter-heavy       --> PERIMETER
Good at both          --> BALANCED
Extreme one-dimension --> FULL OPPOSITE + double-down emphasis

Emphasis Selection (pick 2 from this mapping)

If opponent has high... Pick this checkbox
Free throw attempts No Fouls
Assists Disrupt Passing Lanes
Transition points Limit Fast Breaks
Offensive rebounds Box Out
Single-player scoring Stifle Star Players
Three-point attempts Challenge Every Shot
Points in paint Protect The Rim
Screen assists / PnR Defensive Communication

Scouting Checklist (Pre-Game)

  1. What is opponent's pace and transition rank?
  2. What share of their scoring comes from paint vs. perimeter?
  3. Is scoring concentrated (alpha) or diversified?
  4. What are the alpha's efficient play types (not just volume)?
  5. Do they draw fouls? (high FTA = consider No Fouls)
  6. Are they screen-heavy? (high screen assists = Defensive Communication)
  7. (Big games) What do their losses have in common?

F. Case Studies

1. UConn -- Defense-First, Transition-Dependent

Profile: #53 pace, #35 transition, high efficiency, #200 FTA, starter-dependent, elite defense Threats: Monroe (pull-up 3s in transition), Billings (PnR roll man, foul drawing), Talbert (transition only, inefficient in half-court) Not threats: Larranaga (volume scorer, mostly inefficient post-ups) Settings: Conservative / Balanced / Limit Fast Breaks + Defensive Communication Logic: Can't go Perimeter because of PnR threat from Billings. Balanced + checkboxes covers both threats.

2. Virginia Tech -- Screen-Heavy, Co-Alphas

Profile: #37 pace, #93 transition (pace != transition), high efficiency, #19 screen assists, starter-dependent Threats: Parson (PnR roll man, paint), Eberly (off-screen, transition), Morozov (off-screen) Not threats: Richburg (high volume, horrible 3PT%, only value is transition) Settings: Conservative / Balanced / Defensive Communication + Disrupt Passing Lanes Logic: Co-alphas (Parson paint + Eberly perimeter) = Balanced. Screen-heavy + high assists = Communication + Passing Lanes.

3. Notre Dame -- Three-Shooter Heavy

Profile: #238 pace (slow), #120 3PA, #60 3PT%, #71 FTA, bad defense (#241 DRtg) Threats: Cotal (pull-up 3s, playmaker), Hein (1.31 PPP spot-up -- elite kick target), Island (transition) Not threats: Crooks (only inside threat -- dare him to beat us) Settings: Conservative / Perimeter / Limit Fast Breaks + Disrupt Passing Lanes Logic: 3 players with 50%+ 3PA = Perimeter. Transition + drive-kick = Limit Breaks + Disrupt Lanes. Plan: make Crooks beat us.

4. Elon -- Extreme Paint (The Textbook Case)

Profile: #1 pace, #1 points in paint, #336 3PA (no threes), #179 3PT% (can't make them), #21 FTA Settings: Conservative / Interior / No Fouls + Protect The Rim Logic: Most one-dimensional team possible. Full paint denial. Dare them to shoot threes they never take and can't make. No Fouls because they get to the line at high rate.

5. Toledo -- Transition Pull-Up Threes

Profile: #34 pace, #32 transition, #105 3PA (low for pace), slightly perimeter-tilted, starter-dependent Threats: Fuller (pull-up 3s in transition, spot-up is terrible), Folston (PnR roll man), Opara (transition) Key insight: Good 3PT% overall but bad spot-up efficiency = the threes are pull-ups in transition Settings: Conservative / Perimeter / Limit Fast Breaks + Defensive Communication Logic: Perimeter fights pull-up 3s. Limit Breaks kills transition. Communication handles PnR with Folston.

6. North Florida -- Screen-Heavy Perimeter (Validation Test)

Profile: #114 pace, #215 transition (NOT a transition team), #33 assists, #26 screen assists, #239 paint, #60 3PT% Threats: Elliss (1.27 PPP off-screen -- elite), three efficient PnR roll men, Elliss spot-up (1.00 PPP) Not threats: Brinkley (high volume, inefficient everywhere -- let him shoot) Settings: Conservative / Perimeter / Defensive Communication + Disrupt Passing Lanes Logic: Low transition (215th) rules out Limit Breaks. Perimeter-tilted with good 3PT%. Screen-heavy (26th) + high assists (33rd) = Communication + Passing Lanes.

CJBL Recruit Data Guide

CJBL Recruit Data Guide

Overview

A searchable database of 22,000+ HS recruits for the 2050 class is available via the /recruit command in the CJBL agent. The data was extracted from the CJBL recruiting page and includes all HS stats.

What's in the Data

Each recruit record contains: - Identity: Name, stars (1/3/4/5), position (with dual combos), height, weight, age - Location: Hometown, state, region (Great Lakes, South Atlantic, etc.) - HS Stats: G, GS, MPG, PPG, RPG, APG, SPG, BPG, TPG - Shooting: FGA, FGM, FG%, 3PM, 3PA, 3P%, FTM, FTA, FT%, EFG, FTr, 3PAr - Advanced: BPM, PER, WS (where available) - Computed: Morey Screening Score (0-100, position-aware)

Critical Limitation

HS stats do NOT predict in-game ratings. Per Strategic Playbook, a recruit's 3P% in high school has no correlation with their OutS (Outside Shooting) rating in CJBL. Similarly, APG does not predict Passing, and PPG does not predict overall talent.

Use this data ONLY for: - Identifying high-volume candidates to Express Interest in - Narrowing the field before spending scouting actions (L1/L5) - Finding home-region targets for cheaper scouting costs

Do NOT use this data for: - Predicting a player's actual in-game ability - Making scholarship offers without L5 scouting - Ranking recruits by talent (only by HS production)

Morey Screening Score (0-100)

A position-aware composite score that estimates how well a recruit's HS stats align with Morey Ball principles (rim or three, no mid-range). Higher = more 3PT volume/efficiency relative to position peers.

Weights by position: - Guards (PG, SG): 3P% (30%), 3PAr (25%), PPG (20%), APG (15%), EFG (10%) - Wings (SF): 3P% (35%), 3PAr (25%), PPG (15%), Defensive stats (10%), EFG (15%) - Bigs (PF, C): 3P% (30%), RPG (20%), BPG (20%), PPG (15%), EFG (15%)

What it measures: HS shooting volume and efficiency relative to same-position peers. What it does NOT measure: Actual in-game shooting ability, BBIQ, intangibles, or scheme fit.

/recruit Command Reference

Command Description Example
/recruit help Show syntax and filters /recruit help
/recruit stats Database summary stats /recruit stats pos=PG
/recruit top N Top N by Morey score /recruit top 20 pos=PG
/recruit search Filter by criteria /recruit search pos=SG region=Great Lakes tppct>.35
/recruit player NAME Fuzzy name lookup /recruit player Hardtke
/recruit ask QUESTION Claude analyzes last results /recruit ask which fit Morey Ball?
/recruit pipeline SIUE tracked recruits (tiers 1-3) /recruit pipeline
/recruit pipeline all Include dropped recruits /recruit pipeline all
/recruit pipeline tier=1 Filter pipeline by tier /recruit pipeline tier=1.5
/recruit phase3 Phase 3 action checklist + contingency /recruit phase3
/recruit phase3 immediate IMMEDIATE priority actions only /recruit phase3 immediate

Pipeline Annotations

When searching the 22K recruit database, any player who appears in the SIUE pipeline is automatically tagged in search results. For example:

  • [T1 CLEAN schol] — Tier 1, clean character, scholarship target
  • [T1.5 CLEAN backup] — Tier 1B, clean character, backup scholarship
  • [T3 MIXED] — Tier 3, mixed character, maintain contact
  • [DROPPED FAIL] — Dropped from pipeline due to character fail

The /recruit player NAME detail view also shows a full "SIUE Pipeline" section with tier, status, character, promise, competition, and next Phase 3 action.

Filter Syntax

  • pos=PG or pos=PG,SG (comma = OR, matches dual positions too)
  • stars=1 or stars=1,3
  • region=Great Lakes (case-insensitive contains)
  • state=IL or state=IL,OH,IN
  • Numeric: ppg>10, tppct>.35, apg>3, efg>.50, morey>60, hgt>76
  • limit=25 (default), sort=morey (default, or ppg/tppct/apg/efg)

Live Recruit Page Fetching (MCP)

Claude Code can fetch authenticated recruit pages directly from jblfl.com via the CJBL MCP server. All recruit tools return structured markdown with parsed sections (bio, ratings, tendencies, intangibles, personality, scouting report, recruiting status, top schools, initial list). Pipeline recruits are automatically cross-referenced with siue_pipeline_2050.json.

Available MCP tools:

Tool Description Example
cjbl_recruit_eval("Haug") Full evaluation — structured profile + pipeline + Morey Ball fit table + character assessment + HS database stats Best for decision-making
cjbl_recruit_by_name("Haug") Structured profile by name (fuzzy search). Disambiguation list with pipeline tags if multiple matches. Quick profile lookup
cjbl_recruit(550769) Structured profile by player ID (pid) When pid is known
cjbl_fetch(url) Raw text extraction from any jblfl.com page Non-recruit pages

cjbl_recruit_eval Output Sections

The eval tool returns a complete evaluation with:

  1. Player Profile — Bio, physical, rankings, headline stats
  2. Scouting Ratings — All categories: Shooting, Creation, Execution, Playmaking, Defense, Rebounding, Physical
  3. Tendencies — Driving, Roll:Pop, C&S:Pull-Up, Corner3:ATB
  4. Intangibles — WE, Leadership, Discipline, Prof, Aggression, Toughness, Hustle
  5. Personality — Type (e.g., Introverted, Workhorse, Troublemaker)
  6. Scouting Report — Description, college ability stars, pro potential, strengths/weaknesses lists, accuracy
  7. Recruiting Status — Mood, commit timeline, interest, distance, scouted/called/visited/scholarship/pitch status
  8. Top Schools — Ranked list with interest levels
  9. Initial List — Phase 1 school rankings
  10. SIUE Pipeline — Tier, status, character, personality, promise, competition, commit phase, philosophy, notes
  11. Morey Ball Fit — Position-specific threshold table (Min/Ideal vs actual rating, PASS/BELOW/IDEAL status)
  12. Character Assessment — WE/Disc/Prof vs 15/12/12 thresholds, personality signal, CLEAN/BORDERLINE/FAIL verdict
  13. HS Database Stats — Morey score, 3P%, 3PAr, EFG from recruits_2050.json

Morey Ball Thresholds (Embedded in Eval Tool)

Position Attributes Checked
PG Ball Handling 15+, 3PT 15+, Passing 14+, Gravity 14+, Speed 12+
SG 3PT 15+, Self-Creation 14+, Gravity 14+, Shoot Off Dribble 12+, On-Ball Perimeter D 12+
SF 3PT 15+, Gravity 14+, Self-Creation 14+, On-Ball Perimeter D 12+, Passing 11+
PF 3PT 14+, Gravity 13+, Help Defense 12+, Defensive Rebounding 12+, On-Ball Perimeter D 10+
C Help Defense 14+, Rim Protection 12+, Athleticism 13+, Defensive Rebounding 12+, Endurance 13+

Character Thresholds (Embedded in Eval Tool)

  • Clean: WE 15+, Disc 12+, Prof 12+, no Troublemaker personality
  • Borderline: Fails within 3 of threshold (e.g., Disc 11 or Prof 10)
  • Fail: Any value more than 3 below threshold, or Troublemaker personality

The pid for each pipeline recruit is stored in siue_pipeline_2050.json. The id field in recruits_2050.json is the same as the pid URL parameter.

Note: Requires a valid PHPSESSID cookie configured in .claude/settings.local.json. If tools return a "cookie expired" error, the user needs to update the cookie value.

SIUE-Specific Guidance

  • Prestige: Very Low = focus on 1-star and 3-star recruits (4-5 star unrealistic)
  • Home region: Great Lakes = cheapest scouting actions
  • Position needs: Varies by season; check current roster depth chart
  • Workflow: /recruit top 30 pos=PG stars=1,3 region=Great Lakes -> Express Interest -> L1 scout -> L5 scout if promising -> scholarship offer
CJBL Recruiting Guide

CJBL Recruiting Guide

Source: Official CJBL documentation (Last edited: 3 August 2025, 2048 Season)

Recruiting is the most important aspect of the CJBL. It takes a few seasons to perform confidently. This guide presumes knowledge of the Recruiting rules (Rules > Recruiting tab). Ask for help in #help-and-tips on Slack.


0. Background to Recruiting

  • Recruiting takes place during the offseason for the upcoming year
  • Recruit senior high school players from 1,350+ real schools in North America
  • Recruited players play immediately in the upcoming season (no future commitments)
  • Max 13 scholarship players, minimum 9. With walk-ons, max 15 players, minimum 12 at season opening.
  • Balance classes: Aim for 3-4 players per class (Fr/So/Jr/Sr) with 13 scholarships
  • Prospects play high school games during the season; rankings update based partly on stats

Phase 0: In-Season Recruiting

Phase 0 opens midway through the regular season and runs through postseason into the first week of offseason.

Key Features

  • Player Rankings: Only top 200 (5-star + some 4-star) are ranked initially. All others are 1-star. Full class re-ranked before Phase 1 based partly on Phase 0 interest.
  • Available Actions: Scouting (up to Level 4) and Express Interest only. No calls, visits, or scholarships.
  • Budget: Special $100,000 budget (two $50K installments during season). Does NOT affect normal recruiting budget for Phases 1-10.
  • If you change teams, Phase 0 resets — previous AD's actions wiped, fresh 30 Express Interest actions.

Strategy

  • Spend the entire $100K — unspent funds are wasted (no rollover)
  • Focus locally first — free Level 1 scouting on all players in your region. Local scouting is cheapest.
  • Finding all prospects: Change filter from default (4 & 5 stars) to show 1-star players. Navigate to High Schools page to find out-of-region prospects.
  • Express Interest (30 actions): Informal call letting a recruit know they're on your radar. Significantly increases chance of being on their initial interested schools list in Phase 1. Use wisely — save a few for later as rankings shift.
  • Use tags from the start (Top Target, Potential Target, Not Interested, etc.) to organize the 11,000+ prospects.

1. Recruiting Actions

How to Perform Actions

  • Main screen: Recruiting Menu > Recruit Prospects
  • Default filters show top 200 + 3-stars in your region
  • Click player name > Profile > Recruiting tab for action buttons
  • Actions: Call, Scout, Head Coach Visit, Assistant Coach Visit, Offer Campus Visit, Offer Scholarship
  • Add to Watchlist or Call List from player profile header
  • View your prospects: My Team > Recruiting (or Recruiting > My Prospects)

Actions Per Phase (Replenish Each Phase)

Action Per Phase Notes
Calls 50 (Ph 1-2), 30 (Ph 3+) Free. Reveal player values on first call.
Coach Visits 7 total (HC + AC combined) Cost budget based on distance. HC visits more effective.
Campus Visits 3 Cost budget. Player can only visit 5 schools total.
Scouting Max 2 levels per phase Cost budget based on distance. 5 levels total.
Budget Phase-specific allocation See budget unlock schedule below.

Actions do NOT roll over. Use all actions every phase.

Budget Unlock Schedule

You do not get access to your full recruiting budget immediately. It unlocks progressively:

Phase Budget Access Example ($100K total)
Phase 1 25% of total budget $25,000
Phase 2 25% of remaining $18,750
Phase 3 25% of remaining $14,063
Phase 4 20% of remaining $8,438
Phase 5 20% of remaining $6,750
Phase 6-10 Full remaining budget unlocked $27,000

Implication: Front-load scouting in Phases 1-2 while budget is fresh. By Phase 6, whatever remains is fully available — save expensive coach/campus visits for mid-to-late phases when you've identified your shortlist.

Action chains: Must call before visiting. Must call before offering scholarship.

Warning: Actions on a player may leak on Swatter (in-game Twitter). Weighted toward higher-ranked players.


2. Identifying Target Recruits

Team Identity First

  • You must play your HC's Philosophy (can't change without firing)
  • Identify what positions and player types you need based on existing roster + philosophy
  • Don't waste actions on players that don't fit your system
  • Example: If your philosophy needs shooters, don't recruit guards with 12 or worse shooting

Early Phases = Information Gathering

  • Phase 1 (72 hrs) and Phase 2 (48 hrs): Scout as much as possible. Can't offer scholarships yet.
  • Start with Local/Regional distance, 3-4 stars, Not Committed
  • Closer players = cheaper scouting
  • Each scouting level gets closer to truth; early levels may have missing ratings
  • Max 2 scouting levels per phase — start at L1, reach L5 by Phase 2
  • For lower-ranked players (3-star and below), scouting may slightly increase their interest

Calling Reveals Player Values

First call reveals what the player values most: - Prestige - Distance From Home - Academics - Playing Time - Team Success - Favoured Philosophy

If player doesn't value anything your program offers, and multiple other programs are competing — cut losses early rather than sinking more resources.

Call mechanics: - Clicking "Call Player" uses one of your calls even if you hang up immediately (prevents fishing) - Must call before coach visit (it's rude to show up uninvited) - Answers and choices matter — wrong answers or insults may kill interest - Player personality affects which answers work - Call max 3 times per recruit (hard cap). Not calling = knocked down when phase runs. (Source: Secret - UNL/UWGB)

Target Players That Fit Philosophy

Determine disqualifiers and stick to them. Example: "No guards under 10 shooting" = tag as Not Interested immediately.

Role Players Are Useful

Players don't need to do everything well. Add players who fill roles. A decent 3PT shooter + defender with weak rebounding may be fine for your system. You'll need to concede some positions as role players since bigger teams compete for better players.

Get Comfortable Abandoning Players

Monitor Top Schools list — other programs' actions signal who's serious. You may need to abandon a perfect-fit player because bigger programs are outbidding you. Switch to slightly worse but unopposed prospects, especially if you scouted deeply early.


3. Narrowing the Target List

After Local, Target Neighbouring Regions

  • Neighbouring/Nearby regions: more expensive scouting and visits
  • May lose players to closer-to-home schools
  • Target second/third tier talent (100-200 rankings)
  • If a team has full green interest bar, they may be uncatchable

Use Call Lists and Tags

  • Essential for organizing 11,000+ prospects
  • Suggested tags: Top Target, Priority Target, Potential Target, Not Interested, Backup, Wait and See
  • Tag every scouted player to enable filtering

Scout Further for Accuracy

  • 5 scouting levels. Precision narrows with each level.
  • Early: coarse approximations are fine. A player showing 10 Def Reb won't actually be 18.
  • If you see a player at 14 On-Ball D and need 12-16, scout further to confirm.
  • Worth scouting to L2 before calling to avoid wasting calls.

4. Visits and Promises

Coach & Campus Visits

  • Expensive — gain immediate interest but use judiciously
  • Early phases: better to spend on scouting than visits (you're starting from near-zero info)
  • Save visits for when you've identified your shortlist and need to beat rivals
  • Phase 2: School prestige letters revealed next to names (EH, VH, H, AA, A, BA, L, VL, EL)

Coach Visit Details

  • 7 total per phase (any mix of HC + AC visits)
  • HC visit: Higher success chance, lower failure chance
  • AC visit: Lower success chance, higher failure chance, cheaper
  • Coaching ratings factor in (Discipline, Motivation, Tactics, Technique, Rapport, position coaching, style)
  • Visit occurs immediately. Feedback on success/failure provided.
  • Stack visits on top targets rather than spreading across many recruits. Community consensus is to concentrate as many actions as possible on priority targets each phase. (Sources: Mike - Lightning3, Rubesicon - Saints1)

Campus Visit Details

  • Player must be in "Deciding" to attend (visits happen following phase)
  • Instant interest increase when offered (shows you're serious)
  • Can go well or poorly
  • Player can only visit 5 schools total — highly competitive action
  • Available starting Phase 2
  • Early phase bonus: Offering a campus visit in an earlier phase gives a bonus over schools that offer in a later phase. Front-load campus visit offers on priority targets.

5. Scholarship Offers and Commitments

Hard vs Soft Offers

Type Details
Hard Offer To one player only. Reserved spot — guaranteed unless revoked. Player views much more favorably.
Soft Offer To up to 3 players. First to accept gets it. Other 2 lose all interest in you. Use only in Phase 9-10 as last resort.
  • Can convert soft to hard offer (Click "Convert to Hard Offer")
  • Converting revokes soft offer for other players on that scholarship

Revoking Offers

  • Calls and visits: cannot be revoked — budget removed and action performed immediately
  • Scholarship: can revoke within same phase with no penalty. After phase = impacts Recruiting Trust.
  • Cannot revoke once player enters Deciding (prevents manipulation)

Promises and Recruiting Trust

  • Third and final call: Must offer minutes and role indication
  • Without promise, player won't consider your offer highly
  • Underpromise and overdeliver — don't promise every recruit Marquee + 32 minutes
  • Breaking promises reduces Recruiting Trust (impacts future recruiting + locker room)
  • Trust hit is same whether you miss by 1 minute or 15 minutes
  • Cannot avoid hit by cutting the player

Available Promises

Promise Expectation
Marquee Player Start, 28-32+ MPG. Expects to keep role entire career. Very likely to transfer if demoted.
Important Player Start every game, 24+ MPG
Starter Start every game, 20+ MPG
Sixth Man First off bench, 14-18 MPG
Future Starter Starter by sophomore year. Minute promise is freshman year only.

Community insight — Marquee demotion: Offering a new recruit Marquee when an existing player already has it will upset the current Marquee player, even if a different AD made the original promise. Marquee players always expect to remain Marquee once promised — even if they lower their visible expectation to "Important" or "Starter," they will still be upset by the demotion. Whether it triggers a transfer depends on the player's personality, but expect them to be offended regardless. (Sources: Peyton - UNLV/UMES, Calvin - Wolves1, THS - OKC2)

Pitches (pitches)

Each team gets 2 pitches per offseason (do not roll over — use them or lose them).

Uses: - Convince underclassman to return from draft: Guaranteed success (but not for high draft projections) - Retain your own transfer (Phase 1): 100% success. Must choose pitch OR discuss — can't try discuss first then pitch. - Recruit a transfer: Not guaranteed but significant boost. - Recruit high school prospect: Not guaranteed but major boost.

Key rules: - Pitches consumed on use regardless of outcome (for external targets) - Offers are blind — you don't know if competitors used pitches - Prestige-weighted: Higher prestige pitch = more valuable. But low-prestige pitch still beats no pitch. - Pitches enhance, don't replace, a good offer — still need competitive promise - Pitch timing is risk/reward: Earlier pitches have bigger impact on the recruit's decision. Waiting gives more time to confirm the target is right and avoid wasting against stiff competition. Since pitches don't roll over, the only reason to wait is uncertainty — not saving for later. (Source: James - Jaguars1)


6. Player Commitments

Commit Phases

Commit Phase Player Enters Deciding Can Commit From
Early Phase 5 Phase 6+ (if sufficient interest)
Mid Phase 7 Phase 7+
Late Phase 9 Phase 9+
  • Commits are not phase-locked — once a recruit has sufficient interest, he can commit at any time from his deciding phase onward. The phases above are when each tier starts deciding, not the only window they can commit. (Source: THS - OKC2)
  • Most 5-star players are Early committers
  • If player doesn't receive enough offers, they may defer
  • Don't put all scholarships on Early players — by the time you hear back, Mid players may be gone
  • Lower prestige: lock away targets early before bigger schools come knocking later

How Deciding Works

  1. Player enters "Deciding" based on Commit Phase
  2. Eliminates some schools, narrows to Top 5 (or Top 4/3/2)
  3. Next phase: reduces to Top 3
  4. Phase after: makes final commitment
  5. Once Deciding, no new teams can perform actions — locked to teams already involved
  6. Teams in the final list CAN still call, visit, offer within that group

How Players Choose

  • Player compares teams on their personal factor weights (Prestige, Distance, Academics, Playing Time, Team Success, Philosophy)
  • #1 school is NOT guaranteed — factor weights vary per player, and there's inherent randomness
  • Making Top 3 = real shot even at #3
  • Player traits (dream schools, girlfriend nearby, family) give slight final bonus
  • Think of randomness as an 18-year-old's unpredictable decision-making

Interest and Rankings

  • During a phase, interest changes as teams perform actions
  • Teams aren't compared against each other until phase runs (fairness — different timezones)
  • After phase runs, player re-ranks all teams. Beginning of next phase = most accurate ranking.
  • Multiple teams at max interest bar doesn't mean they're equal — comparison happens at phase run

7. Walk-Ons

  • Available after recruiting ends
  • All walk-ons are freshmen, allocated from local high schools
  • Can't choose specific players, only request positions (My Team > Scholarships)
  • If roster < 12, walk-ons allocated automatically
  • Can add up to 3 walk-ons (max 15 roster)
  • Walk-ons stay max 2 years — must offer scholarship or cut before junior year
  • Cutting walk-ons does NOT count toward your 4 scholarship cuts per year

Recruiting Timeline Summary

Phase Duration Available Actions
0 Mid-season through early offseason Scout (to L4), Express Interest (30)
1 72 hours Scout, 50 Calls, 7 Coach Visits
2 48 hours + 3 Campus Visits
3 24 hours + Scholarship Offers. Pitches available (earlier = bigger impact).
4 24 hours Continued recruiting actions.
5 24 hours Early deciders enter Deciding phase. Can commit Phase 6+ if interest sufficient.
6 24 hours Early deciders can commit. Mid deciders narrowing.
7 24 hours Mid deciders enter Deciding. Can commit Phase 7+.
8 24 hours Late deciders narrowing.
9 24 hours Late deciders enter Deciding. Can commit Phase 9+.
10 24 hours Fill remaining scholarships

Phase timer is a deadline — may run any time after.


Common Questions

Scouting and Ratings

  • Star ratings: Quick filter only, not definitive. Weighted sum of visible + hidden ratings. Common to find 1-star "gems" who scout as 4-5 star ability.
  • Hidden ratings: Clutch, Big Game, Positioning, Off/Def Awareness. Make players unique.
  • Basketball IQ (BBIQ): Used in nearly every court decision. Low BBIQ is debilitating. Only improves 2-3 points over entire career — essentially innate.
  • HS to college transition: Ratings re-scale (HS scale to college scale). Expect ~2 point drop on average (range 0-5). Not actual ability change, just viewing scale.
  • Pro potential ≠ college ability: Pro potential is ceiling (what they could become with ideal development), not current skill. No correlation with CJBL performance. Not relevant for CJBL recruiting — focus on current ratings. (Source: Brandon - Indiana University)

Recruiting Mechanics

  • Initial interest: Based on player factors (family nearby, girlfriend, traits) + random element. Some players have shortlists giving certain schools a head start.
  • Multiple teams maxed interest: Not equal — comparison happens at phase run, not during phase.
  • Dream schools: Weighted toward higher prestige and closer schools, but can be anywhere.
  • Traits matter: Impact initial interest + slight final bonus at commitment decision.
  • Offer leaks: Small chance your offers leak to other prospects at same position. May reduce their interest.
  • Pipelines: Small initial bump to player interest from repeatedly recruiting at same high school. Commitments build strongest pipelines. The boost is "probably quite small." (Source: ChrisSpi - Stetson)
  • Actions in early phases have greater impact than same actions in later phases. This is specifically confirmed for pitches ("bags"), campus visits, and most recruiting actions — earlier deployment = stronger effect. (Sources: THS - OKC2, Morgado - USC)
  • Upgrading offer importance late (e.g., Starter → Marquee during deciding) still helps but less than if done with the initial promise. Front-load your best offer. (Sources: PDM - St. Francis, THS - OKC2)
  • Early deciders with no offers don't enter deciding until the next phase flip. A good offer triggers deciding at the next phase change. (Source: Coach Martin)
  • Always do all actions on top targets — even with full interest bar and no competition. Don't redeploy resources until player is committed. (Community consensus)
  • Revoking a scholarship: Lose interest in that recruit. If revoked in a later phase than when offered, also lose a small amount of AD trust. (Source: Calvin - Michigan State/Campbell)

Strategy for Low-Prestige Schools

  • Focus on gradually improving roster talent level
  • Accept players with flaws — identify which deficiencies you can work with
  • Build roster of "decent basketball players" before optimizing for system fit
  • Balanced and Grit & Grind are easiest to recruit for at low prestige (don't need specialized player types)

Soft Offers Caution

Strongly recommended to avoid soft offers until Phase 9-10. Players view soft offers as less valuable than hard offers. The two players who don't accept lose all interest — risky.

CJBL Strategic Playbook

CJBL Strategic Playbook

Transfer Market Strategy

Core Principle

Transfer market is DEPTH ARBITRAGE, not talent upgrade.

What works: - Acquiring role players willing to accept reduced minutes - High prestige programs convincing players to accept bench roles for exposure - Filling specific gaps (backup PG, stretch 4, etc.)

What doesn't work: - Expecting to acquire starters via transfers - Building your team around transfer acquisitions - Treating transfers as recruiting replacement

Prestige Arbitrage

Mechanism: Players want to "move up" in prestige and will accept LESS playing time at HIGHER prestige schools.

For high prestige ADs: - Can under-promise playing time (prestige compensates) - Target players seeking exposure over minutes

For low prestige ADs: - Must over-promise to compete - Expect to LOSE talent to prestige-chasing transfers - Over-recruit freshmen to compensate for annual losses

Transfer Portal Phases

Phase 1 (Discussion): - Talk to YOUR players considering transfer - Use Discuss OR pitch (not both) - Cannot offer external players yet - Junior walk-ons must get scholarship offer or auto-released

Phase 2 (Live Bidding): - Offer external transfer players - ONE-SHOT OFFERS: Cannot upgrade promises after initial offer - Can revoke if not in player's Top 3

Phase 3+: - Actions refresh each phase - pitch does NOT refresh (2 total per offseason)


Recruiting Strategy

Recruiting by Prestige Tier

Extremely Low / Very Low: - Target: 2-3★ local players - Strategy: Scout entire local region, express interest in 25-30 locals - Goal: Find hidden 1★ players who become 2-3★ after re-ranking - Reality: You're not getting 4-5★ players - Don't recruit above your weight class: If programs 3-4 prestige levels above you are committing major resources (Marquee/Important offer, 24-28+ min, pitch) to a recruit, competing is a losing play. Focus on realistic targets where prestige gap is manageable. (Source: Morgado - USC) - Player values are FUNCTIONAL, not flavor: Prioritize recruits with LOW "Prestige" value (won't gravitate to bigger schools) and HIGH "Distance" or "Playing Time" value (your local/minutes advantage matters more to them). (Source: THS - OKC2) - Target early deciders + uncontested recruits: Lock them up before higher prestige teams can pivot. Finding early-deciding guys you can win quickly and uncontested is the safest EL/VL strategy. (Source: Kevin - Miami OH, Bill - Manhattan) - When contesting above your weight: Need a significantly better offer (Marquee/28+ vs their Rotation/12), scout their roster to gauge their likely offer, and stack distance + pitch. Still expect to lose most contested battles — budget pitches accordingly.

Low / Below Average: - Target: 3★ realistic, 4★ reaches - Strategy: Mix of local 3★ (high conversion) + national 4★ gambles - Timeline: 3-4 years to tournament contention

Average / Above Average: - Target: 3.5-4★ core, 4.5★ reaches - Strategy: National recruiting, compete for top-50 recruits - Timeline: 2-3 years to conference championship contention

High / Very High / Extremely High: - Target: 4-5★ elite players - Strategy: National recruiting, leverage prestige advantage - Timeline: Must win NOW (hot seat pressure)

Phase 0 Scouting

Budget: $100k to spend on scouting

Scouting limits: Can only scout TWICE per phase. Scout straight to Level 4 on top targets.

Optimal workflow: 1. Filter by position + region 2. Scout Coach Rec list 3. Scout to Level 4 before expressing serious interest 4. Character filter eliminates ~77% — do this FIRST

Ratings vs. Statistics Disconnect (CRITICAL)

Finding: Advertised high school 3PT% does NOT correlate with OutS (Outside Shooting) rating.

Player 3PT% OutS Rating Reality
OD Pannell 40.9% 13 Average shooter
Denham Lamin 38.7% 8 Poor shooter
Dain Friel 38.2% 9 Poor shooter

Solution: Always scout to Level 4 before expressing serious interest. OutS rating is the only reliable shooting indicator.

Character Evaluation (UPDATED)

"Maddeningly inconsistent" — ⚠️ YELLOW FLAG (changed from auto-reject) - Appeared in 77% of Level 4 scouts - Weigh against other factors — NOT automatic elimination - Players with elite intangibles elsewhere may overcome this - Example: Player with WE 20, Discipline 20, Leadership 18 + "inconsistent" tag warrants consideration

Strong Concern (Need Compensating Factors): - "Intensity fluctuates throughout games" - "Disappears for stretches" - "Questions about motor and effort" - "Could be more dedicated"

Clean character indicators: - "Reliable leadership presence" - "Extension of coaching staff on court" - "Maintains composure under pressure" - "Consistent effort" language - Work Ethic 17+, Discipline 15+, Professionalism 15+


Positional Flexibility in Roster Building

There is no positional proficiency mechanic. Players can be plugged into any position immediately if they have the right ratings — no learning curve, no development time. System position labels (PG, SG, SF, PF, C) are just labels, not constraints. (Source: Morgado - USC)

  • Example: 2049 POTY (Craig at UNC) was a "SF" playing PG. SFs at PF, PGs at SG all work fine.
  • Recruit by ratings, not position label. A 6'5 "PG" with strong defensive ratings can play SG day one.
  • Roster construction benefit: Position versatility is free. Evaluate skill profiles against philosophy needs rather than filtering by listed position.
  • Morey Ball application: If your best available prospect is a "SF" with OutS 17, Gravity 15, and strong ball handling — he can play SG or even PG regardless of what the system calls him. Don't pass on talent because of a position label.

Philosophy-Specific Roster Building

Morey Ball (SIUE Current Philosophy)

Core Principle: Efficiency-obsessed. Rim or three ONLY. No mid-range.

Primary Actions: - Spot Up (three-point focused) - Pick & Roll (rim attacks) - Transition (early threes, rim attacks) - Isolation (for elite creators)

Pace: Average (4-6) — not as fast as Seven Seconds

Recruit for: - 3PT shooting (OutS 15+ minimum, 17+ preferred) - Isolation ability (SlfC 15+) - Ball handling (Hndl 15+) - Athletic rim runners (for C only) - Self-creation - High gravity (Grav 14+)

Avoid: - Traditional post players who can't shoot - Mid-range specialists (wasted skill points) - Wings with OutS below 12 - Slow-footed bigs who can't roll

Spacing threshold: ~18 3PA/100 possessions at 37%+

Lineup construction: - Ideal: 5 shooters (rare) - Acceptable: 4 shooters + 1 rim-runner - Minimum: 3 shooters + 2 rim-runners (spacing suffers)

Tendency settings: - "Shoot More Threes" - "Shoot Less Mid-Range" - "Drive More"

Morey Ball Big Man Profile: Two acceptable archetypes:

  1. Stretch Big (preferred)
  2. OutS 14+
  3. Can space floor
  4. Pop threat in PnR

  5. Athletic Rim-Runner (acceptable)

  6. OutS doesn't matter
  7. Roll threat in PnR (screen, roll, finish lobs)
  8. Rim protection anchor
  9. Example: Canyon Jihad (3PT 6 but elite rim protection)
  10. ⚠️ Offensive rebounding is NEGATIVELY correlated (r=-0.347) — don't recruit for OREB
  11. ⚠️ Post scoring (insideShot) is NEGATIVELY correlated (r=-0.631) — stretch or rim-run ONLY
  12. ⚠️ Lighter bigs preferred (avg weight r=-0.568 at low prestige)

Morey Ball Wing Profile: - OutS 15+ (non-negotiable) - Gravity 14+ - Shoot Off Dribble 12+ - Self-Creation 14+

Morey Ball Guard Profile: - OutS 15+ - Ball Handling 15+ - Passing 14+ - Gravity 14+

Common Morey Ball Mistakes: 1. Playing non-shooters at wing positions 2. Running mid-range heavy players 3. Post-up actions (unless dunking) 4. Slow pace when you have athletes 5. Not pushing transition for early threes 6. Stacking creators without shooters (r=-0.443 at low prestige) 7. Recruiting post-scoring bigs (insideShot r=-0.631 at low prestige) 8. Prioritizing offensive rebounding in bigs (r=-0.347 partial) 9. Building heavy/slow rosters (avg weight r=-0.568 at low prestige) 10. Recruiting broadly for PnR Execution — it's role-specific (PG + C only), negative at team-average level in college (College Ridge: -0.0077). Wings/SGs should have shooting and spacing, not PnR skills.

Morey Ball Data-Driven Findings (Corrected Analysis, 31 Teams)

Source: Cross-league analysis of 31 confirmed Morey Ball teams. See CJBL_Morey_Ball_Corrected_Analysis.md for full methodology.

Key Correlations with Win % (low prestige, n=13):

Factor r Direction
Guard outsideShot +0.688 #1 predictor
Non-shooters on roster -0.652 Every non-shooter (OutS <13) actively hurts
Big insideShot -0.631 Post scorers toxic
Shooters 15+ count +0.576 Stacking shooters works
Big outsideShot +0.452 Stretch bigs help
Creator count (SlfC 15+) -0.443 Too many creators hurt
Big finish -0.364 Post-oriented bigs counterproductive
Big offensive reb -0.347 OREB bigs clog spacing

Physical correlations (low prestige):

Factor r Direction
Avg weight -0.568 Lighter rosters win
Avg first step -0.414 Creator trap proxy
Max height +0.385 Need one rim protector
Avg height -0.324 Shorter on average is better

Benchmarks: - Winners (WP>.55): Guard OutS 15.5 avg, 4.2 shooters 15+ - Losers (WP<.45): Guard OutS 13.6 avg, 3.7 shooters 15+ - Low prestige mean: 4.3 shooters 15+, guard OutS 14.6

Model teams: UIC (P4, .570) -- 6 shooters 15+, one elite creator, roll-and-cut bigs, 213 lbs avg. Harvard (P4, .536) -- 6 shooters 15+, guard depth, 206 lbs avg.

Anti-model: Lipscomb (P2, .372) -- 6 creators (SlfC 15+), only 3 shooters 15+, athletic but no shooting discipline.

Pace & Space

Recruit for: - Shooters - Passers - Off-ball movement - Roll gravity bigs

Spacing threshold: ~15 3PA/100 at 35%+

Seven Seconds or Less

Recruit for: - Ball handlers - Athletes (speed, quickness) - Rim runners - Fearless transition shooters - Endurance (fast pace = fatigue)

Priority: Athleticism over skill (speed > technique)

Post Centric

Recruit for: - Post scorers - Face-up forwards - Distributing guards - PnR execution

Spacing threshold: ~10 3PA/100 at 33%+ (can survive with 1-2 non-shooters)

Triangle

Recruit for: - Passing bigs - Isolation wings - High BBIQ players - Veterans (cohesion requirement)

Reality: Avoid this philosophy for rebuilds


Offensive Settings Best Practices

Universal Principle: Use Screens More

"All great teams are checking 'Use Screens More' every time."

Screen action is foundational — it generates both rim AND three opportunities. This is the primary mechanism for creating advantages in modern basketball.

Why screens matter: - Creates advantages before defense can react - Opens driving lanes AND three-point looks - Generates assisted baskets (higher efficiency than isolation) - Works regardless of philosophy

Recommended emphasis combination: - Use Screens More (creates advantages) - Get To The Rim OR Look For Early Threes (finishes advantages)


Big Game Scouting: Loss Pattern Analysis

For important games, study opponent losses — find the blueprint.

Instead of just profiling what they do well, look at: - Games they lost - What the winning team did differently - What statistical patterns appear in their losses vs. wins

Questions to answer: - Did they lose when forced to shoot threes? - Did they lose when transition was taken away? - Did they lose to teams that packed the paint? - Did they lose when their alpha was contained? - Did they lose the rebound battle in every loss?

Workflow: 1. Pull their losses 2. Identify common statistical patterns across those losses 3. Build your settings to replicate those conditions

Key insight: "Other teams give you the roadmap." If 4 different teams beat them the same way, that's a proven formula, not a theory.


Role Player Efficiency Principle (NEW)

Core Insight: Role players must be more efficient than stars.

Role Usage TS% Threshold Rationale
Star 25%+ USG 54%+ acceptable Takes difficult shots, creates offense
Secondary 15-25% USG 56%+ expected Mix of creation and catch-and-shoot
Role Player <15% USG 58%+ required Should ONLY take good shots

Why this matters: - Stars earn lower efficiency through shot difficulty - Role players who are inefficient = wasted possessions - A role player shooting 52% TS on 10% usage is actively hurting the team - Better to give those possessions to a star at 54%

Application: - Evaluate role players by efficiency, not counting stats - Bench inefficient role players even if they "produce" - Role player value = floor spacing + defensive contribution + high-efficiency finishing


Coaching Attribute Priority (NEW — COMMISH CONFIRMED)

Attribute Hierarchy

Per Commish intel, coaching attributes rank in this order:

  1. Tactics — Most critical. Determines scheme execution quality.
  2. Coach Offense / Coach Defense — Core competency for their side of the ball.
  3. Adaptability — Prevents "locked state" where coach can't adjust mid-game.
  4. Position-specific coaching (Guards, Forwards, Centers)
  5. Culture, Patience, etc.

What Adaptability Does

  • Low Adaptability = coach gets stuck in predictable patterns
  • Cannot adjust to opponent's halftime changes
  • Cannot counter when game plan isn't working
  • Creates ceiling regardless of roster talent

Implications for Staff Construction

HC Weakness AC Priority
Tactics <12 Consider whether AC can compensate (unclear if AC Tactics affects games)
Defense <12 Defense 15+ AC
Adaptability <10 Accept tactical ceiling, win with execution

SIUE Reality: Both Allen (Tactics 10, Adapt 8) and Davis (Tactics 10, Adapt 9) are weak in the most critical attribute. Staff cannot adapt mid-game. Must win with talent, not schemes.


Roster Management

Playing Time Thresholds

~16 MPG: Critical threshold. Players below this generally want more minutes. - 20+ MPG = very safe retention zone - 16-20 MPG = acceptable, monitor mood - Below 16 MPG = always a transfer risk

Promise Management

Promise threshold: 85% of promised minutes = "safe" - Below 85% = risk of breaking promise - 95%+ fulfillment = definitely safe

Breaking promise consequences: - Immediate high transfer chance (~95-98%) - Major AD Trust hit (long-term recruiting damage)

One broken promise ≠ death sentence. Pattern of breaks = severe trust damage.

Roster Retention After Philosophy Change

Expect 3-5 players to carry over. 7-8 staying = great outcome.

Why retention is low: - Scheme fit changes - Role changes (starters become backups) - Coach personality conflicts


Training Camp Focus Strategy

By Player Role

Primary ball handler (PG): - Focus 1: Playmaking (50-75%) - Focus 2: Shooting OR Offensive Execution (25-50%)

Wing scorer (SF/SG): - Focus 1: Shooting (60-80%) - Focus 2: Perimeter Defense OR Athletics (20-40%)

Stretch big: - Focus 1: Shooting (50-70%) - Focus 2: Rebounding OR Post Defense (30-50%)

Traditional big: - Focus 1: Rebounding (50-70%) - Focus 2: Post Defense OR Offensive Execution (30-50%)

Tendency Work by Philosophy

Philosophy Tendency Focus
Morey Ball Shoot More Threes, Shoot Less Mid-Range, Drive More
Pace & Space Shoot More Threes, Catch & Shoot More
Seven Seconds Shoot More Threes, Drive More
Post Centric Shoot Less Threes, Shoot More Mid-Range

Morey Ball Specific Development

For non-shooters you must keep: - 100% training camp focus on 3PT Shooting - Only works if Work Ethic 15+ - Accept 1-2 seasons of development time

For mid-range specialists: - Tendency work to shift shots to three - May never fully convert — consider replacement


Offensive Hierarchy & Lineup Configuration

Source: Vince (Rockets1 CM, Houston/Fordham) — experienced manager running heliocentric pro builds.

Hierarchy Function

The offensive hierarchy is primarily a usage control lever (~95% of its function). It determines who terminates possessions and at what volume.

Secondary effects of the #1 slot: - Gets ball in broken plays / shot clock expiring → must have Iso/Self-Creation skills - Interacts with Clutch "Prefer Star Iso" setting → more late-game touches even without that setting checked - End-of-game touch distribution skews toward #1

Rule: Never put a player at #1 who lacks SlfC/Iso ability. Broken plays and late-clock situations will find them.

Heliocentric Setting — Tiered Usage Bumps

When Heliocentric is checked, usage bumps are applied in three tiers:

Hierarchy Position Usage Bumps Received
#1 Three bumps (Top 6 + Top 3 + Top 1)
#2–#3 Two bumps (Top 6 + Top 3)
#4–#6 One bump (Top 6 only)
#7+ No bump

Implication: To suppress a ball-dominant player's usage, place them below #3. Even at #4, high-BD players still fight for touches (see Ball Dominance interaction in Meta Knowledge).

Hierarchy Tuning — Practical Example

Vince's Austin Rockets (Prime) — full hierarchy and resulting usage:

Hierarchy Slot Player Role / BD Profile Final USG%
#1 Rigaux (C) Low BD (11), not a natural shot-seeker 24%
#2 Wills (PF) Starting PF 19%
#3 Link 6th man (low hierarchy most of year) 16%
#4 Briggs Greenaway Heliocentric monster, was 33% USG at #1 29%
#5 Cardez Moss (SF) Career 27% USG, ball-dominant 21%
#6 DPJ -- 18%

Key observations: - Even at #4, Briggs only dropped from 33% to 29% USG — BD creates a usage floor that hierarchy can't fully override - Rigaux (BD 11) at #1 was pulled UP to 24% — low-BD players placed high accept more touches than they'd naturally seek - Moss at #5 dropped to 21% and had the most efficient season of his career — proof that reduced usage can improve shot selection for the right players - Vince's next-season goal: Briggs and Rigaux both 25-26%, DPJ and Wills up to 20% — gradual rebalancing

Takeaway: Hierarchy placement and Ball Dominance interact. You can't fully suppress a high-BD player through hierarchy alone. Conversely, low-BD players placed high will accept more touches than their nature would otherwise generate. Use the #3 threshold (Heliocentric tiers) as your primary lever.

Reduced Usage and Efficiency

Not all players respond the same to reduced usage: - Some become more efficient with fewer touches — better shot selection, only taking good looks (Moss example above) - Others struggle with rhythm disruption — "no-chill chuckers" who can't take a hint may force bad shots regardless - Monitor over 10-game windows after hierarchy changes - If a player's efficiency drops after being moved down, they may not be suited to a reduced role

Sample Size for Setting Changes

Analysis Type Directional Signal Trustworthy Data
Hierarchy/setting tweaks 10 games 20 games
Lineup combinations 100 minutes 200 minutes

Rule: Don't overreact to small samples. Make one change, run 10 games, evaluate. Commit or revert at 20 games.

SIUE Application Notes

With Allen's Tactics 10 and Adapt 8, we have limited ability to run complex hierarchy schemes. Priorities: - Keep our best SlfC player at #1 (currently Coddington or Robinzine depending on roster) - If running Heliocentric, ensure #1-3 are players we WANT taking high-volume shots - Suppress non-creators by placing them #4+ - Monitor over 10-game windows before adjusting


Defensive Gameplan Guidelines

Against High Spacing Opponent

  • Zone Frequency: Rarely (zone vulnerable to shooting)
  • Screen Defense: Over (prevent open threes)
  • Defensive Focus: Perimeter

Against Low Spacing Opponent

  • Zone Frequency: Often (pack paint, dare them to shoot)
  • Screen Defense: Under (let them shoot)
  • Defensive Focus: Interior

Common Mistakes to Avoid

1. Over-recruiting upperclassmen in Year 1 - Juniors/seniors give 1-2 years max - Prioritize younger players UNLESS on hot seat

2. Breaking promises to "win now" - AD Trust follows you to next job - Better to lose games Year 1 than destroy trust

3. Scheduling too hard too early - Save 2-3 OOC schedule slots until post-recruiting

4. Treating transfer market as recruiting replacement - Transfers are depth, not starters

5. Keeping wrong philosophy too long - Firing inherited coach = 0% penalty - Eating one year of bad philosophy costs more

6. Trusting statistics over ratings - HS 3PT% ≠ OutS rating - Always scout to Level 4

7. Running non-shooters in Morey Ball - Each non-shooter collapses spacing - Maximum 1 non-shooter on floor (rim-runner C only)

8. Treating character flags as binary - "Maddeningly inconsistent" = yellow flag, not auto-reject - Weigh against other factors, especially Work Ethic and Discipline

9. Chasing favorites over gettable targets - You will generally NOT sign your favorite recruits — the odds of them wanting your specific program are small - A 95% chance at a very good player who fills your needs > 33% chance at an outstanding player who's perfect - Excellent classes are built from gettable alternatives, not dream targets (Source: Morgado - USC, Vince - Rockets1)

10. Using pitches in crowded fields - If 4+ teams at similar prestige are all pitching, there's no tiebreaker advantage — it's pure RNG - Deploy pitches for overwhelming superiority in thin competition, not hopeful swings in crowded races (Source: THS - OKC2, Morgado - USC)


Play-by-Play Analysis Insights

Action Type Efficiency Benchmarks

Metric Good Average Bad
Transition PPP 1.10+ 0.90-1.10 <0.90
PnR Ball Handler PPP 0.90+ 0.75-0.90 <0.75
ISO PPP 0.85+ 0.70-0.85 <0.70
Spot-up PPP 0.75+ 0.60-0.75 <0.60
vs Zone PPP 0.70+ 0.50-0.70 <0.50

Red Flags in PBP Data

  • 30%+ efficiency drop vs zone -- Spacing problem (note: SIUE's v2-corrected zone drop was only 0.5%, not the originally reported 32%)
  • 20+ point home/away margin swing — Mental toughness issue (often defensive)
  • Spot-up PPP below 0.60 — Can't space the floor
  • Star player efficiency below role players — Usage optimization needed
  • Role player TS% below 55% — Wasted possessions, bench them

Assistant Coach Strategy

Complementary Hiring

Hire assistants who patch head coach weaknesses:

If HC is weak in... Hire AC with...
Defense Defense 15+
Offense Offense 15+
Guards Coach Guards 14+
Forwards Coach Forwards 14+
Centers Coach Centers 14+
Culture Culture 12+

Morey Ball Specific

Morey Ball HCs typically have: - High Offense (15+) - Lower Defense (8-13) - Lower Culture (7-10)

Compensate with: Defensive specialist AC (Defense 15+, Culture 12+)

SIUE Example: Taavi Allen (Off 17, Def 13, Cul 8) + Noah Davis (Def 17, Cul 10) - Defense covered ✓ - Culture still weak — rely on player leadership (Totten) - Tactics gap NOT solved — both coaches have Tactics 10


Coach Visit Strategy (NEW)

Visit Type Success Rate Failure Risk Cost Use For
HC Visit Higher Lower Higher Tier 1 targets — can't afford to fail
AC Visit Lower Higher Lower Tier 2-3 targets, budget constraints

Rule: Use HC visits on must-get targets. The reduced failure risk is worth the cost premium.

CJBL Team Context SIUE

Team Context: Southern Illinois-Edwardsville (Last Updated: 2050 Offseason - Phase 1 Recruiting)

Team

  • Name: Southern Illinois-Edwardsville (SIUE)
  • Conference: OVC (Ohio Valley Conference)
  • Prestige: Very Low
  • Academics: Low
  • Location: Edwardsville, Illinois
  • Record (2049): 16-20 (36 games; Conference: 9-11, 4th place)
  • Championships: 0
  • Conference Titles: 1

Board Expectations

  • Expectation: Finish with .500 Conference Record
  • Job Security: Safe
  • Status: Missed expectation (9-11 vs 10-10 needed) but still safe

2049 Season Performance Analysis (PBP Findings)

The Road Problem

Venue Record Win%
Home 15-7 68.2%
Away 1-13 7.1%
  • 17-point margin swing between home and away
  • Many road losses were close (1-7 point margins)
  • Root cause: Defensive breakdown on road, not offensive issues
  • Road offense actually scored MORE than home offense
  • Mental toughness manifests defensively

Zone Defense Vulnerability (CORRECTED — v2 Parser)

Original analysis claimed catastrophic -32% PPP vs zone. Corrected v2 parser (fixed foul/shot ordering, 3PT attribution, turnover logic) found:

Defense Faced Poss SIUE PPP Assessment
Man-to-Man 2,040 0.966 Baseline
Zone (all) 462 0.961 Nearly identical
Diff -0.5% No real vulnerability

The original -32% was a parsing methodology error (shots with "draws foul" swallowed by foul handler, all 3PT counted as 2PT). Zone offense is not a weakness overall — but specific zone types matter:

Zone Type Poss PPP Verdict
1-2-2 Zone 20 1.400 Dominant
1-3-1 Zone 40 1.150 Strong
2-3 Zone 142 1.049 Good
3-2 Zone 101 0.960 Average
2-1-2 Zone 131 0.794 Struggle
2-3 Matchup 28 0.714 Struggle

Zone Performance by Player (v2 Corrected Data)

Player Man PPP Zone PPP Diff Assessment
Tabari Totten 0.907 1.312 +0.405 Elite zone killer
Arkel Topercer 1.096 1.234 +0.138 Zone killer (graduating)
Cyncier Levett 1.172 1.250 +0.078 Great both (graduating)
Canyon Jihad 1.172 1.195 +0.023 Neutral
Arlon Sanderson 1.009 0.964 -0.045 Slight zone weakness
Izayah Coddington 1.032 0.970 -0.062 Slight zone weakness
Quion Sliwoski 1.054 0.975 -0.079 Slight zone weakness
Donchevell Ross 1.093 0.800 -0.293 Zone liability

Action Type Efficiency (v2 Corrected)

Action Poss PPP Share Assessment
Transition 364 1.261 13.9% ELITE — push pace aggressively
Cut 261 1.034 10.0% Underused, efficient
PnR Roll Man 229 0.987 8.8% Solid
Off Screen 170 0.982 6.5% Solid
ISO 188 0.973 7.2% Average
PnR Ball Handler 452 0.918 17.3% Below benchmark
Post-Up 78 0.910 3.0% Below average
Hand Off 133 0.872 5.1% Below average
Spot-Up 693 0.802 26.5% WORST — highest volume, lowest efficiency

Hidden Efficiency Finding: Mark Mimms

  • 46.0% FG, 43.6% 3P (best percentages on team)
  • Spot-Up 1.156 PPP (only player above average), PnR BH 1.032 PPP
  • 84 assists, +53 duo differential with Jihad
  • Severely underutilized — should get increased minutes in 2050

SIUE Offensive Settings (Morey Ball)

Setting Value Rationale
Pace Faster Transition (1.261 PPP) is best action
Focus Drive Pushes rim attacks, creates kick-out opportunities
Freedom Freelance Morey Ball thrives on read-and-react
Clutch Prefer Star Iso Coddington (Iso 17, Ball Dom 18) in crunch time
Emphasis 1 Use Screens More "All great teams check this" — creates advantages
Emphasis 2 Get To The Rim Finishes screen advantages at rim
Playmakers Sanderson, Coddington, Mimms Ball handlers who create
Ball Distribution Heliocentric Build around Coddington

Key Insight: "Use Screens More" is foundational — it generates both rim AND three opportunities. Screen action is the primary mechanism for creating advantages in modern basketball.

Defensive Settings Recommendations

Setting Current Recommendation Rationale
Zone Frequency Always Often or Situational Flexibility based on opponent
Zone Type 2-3 Keep Jihad anchors this well
Emphasis Limit Fast Breaks, Box Out Keep Sound choices

Coaching Staff (UPDATED - 2050)

Head Coach: Taavi Allen (NEW)

Attribute Value
Age 55
Salary TBD
Philosophy Morey Ball
Style Psychologist

Ratings:

Skill Rating Assessment
Coach Offense 17 Excellent
Coach Guards 14 Good
Coach Centers 12 Adequate
Patience 12 Good
Leadership 12 Good
Coach Defense 13 Adequate
Motivation 10 Average
Discipline 10 Average
Tactics 10 ⚠️ WEAK — Critical attribute
Adaptability 8 ⚠️ WEAK — Cannot adjust mid-game
Culture 8 Weak
Coach Forwards 8 Weak
Temperament 8 Below average
Media Handling 9 Below average

Assessment: Elite offensive mind (Off 17) who will implement Morey Ball. However, Tactics 10 and Adaptability 8 are critical weaknesses. Per Commish intel, Tactics is the most important coaching attribute. Allen cannot adapt mid-game or counter opponent adjustments effectively. Must win with talent execution, not scheme complexity.

Assistant Coach: Noah Davis (NEW)

Attribute Value
Age 37
Philosophy Morey Ball
Style Instructor

Ratings:

Skill Rating Assessment
Coach Defense 17 Excellent — patches Allen's weakness
Coach Forwards 12 Good
Discipline 12 Good
Media Handling 12 Good
Coach Offense 11 Adequate
Patience 11 Adequate
Culture 10 Average
Tactics 10 ⚠️ WEAK — Same as Allen
Adaptability 9 ⚠️ WEAK — Same tier as Allen
Motivation 9 Below average
Temperament 8 Below average
Leadership 8 Below average
Coach Guards 8 Weak
Coach Centers 8 Weak

Assessment: Same philosophy alignment (Morey Ball). Elite Defense (17) patches Allen's defensive weakness. However, does NOT solve Tactics/Adaptability gap — both coaches are weak in the most critical attribute.

Combined Staff Profile

Attribute Allen (HC) Davis (AC) Combined Assessment
Philosophy Morey Ball Morey Ball ✅ Perfect alignment
Offense 17 11 ✅ Allen carries
Defense 13 17 ✅ Davis covers
Tactics 10 10 MAJOR GAP
Adaptability 8 9 MAJOR GAP
Coach Guards 14 8 Allen carries
Coach Forwards 8 12 Davis covers
Coach Centers 12 8 Adequate
Culture 8 10 ⚠️ Weak — rely on Totten

Strategic Implication: Staff cannot adapt mid-game or counter opponent adjustments. Simplify schemes, win with talent execution. Don't expect tactical outmaneuvering.


Roster Overview (UPDATED - Including Transfers)

Team Chemistry & Culture

  • Team Chemistry: Very Harmonious
  • Team Continuity: 94.7%
  • Captain: NOT SELECTED (Recommend: Tabari Totten)

Roster by Class

Seniors (4) — Graduating: - Arkel Topercer (SG) — Important Player, Lazy, LEADING SCORER - Cyncier Levett (SF) — Rotation, Friendly, Leadership Group - Ramel Brink (PG) — Rotation, 0 minutes played - Loyd Robinzine (SF) — TRANSFER ACQUISITION

Juniors (6) — Core for 2050: - Canyon Jihad (C) — Important Player, anchor - Quion Sliwoski (PF) — Starter, low morale - Tabari Totten (PF) — Rotation, LEADER, High Influence - Arlon Sanderson (SF) — Rotation, Leadership Group - Mark Mimms (PG) — Rotation, 30% transfer risk - Donchevell Ross (C) — Rotation, stable

Sophomores (2) — Long-term core: - Izayah Coddington (SG) — Marquee, franchise player - Stanislav Kunchik (SG/PG) — Rotation, highest aptitude (15), 30% transfer risk

Scholarship Situation

  • Current: 12/13 used (after Robinzine)
  • Available: 2
  • Walk-ons: 0/3
  • pitches: 2/2 (unused)

Transfer Portal Acquisition: Loyd Robinzine ⭐

Attribute Value
Position SF
Height 6'6"
Year Senior (1-year rental)
Hometown Indianapolis, IN
Previous School DePaul

Key Ratings

Strength Rating Weakness Rating
3PT Shooting 18 Passing 5
Gravity 18 Help Defense 8
Finishing 19 Rim Protection 7
On-Ball Defense 19 Endurance 9
Free Throws 19 Off Rebounding 3

What He Solves

Problem Before After
SF Shooting Sanderson (OutS 9) Robinzine (OutS 18)
Morey Ball Spacing 2 shooters 3+ shooters in starting lineup
Zone Vulnerability Overstated (0.5% drop, not 32%) Non-issue overall

Character Concerns

  • Work Ethic: "Minimal appetite for extra work"
  • Consistency: "Consistency is rare"
  • Assessment: Won't develop as a senior, but elite shooter for one year

Promise Made

  • Role: Starter
  • Minutes: 28 MPG
  • pitch: Not used

Complete Player Profiles

1. Ramel Brink (PG) — Senior, Graduating

Attribute Value
Physical 5'11, 197 lbs, 6'2 WS
Age 23
Hometown Villa Park, Illinois
Role Rotation (0 MPG actual)
Morale 63%
Transfer Risk 0% (Graduating)
Personality Driven
Offensive Archetype Primary Ballhandler
Defensive Archetype Point of Attack
College Ability 3/5 stars
Pro Potential Longshot

2049 Stats: 0.0 PPG, 0.0 RPG, 0.0 APG (Did not play)

Ratings:

Category Strengths (15+) Weaknesses (<10)
Shooting Mid 17, 3PT 17, FT 15 Inside 13
Creation Self-Creation 19, Shoot Off Dribble 18 Ball Dom 7, Foul Draw 8, Finishing 9
Execution Post 5, BBIQ 7
Playmaking Passing 17 Spacing 9
Defense On-Ball 19, Off-Ball 17, Stealing 18 Post Def 3, Rim Prot 3
Physical Ath 17, First Step 17, Speed 15, End 15 Strength 4
Intangibles Work Ethic 20, Prof 19, Hustle 19, Tough 16 Leadership 6

Assessment: Elite perimeter defender with shooting and creation ability. Criminally underused (0 MPG). Listed as playmaker despite no minutes. Graduating — no action needed.


2. Izayah Coddington (SG) — Sophomore, MARQUEE PLAYER ⭐

Attribute Value
Physical 6'7, 195 lbs, 7'3 WS
Age 20
Hometown Skokie, Illinois
Role Marquee Player
Morale 78%
Transfer Risk 5% (Very Low)
Personality Friendly
Offensive Archetype Secondary Creator
Defensive Archetype Chaser
College Ability 3/5 stars
Pro Potential Fringe prospect, possible rotation upside

2049 Stats: 15.5 PPG, 3.2 RPG, 2.4 APG, 1.21 SPG, 29.0 MPG

PBP Insight: Slight zone weakness (0.970 zone vs 1.032 man, -6%). Not catastrophic. Elite cutter (1.128 PPP, 57.6% FG) — use off-ball more than PnR BH (0.774).

Morey Ball Fit: ⭐⭐⭐⭐⭐ Perfect — OutS 18, Gravity 18, Ball Dom 18

Ratings:

Category Strengths (15+) Weaknesses (<10)
Shooting 3PT 18, FT 17, Inside 15 Mid 11
Creation Ball Dom 18 Foul Draw 10, Finishing 10
Execution Iso 17, PnR 18 Post 3, BBIQ 8
Playmaking Gravity 18, Ball Handling 16
Defense On-Ball 10, Off-Ball 10, Help 8, Stealing 6, Post 3, Rim 3
Rebounding Off 14, Def 14
Physical End 17, Quickness 15 Ath 10, First Step 10, Strength 5
Intangibles Work Ethic 16 BBIQ 8, Disc 8, Prof 7, Hustle 7

Assessment: FRANCHISE PLAYER. Elite 3PT shooter (18) with length (7'3 WS). High gravity (18) creates space. Weak defender but only a sophomore — developable. Ball dominant (18) so heliocentric offense makes sense. Build around him.

Development Focus: Defense (On-Ball, Help) + Finishing


3. Canyon Jihad (C) — Junior, IMPORTANT PLAYER ⭐

Attribute Value
Physical 7'1, 239 lbs, 7'6 WS
Age 21
Hometown New Concord, Ohio
Role Important Player
Morale 64%
Transfer Risk 10% (Very Low)
Personality Professional
Offensive Archetype Roll-and-Cut Big
Defensive Archetype Anchor Big
College Ability 3/5 stars
Pro Potential Average prospect, solid bench contributor

2049 Stats: 5.8 PPG, 8.6 RPG, 0.6 APG, 2.13 BPG, 25.5 MPG

Morey Ball Fit: ⭐⭐ Rim-runner only — OutS 6, but elite rim protection makes him viable

Ratings:

Category Strengths (15+) Weaknesses (<10)
Shooting Mid 5, 3PT 6
Creation Foul Draw 19 Self-Creation 9, Shoot Off Dribble 3, Ball Dom 5, Finishing 8
Execution Post 11 Iso 6, PnR 6, BBIQ 8
Playmaking Playmaking 1, Passing 3, Gravity 4, Spacing 3, Ball Handling 4
Defense Post Def 19, Rim Prot 16, Help 15 On-Ball Peri 2, Off-Ball 5
Rebounding Def 18, Off 14
Physical End 19 First Step 9, Quickness 9, Speed 6
Intangibles Prof 20, Disc 18, Leadership 17 Work Ethic 8, Hustle 8

Assessment: ELITE RIM PROTECTOR. 7'1 with 7'6 wingspan, Post Defense 19, Rim Protection 16, 2.13 BPG. Rebounds (8.6 RPG) and draws fouls (19). Zero shooting but fills rim-runner role in Morey Ball.

Development Focus: Finishing + PnR Execution


4. Stanislav Kunchik (SG/PG) — Sophomore

Attribute Value
Physical 6'3, 188 lbs, 6'6 WS
Age 21
Hometown Berezan, Ukraine
Role Rotation
Morale 56%
Transfer Risk 30% (Average)
Personality Confident
College Ability 2/5 stars

2049 Stats: 2.2 PPG, 0.8 RPG, 1.19 APG, 11.7 MPG

Morey Ball Fit: ⭐⭐⭐ Mixed — OutS 12 (below threshold), but elite intangibles

Assessment: HIGH UPSIDE SOPHOMORE. Highest aptitude on team (15). Elite first step (18), passing (18), self-creation (18). High BBIQ (17). Wants more minutes (30% transfer risk). Work ethic 19 = will develop.

Development Focus: 3PT Shooting + Defense


5-11. [Remaining players same as original document]


Recruiting: Current Status (Phase 1)

Resources Available

Resource Status
Scholarships 2
pitches 2/2 (unused)
Scouting Budget ~$56K remaining

Primary Targets (UPDATED -- Corrected Analysis Applied)

Strategy shift: Guard shooting is now #1 predictor (r=+0.688 at low prestige). SF-specific urgency was overstated.

Priority Player Pos Interest OutS Character Commit Notes
#1 Clent Seumalo G Very High 19 TBD (L3) Mid ELEVATED -- highest OutS, guard, addresses #1 predictor
#1 Emanuel Delaney PG High 18 Elite (WE 20) EARLY Ph5 IRREPLACEABLE -- every other PG failed character
#3 Mateo Mangrum SF Above Avg 18 Clean (Prof 20) EARLY Ph5 Good but SF urgency overstated
#4 Jon Haug SG High 17 Clean (WE 20) Mid Ph7 ELEVATED -- local guard shooter
#5 Tshiefu Cannon SF High 17 Clean EARLY Ph5 Local, solid shooter
#6 Antwine W High 16 Clean EARLY Safety option
#7 Pfister G High 15 Clean EARLY Safety option
#8 Kalil -- Average 16 Clean Late Low priority

Scholarship Allocation (2 available): - Option A (Guard-First): Delaney + Seumalo -- NOW CLEARLY BEST per corrected data. Two guards OutS 18+/19 directly attacks strongest predictor. - Option B (Balanced): Delaney + Mangrum -- Previously Option A. Still viable but SF urgency based on contaminated data. - Option C (If Delaney lost): Seumalo + Haug -- Two elite guard shooters.

Pending actions: - Scout Seumalo to L4 (elevated priority) - Scout Cannon to L4 - Offer Delaney scholarship Phase 3 - Deploy pitches Phase 4+

Commit Timeline

  • Phase 5: EARLY players commit (Delaney, Mangrum)
  • Phase 7: MID players commit (Mundy, Haug)

Character Filtering Update

"Maddeningly inconsistent" is now a YELLOW FLAG, not auto-reject.

Previous analysis was overcautious. The trait should be weighed against other factors rather than eliminating prospects outright.


Season Goals (2050)

  • Realistic: .500 conference record (10-10), improve from 9-11
  • Stretch: Make OVC tournament, compete for 3rd place
  • Floor: Maintain roster stability, develop Coddington/Kunchik/Sanderson

Immediate Action Items

  1. Select Captain: Tabari Totten (Leader 19, High Influence)

  2. Fix Playmaker Settings: Remove Ramel Brink (0 MPG), add Izayah Coddington

  3. Add Offensive Emphasis: "Get To The Rim" + "Look For Early Threes"

  4. Recruiting Phase 1: Prioritize Delaney (irreplaceable PG) and Mangrum (complete wing)

  5. Increase Mark Mimms Minutes: 1.090 PPP efficiency warrants larger role

  6. Monitor Transfer Risks: Mimms (30%), Kunchik (30%), Totten (25%)


Strategic Summary

Corrected Morey Ball Analysis (NEW)

Cross-league data (31 Morey teams) reveals key strategic corrections: - Guard shooting is #1 predictor at low prestige (r=+0.688), not PnR or SF shooting - Non-shooters actively harmful (r=-0.652) -- every roster spot with OutS <13 hurts - Post-scoring bigs toxic (r=-0.631) -- stretch or rim-run only - Creator trap: too many SlfC 15+ hurts (r=-0.443) - Physical: lighter/faster rosters win (weight r=-0.568) - Shooter stacking effect confirmed (r=+0.576) - Full analysis: CJBL_Morey_Ball_Corrected_Analysis.md

Strengths

  • Very Harmonious chemistry (94.7% continuity)
  • Safe job security (no pressure)
  • Morey Ball philosophy (modern, efficient)
  • Franchise player in Coddington (So, Marquee, 3PT 18)
  • Elite rim protector in Jihad (Jr, 7'1, Post Def 19)
  • Transfer acquisition Robinzine solves SF shooting for 2050
  • Transition offense is elite (1.261 PPP)

Weaknesses

  • Very Low prestige (recruiting handicap)
  • Coaching staff weak in Tactics (10) and Adaptability (8-9) — cannot adjust mid-game
  • Scholarship cliff in 2052 (down to 2 players)
  • Losing leading scorer (Topercer, 17.4 PPG)
  • Several players want larger roles
  • 1-13 road record (defensive effort issue)
  • Robinzine is 1-year rental — SF need returns in 2051

Philosophy Fit

Morey Ball requires shooters. Current shooters meeting OutS 15+ threshold: - Coddington (18) ⭐⭐⭐⭐⭐ - Robinzine (18) ⭐⭐⭐⭐⭐ (1 year) - Mimms (18) ⭐⭐⭐⭐ - Totten (16) ⭐⭐⭐ - Sliwoski (15) ⭐⭐⭐ - Kunchik (12) ⚠️ Below threshold - Sanderson (9) ❌ Cannot play Morey Ball - Jihad (6) ⭐⭐ Rim-runner only

3-Year Outlook

2050: Transition year. Lose Topercer's scoring, integrate Robinzine and recruits. Target: .500 conference.

2051: Coddington Jr year, Jihad Sr year. Peak window opens. Need to replace Robinzine at SF. Target: Tournament.

2052: Cliff year. Only Coddington (Sr) and Kunchik (Sr) remain from current roster. Must have recruited 8+ players by now.

JBL Pro Win Predictors

JBL Pro Win Predictors

Cross-project knowledge doc from JBL Analytics (Denver Dragons). Source: 36-team regression analysis, 516 players, 3,180 games (2049 season).

Top Skills That Predict Pro Wins

From bivariate OLS regression (each rating vs team win%, minutes-weighted):

Rank Skill What It Means
1 Inside Shooting 0.522 Strongest single predictor. Teams with better inside shooters win more.
2 Defensive Rebounding 0.428 Boards = possessions = wins.
3 Pick & Roll Execution 0.418 The engine of modern JBL offense.
4 Help Defense 0.398 Team defense matters more than on-ball.
5 3PT Shooting 0.373 Spacing opens everything else.
6 Offensive Rebounding 0.303 Second-chance points add up.
7 Shoot Off Dribble 0.232 Shot creation under pressure.
8 On-Ball Perimeter Defense 0.228 Matters, but less than help defense.
9 Endurance 0.220 Minutes = impact. Low endurance = diminished role.
10 Speed 0.110 Transition + defensive rotation.

Best Win-Prediction Model (Stepwise)

4-variable model: adj-R² = 0.642 (explains 64% of team win% variance)

Variable Coefficient p-value
Pick & Roll Execution 0.0383 0.024
Defensive Rebounding 0.0632 0.001
Inside Shooting 0.1002 0.006
Speed 0.0414 0.026
(Intercept) -2.9131 0.000

Interpretation: A +1 point increase in team-average Inside Shooting rating predicts +0.10 in win% (roughly +8 wins over 82 games). Inside Shooting is 2.5x more impactful per unit than PnR Execution.

Ridge Regression Coefficients (Top 10, Standardized)

These account for collinearity and are the most robust importance weights:

Skill Ridge Coef Tier
Defensive Rebounding 0.0273 Elite
Pick & Roll Execution 0.0252 Elite
Help Defense 0.0202 High
3PT Shooting 0.0201 High
Offensive Rebounding 0.0165 Moderate
Inside Shooting 0.0140 Moderate
Endurance 0.0126 Moderate
Shoot Off Dribble 0.0122 Moderate
Speed 0.0113 Moderate
On-Ball Perimeter Defense 0.0111 Moderate

College vs Pro Priority Comparison

Priority CJBL (College) JBL (Pro)
#1 Basketball IQ (2x importance via GB model) Inside Shooting (R²=0.522)
#2 Help Defense (r=+0.51) Defensive Rebounding (R²=0.428)
#3 Outside Shooting Pick & Roll Execution (R²=0.418)
#4 Off-Ball Defense Help Defense (R²=0.398)
#5 Endurance (minutes = development) 3PT Shooting (R²=0.373)

Key Differences: - BBIQ is the #1 predictor in college (2x any other) but not in the JBL top-5 for wins. This suggests BBIQ enables role execution rather than directly driving wins. - Inside Shooting dominates at pro level, possibly because pro defenses are better and finishing through contact matters more. - PnR Execution is top-3 in both leagues, confirming it as a universal skill.

Recruiting Weight Adjustments for SIUE

Based on JBL pro data, SIUE should upweight these skills when evaluating recruits with pro potential:

  1. PnR Execution — Already valued in CJBL. Keep as top priority. Translates directly to pro wins.
  2. Inside Shooting — Not explicitly in CJBL Morey Ball thresholds. Consider adding a soft threshold (12+) for scholarship targets.
  3. Defensive Rebounding — Highly win-correlated at pro. Bigs with Def Reb 14+ have outsized pro value.
  4. Speed — In the stepwise model. Fast players contribute to transition and defensive rotation.
  5. 3PT Shooting — Already in Morey Ball thresholds. Confirmed as pro win-driver.

CA Bias Warning

JBL's Current Ability (CA) rating is systematically biased toward bigs with inside scoring:

  • CA favors: Inside Shooting (+0.449), Ball Dominance (+0.434), Big Fit (+0.451)
  • CA ignores: Spacing (+0.014), Free Throws (+0.014), Ball Handling (-0.016), Speed (-0.062)

Implication for Draft: College guards/wings with elite spacing and perimeter skills will be undervalued by their CA rating. Look for pro upside that CA misses, especially in: - High Spacing + High 3PT + High Speed guards - Defensive specialists with elite Help Defense + Rim Protection

JBL Salary Cap & Exceptions

JBL Salary Cap & Exceptions

Max Level Exception (XLE)

Official rules from the commissioner. Also called "Max Level Extension."

What It Does

Allows a team to sign an unrestricted free agent from another team (or a renounced player) at the applicable maximum salary level, even if the team is over the salary cap (up to the Hard Cap).

Eligibility

Who can USE the XLE: - Any team, subject to the 2-season cooldown - Must have re-signed all own free agents the team intends to retain first - Upon submitting an XLE offer, the team immediately forfeits all remaining tenure rights (Tenure, Early Tenure, or Unrestricted) on all unsigned players

Who can be SIGNED with the XLE: - Unrestricted free agents from other teams - Renounced (free agent) players

Who CANNOT be signed: - The team's own free agents (including Bird, Early Bird, Non-Bird rights) - Restricted free agents - A player renounced from that team in the same offseason - Sign-and-trade players

Timing

XLE offers may only be made from Day 2 of Free Agency onwards (not during Moratorium, Days 1-4).

Contract Structure

Max Salary Level Contract Length
$25,000,000 2 years
$30,000,000 3 years
$35,000,000 4 years

Age exception: If the player is 32 or older, the offer must be $35M/3.

No-Trade Clause: All XLE contracts include a two-year no-trade clause: - Year 1: Hard no-trade clause - Year 2: Player has the right to waive the clause

Restrictions

  • Cap exception — can exceed salary cap but NOT the Hard Cap ($150M)
  • No MLE — a team that uses the XLE cannot use the MLE in the same season
  • May still use Minimum Salary Exception and Rookie Exception
  • Sign-and-trade prohibited for XLE contracts
  • Cooldown: Once every two seasons

Pre-Offer Requirements

Before offering an XLE contract: 1. Team must have re-signed all own FAs it intends to retain 2. Must use available cap space to sign own players first 3. Upon submitting the XLE offer, all remaining tenure rights on unsigned players are immediately forfeited

This makes the XLE a "burn the boats" move — you must be done with your own FA business first.


Replacement Player Exemption (RPE)

Granted to a team that loses a player to another team's XLE signing.

Amount

$15,000,000 — one-time exemption.

Usage Options (choose ONE)

  1. Signing exception (same offseason) — sign an unrestricted FA from another team or a renounced player, up to $15M, exceeding the salary cap up to the Hard Cap. Cannot be used to re-sign own free agents.

  2. Trade exception (TPE) of $15M — valid until the next trade deadline. Can absorb incoming salary in a trade without matching outgoing salaries.

RPE Rules

  • Cannot be split between uses (must be wholly applied as signing exception, trade exception, or deferred in full)
  • A team CAN use both the MLE and RPE in the same offseason
  • If converted to TPE, expires at the next trade deadline
  • The RPE is not a tradeable asset

Trade Salary Matching Rules

Confirmed by commissioner. Applies to all teams, always — no cap-based exceptions.

Three Rules

  1. Player-for-pick (exactly 1 player in the trade, rest are picks): No salary matching required. Any salary works — you can trade a $40M player for a future first straight up.

  2. All other trades (player-for-player, multi-player packages): 50% rule — incoming salary must be within 150% of outgoing salary.

  3. Hard Trading Limit ($135M):

  4. Under $135M: Trade freely, but cannot cross $135M via trade (post-trade salary must stay ≤ $135M)
  5. Over $135M: Can add up to $3.25M net salary via trade (incoming salary - outgoing salary ≤ $3.25M)

50% Rule Formula

Incoming salary ≤ Outgoing salary × 1.50
Outgoing salary ≥ Incoming salary / 1.50 (i.e., ≥ 66.7% of incoming)

Examples

Sending out Can receive up to Minimum to send for $30M player
$10M $15M
$20M $30M $20M
$25M $37.5M
$30M $45M
$20M (= $30M / 1.5)

Key Implications for Denver

  • At $142.75M, Denver is over $135M — can only add up to $3.25M net salary via trade
  • The 50% rule still applies on top of the $3.25M limit (both constraints must be satisfied)
  • Player-for-pick exception: A single-player-for-picks trade bypasses 50% matching entirely (useful for salary dumps)
  • Net salary math: Denver sends $20M, receives $23.25M = legal ($3.25M net). Denver sends $20M, receives $24M = illegal ($4M net exceeds $3.25M limit).
  • To make a big acquisition, Denver must shed salary first (e.g., dump Jurkin's contract for picks, then trade for the target)
  • The RPE trade exception ($15M) can absorb incoming salary without matching (see RPE section above)

Key Salary Cap Thresholds (2050)

Threshold Amount Notes
Hard Cap $150,000,000 Absolute ceiling — cannot exceed
Hard Trading Limit $135,000,000 Under: can't cross via trade. Over: can add up to $3.25M net.
Over-Limit Trade Buffer $3,250,000 Max net salary increase per trade when over $135M
Salary Cap TBD Soft cap — can exceed with exceptions

Community Strategy Notes

"Terroristic Bidding" (from 2049 Offseason Primer by Jesse Goehl)

The XLE enables aggressive FA poaching. Key strategic considerations:

  • Detroit cautionary tale: Lost Mazalovic and Zouita to XLE in Year 1 of the rule. A year later, Detroit was "certainly better off not paying Mazalovic $35m/4 or Zouita $30m/3."
  • Timing tension: The "re-sign your own guys first" rule means XLE users must complete their own FA business before making the offer. This creates a window where the target's current team can make their own offer.
  • Vipers used XLE (2049): "Vipers Trade Second to Cleveland for Salary, Intend to use XLE" — teams actively create cap room to deploy XLE.
  • Defensive implication: If you have a star FA, another team's XLE is a real threat. Consider re-signing early to remove the risk.
Morey Ball Role-Based Evaluation Framework

Morey Ball Role-Based Evaluation Framework

Philosophy

Define the role FIRST → Identify enabling ratings → Validate with advanced stats

Stats are context, not the starting point. A player who doesn't fit the role is a problem regardless of efficiency.


PnR Execution: Role-Specific, Not Roster-Wide

Critical insight (from College Ridge model): PnR Execution is NEGATIVE at the team-average level in college (-0.0077) because it's a role-specific skill. Only 2 of 5 players on the floor need it (ball handler + roll man). High roster-wide PnR Execution signals a misbuilt roster over-indexed on bigs lacking shooters.

Recruiting rule: Only evaluate PnR Execution for PG (Role 1) and Center (Role 5A). For wings and shooting guards, ignore PnR Execution entirely — prioritize OutS, Gravity, BBIQ, and Help Defense instead.


ROLE 1: Primary Ball Handler (PG)

What We Need

  • Run PnR as ball handler
  • Attack rim or pull-up 3 off the pick
  • Create for others when defense commits
  • Control pace and tempo
  • Limit turnovers in half-court

Required Ratings (Scouting Priority)

Attribute Minimum Ideal Why
Ball Handling (Hndl) 15 17+ PnR execution, pressure handling
Outside Shooting (OutS) 15 17+ Pull-up 3 threat off pick
PnR Execution (PnR) 13 15+ Core PnR ball handler skill — role-specific, recruit for it HERE
Passing (Pass) 14 16+ Hit roll man, kick to corners
Gravity (Grav) 14 16+ Defense must respect
Speed (Spd) 12 14+ Attack downhill off pick

Validation Stats

Stat Target Red Flag
PnR BH PPP > 0.85 < 0.70
3PAr > 35% < 25%
AST/TOV > 2.0 < 1.0
TS% > 55% < 50%
TOV% < 15% > 20%

Role Player vs Star Threshold

  • Star PG (20%+ USG): TS% 54%+ acceptable, can take tough shots
  • Role PG (<15% USG): TS% 58%+ required, should only take good shots

ROLE 2: Scoring Guard (SG)

What We Need

  • Primary or secondary scoring option
  • Spot-up 3PT threat off kick-outs
  • ISO ability when play breaks down
  • Can run PnR as secondary ball handler
  • Reliable in late-clock situations

Required Ratings (Scouting Priority)

Attribute Minimum Ideal Why
Outside Shooting (OutS) 15 18+ Non-negotiable for spacing
Self-Creation (SlfC) 14 16+ ISO scoring, shot creation
Gravity (Grav) 14 16+ Pulls defenders, opens lanes
Shoot Off Dribble (ShOD) 12 15+ Pull-up game
Perimeter D (PerD) 12 14+ Must not be liability

Validation Stats

Stat Target Red Flag
Spot-Up PPP > 0.80 < 0.60
ISO PPP > 0.80 < 0.65
3PAr > 40% < 30%
MidAr < 20% > 30%
TS% > 56% < 52%

Role Player vs Star Threshold

  • Star SG (25%+ USG): TS% 55%+ acceptable
  • Role SG (<15% USG): TS% 58%+ required

ROLE 3: Shooting Wing (SF)

What We Need

  • Floor spacer — must shoot 3s at volume
  • Secondary playmaker off closeouts
  • Switchable defender (guard 2-4)
  • Transition finisher
  • Connective passer in half-court

Required Ratings (Scouting Priority)

Attribute Minimum Ideal Why
Outside Shooting (OutS) 15 17+ NON-NEGOTIABLE
Gravity (Grav) 14 16+ Must be respected
Self-Creation (SlfC) 14 16+ Attack closeouts
Perimeter D (PerD) 12 15+ Switch onto guards
Passing (Pass) 11 13+ Make simple reads

Validation Stats

Stat Target Red Flag
3PAr > 45% < 35%
MidAr < 15% > 25%
Spot-Up PPP > 0.75 < 0.55
TS% > 55% < 50%
Corner 3 % > 38% < 32%

Role Player vs Star Threshold

  • Star SF (20%+ USG): TS% 55%+ acceptable
  • Role SF (<15% USG): TS% 58%+ required

IMPORTANT NOTE

A wing with OutS < 15 collapses the entire offense. Non-shooters at SF = 4-on-5 spacing. However, corrected cross-league analysis (31 Morey teams) shows SF-specific OutS is NOT the strongest positional predictor — guard shooting (r=+0.688 at low prestige) matters more. SF shooting is table stakes, not the primary differentiator. Prioritize guard OutS in recruiting, then ensure SF meets minimum thresholds.


ROLE 4: Stretch Big (PF)

What We Need

  • Floor spacing from 4 position
  • Pop threat in PnR (forces help decisions)
  • Can switch onto perimeter players
  • Rim protection as help defender
  • Offensive rebounding in scramble situations

Required Ratings (Scouting Priority)

Attribute Minimum Ideal Why
Outside Shooting (OutS) 14 16+ Space the floor
Gravity (Grav) 13 15+ Defense respects shot
Interior D (IntD) 12 14+ Rim protection
Rebounding (Reb) 12 14+ Crash offensive glass
Perimeter D (PerD) 10 12+ Switch capability

Validation Stats

Stat Target Red Flag
3PAr > 35% < 20%
MidAr < 20% > 30%
Spot-Up PPP > 0.70 < 0.50
TS% > 55% < 50%
DRB% > 15% < 10%

Role Player vs Star Threshold

  • Star PF (18%+ USG): TS% 54%+ acceptable
  • Role PF (<15% USG): TS% 57%+ required

ROLE 5A: Rim-Running Center (C) — Athletic Archetype

What We Need

  • Roll threat in PnR (finishes at rim)
  • Rim protection anchor
  • Vertical spacing (lob threat)
  • Doesn't need the ball to impact game
  • Screens and stays out of the way when not rolling

Required Ratings (Scouting Priority)

Attribute Minimum Ideal Why
Interior D (IntD) 14 17+ Anchor the defense
Block (Blk) 12 15+ Rim deterrence
PnR Execution (PnR) 13 15+ Core roll man skill — role-specific, recruit for it HERE
Athleticism (Ath) 13 15+ Rim running, lobs
Rebounding (Reb) 12 14+ Defensive glass control
Endurance (End) 13 16+ Pace demands

Note: OutS not required for this archetype. They create value through rolling, screening, and defending.

DATA-DRIVEN WARNINGS (Corrected Analysis, 31 Morey Teams): - Offensive rebounding is NEGATIVELY correlated with winning (r=-0.347 partial, r=-0.364 low-prestige). Bigs who crash boards pull spacing apart. Rim-runners should get back or space, not hunt OREBs. - High insideShot is NEGATIVELY correlated (r=-0.631 low-prestige). Rim-runners should finish lobs/rolls, not post up. - High finish is NEGATIVELY correlated (r=-0.364 low-prestige). Likely reflects bigs who demand post touches. - Lighter rosters win at low prestige (avg weight r=-0.568). Prefer mobile rim-runners over heavy plodders. - Strength is NOT an asset at low prestige (r=-0.219). Correlates with weight, which correlates with losing.

Validation Stats

Stat Target Red Flag
RimAr > 60% < 45%
Rim FG% > 60% < 50%
MidAr < 15% > 25%
PnR RM PPP > 1.00 < 0.80
DRB% > 20% < 12%
BLK/75 > 1.5 < 0.5

Efficiency Threshold

  • TS% > 58% (should only take layups/dunks)
  • If TS% < 55%, they're taking bad shots or can't finish

ROLE 5B: Stretch Center (C) — Shooting Archetype

What We Need

  • Floor spacing from 5 position (rare, high value)
  • Pop threat in PnR
  • Still provides some rim protection
  • Allows 5-out lineups
  • Opens driving lanes for guards

Required Ratings (Scouting Priority)

Attribute Minimum Ideal Why
Outside Shooting (OutS) 14 16+ Must hit 3s
Interior D (IntD) 12 14+ Still need some rim protection
Rebounding (Reb) 12 14+ Can't give up glass
Gravity (Grav) 13 15+ Defense respects shot
Height 6'9"+ 6'11"+ Size for rebounding

Validation Stats

Stat Target Red Flag
3PAr > 30% < 15%
3P% > 35% < 30%
Spot-Up PPP > 0.70 < 0.50
TS% > 55% < 50%
DRB% > 15% < 10%

LINEUP CONSTRUCTION RULES

Spacing Requirements

Configuration Viability Notes
5 shooters (OutS 14+) Ideal Rare, maximum spacing
4 shooters + 1 rim-runner Good Standard Morey Ball
3 shooters + stretch 4 + rim-runner Acceptable Need 4 to be reliable
3 shooters + 2 non-shooters Poor Spacing collapses

Position Priorities for Shooting (UPDATED — Corrected Analysis)

  1. Guards (PG/SG) — Most critical (guard OutS r=+0.688 at low prestige — #1 predictor)
  2. SF — Must shoot (non-shooter = 4-on-5 spacing, but SF-specific OutS is NOT the top differentiator)
  3. PF — Strong preference for shooting (stretch bigs r=+0.452 at low prestige)
  4. C — Can survive with rim-runner (max 1 non-shooter on floor)

Shooter Stacking Effect (DATA-DRIVEN)

More players with OutS 15+ = more wins (r=+0.315 partial, r=+0.576 low-prestige).

Shooters 15+ Assessment
6+ Elite spacing — UIC model (.570 WP%)
5 Strong
4 League average (mean 4.1)
3 or fewer Below average — losing teams average 3.7

The Creator Trap (DATA-DRIVEN — r=-0.443 at low prestige)

Too many self-creation specialists (SlfC 15+) is NEGATIVELY correlated with winning.

Why: Players with high SlfC but low OutS take possessions out of the system, generate mid-range shots, reduce ball movement, and don't provide spacing.

Rule: One franchise creator (Coddington-type), surround with system shooters. Not 2-3 ISO wings.

Case study: Lipscomb (P2, .372 WP%) had 6 players with SlfC 15+ — worst record in low-prestige Morey subset.

Physical Profile for Morey Ball (DATA-DRIVEN)

Winning low-prestige Morey teams are lighter, shorter on average, and faster.

Metric Low-Prestige r Implication
Avg weight -0.568 Heavier rosters lose
Avg height -0.324 Shorter rosters win on average
Max height +0.385 Need one rim protector
Avg first step -0.414 Creator trap proxy — driving/ISO attribute

The ideal: Light, mobile roster of shooters (~sub-215 lbs avg) with one tall rim protector.

Usage Distribution Principle

  • Stars (25%+ USG): 1-2 players max
  • Secondary (15-25% USG): 1-2 players
  • Role players (<15% USG): Everyone else

Role players with USG > 15% but TS% < 55% are actively hurting the team.


EVALUATION CHECKLIST

For Any Prospect/Player:

Step 1: Define Role - [ ] What position will they play? - [ ] What role archetype fits? (Star/Secondary/Role) - [ ] What specific actions do we need from them?

Step 2: Check Ratings - [ ] Do they meet minimum thresholds for role? - [ ] Any critical gaps? (OutS < 15 for wings = disqualifying) - [ ] Character clean? ("Maddeningly inconsistent" = ⚠️ yellow flag, weigh against WE/Discipline/Prof)

Step 3: Validate with Stats (if available) - [ ] TS% appropriate for usage level? - [ ] Shot distribution fits Morey Ball? (3PAr, MidAr, RimAr) - [ ] Play type efficiency? (PnR, Spot-Up, ISO PPP) - [ ] Any red flags? (Zone PPP collapse, high TOV%)

Step 4: Contextual Factors - [ ] ORTG/DRTG from previous system? (discount if philosophy changed) - [ ] Sample size sufficient? (50+ possessions minimum) - [ ] Lineup effects? (who did they play with)


QUICK REFERENCE: DISQUALIFYING TRAITS

Auto-Reject (Regardless of Other Attributes)

  • Wing with OutS < 12
  • Primary ball handler with Hndl < 13
  • Role player with TS% < 50% on 50+ possessions
  • Guard with OutS < 13 (actively harmful per r=-0.652 non-shooter correlation)

Yellow Flag (Need Strong Compensating Factors)

  • "Maddeningly inconsistent" character — weigh against WE, Discipline, Professionalism

Strong Concern (Need Compensating Factors)

  • Wing with OutS 12-14 (only if elite elsewhere)
  • MidAr > 30% (tendency work may not fix)
  • Role player with TS% 50-55%
  • ORTG 15+ points below team average

APPLYING TO SIUE 2050 ROSTER

Player Role Ratings Fit Stats Fit Verdict
Coddington Star SG ✓ OutS 18 ⚠ MidAr 30.5% Tendency work needed
Topercer Secondary SG ✓ Fits ✓ Fits Keep, increase role
Mimms Role PG ✓ Fits ✓ 65.4% TS More minutes
Q. Sliwoski Role PF ✓ Fits ⚠ 52.9% TS Monitor efficiency
Totten Role PF ⚠ OutS 16 ⚠ 52.6% TS Acceptable, not ideal
Jihad Rim-Runner C ✓ Fits ✓ 58.3% TS Anchor — but OREB 14 is a yellow flag per corrected data
Sanderson ??? ✗ OutS 9 ✗ 53.5% TS Cannot play Morey Ball — non-shooter actively harmful (r=-0.652)
SIUE 2049 Season Analytics

SIUE 2049 Season Analytics

Source: Play-by-play parsing of 36 games (2049 season), validated 33/36 within 3 points Parser: v2 with corrected shot detection, 3PT attribution, turnover team logic, and foul-draw handling Last Updated: 2050 Offseason


Season Summary

Metric Value
Record 16-20 (36 games)
Home 15-7 (68.2%)
Away 1-13 (7.1%)
Avg Score SIUE 70.2 - OPP 72.2
Avg Margin -2.0
Total Possessions Tracked 2,615

Only road win: at Tennessee-Martin (Game 33, Day 119)

Diagnosis: Same opponents, dramatically different outcomes at home vs away. This is a mental toughness/effort problem, not a talent gap. Home crowd masks effort deficiencies.


Game Log

# H/A Opponent W/L Score Day
1 H Bellarmine L 81-87 2
2 H Detroit Mercy W 76-66 4
3 H Coppin State W 77-62 6
4 A Ball State L 73-84 9
5 H North Alabama W 63-52 13
6 A Lafayette L 65-88 18
7 H Mercer W 73-67 21
8 H Quinnipiac L 55-67 23
9 H South Alabama W 83-72 26
8b H Western Illinois W 75-74 23
11 H Western Illinois L 64-77 32
12 A Lamar L 67-77 34
13 H Hampton W 72-69 40
14 A Campbell L 59-88 44
15 H Robert Morris L 60-75 47
16 A Binghamton L 62-66 51
17 H IUPUI W 68-63 56
18 H Austin Peay L 72-78 59
19 A Tennessee Tech L 50-67 63
20 A Eastern Illinois L 53-63 67
21 H Southern Indiana W 82-72 71
22 A SE Missouri State L 65-66 75
23 A Little Rock L 80-90 79
24 H Tennessee-Martin W 75-73 83
25 A Tennessee State L 80-87 87
26 H Lindenwood W 72-67 91
27 H Tennessee Tech W 60-58 95
28 A Morehead State L 83-84 99
29 H Eastern Illinois W 81-61 103
30 A Southern Indiana L 101-105 107
31 H SE Missouri State W 81-67 111
32 H Little Rock L 69-84* 115
33 A Tennessee-Martin W 77-66 119
34 H Tennessee State L 81-93 123
35 A Lindenwood L 56-68 127
36 H Morehead State W 92-82 133

*Game 32 header score (14-18) is a data error; tracked score is 69-84.

Data anomalies: duplicate Game 8 numbering (Quinnipiac + Western Illinois), Game 10 and Game 16 missing from original numbering.


Shot Distribution

By Distance Zone

Zone FGA FGM FG% Pts Share
Rim (0-3ft) 803 402 50.1% 804 34.6%
Short Mid (4-10ft) 342 100 29.2% 200 14.7%
Long Mid (11-20ft) 377 124 32.9% 248 16.2%
3PT (21+ft) 802 281 35.0% 836 34.5%

Morey Ball 3-Zone

Zone FGA FGM FG% Pts Share
Rim (0-5ft) 974 456 46.8% 912 41.9%
Mid-Range 558 177 31.7% 354 24.0%
Three 792 274 34.6% 822 34.1%

Assessment: Mid-range share at 24.0% is too high for Morey Ball. Target: reduce to <15% by eliminating long twos. Rim and three share combined is 76.0% — goal should be 85%+.


Play Type Efficiency

Play Type Poss Pts PPP FG% TO FD Share
Transition 364 459 1.261 60.0% 44 17 13.9%
Putback 41 46 1.122 47.1% 7 11 1.6%
Cut 261 270 1.034 41.5% 27 53 10.0%
PnR Roll Man 229 226 0.987 40.0% 29 49 8.8%
Off Screen 170 167 0.982 36.1% 15 15 6.5%
ISO 188 183 0.973 35.8% 12 22 7.2%
PnR Ball Handler 452 415 0.918 35.7% 51 55 17.3%
Post-Up 78 71 0.910 37.3% 11 16 3.0%
Hand Off 133 116 0.872 35.0% 10 10 5.1%
Spot-Up 693 556 0.802 31.3% 79 46 26.5%

Key Takeaways: - Transition is elite (1.261 PPP) — push pace aggressively, this is the #1 action - Spot-Up is the biggest problem — highest-volume action (26.5% share) at worst efficiency (0.802 PPP). Reducing volume and improving personnel here is the single highest-leverage fix. - Cuts are underused at 10.0% — 1.034 PPP justifies more off-ball movement - PnR BH at 0.918 is below the 1.0 benchmark — ball handler creation needs improvement


Zone vs Man-to-Man Efficiency

Scheme Faced Poss Pts PPP FG% TO
Man-to-Man 2,040 1,970 0.966 39.0% 201
Zone 462 444 0.961 38.0% 54

Zone vs Man PPP diff: -0.5%

CORRECTION: The previous analysis claimed a -32% efficiency drop vs zone. This was a parsing error in the original PBP extraction (shots with "draws foul" were being swallowed by the foul handler, and all three-pointers were counted as two-pointers). The corrected analysis shows essentially no difference between zone and man offense.

Zone Type Breakdown

Zone Type Poss PPP FG%
1-2-2 Zone 20 1.400 57.9%
1-3-1 Zone 40 1.150 40.0%
2-3 Zone 142 1.049 41.1%
3-2 Zone 101 0.960 39.3%
2-1-2 Zone 131 0.794 32.2%
2-3 Matchup Zone 28 0.714 26.1%

Actionable: SIUE struggles significantly against 2-1-2 Zone (0.794) and 2-3 Matchup Zone (0.714). These specific zone types are where preparation is needed, not zone defense in general.

Player Man vs Zone Performance (min 20 poss each)

Player Man Poss Man PPP Zone Poss Zone PPP Diff
Tabari Totten 118 0.907 32 1.312 +0.405
Arkel Topercer 417 1.096 94 1.234 +0.138
Cyncier Levett 116 1.172 40 1.250 +0.078
Canyon Jihad 128 1.172 41 1.195 +0.023
Arlon Sanderson 109 1.009 28 0.964 -0.045
Izayah Coddington 437 1.032 67 0.970 -0.062
Quion Sliwoski 167 1.054 40 0.975 -0.079
Donchevell Ross 97 1.093 25 0.800 -0.293

Zone Killers: Totten (+0.405) and Topercer (+0.138) — Topercer is graduating, so Totten becomes the primary zone weapon. Zone Struggles: Ross (-0.293) and Coddington (-0.062) — shield these players from primary usage vs zone.


Shot Quality: Defender Proximity

Proximity FGA FGM FG% Pts
Wide Open 667 312 46.8% 779
Open 449 160 35.6% 389
Tight 600 289 48.2% 623
V Tight 608 146 24.0% 297

Notable: "Tight" FG% (48.2%) is higher than "Open" (35.6%). This suggests SIUE's best scorers are effective finishers at the rim where tight contests are common, but perimeter shooters are missing open looks. This reinforces the need for better shooting personnel.


Player Statistics (min 10 FGA)

Player Pts FGA FG% 3PA 3P% FTM AST TO FD
Arkel Topercer 591 524 38.0% 267 38.2% 91 48 49 58
Izayah Coddington 545 529 38.4% 196 33.7% 73 88 74 52
Quion Sliwoski 223 216 36.1% 78 29.5% 44 22 20 33
Canyon Jihad 202 172 41.9% 1 0.0% 58 20 9 49
Cyncier Levett 194 162 42.6% 62 43.5% 29 7 15 19
Tabari Totten 159 157 39.5% 60 26.7% 19 31 19 14
Arlon Sanderson 145 143 39.2% 0 0.0% 33 171 16 22
Donchevell Ross 129 125 40.8% 0 0.0% 27 10 12 20
Mark Mimms 129 100 46.0% 55 43.6% 13 84 16 7
Melo McMahon 114 119 37.0% 48 20.8% 16 67 17 9
Stanislav Kunchik 78 77 35.1% 25 24.0% 18 63 15 11

Player Notes: - Sanderson leads in assists (171) by a massive margin — true floor general - Topercer is the volume scorer (591 pts) with decent 3P% (38.2%) — graduating - Coddington has highest FGA (529) but mediocre efficiency, especially from 3 (33.7%) - Levett is the most efficient shooter (42.6% FG, 43.5% 3P) — graduating - Mimms is the best shooter by percentage (46.0% FG, 43.6% 3P) with 84 assists — underutilized - Jihad draws the most fouls (49) with only 9 turnovers — elite ball security for a big


Player Play Type Breakdowns (min 20 possessions)

PnR Ball Handler

Player Poss PPP FG%
Cyncier Levett 34 1.235 43.8%
Mark Mimms 31 1.032 39.3%
Arkel Topercer 131 1.015 37.5%
Arlon Sanderson 31 0.968 31.0%
Stanislav Kunchik 23 0.913 38.9%
Melo McMahon 32 0.781 28.6%
Izayah Coddington 159 0.774 33.1%

Critical: Coddington is the highest-volume PnR ball handler (159 poss) but ranks LAST in efficiency (0.774). This is the team's most-run action with its worst performer. Topercer (1.015) is far better but is graduating. Finding a PnR-capable guard is the top offensive priority.

Isolation

Player Poss PPP FG%
Cyncier Levett 22 1.364 40.9%
Arkel Topercer 45 1.111 34.9%
Izayah Coddington 60 0.783 32.1%

Spot-Up

Player Poss PPP FG%
Mark Mimms 32 1.156 44.0%
Canyon Jihad 22 0.955 33.3%
Arlon Sanderson 32 0.906 31.0%
Arkel Topercer 162 0.852 31.2%
Izayah Coddington 154 0.818 31.6%
Cyncier Levett 53 0.792 34.0%
Melo McMahon 36 0.778 33.3%
Quion Sliwoski 70 0.743 25.4%
Tabari Totten 87 0.690 31.6%
Stanislav Kunchik 20 0.600 23.5%

Spot-up is the team's biggest weakness. Only Mimms (1.156) is above average. The highest-volume spot-up shooters (Topercer 162 poss at 0.852, Coddington 154 poss at 0.818) are both below benchmark.

Transition

Player Poss PPP FG%
Arkel Topercer 86 1.419 60.3%
Cyncier Levett 25 1.320 60.9%
Izayah Coddington 87 1.287 66.2%
Melo McMahon 25 1.280 68.2%
Arlon Sanderson 43 1.070 55.3%
Stanislav Kunchik 21 0.667 35.3%

Everyone except Kunchik is effective in transition. This is a team-wide strength.

PnR Roll Man

Player Poss PPP FG%
Tabari Totten 29 1.241 44.4%
Canyon Jihad 66 1.106 39.3%
Donchevell Ross 51 0.902 43.2%
Quion Sliwoski 72 0.889 37.5%

Jihad is a solid roll man (1.106 PPP) — better than his old legacy numbers suggested (0.740 was from the bugged parser).

Off Screen

Player Poss PPP FG%
Arkel Topercer 42 1.024 35.0%
Izayah Coddington 52 0.942 31.4%

Cut

Player Poss PPP FG%
Quion Sliwoski 36 1.222 43.3%
Arkel Topercer 47 1.170 36.4%
Izayah Coddington 39 1.128 57.6%
Canyon Jihad 40 0.975 41.0%
Donchevell Ross 37 0.973 36.1%

Cutting is efficient across the board. Coddington has elite cut FG% (57.6%) — run more off-ball actions for him rather than PnR creation.


Set Play Analysis

Category Poss Pts PPP FG%
Freelance 1,268 1,230 0.970 40.3%
Structured 1,296 1,231 0.950 37.6%
Broken Play 41 46 1.122 47.1%

Freelance slightly outperforms structured sets. Not a major gap.

Named Set Efficiency (min 15 uses)

Set Uses PPP Verdict
Zipper 24 1.292 Elite
Iverson 91 1.154 Elite, high volume
Wide 19 1.105 Strong
Pop 18 1.056 Good
Cross Punch 25 1.000 Average
Pistol 21 84 1.000 Average, high volume
Ram 84 0.976 Average, high volume
Delay 117 0.932 Below avg, high volume
Rip 32 0.938 Below avg
Tight Pin 56 0.929 Below avg
Horns 156 0.885 Poor, HIGHEST volume
Weave 45 0.889 Poor
Double Drag 99 0.788 Poor, high volume
Wide Pin 34 0.676 Terrible

Action Items: - Run more Iverson (1.154, 91 uses) and Zipper (1.292) — both elite - Reduce Horns (0.885, 156 uses) — most-used set but below average - Eliminate Double Drag (0.788, 99 uses) and Wide Pin (0.676) — these are destroying possessions


PnR Coverage Analysis (How Opponents Defend SIUE PnR)

Guard Coverage Big Coverage Poss PPP FG%
Over None 16 1.062 53.8%
Over Drop 188 1.059 39.8%
Over Blitz 8 1.125 37.5%
Over Hard 15 1.000 35.7%
Under Switch 51 1.098 43.2%
Under Drop 62 0.935 37.7%
Over Switch 86 0.872 34.6%
Zone Step 16 0.875 33.3%
Zone Contain 84 0.810 31.9%
Over Soft 65 0.769 32.8%
Ice Switch 28 0.786 30.4%
Ice Drop 35 0.743 35.7%

Key Findings: - SIUE handles Over/Drop well (1.059, most common coverage) — standard coverage isn't a problem - Under/Switch is exploitable by opponents but SIUE scores well (1.098) — likely because switches create mismatches - Ice coverages are SIUE's kryptonite — Drop (0.743) and Switch (0.786) both struggle. Opponents who ice the PnR effectively shut SIUE down. - Zone/Contain (0.810) is also a struggle — reinforces need for better perimeter creation against help schemes - Over/Soft (0.769) — SIUE can't punish soft drops, suggesting the ball handler can't pull up effectively from mid-range (expected with Morey Ball)


Lineup Analysis (Point Differential)

Starting Lineup

Lineup Diff Pts For Pts Agst
Topercer / Sanderson / Coddington / Sliwoski / Jihad -94 965 1,059

The starting five was the single worst lineup on the team by total differential. This unit played the most minutes and was consistently outscored.

Best Lineups

Diff Lineup
+22 Jihad / Mimms / McMahon / Kunchik / Totten
+19 Topercer / Sanderson / Jihad / Kunchik / Totten
+17 Jihad / Levett / Coddington / Mimms / Totten
+15 Sanderson / Jihad / Levett / Sliwoski / Kunchik
+13 Topercer / Sanderson / Coddington / Ross / Totten
+11 Topercer / Jihad / Coddington / McMahon / Sliwoski

Pattern: The best lineups consistently feature Jihad + Totten and/or Mimms + Kunchik. Removing Sanderson-Coddington from the same unit improves results.

Worst Lineups

Diff Lineup
-94 Topercer / Sanderson / Jihad / Coddington / Sliwoski
-51 Levett / Ross / Mimms / Kunchik / Totten
-29 Sanderson / Coddington / Ross / McMahon / Totten
-17 Jihad / Mimms / McMahon / Sliwoski / Kunchik

Two-Man Combinations

Best Duos (by point differential)

Diff Duo
+76 Canyon Jihad + Tabari Totten
+53 Canyon Jihad + Mark Mimms
+47 Canyon Jihad + Cyncier Levett
+43 Izayah Coddington + Mark Mimms
+43 Arkel Topercer + Tabari Totten
+39 Canyon Jihad + Stanislav Kunchik
+27 Arlon Sanderson + Stanislav Kunchik
+25 Cyncier Levett + Izayah Coddington

Jihad is the connective hub — his best pairings (Totten +76, Mimms +53, Levett +47, Kunchik +39) are all among the team's best duos.

Worst Duos

Diff Duo
-99 Arlon Sanderson + Izayah Coddington
-90 Arkel Topercer + Quion Sliwoski
-87 Canyon Jihad + Quion Sliwoski
-85 Arlon Sanderson + Quion Sliwoski
-70 Donchevell Ross + Tabari Totten
-66 Izayah Coddington + Quion Sliwoski
-62 Arkel Topercer + Arlon Sanderson
-60 Arlon Sanderson + Canyon Jihad
-58 Arkel Topercer + Canyon Jihad
-56 Donchevell Ross + Stanislav Kunchik

Sliwoski is a lineup poison pill — appears in 3 of the 4 worst duos. His lack of shooting (29.5% 3P) collapses spacing. Sanderson + Coddington (-99) is the single worst pairing. These two should not share the floor together. This has massive implications — they were both starters.


Home vs Away Play Type Splits

Home (1,588 poss, 0.981 PPP overall)

Play Type Poss PPP
Transition 218 1.174
PnR BH 272 0.974
ISO 114 1.053
PnR RM 138 1.014
Off Screen 113 1.000
Spot-Up 412 0.847
Hand Off 88 0.841

Away (1,027 poss, 0.926 PPP overall)

Play Type Poss PPP
Transition 146 1.390
Cut 105 0.971
PnR RM 91 0.945
Off Screen 57 0.947
PnR BH 180 0.833
ISO 74 0.851
Spot-Up 281 0.737

Away transition is the best road play (1.390 PPP). On the road, SIUE should push pace even harder rather than trying to execute in the half court where they collapse (spot-up 0.737 road vs 0.847 home).


Strategic Recommendations for 2050

Offense

  1. Push transition relentlessly — 1.261 PPP overall, 1.390 on the road. This is SIUE's best action regardless of venue.
  2. Run more Iverson and Zipper sets — Both elite PPP (1.154 and 1.292). Reduce Horns (0.885) and eliminate Double Drag (0.788) and Wide Pin (0.676).
  3. Get Coddington off the ball — His PnR BH (0.774) is team-worst. His cutting (1.128 PPP, 57.6% FG) is elite. Use him as an off-ball scorer, not primary creator.
  4. Increase Mimms usage — Best shooter (43.6% 3P), best spot-up scorer (1.156 PPP), 84 assists. He should be getting starter minutes with better lineups around him.
  5. Separate Sanderson and Coddington — Their -99 differential as a duo is catastrophic. Stagger their minutes.
  6. Use Totten as primary zone weapon — +0.405 PPP differential vs zone. Only remaining elite zone killer after Topercer graduates.
  7. Target mid-range elimination — 24% of shots from mid-range at 31.7% FG is destroying efficiency. Morey Ball demands this drops below 15%.

Recruiting Priorities (Validated by Analytics)

Need Analytics Justification
PnR-capable guard Coddington (0.774 PnR BH) and McMahon (0.781) can't run PnR. Topercer (1.015) and Levett (1.235) are graduating.
Spot-up shooters 26.5% of possessions at 0.802 PPP. Only Mimms (1.156) is above average.
Players who can punish Ice PnR 0.743-0.786 PPP vs Ice coverage is the exploitable gap opponents will target.
Mental toughness 1-13 road record with same opponents means effort, not talent, is the gap.

Rotation Changes

Player 2049 Role Recommended 2050 Role Rationale
Mimms Bench only Mix with starters Best shooter, best lineup pairings with Jihad (+53)
Totten Rotation Expanded, zone specialist +0.405 zone diff, +76 with Jihad
Coddington Primary creator Off-ball scorer 0.774 PnR BH vs 1.128 Cut PPP
Sliwoski Starter Reduced role 29.5% 3P collapses spacing, -90/-87/-66 worst duos

Data Notes

  • Parser version: v2 with corrected shot/foul ordering, 3PT detection, turnover attribution
  • Validation: 33/36 games within 3 points of header score; Game 26 (off 4), Game 30 (off 4), Game 32 (header error)
  • Possessions: 2,615 total tracked from play-by-play chains
  • Previous analysis errors corrected: Zone vulnerability was -0.5%, not -32%. Record was 16-20, not 17-21. Jihad roll man PPP was 1.106, not 0.740.
  • PBP data anomalies: Duplicate Game 8 numbering, missing Games 10/16 from numbering, Game 32 header score (14-18) is clearly wrong vs tracked (69-84)
SIUE Recruiting Phase 1 Summary 2050

SIUE Recruiting: Phase 1 Complete Summary (2050 Class)

Last Updated: Phase 1 End Scholarships Available: 2 Budget Remaining: $7,747 Calls Remaining: 0 (phase over) Visits Used: 7 HC visits (Delaney, Mangrum, Haug, Cannon, Seumalo, Kalil, Pfister)


Phase 1 Resource Usage

Resource Available Used Remaining
Budget ~$25,900 ~$18,175 $7,747
Phone Calls 50 50 0
HC Coach Visits 7 7 0
AC Coach Visits 0 0 0
Campus Visits 0 N/A N/A (Phase 1 has none)
Scholarships 2 0 2
pitches 2 0 2

Complete Target Board (41 Players)

GROUP A: Original 8 Targets (Called + Visited, Character Confirmed)

All 8 have clean character. All were called in Phase 1. HC visited: Delaney, Mangrum, Haug, Cannon, Seumalo, Kalil, Pfister (7 of 8). Antwine not visited.

Tier Name Pos Ht/WS OutS Key Ratings WE Interest Commit Distance Scout Lvl Notes
1 Clent Seumalo G 19 ? Very High Mid 91 mi L5 Best shooter + highest interest. Near-lock.
1 Emanuel Delaney PG 18 OnBD 20, End 13 20 High EARLY L5 Irreplaceable PG. Elite character. EARLY = Phase 5 deadline.
2 Jon Haug SG 17 Fnsh 18 20 High Mid 13 mi L5 Local. Reliable.
2 Mateo Mangrum SF 18 SlfC 20, End 10 11 Above Avg EARLY L5 ⚠️ Motor/effort yellow flag. Best SF creator.
2 Tshiefu Cannon SF 6'7/6'11 17 OnBD 17, End 18, Quck 19 ? High EARLY 14 mi L5 Local. Low Grav (5-9), low SlfC (11). Catch-and-shoot only.
3 Zeek Antwine W 6'4 16 OnBD 15, Playmaking 16-17, Pass 15 ? Above Avg EARLY 12 mi L4 Local. Undersized. Point-forward potential.
3 Jamauri Pfister G 15 End 6-9 ? High EARLY 453 mi L4 Can't play starter minutes (End 6-9).
3 Jayron Kalil ? 16 ? Average Late L3 Low priority. Average interest.

GROUP B: 33 New Phase 1 Scouts (Called, Need L5 Character Check)

All scouted to L3 and called in Phase 1. Need L3 to L5 in Phase 2 for character data.

B1: No Character Red Flags (Clean Signals) — PRIORITY FOR PHASE 2 L5

Name Pos Ht/WS St Dist OutS End OnBD SlfC Hndl Pass PnRE Grav Key Notes
Jules Toone PG/SG 6'0/6'3 WI 340 17 11 18 15 15 10 9 13 ✅ CONFIRMED: WE 20, Disc 19, Prof 20. "Workhorse." L5 scouted. Best character in pool. End 11 limits minutes.
Matiss Glasford SG 6'0/6'4 IL 91 16 18 19 12 10 11 13-15 11-15 No flags. Elite D + endurance + finishing (18). IL. Role player ceiling. FrT 18, FlDr 16-20.
Sarp Anderton PG/SG 6'0/6'3 MI 570 16 19 18 11 13 18 4-6 11-15 No flags. "Floor general who runs the show." Elite durability + defense + passing. ShOD 2, SlfC 11 = can't create.
Admassu Abdallah SF 6'4/6'8 OH 341 16 17 16 17 13 9 14-16 10-14 No flags. "Obsessive dedication to improvement" + "Motor runs consistently for 40 min." Fnsh 19. BD 16-20 concern. Playmaking 5.
Ikenna Ranson SF 6'4/6'7 WI 279 15 19 17 14 9 6 7-9 7-11 No flags. "Motor never slows down" + "Relentless pursuit of perfection." Energy/D specialist. OutS 15 = floor threshold.
LD Gates SG/PG 5'11/6'2 IL 249 14 18 15 14 15 15 15-17 14-18 No flags. "Sets the tone in practice." Best true PG profile in Tier 3. OutS 14 = below threshold. Age 17.

B2: Positive WE Signals Despite Inconsistent Flag

Name Pos Ht/WS St Dist OutS End OnBD SlfC Hndl Pass PnRE Grav Key Notes
Dima Bingaya SF 6'5/6'8 IL 246 17 17 13 6 5 9 7-9 7-11 Inconsistent BUT "obsessive dedication + motor runs 40 min." Pure catch-and-shoot. Quck 19.
Sai Iruafemi SG/SF 6'6/7'1 IL 73 16 16 15 10 8 7 13-15 13-17 Inconsistent BUT "Work ethic separates him from peers." Also "far too inconsistent." Local (73 mi). BD 15-19.

B3: High Upside, Character Flags (Inconsistent / Motor / Effort)

Name Pos Ht/WS St Dist OutS End OnBD SlfC Hndl Pass PnRE Grav Flags Key Notes
Jovan Hauser PG 6'2/6'7 OH 477 19 16 18 17 13 16 12-14 11-15 Inconsistent Best shooter in pool (tied Seumalo). ShOD 16. Senior. Hndl 13 below PG threshold.
Jabe Raymond SG/SF 6'3/6'6 MI 447 16 18 17 19 18 18 6-8 15-19 Inconsistent "Sublime playmaker." Elite creator. OffR 15, DefR 16. ShOD 4 contradicts SlfC 19. BBIQ 5-8. Highest upside wing.
Rashaad Herndon PG 6'5/6'10 OH 486 18 9 18 17 15 17 17-19 12-16 Double: motor + inconsistent Rarest PnR skill in pool. "Starter potential" pro ceiling. End 9 is crippling. Great size.
Dekedran Stockton SG/SF 6'8/7'3 IN 299 17 14 13 15 15 5 16-18 16-20 Inconsistent + "lacking intangibles" Best physical tools (size/WS). PnRE + Grav elite. BBIQ 3-6, Pass 5. "Rotation player upside."
Matur Marx SG 6'8/7'0 IN 216 16 18 17 13 9 13 5-7 10-14 Inconsistent + "disappears" Great size for SG. Elite endurance + D. Catch-and-shoot. ShOD 3.
Jaxson Nickelson SG/PG 6'0/6'3 IN 200 15 15 17 15 16 10 17 14 Inconsistent, WE 17 Disc 9 L5 confirmed. PnRE 17 rare. Disc 9 + inconsistent = yellow. "Friendly."
Jamion Dunklin SF 6'7/6'10 OH 498 18 17 14 5 5 8 3-5 7-11 Inconsistent + "poor decision maker" Pure spot-up. Zero creation. "Alpha personality who sets the tone." Spacing 16-20.
Nolen Ratcliff SF 6'8/6'11 IL 239 16 19 17 11 8 5 7-9 9-13 Inconsistent Very High interest. Elite End + athleticism. FstS 18. Str 3. IL.
Jaison Ames PF 6'11/7'3 IL 19 15 16 3 7 5 7 7-9 6-10 Inconsistent, "dedication in practice" positive Only stretch big. Very High interest. Ultra local. RimP 15, PostD 13, OffR 13, DefR 14. OnBD 3 = can't switch.
Kuba Haralson PG 5'9/6'0 WI 370 16 18 17 15 17 10 8-10 9-13 "Could be more dedicated" Complete tools but tiny and WE concern. Fnsh 17. "Rotation upside."
Dyrbe Olsen SG/SF 6'2/6'5 IN 225 18 13 18 5 15 7 4-6 7-11 Inconsistent 3&D specialist. Stl 17, Spd 19. Zero creation. 4.6 MPG in HS. Late commit.
Haboubacar LaGrone SG 6'2/6'4 MI 410 18 16 14 14 12 6 10-14 11-15 "Fades from impact" ShOD 16, Spd 19, Str 16. Playmaking 15 but Pass 6 contradicts.
Frederick Crossland SG/SF 6'5/6'8 IL 264 15 14 20 13 13 10 10-12 8-12 Inconsistent, "relentless pursuit of perfection" Very High interest. Best perimeter defender in pool. OutS 15 = floor. Fnsh 7. IL.

B4: PASS — Not Worth Phase 2 Investment

Name Pos Why Pass
Jontray Easton PG BBIQ 6-9, 5'10/5'9 WS, inconsistent, "poor decision maker."
LaRond Piepoli PG BBIQ 4-7, PnRE 7-9, inconsistent. ShOD 19 interesting but can't run offense.
Zep Golladay PG PnRE 5-7, BBIQ 7-10, inconsistent. Good size (6'5) but can't execute PnR.
Issiah Wnuk SG/PG Double flag: motor + intensity. ShOD 3 despite Hndl 18.
Traver Leeks PG/SG End 10, Str 2, 4.4 MPG in HS, inconsistent. BD 16-20 (ball hog).
Benjamin Voelkel PG Motor flag. PnRE 9-11, Playmaking 9. Can't run offense despite Very High interest.
Marque Berg SG/PG OnBD 11, SlfC 9. Can't defend or create. Inconsistent.
Darian Amis SG "Inconsistent work ethic" explicit in scouting text. OnBD 11.
Evann Burrus PG WE 8. Disqualifying. PnRE 18 + 6'6 is dream profile but motor kills it. Disc 20, Prof 20 can't save WE 8.
Sedric Dunaway SF "Work ethic won't separate him" + "intensity varies." IL. SlfC 18, Grav 15-19 elite but character too risky.
Joevan DeLord SF OutS 14 + "could be more dedicated" + "disappears." Multiple character flags.
O'Karo Isham SF OutS 14, OnBD 11, Str 2. Below Morey thresholds. Inconsistent + "intensity varies."

Scholarship Strategy

  1. Emanuel Delaney — EARLY commit, offer in Phase 3. Irreplaceable PG.
  2. Clent Seumalo — Mid commit, offer in Phase 3. Best shooter + Very High interest.

Plan B (If Delaney falls through)

  1. Clent Seumalo
  2. Jon Haug OR Tshiefu Cannon (both local, both clean)

Plan C (If Seumalo falls through)

  1. Emanuel Delaney
  2. Jon Haug (safe) OR Tshiefu Cannon (local SF need)

Phase 2 Priorities

1. L5 Character Checks (Budget: $7,747)

Priority order for L3 to L5 scouting ($375/each Great Lakes):

  1. Matiss Glasford — No flags, elite role player profile
  2. Sarp Anderton — No flags, floor general
  3. Admassu Abdallah — No flags, "obsessive dedication"
  4. Ikenna Ranson — No flags, "relentless pursuit of perfection"
  5. LD Gates — No flags, true PG (OutS 14 borderline)
  6. Dima Bingaya — Positive WE signals despite inconsistency
  7. Sai Iruafemi — "WE separates him from peers" despite inconsistency
  8. Jovan Hauser — OutS 19, best shooter
  9. Jabe Raymond — Elite creator, highest upside wing
  10. Rashaad Herndon — PnRE 17-19, rarest skill
  11. Dekedran Stockton — Best physical tools
  12. Nolen Ratcliff — Very High interest, End 19, IL
  13. Jaison Ames — Only stretch big, Very High interest, ultra local
  14. Frederick Crossland — Very High interest, OnBD 20, IL

At $375/each, 14 checks = $5,250. Budget allows ~20 checks ($7,500).

2. Visit Allocation (Phase 2: 7 HC visits + 3 campus visits — ALL REFRESH)

Visits refresh each phase. This gives us 7 HC + 3 campus every phase.

Campus visits (3) — Tier 1 scholarship targets (prerequisite for Phase 3 scholarship offer): - Seumalo - Delaney - Haug OR Cannon (whichever is backup scholarship)

HC visits (7) — Split between original targets and best new character-clean survivors: - Re-visit: Mangrum, Cannon/Haug (whichever didn't get campus), Antwine (never visited) - New: Top 4 clean character survivors from Phase 2 L5 checks

3. Call Strategy

  • Re-call all 8 original targets
  • Call clean Phase 2 character check survivors

Key Learnings from Phase 1 Scouting

  • 77% character failure rate confirmed across all tiers. "Maddeningly inconsistent" dominates the prospect pool.
  • Positive WE language in scouting text is predictive. Players with "obsessive dedication," "relentless pursuit," "work ethic separates him" tend to have clean character even with inconsistent flags.
  • Explicit WE concerns in scouting text are disqualifying. "Could be more dedicated," "inconsistent work ethic," "work ethic won't separate him" = fail.
  • OutS and End are the two ratings that most determine usability. Elite shooters with End < 10 can't play starter minutes.
  • PnR Execution is the rarest elite skill. Only Herndon (17-19), Burrus (18), Nickelson (17), Stockton (16-18), and Duckworth (18) showed elite PnRE. Most prospects score 4-10.
  • BBIQ correlates poorly with other skills. Many elite shooters/defenders have BBIQ 3-8. Either the rating is unreliable at L3 or the sim devalues it.
  • International recruiting is unviable at Very Low prestige. Zero internationals with High+ interest. Scouting costs 16x domestic ($9,200 vs $575 for L3).
  • Scouting report text provides character signals not captured in ratings. Always read the strengths/weaknesses section for WE/motor/dedication language.

Scouting Cost Reference

Region L2 L3 L4 L5 (from L3)
Great Lakes (home) $275 $575 $950 $375
Great Plains (neighbor) $550 $1,150 $1,900 $750
South Atlantic (moderate) $1,320 $2,760 $4,560 $1,800
International $4,400 $9,200

Max 2 levels per player per phase.

SIUE Recruiting Phase 2 Plan 2050

SIUE Recruiting: Phase 2 Plan (2050 Class) -- POST-HTML EXTRACTION UPDATE

Last Updated: Phase 2, mid-phase (HTML page extractions complete for 6 players) Phase Duration: 48 hours (deadline ~Feb 10, 2026 02:00 UTC) Phase Type: Last pure information-gathering phase before scholarships open (Phase 3) Status: Actions nearly exhausted. Awaiting phase run. Remaining: 6 calls only.


Phase 2 Resources -- CURRENT (from Cannon page header)

Resource Total Used Remaining
Budget (overall) $275,968 ~$90,820 $185,148
Budget (Phase 2) -- -- $24,105
Scholarships 2 0 2 (Phase 3)
pitches 2 0 2
Phone Calls 30 24 6
HC Coach Visits 8 8 0
Campus Visits 3 3 (Delaney, Seumalo, Haug) 0
Scouting Max 2 levels/player/phase Complete --

COMPETITIVE INTELLIGENCE -- HTML Page Extractions (Phase 2)

Extracted from full HTML source of player recruiting pages. Contains Top Schools interest bars, player values, philosophy preferences, and competitive landscape data not visible in normal scouting.

Emanuel Delaney (PG) -- Priority 1

Metric Value
SIUE Rank 4th of 10 schools
Interest Bar 83% (success)
#1 School Green Bay at 100% (gold bar, campus visit offered)
Gap to #1 17 points
Interest Raw High
Last Change Positive
Philosophy Pref Balanced (not Morey Ball)
Prestige Weight Medium (risky for VL)
Girlfriend Intel "Girlfriend going to Purdue"
Commit Phase EARLY (decides Ph 3-4, commits Ph 5)
Decision Mode Accepting New Offers

Character (L5 confirmed): WE 20, Disc 15, Prof 16 -- CLEAN Key ratings: OutS 18, OnBD 20, End 13, Pass 17, BBIQ 14, Spd 17

Assessment: Most important target but most at-risk. 4th of 10 with Green Bay dominant. EARLY commit means Phase 3-4 is the decision window. Must deploy scholarship + possibly pitch immediately in Phase 3. Only clean-character PG in the entire pool.

Clent Seumalo (G) -- Priority 2 (CHARACTER DECISION PENDING)

Metric Value
SIUE Rank 2nd of ~5 schools
Interest Bar 90% (success)
#1 School Unknown at 100%
Gap to #1 10 points
Interest Raw Very High (903, last change +37)
Philosophy Pref Balanced (not Morey Ball)
Prestige Weight Medium
Girlfriend Intel "Girlfriend going to Arizona State" (ASU is #1 on his Initial List)
Commit Phase MID (decides Ph 5-6, commits Ph 7)
Decision Mode Accepting New Offers

Character (L5 confirmed): WE 20, Disc 8, Prof 9 -- FAIL (both below 10 floor) Key ratings: OutS 19, OnBD 19, Spd 18, Pass 18, End 10 (stamina limitation)

CRITICAL UPDATE: Seumalo was listed as "CLEAN (Group A)" in all previous plans. L5 HTML extraction reveals Disc 8 and Prof 9, both below the "disqualifying regardless of WE" threshold of 10. By our own character rules, Seumalo is disqualified.

The dilemma: OutS 19 is the best in the entire pool. 2nd place with huge interest momentum (+37). At Very Low prestige, can we afford to enforce the character threshold on our best on-court talent? Decision deferred until all HTML extractions complete.

Jon Haug (SG) -- Elevated to Top Target

Metric Value
SIUE Rank 2nd of 3 schools (very low competition!)
Interest Bar 83% (success)
#1 School Unknown at 90%
Gap to #1 7 points (closest of any target)
Interest Raw Very High (842, last change +40)
Philosophy Pref MOREY BALL (PhilosophyValue 8) -- ONLY Morey Ball match in pool!
Prestige Weight Low (47) -- favorable for VL SIUE
Distance 13 miles (ultra-local)
Academics Weight Low (49)
Commit Phase MID (decides Ph 5-6, commits Ph 7)
Decision Mode Accepting New Offers
Personality Introverted / Neutral

Character (L5 confirmed): WE 20, Disc 11, Prof 10 -- BORDERLINE - Disc 11: 1 below 12+ threshold, but above 10 floor - Prof 10: Exactly at "disqualifying" floor (our rule: "below 10 is disqualifying") - No Troublemaker personality

Key ratings: OutS 17, Fnsh 18, Spac 17, Spd 19, Quick 18, Stl 18, End 16 Tendencies: C&S:PullUp 91:9 (pure catch-and-shoot), Corner:ATB 46:54, Roll:Pop 28:72 Weaknesses: SlfC 5, Pass 5 (pure specialist -- can't create or distribute)

Why Haug is now a top target: 1. Only Morey Ball philosophy match in the entire recruiting pool 2. Lowest competition (3 schools vs Delaney's 10) 3. Closest to #1 (7-point gap, closeable) 4. Ultra-local (13 mi) with Low prestige weight -- checks every box SIUE can offer 5. Massive +40 interest jump shows our Phase 2 investment paid off 6. Pure catch-and-shoot profile (91% C&S) is textbook Morey Ball role player 7. All Phase 2 actions already complete (L5 scouted, 2x called, HC visit, campus visit)

Tshiefu Cannon (SF/SG) -- DROPPED (CHARACTER HARD FAIL)

Metric Value
SIUE Rank 1st of 3 (tied at 72%)
Interest Bar 72% (midsuccess) -- all 3 schools tied
Last Change +0 (flat -- Phase 2 actions had no effect)
Philosophy Pref Balanced (not Morey Ball)
Prestige Weight Medium (62) -- unfavorable for VL
Academics Weight High (85) -- SIUE academics only 30
Distance 14 miles (local)
Commit Phase EARLY

Character (L5 confirmed): WE 18, Disc 6, Prof 20 -- CATASTROPHIC FAIL - Disc 6 is well below the "disqualifying regardless of WE" floor of 10 - Confirms meta pattern: "Leader = high Prof but Disc varies" (Lead 20, Prof 20, but Disc 6) - Confirms "Prof 20 is NOT a strong positive signal"

On-court Morey Ball fit also weak: Spac 11 (below 15+ threshold), Fnsh 8 (can't finish at rim), Grav 8. Breaks the rim-or-three profile. Defense is strong (OnBD 17, OffBD 16) but doesn't compensate.

Verdict: DROP from all scholarship plans. Stop all investment. Do not call.

Sai Iruafemi (SG/SF) -- CLEAN Character, Backup Scholarship Target

Metric Value
SIUE Rank 1st of 1 school (SIUE is ONLY school in Top Schools!)
Interest Bar 100% (gold-level)
Interest Raw Very High (897, last change positive)
#1 School SIUE (no competition visible)
Philosophy Pref Unknown (not displayed)
Prestige Weight Unknown
Distance 73 miles (local -- Great Lakes region)
Commit Phase MID (decides Ph 5-6, commits Ph 7)
Decision Mode Accepting New Offers
Personality Workhorse

Character (L5 confirmed): WE 20, Disc 20, Prof 20 -- PERFECT Key ratings: OutS 15, 3PT 15, PnRE 16, Grav 16, Spac 13, OnBD 16, End 16 Wingspan: 7'1 (elite for 6'6 SG/SF)

Initial List (10 schools): Bradley, Ohio State, Northwestern, Notre Dame, SMU, Kent State, Southern Illinois, Michigan State, Western Illinois, DePaul

Assessment: Safest target on the board. SIUE is the ONLY school recruiting him -- zero competition visible. Perfect character trifecta (20/20/20) with Workhorse personality. Interest is Very High and trending positive. OutS 15 is below Delaney/Haug tier but the combination of zero competition + perfect character + local + 7'1 wingspan makes him an extremely high-probability signing. Should be scholarship target if any higher-priority target falls through.

Phase 2 actions completed: L5 scouted, 2x called (Ph1 Academics, Ph2 Playing Time), HC visit.

Kenny Danridge (PG) -- CHARACTER FAIL + POOR MOREY BALL FIT

Metric Value
SIUE Rank 1st of 1 school (SIUE is ONLY school in Top Schools)
Interest Bar 100% (gold-level, highest raw interest in pool: 954)
Interest Raw Very High (954)
#1 School SIUE (no competition visible)
Philosophy Pref Balanced (not Morey Ball)
Prestige Weight Unknown
Distance 131 miles (Great Lakes region, has family in region)
Commit Phase MID (decides Ph 5-6, commits Ph 7)
Decision Mode Accepting New Offers
Personality Friendly (neutral signal)

Character (L5 NOW CONFIRMED -- was PENDING): WE 19, Disc 9, Prof 8 -- FAIL (both below 10 floor) L5 Scouting ratings (full): - Shooting: 3PT 18 (excellent), SlfC 16 - PnR/Creation: PnRE 9 (average -- BAD for Morey Ball PG), BallD 8, Play 9 - Defense: OnBD 18, OffBD 17, Stl 16 (elite perimeter D) - Physical: End 18, Ath 15, Quck 15, Spd 15, Fnsh 7 (can't finish at rim) - Size: 5'9/6'0 WS (very small)

HS Stats: 10.9 PPG, 4.2 APG, 2.21 SPG, .265 3P%, .325 FG%

Initial List (10 schools): Northern Illinois, Bradley, Illinois State, Northwestern, DePaul, Loyola Chicago, Evansville, Western Michigan, Kent State, Wisconsin

Assessment: DROP from scholarship plans. 1. Character FAIL: Disc 9 and Prof 8 both below the disqualifying floor of 10. 2. Poor Morey Ball PG fit: PnRE 9 means he can't run PnR (the core Morey Ball action). Fnsh 7 means he can't finish at rim. BallD 8 and Play 9 are low for a primary ball-handler. 3. Shooting paradox: 3PT rating 18 but actual 3P% is .265 -- suggests BBIQ/shot selection issues (consistent with "disappears for stretches" scouting text). 4. Size risk: 5'9/6'0 WS is undersized even at Very Low prestige. 5. Despite highest interest in the pool (954), the character fail alone disqualifies him under our rules.

Phase 2 actions completed: L5 scouted (L3->L5), 1x called (Ph2 Playing Time), HC visit.


Phase 2 Scouting Results -- Complete Character Data

80% character fail rate confirmed (worse than 77% historical average)

CLEAN (3 of 16 L5 scouted in Phase 2 batch)

Player Pos Ht/WS OutS End OnBD WE Disc Prof Personality Key Value
Jules Toone PG/SG 6'0/6'3 17 11 18 20 19 20 -- Best character in pool. End 11 limits minutes.
Sai Iruafemi SG/SF 6'6/7'1 15 16 16 20 20 20 Workhorse Perfect character trifecta. 7'1 WS. Local (73 mi). PnrE 16, Grav 16.
Ikenna Ranson SF 6'4/6'7 15 19 18 20 13 14 Introverted Iron man 3&D. Pure C&S (77:23). WE maximum. Str 12 decent.

GROUP A CHARACTER RESULTS (from HTML Page Extractions)

Player Pos OutS WE Disc Prof Character Philosophy Match Competition
Delaney PG 18 20 15 16 CLEAN Balanced 4th of 10 (at risk)
Seumalo G 19 20 8 9 FAIL Balanced 2nd of ~5
Haug SG 17 20 11 10 BORDERLINE Morey Ball 2nd of 3
Cannon SF 16 18 6 20 HARD FAIL Balanced 1st of 3 (tied)
Iruafemi SG/SF 15 20 20 20 CLEAN (PERFECT) Unknown 1st of 1 (no competition)
Danridge PG 18 19 9 8 FAIL Balanced 1st of 1 (no competition)

Key finding: Of the 6 targets with HTML-confirmed data, only Delaney and Iruafemi have clean character. Seumalo, Cannon, and Danridge are disqualified by the Disc <10 rule. Haug is borderline.

MIXED -- Usable with Risk Management (6 of 16)

Player Pos Ht/WS OutS End OnBD WE Disc Prof Issue Best Trait
Nolen Ratcliff SF 6'8/6'11 16 19 17 10 20 20 Low WE = slow dev Professional. Best SF size. Disc/Prof perfect.
Sarp Anderton PG/SG 6'0/6'3 16 19 19 17 11 20 Disc 11 Leader personality. End 19 + OnBD 19. Pass 17.
Dima Bingaya SF 6'5/6'8 17 17 14 19 11 20 Disc 11 Driven. Pure C&S (94:6). FlDr 19, Quck 19.
Jabe Raymond SG/SF 6'3/6'6 15 18 17 20 8 15 Disc 8 Confident. Play 19, SlfC 19, Grav 18, Hndl 18. Creator trap risk.
Matiss Glasford SG 6'0/6'4 16 18 19 19 8 20 Disc 8 Leader. Best two-way SG. Fnsh 18, FlDr 19, PnrE 16.
Dekedran Stockton SG/SF 6'8/7'3 17 14 14 20 9 20 Disc 9, End 14 Driven. PnrE 18, Grav 19. Best PnR/Gravity combo.

FAIL -- Drop Immediately (9 of 16 + Cannon from Group A)

Player WE Disc Prof Personality Disqualifier
Tshiefu Cannon 18 6 20 Leader Disc 6. Group A target -- resources freed. Morey fit also weak (Spac 11, Fnsh 8).
Zeek Antwine 17 5 11 Friendly Disc 5 critically low. Group A target -- resources freed.
Frederick Crossland 19 20 7 Demanding Prof 7 + OutS 14 (below Morey Ball threshold).
LD Gates 20 9 8 Introverted Double fail. OutS 14 also below threshold.
Admassu Abdallah 19 7 7 Introverted Double fail. BalD 19 + Disc 7 = ball hog with bad decisions.
Jovan Hauser 7 5 5 Troublemaker Triple fail. Worst character in class despite best on-court Morey fit.
Jaison Ames 16 12 10 Friendly On-court fail: OnBD 3, Quck 8, Spd 5. Can't play Morey Ball.
Jayron Kalil 19 9 11 Introverted Disc 9 + OnBD 9 + Average interest + Late commit. Elite offense wasted.
Kenny Danridge 19 9 8 Friendly Disc 9/Prof 8 double fail. PnRE 9 + Fnsh 7 = can't run Morey Ball PnR. 3PT 18 but .265 3P%. Size 5'9/6'0.

Tier 4 L3 Results -- Danridge L5 Now Complete

Player Pos OutS Morey Ball Fit Verdict
Kenny Danridge PG 18 L5 COMPLETE -- CHARACTER FAIL (Disc 9, Prof 8). DROPPED. PnRE 9 + Fnsh 7 = can't run Morey Ball PnR. L5 COMPLETE -- CHARACTER FAIL (Disc 9, Prof 8). DROPPED.
Deriece Hockman SG/PG 18 Marginal -- ShtD 3, OnBD 13, age 20 Low priority monitor
Keon Culpepper SF 16 Fail -- BBIQ 2, SlfC 5, ShtD 4 DROP
Jakeem Flahive PG 13 Fail -- OutS below threshold DROP
Barron Phillips SG 12 Fail -- OutS below threshold DROP
Burton Davison SG 10 Fail -- OutS below threshold DROP
Dan Curnell C 8 Fail -- OutS below threshold, traditional C DROP

Critical Meta Learnings from Phase 2 Scouting

  1. "Inconsistent" in scouting text has ZERO correlation with Discipline. Iruafemi has "Maddeningly inconsistent" AND "Far too inconsistent" yet Disc 20 (maximum). Ratcliff has "Maddeningly inconsistent" yet Disc 20. Crossland has "Maddeningly inconsistent" yet Disc 20. The inconsistency flag describes on-court performance variance (likely driven by low BBIQ), NOT character discipline.

  2. Positive WE language predicts WE specifically, NOT Disc/Prof. "Obsessive dedication" predicted WE 19 for Abdallah, but his Disc was 7 and Prof was 7. "Sets the tone in practice" predicted WE 20 for Gates, but his Disc was 9 and Prof was 8. WE, Disc, and Prof are fully independent ratings.

  3. 80% fail rate is the new baseline (worse than 77% historical). Only 3 clean out of 16 L5 scouted. Budget for volume scouting in future phases.

  4. Personality types provide signal. Workhorse (Iruafemi) = clean. Professional (Ratcliff) = reliable Disc/Prof. Troublemaker (Hauser) = catastrophic. Leader (Anderton, Glasford, Cannon) = high Prof but Disc varies wildly. Driven (Bingaya, Stockton) = high WE/Prof but Disc varies.

  5. Prof 20 is the most common elite intangible. Many players with poor Disc still have Prof 20 (Anderton, Bingaya, Glasford, Stockton, Crossland, Cannon). Prof appears easier to max than Disc or WE. Don't treat Prof 20 as a strong positive signal by itself.

  6. Philosophy match matters. Haug is the ONLY target in the pool who prefers Morey Ball. All other major targets (Delaney, Seumalo, Cannon) prefer Balanced. Philosophy match may provide a recruiting edge at low prestige.

  7. One call per player per phase. Cannot stack calls. Burn remaining calls on any player with interest to prevent knockdowns -- there is no downside to calling a player you don't want.


Phase 2 Actions -- ALL COMPLETE

Delaney (Priority 1 -- ALL IN)

  • [x] Campus visit (matching Green Bay's offer)
  • [x] HC visit
  • [x] Call
  • Phase 3: Scholarship offer + pitch #1 if needed

Seumalo (Priority 2 -- CHARACTER DECISION PENDING)

  • [x] Campus visit (claiming visit slot)
  • [x] HC visit
  • [x] Call
  • Phase 3: Scholarship offer contingent on character decision
  • CHARACTER FAIL: Disc 8, Prof 9 -- both below 10 floor

Haug (ELEVATED -- Plan B Scholarship Target)

  • [x] Campus visit (3rd visit -- positioning for Phase 3 scholarship)
  • [x] HC visit
  • [x] Call
  • Phase 3: Scholarship if Seumalo dropped from plans
  • ONLY Morey Ball philosophy match in pool. Borderline character (Disc 11, Prof 10).

Character-Clean Discoveries

  • [x] HC visit + call: Iruafemi (CLEAN 20/20/20)
  • [x] HC visit + call: Ranson (CLEAN 20/13/14)
  • [x] HC visit + call: Toone (CLEAN 20/19/20)

Mixed Tier (maintaining contact)

  • [x] HC visit + call: Glasford (Disc 8 but best two-way SG)
  • [x] HC visit + call: Danridge (L5 CHARACTER FAIL -- Disc 9, Prof 8. DROPPED.)
  • [x] Call: Ratcliff, Stockton, Raymond, Hockman

Calls (30 total, 24 used, 6 remaining)

  • [x] ~20 priority calls to board targets
  • [x] ~4 additional calls for knockdown prevention
  • [ ] 6 remaining -- use on any uncalled interested prospects before phase runs

COACH VISITS COMPLETED (8 of 8)

Slot Target Type Rationale
1 Delaney HC ALL IN. 4th of 10, must climb.
2 Seumalo HC Reinforce 2nd place.
3 Haug HC Plan B scholarship target. Local (13 mi). Morey Ball match.
4 Iruafemi HC CLEAN (20/20/20). Local (73 mi). Best character discovery.
5 Ranson HC CLEAN (20/13/14). Iron man 3&D wing. End 19 + OnBD 18.
6 Toone HC CLEAN (20/19/20). Never visited before. Best character in pool.
7 Glasford HC Mixed but best two-way SG. WE 19 + Prof 20 offset Disc 8 partially.
8 Danridge HC Tier 4 breakout. OutS 18/End 18/OnBD 18. Character pending. Size concern (5'9/6'0).

MAJOR PITCHES -- UPDATED STRATEGY

Pitch #1: RESERVE FOR DELANEY (Phase 3-4) - Still the correct use. He's the only clean-character PG in the pool. - Deploy if not in his Top 3 after Phase 2 runs. - 4th of 10 with Green Bay dominant -- likely needed.

Pitch #2: CONDITIONAL - If Seumalo stays in plans despite character: save for Seumalo if competition intensifies. - If Seumalo dropped: save for Haug if a new competitor enters his small field. - Or hold for surprise late-phase need. - Do NOT use both on the same player.


SCHOLARSHIP STRATEGY (Phase 3) -- REVISED

KEY CHANGE: Seumalo (Disc 8, Prof 9) and Cannon (Disc 6) both fail character threshold. Previous plans had Seumalo in ALL 6 scholarship options and Cannon as Plan C. Both are now invalidated.

Option 1: ENFORCE CHARACTER THRESHOLD STRICTLY

Drop Seumalo from scholarship plans. Haug becomes the primary Plan B.

Plan Target 1 Target 2 Trigger
A (Best) Delaney (PG, OutS 18, CLEAN, EARLY) Haug (SG, OutS 17, BORDERLINE, Morey Ball, MID) Both available. pitch on Delaney.
B Delaney Iruafemi (SG/SF, OutS 15, CLEAN 20/20/20, MID) Haug lost.
C Delaney Ranson (SF, OutS 15, CLEAN 20/13/14, EARLY) Haug + Iruafemi lost.
D Haug Iruafemi Delaney lost. Two Morey-compatible wings.
~~E~~ ~~Haug~~ ~~Danridge~~ ~~INVALID -- Danridge CHARACTER FAIL (Disc 9, Prof 8). L5 confirmed.~~
E Iruafemi Ranson Delaney + Haug lost. All-clean, lower ceiling.

Note: Haug's borderline character (Disc 11, Prof 10) is accepted in this scenario because he's the only Morey Ball philosophy match, has minimal competition (3 schools), and his Disc 11/Prof 10 are above the "disqualifying" 10 floor (even if below the 12+ "preferred" threshold).

Option 2: ACCEPT SEUMALO DESPITE CHARACTER FAIL

Keep Seumalo in plans, acknowledging Disc 8 and Prof 9 as a known risk for the sake of OutS 19 talent.

Plan Target 1 Target 2 Trigger
A (Best) Delaney (PG, CLEAN, EARLY) Seumalo (G, OutS 19, FAIL character, MID) Both available. pitch on Delaney. Accept character risk on Seumalo.
B Delaney Haug (SG, BORDERLINE, Morey Ball, MID) Seumalo lost or character concern too high.
C Seumalo Haug Delaney lost. Best on-court talent + best system fit.
D Seumalo Iruafemi Delaney + Haug lost.
E Haug Iruafemi Delaney + Seumalo lost.

Risk: Disc 8 means potential discipline issues (locker room, effort inconsistency). Prof 9 compounds. At Very Low prestige with a young roster, character problems can cascade. But OutS 19 may be worth the gamble when our alternative is OutS 15-17 players.

DECISION RESOLVED: Option 1 -- Enforce character threshold strictly. Seumalo DROPPED. See Phase 3 Plan.


COMPLETE PROSPECT BOARD -- CORRECTED RANKINGS

Tier 1: Scholarship Targets (2 offers available Phase 3)

Rank Player Pos OutS End OnBD Character Interest Commit Status
1 Delaney PG 18 13 20 CLEAN (20/15/16) 4th of 10 EARLY ALL IN. pitch reserved. Only clean PG.
2 Haug SG 17 16 14 BORDERLINE (20/11/10) 2nd of 3 MID ELEVATED. Only Morey Ball match. Lowest competition. Ultra-local.

Tier 1B: Conditional (pending character decision)

Rank Player Pos OutS End OnBD Character Interest Commit Status
2B Seumalo G 19 10 19 FAIL (20/8/9) 2nd of ~5 MID Best on-court talent. Character decision pending.

Tier 2: Backup Scholarship Targets

Rank Player Pos OutS End OnBD Character Interest Commit Notes
3 Iruafemi SG/SF 15 16 16 CLEAN (20/20/20) Very High MID Perfect character. 7'1 WS. Local (73 mi). PnrE 16, Grav 16. SIUE #1 (no competition).
4 Ranson SF 15 19 18 CLEAN (20/13/14) High EARLY Iron man. Pure 3&D. WE maximum.
5 Toone PG/SG 17 11 18 CLEAN (20/19/20) -- -- Best character. End 11 limits to 15-20 MPG.

Tier 3: Mixed Character -- High Upside Gambles

Rank Player Pos OutS End OnBD Character Issue Upside
7 Ratcliff SF 16 19 17 WE 10 (slow dev) 6'8/6'11 size. Disc 20/Prof 20.
8 Anderton PG/SG 16 19 19 Disc 11 End 19, OnBD 19, Pass 17.
9 Glasford SG 16 18 19 Disc 8 Best two-way SG. Fnsh 18, FlDr 19.
10 Bingaya SF 17 17 14 Disc 11 OutS 17, pure C&S.
11 Stockton SG/SF 17 14 14 Disc 9, End 14 PnrE 18, Grav 19. Best PnR combo.
12 Raymond SG/SF 15 18 17 Disc 8 Play 19, SlfC 19. Most talented overall.

Tier 4: Dropped (stop all investment)

Mangrum, Cannon, Antwine, Hauser, Gates, Abdallah, Crossland, Ames, Kalil, Pfister, Flahive, Davison, Phillips, Culpepper, Curnell, Danridge (L5 CHARACTER FAIL -- Disc 9, Prof 8)


PHASE 3 PREPARATION -- FIRST ACTIONS

  1. Offer scholarship to Delaney -- first available. He's EARLY commit, entering Deciding Phase 3-4.
  2. Decide on Seumalo character question -- enforce threshold or accept risk?
  3. Offer second scholarship based on decision: Seumalo (if accepted) or Haug (if enforcing threshold).
  4. Assess: pitch #1 on Delaney? -- if still not in his Top 3 after Phase 2 runs, deploy immediately.
  5. Call all board targets -- refresh calls on all remaining targets + burn remaining calls on interested prospects.

Phase 3 Contingency Planning

  • If Delaney rejects scholarship: shift to Haug as primary + Iruafemi for #2.
  • If Haug competition intensifies: consider pitch #2 on Haug (his small field makes it high-impact).
  • If both Delaney AND Haug are lost: crisis mode. Iruafemi + Ranson are clean but limited ceilings (OutS 15).
  • Danridge L5 confirmed CHARACTER FAIL (Disc 9, Prof 8). No longer a contingency option. Removed from all plans.

Remaining HTML Extractions Needed

Player Priority Why
Toone Medium CLEAN character. Lower priority -- End 11 limits role.
Ranson Medium CLEAN character. Could be useful if higher targets fall through.

Iruafemi and Danridge HTML extractions COMPLETE (see sections above).


Scouting Cost Reference

Region L2 L3 L4 L5 (from L3)
Great Lakes (home) $275 $575 $950 $375
Great Plains (neighbor) $550 $1,150 $1,900 $750

Max 2 levels per player per phase.

SIUE Recruiting Phase 3 Plan 2050

SIUE Recruiting: Phase 3 Plan (2050 Class)

Last Updated: 2026-02-10 (Pitch timing, commit timeline, and call limit corrections from community intel) Phase Duration: 24 hours (shortest phase -- immediate action required) Phase Type: FIRST phase where scholarship offers can be made Status: Pre-phase plan. Execute after Phase 2 results confirmed.

Reference: See SIUE Recruiting Phase 2 Plan 2050.md for complete L5 character data, HTML competitive intelligence, scouting results, and meta learnings.


Analytical Audit (2026-02-09) -- Cross-Project Data Review

All 5 active targets evaluated against college win predictors (368-team GB models), pro Ridge regression (36-team, adj-R²=0.642), and Morey Ball fit thresholds.

Key findings: - Delaney confirmed #1 — only CLEAN PG, 5/5 Morey Ball. BBIQ 8 is a concern (college predictor #1) but no PG in the pool has better BBIQ. Positional necessity overrides. - Haug confirmed #2 — best college predictor profile (Help Def 13, only one above 9). C&S 91% = ideal Morey Ball wing. BORDERLINE character accepted. - Iruafemi reframed — highest pro Ridge score in the group (2.110) driven by PnRE 16 + BallDom 18. Workhorse character guarantees development ROI. College impact limited by BBIQ 5 + 3PT 13 (below Morey Ball SG minimum). Long-term asset, not Year 1 contributor. - Toone elevated to Tier 1.5 — analytically stronger than Ranson: 3PT 17 vs 15, SelfC 15 vs 14, ShotOD 15 vs 3 (massive), Workhorse vs CLEAN. Only weakness is End 11 (limits minutes). Replaces Ranson in contingency Plans C and E. - BBIQ 8 is universal across all targets — pool constraint at Very Low prestige, not a target selection failure.

Pro Ridge scores: Iruafemi 2.110 > Haug 2.079 > Ranson 1.813 > Toone 1.755 > Delaney 1.377


Seumalo Decision -- RESOLVED

Decision: Option 1 -- Enforce character threshold strictly. Seumalo DROPPED.

Seumalo's L5 character (WE 20, Disc 8, Prof 9) has both Disc and Prof below the disqualifying floor of 10. Despite OutS 19 being the best in the pool and strong interest momentum (2nd of ~5 schools, +37 interest jump), the character threshold is enforced consistently. At Very Low prestige with a young roster, Disc 8 + Prof 9 risks locker room issues that compound.

Seumalo is removed from all scholarship plans, call lists, visit allocations, and contingency options. His campus visit (Phase 2) is a sunk cost.


Phase 3 Resources (Fresh Allocation)

Resource Phase 3 Allocation Notes
Calls 30 (1/player/phase, max 3 total per recruit) All scholarship targets on 3rd/FINAL call = promise required
Coach Visits 7 (HC + AC combined) HC preferred for scholarship targets
Campus Visits 3 Delaney/Seumalo/Haug already visited in Ph2. Use on Ranson, Iruafemi, Toone.
Scouting 2 levels/player Not needed -- all targets at L5
Scholarships 2 (hard offers only) First phase available. No soft offers.
pitches 2 #1 conditional on Delaney, #2 held
Budget ~25% of remaining (~$40-46K est.) Modest Phase 3 spend (~$5-10K for visits)

Pre-Phase 3 Checklist (Before Phase 3 Opens)

  1. Burn remaining 6 Phase 2 calls on Tier 3 targets (Ratcliff, Anderton, Bingaya, Stockton, Raymond, + 1 more) for knockdown prevention
  2. Pull Ranson HTML page -- EARLY commit, need competition data before Phase 3
  3. Pull Toone HTML page if time permits
  4. Verify all Phase 2 actions submitted -- nothing left unexecuted
  5. Have promise decisions finalized -- no improvising on 3rd calls

Post-Phase 2 Assessment (After Phase 2 Runs, Before Executing Phase 3)

Check these data points immediately when Phase 2 results come in:

  • Delaney ranking: Still 4th of 10? Climbed to Top 3? Dropped?
  • Haug ranking: Still 2nd of 3? New schools appeared?
  • Iruafemi: Still sole school? New competition entered?
  • Ranson: Competition level? (HTML extraction will reveal)
  • EARLY committers: Have Delaney and Ranson entered "Deciding" status?
  • Interest changes: Any major swings (+/- 20+ points) on any target?

Decision Gate

After reviewing Post-Phase 2 data, confirm Plan A or activate contingency:

  • If Delaney in Top 3 and Haug still 2nd of 3 --> Plan A confirmed, execute immediately
  • If Delaney dropped below 5th --> Plan A still holds — pitch is already deploying regardless
  • If Haug gained new competitors (5+ schools) --> Plan A still holds — pitch is already deploying regardless
  • If Iruafemi gained competitors --> Escalate Iruafemi promise to Future Starter

Scholarship Strategy -- FINAL (Option 1 Enforced)

Both scholarships offered as HARD offers in Phase 3 immediately. No reason to hold. No soft offers.

Plan Schol 1 Schol 2 Trigger
A Delaney (PG, CLEAN, EARLY) Haug (SG, BORDERLINE, Morey Ball, MID) Both available
B Delaney Iruafemi (CLEAN PERFECT, MID) Haug lost
C Delaney Toone (CLEAN Workhorse, EARLY) Haug + Iruafemi lost
D Haug Iruafemi Delaney lost
E Iruafemi Toone Delaney + Haug lost (crisis)

Plan A is the default. Only deviate if post-Phase 2 data shows a target is lost (dropped off board, committed elsewhere, or interest collapsed).


Promise Strategy (3rd/Final Call)

All scholarship targets have 2 calls completed (Ph1 + Ph2). Phase 3 call is 3rd/FINAL and MUST include a minutes/role promise. Principle: "Underpromise and overdeliver."

Player Promise Rationale
Delaney Starter (20+ MPG) Starting PG role. 4 seniors graduating frees ~60 MPG. Mimms only PG competition. Achievable and competitive. Escalate to "Important Player (24+)" only if losing Delaney race badly post-Ph2.
Haug Sixth Man (14-18 MPG) SG backup to Marquee Coddington. Can't promise Starter without conflicting Coddington. His natural advantages (Morey Ball match, ultra-local, low prestige weight) mean we don't need to over-promise. End 16 supports this range.
Iruafemi Sixth Man (14-18 MPG) Zero competition = no need to over-promise. Escalate to "Future Starter" only if new schools appear. Robinzine graduates 2050 = SF starting spot opens for his sophomore year.
Toone Sixth Man (14-18 MPG) Matches his End 11 physical ceiling exactly. Honest and deliverable. Elevated to Tier 1.5 — most versatile offensive backup (3PT 17, SelfC 15, ShotOD 15).
Ranson Sixth Man (14-18 MPG) or Future Starter Defer to post-Ph2 data. If low competition: Sixth Man. If competitive: Future Starter. Same Robinzine graduation logic. Moved behind Toone in contingency order.

Promise Conflict Check

  • Coddington: Marquee (28+ MPG) -- SG starter locked. No freshman SG gets Starter promise.
  • Robinzine: Starter (28 MPG) -- SF starter for 2050. Graduates after 2050, opening SF spot for 2051.
  • Jihad: Important Player -- PF minutes secured.
  • Delaney Starter (20+ MPG): No conflict. PG spot open (Mimms is the only other PG, limited).
  • All other freshmen at Sixth Man (14-18 MPG): No conflicts with existing roster promises.

Phase 3 Action Submission (Priority Order)

Immediate (Phase 3 Opens)

# Action Target Details
1 Hard Offer Scholarship 1 Delaney PG, CLEAN, EARLY commit -- highest urgency
2 Hard Offer Scholarship 2 Haug SG, Morey Ball match, lowest competition
3 Call (3rd, FINAL) Delaney Starter 20+ MPG promise
4 Call (3rd, FINAL) Ranson Sixth Man or Future Starter promise -- EARLY commit urgency
5 Call (3rd, FINAL) Haug Sixth Man 14-18 MPG promise
6 Call (3rd, FINAL) Iruafemi Sixth Man 14-18 MPG promise
7 Call (3rd, FINAL) Toone Sixth Man 14-18 MPG promise
8 pitch #1 Delaney Deploy now — earlier = bigger impact, pitches don't roll over
9 pitch #2 Haug Deploy now — same logic, maximize commit probability on #2 target

Priority 2

# Action Target Details
10 HC Visit Delaney Scholarship target, EARLY commit
11 HC Visit Haug Scholarship target, Morey Ball match
12 HC Visit Iruafemi CLEAN PERFECT, backup scholarship
13 HC Visit Ranson CLEAN, EARLY commit
14 HC Visit Toone CLEAN, best character in pool
15 AC Visit Ratcliff Tier 3, maintain contact (6'8 SF, Disc/Prof perfect)
16 AC Visit Anderton Tier 3, maintain contact (End 19, OnBD 19)
17 Campus Visit Ranson Never visited SIUE. EARLY commit -- urgent.
18 Campus Visit Iruafemi Never visited SIUE. CLEAN PERFECT.
19 Campus Visit Toone Never visited SIUE. Best character.

Priority 3

# Action Target Details
20-25 Calls Tier 3 (Ratcliff, Anderton, Bingaya, Glasford, Stockton, Raymond) Maintain contact, knockdown prevention
26-30+ Calls Any interested prospects Burn remaining calls -- no downside to calling

pitch Deployment

Pitch #1: Delaney — DEPLOY IN PHASE 3 - New intel: Earlier pitches have bigger impact. Pitches do NOT roll over year to year — use them or lose them. - Delaney is the only clean-character PG in the pool and an EARLY decider (deciding Phase 5, could commit Phase 6). - If he's our guy, pitch him now. Waiting for Phase 4 only makes sense if we're unsure about the target — and we're not unsure about Delaney. - Deploy pitch #1 on Delaney in Phase 3 regardless of ranking position.

Pitch #2: Haug — DEPLOY IN PHASE 3 - Same logic: pitches expire, earlier = bigger impact, and Haug is our #2 target with Morey Ball philosophy match. - His natural advantages (ultra-local, low prestige weight, low competition) are strong, but adding a pitch maximizes commit probability. - Previous plan held Pitch #2 for Phase 4+. With the new intel that pitches don't roll over and early action is stronger, deploying now is the percentage play. - Never use both pitches on the same player — one each on our two scholarship targets.


Contingency Tree

Branch 1: Delaney Outcome (EARLY -- resolves by Phase 5)

Delaney EARLY commit timeline: Deciding Ph 5, Can commit Ph 6+

IF Delaney accepts scholarship (Ph 5-6+):
  --> Lock Scholarship 1 = Delaney
  --> Scholarship 2 stays with Haug (Plan A)
  --> Hold pitch #2 for Haug insurance

IF Delaney rejects / commits elsewhere:
  --> Redirect Scholarship 1:
      Priority: Iruafemi (CLEAN PERFECT, no competition)
      Fallback: Toone (CLEAN Workhorse, EARLY)
  --> Activate Plan D (Haug + Iruafemi) or Plan E (Iruafemi + Toone)
  --> pitch #1 is either spent or saved -- reassess for Haug

Branch 2: Haug Outcome (MID -- resolves by Phase 7)

Haug MID commit timeline: Deciding Ph 7, Can commit Ph 7+

IF Haug accepts scholarship:
  --> Lock Scholarship 2 = Haug
  --> Monitor character borderline (Disc 11, Prof 10) through season

IF Haug field expands (5+ schools by Ph4):
  --> Consider pitch #2 on Haug
  --> His advantages (Morey Ball match, ultra-local, low prestige weight) may hold

IF Haug rejects / commits elsewhere:
  --> Redirect Scholarship 2:
      Priority: Iruafemi (CLEAN PERFECT)
      Fallback: Toone (CLEAN Workhorse, EARLY)
  --> Activate Plan B (Delaney + Iruafemi)

Branch 3: Worst Case (Both Delaney + Haug Lost)

IF both Delaney AND Haug lost:
  --> Crisis mode. Plan E: Iruafemi + Toone
  --> Both CLEAN Workhorse character — guaranteed development payoff
  --> Toone: 3PT 17, SelfC 15, ShotOD 15 (more offensive versatility than Ranson)
  --> Use remaining pitches to secure at least one
  --> Tier 3 Mixed players become emergency depth options (no scholarship)
  --> Consider: is a walk-on year better than a failed character scholarship?

Explicit DO-NOT List

  • No calls/visits/offers to Seumalo, Danridge, Cannon, or any Tier 4 dropped player
  • No soft scholarship offers -- hard only (2 available, use both)
  • No "Marquee Player" promises to freshmen (conflicts with Coddington)
  • No "Important Player (24+)" for anyone except Delaney (and only as escalation if losing his race)
  • No scouting -- all targets are at L5
  • No holding pitches for later phases -- pitches don't roll over and earlier deployment has bigger impact
  • No scholarship offers to Tier 3 Mixed-character players (Ratcliff, Anderton, Bingaya, Glasford, Stockton, Raymond)
  • No improvising promises on 3rd calls -- follow the promise table above

Remaining HTML Extractions Needed

Player Priority Why
Ranson HIGH EARLY commit. Need competition data before Phase 3 opens.
Toone Medium CLEAN character. Lower priority -- End 11 limits role.

Pull Ranson's HTML page during Phase 2 (before phase runs) if possible. Toone can wait until Phase 3 if needed.


Phase 3 Success Criteria

By end of Phase 3: 1. Both hard scholarship offers submitted (Delaney + Haug, or contingency pair) 2. All 5 CLEAN/BORDERLINE targets called with appropriate promises 3. All 3 campus visits used (Ranson, Iruafemi, Toone) 4. HC visits deployed to all scholarship targets 5. pitch #1 deployed on Delaney 6. pitch #2 deployed on Haug 7. Post-Phase 2 assessment completed and plan confirmed or adjusted

SIUE Recruiting Retrospective 2050

SIUE Recruiting Retrospective: 2050 Class

Status: IN PROGRESS -- fill in outcomes as recruiting concludes Scholarships Available: 2 Previous Class (2049): (fill in for comparison)


1. Roster Outcome

Slot Player Pos Character How Acquired Scholarship?
1 Sai Iruafemi SG/SF CLEAN (20/20/20) Workhorse Committed (Phase ?) Yes
2 (Anderton or TBD)

Scholarships used: /2 Walk-ons added: _ Final roster size: /15

What We Got vs What We Needed

Entering the offseason, SIUE needed: PG (replacing graduating senior), wing shooting, Morey Ball fit players

What we actually got: - (fill in: positions, key ratings, Morey Ball fit assessment)

Gap assessment: Did the class address the roster's biggest needs? What's still missing for 2051?


2. Target-by-Target Review

Tier 1 Targets (Scholarship Offers Made)

Emanuel Delaney (PG) -- LOST

Metric Pre-Recruiting Outcome
Character CLEAN (WE 20, Disc 15, Prof 16) N/A
Interest High (83%, 4th of 10) Lost (Phase 4?)
Competition 10 schools, Green Bay dominant Green Bay won?
Actions Taken 3 calls, HC visits, campus visit, scholarship, pitch #1 All resources deployed
Promise Starter (20+ MPG) N/A

What happened: (fill in: when did we lose him, what was the deciding factor?)

What we'd do differently: We were 4th of 10 from the start -- was this a viable target at Very Low prestige? Green Bay had 100% interest bar and a campus visit. Should we have recognized this earlier and redirected resources?

Key question: Did the pitch move the needle at all? Did we climb from 4th?

Jon Haug (SG) -- LOST

Metric Pre-Recruiting Outcome
Character BORDERLINE (WE 20, Disc 11, Prof 10) N/A
Interest Very High (83%, 2nd of 3, +40 jump) Committed to Western Illinois
Competition Only 3 schools -- low competition Lost despite low field
Actions Taken 3 calls, HC visits, campus visit, scholarship, pitch #2 All resources deployed
Promise Sixth Man (14-18 MPG) N/A
Philosophy ONLY Morey Ball match in pool N/A

What happened: Lost to Western Illinois despite being 2nd of only 3 schools with massive interest momentum (+40).

What we'd do differently: Was Sixth Man promise too low? Western Illinois likely offered more playing time. Should we have promised Starter given he was a scholarship target? The "underpromise and overdeliver" philosophy may have cost us here against a school willing to promise more.

Key question: What was Western Illinois's prestige level? Did they out-promise us?

Sai Iruafemi (SG/SF) -- COMMITTED

Metric Pre-Recruiting Outcome
Character CLEAN PERFECT (20/20/20) Workhorse Committed
Interest Very High (100%, SIUE only school) Committed as expected
Competition Zero (SIUE was only school) None appeared
Actions Taken 3 calls, HC visit, campus visit, scholarship Full investment
Promise Sixth Man (14-18 MPG) N/A

What happened: Committed as projected. Zero-competition target with perfect character.

Lesson: Zero-competition + perfect character = near-guaranteed commit. This should be our archetype for "safe scholarship" targets going forward.

Tier 1.5 / Contingency Targets

Jules Toone (PG/SG) -- (outcome?)

  • Character: CLEAN Workhorse (WE 20, Disc 19, Prof 20)
  • Key limitation: End 11 (can't play starter minutes)
  • Actions taken: 3 calls, HC visit, campus visit
  • Outcome: (fill in)
  • Lesson: (fill in)

Ikenna Ranson (SF) -- (outcome?)

  • Character: CLEAN (WE 20, Disc 13, Prof 14)
  • EARLY commit timeline
  • Actions taken: 3 calls, HC visit, campus visit
  • Outcome: (fill in)
  • Lesson: (fill in)

Tier 2 / Dropped Targets

Clent Seumalo (G) -- DROPPED (Character Fail)

  • Was our original #1 target (OutS 19, Very High interest, 2nd of 5)
  • L5 reveal: Disc 8, Prof 9 -- both below floor
  • Lesson: Don't assume Group A targets are clean until L5 confirmed. Seumalo was listed as "CLEAN (Group A)" through all of Phase 1. Budget for L5 checks on ALL top targets early, not just new discoveries.
  • Cost of the miss: Campus visit + multiple calls + HC visits spent before character was confirmed. ~$1,500-2,000 in scouting + visit costs.

Tshiefu Cannon (SF) -- DROPPED (Character Hard Fail)

  • Disc 6 despite Leader personality and Prof 20
  • Lesson: "Leader" personality predicts high Prof but NOT Disc. Prof 20 is common and not a strong positive signal by itself.

3. Resource Audit

Budget

Phase Budget Available Budget Spent Efficiency Notes
Phase 0 $100,000 (fill in) Was the full $100K spent?
Phase 1 ~$25,900 ~$18,175 50 calls used (all), 7 HC visits, 0 campus
Phase 2 (fill in) (fill in) 24/30 calls, 8/8 HC visits, 3/3 campus
Phase 3+ (fill in) (fill in) (fill in)
Total $275,968 (fill in) (fill in)

Budget efficiency question: Did we leave money unspent? Could we have scouted more players to L5 earlier?

Action Economy

Resource Total Used Wasted on Drops Effective Uses
Calls (3 max/recruit) (fill in) Calls on Seumalo, Cannon, Danridge, etc. after they'd eventually fail character
HC Visits (fill in) Seumalo, Cannon, Danridge visits were sunk costs
Campus Visits (fill in) Seumalo campus visit = sunk cost
pitches (2) 2 used (were they effective?)
Scholarships (2) (fill in)
L5 Scouting (fill in)

Key waste: Campus visit on Seumalo before L5 character was confirmed. In hindsight, L5 scouting should have been prioritized before campus visits.

Scouting ROI

Batch Players Scouted to L5 CLEAN BORDERLINE FAIL Hit Rate
Group A (Phase 1) 8 2 (Delaney, Iruafemi) 1 (Haug) 5 25% clean
Group B (Phase 2) 16 3 (Toone, Iruafemi*, Ranson) 6 mixed 9 19% clean
Total ~24 5 7 14 ~21% clean

80% fail rate confirmed -- worse than the 77% historical average. Budget for volume scouting.


4. Mechanic Learnings (Confirmed This Cycle)

Character Screening

  • [x] 80% fail rate is the baseline at Very Low prestige (worse than 77% historical)
  • [x] "Inconsistent" scouting text has ZERO correlation with Discipline rating. Iruafemi has "Maddeningly inconsistent" AND Disc 20. Ratcliff has "Maddeningly inconsistent" AND Disc 20.
  • [x] Positive WE scouting language predicts WE only, NOT Disc/Prof. "Obsessive dedication" = high WE, but Disc/Prof are fully independent.
  • [x] Prof 20 is the most common elite intangible -- not a strong positive signal by itself.
  • [x] Personality type signals: Workhorse = guaranteed clean. Leader = high Prof but Disc varies wildly. Troublemaker = catastrophic.
  • [x] WE, Disc, and Prof are fully independent ratings. Cannot predict one from the others.

Recruiting Mechanics

  • [x] Max 3 calls per recruit (hard cap across all phases)
  • [x] pitches don't roll over year to year -- use them or lose them
  • [x] Earlier pitch deployment = bigger impact (community intel)
  • [x] One call per player per phase -- cannot stack
  • [x] Philosophy match matters -- Haug was the ONLY Morey Ball match in the entire pool
  • [x] Visits refresh each phase (7 HC + 3 campus per phase)

Targeting

  • [x] PnR Execution is the rarest elite skill in the HS prospect pool
  • [x] International recruiting unviable at Very Low prestige -- zero High+ interest, 16x scouting cost
  • [x] BBIQ 8 is universal across all targets at VL prestige -- pool constraint, not selection failure

New Learnings (fill in as cycle concludes)

  • [ ] (Did promise level affect commit probability? Compare Haug's Sixth Man vs what winner offered)
  • [ ] (Did pitches have measurable impact on interest? Compare pre/post pitch interest bars)
  • [ ] (Did early vs late scholarship timing matter?)
  • [ ] (How did our prestige weight affect outcomes vs schools with higher prestige?)

5. Model Accuracy Review

Character Screening Algorithm

Prediction Correct? Notes
Workhorse = guaranteed CLEAN YES Iruafemi 20/20/20
"Inconsistent" text = character risk NO Zero correlation with Disc
Positive WE language = clean character PARTIAL Predicts WE only, not Disc/Prof
80% fail rate YES Confirmed at our prestige level

Ridge Score vs Actual Contribution

(Fill in after season plays out)

Player College Ridge Score Actual Contribution Accurate?
Iruafemi (fill in) (after season)
(other commits)

Morey Ball Fit Assessment

(Fill in after season)

Player Pre-Season Morey Ball Fit Actual Role Performance Notes
Iruafemi PnRE 16, Grav 16, OutS 15 (borderline) (after season)

Interest/Competition Reads

Target Our Read Actual Outcome Accurate?
Delaney "At risk" (4th of 10) Lost YES -- correctly flagged risk
Haug "Low competition" (2nd of 3) Lost to Western Illinois NO -- underestimated competitor
Iruafemi "Zero competition, safe" Committed YES
Seumalo "High interest, closeable" Dropped (character) N/A -- we dropped him

6. Strategic Takeaways for 2051

What Worked

  1. Character-first screening -- saved us from investing in 80% of the pool
  2. Zero-competition targeting (Iruafemi) -- highest-probability path to a commit at VL prestige
  3. Tiered contingency planning (Plans A-E) -- adapted smoothly when Seumalo and Cannon failed character
  4. HTML page extraction for competitive intel -- revealed interest bars, competition counts, philosophy prefs
  5. (add more as cycle concludes)

What Didn't Work

  1. Assuming Group A character was clean -- Seumalo, Cannon both failed L5. Wasted campus visit + resources.
  2. Targeting 4th-of-10 recruit (Delaney) at VL prestige -- was this ever realistic? Should we set a threshold (e.g., must be top 3 or have <5 competitors)?
  3. "Underpromise" philosophy may have cost Haug -- if Western Illinois promised Starter and we offered Sixth Man, the promise gap could be decisive at similar prestige levels
  4. Late L5 scouting -- character should be confirmed BEFORE campus visits, not after
  5. (add more as cycle concludes)

Process Changes for 2051

Scouting Sequencing (HIGHEST PRIORITY)

Problem: We spent campus visits on Seumalo before confirming L5 character. Fix: New rule -- No campus visit until L5 character is confirmed CLEAN or BORDERLINE. L5 scouting must be Phase 1-2 priority for all scholarship-track targets.

Target Selection Criteria

Problem: We targeted Delaney (4th of 10 schools) knowing VL prestige is a disadvantage. Fix: Establish targeting thresholds: - Green light: Top 2 of <=5 schools, or sole school - Yellow light: Top 3 of <=7 schools (invest cautiously) - Red light: 4th+ of 8+ schools at VL prestige (don't invest major resources unless no alternative)

Promise Calibration

Problem: Haug lost despite low competition -- possible under-promise. Fix: (fill in after confirming what Western Illinois offered) - Research: Do competitors at similar prestige promise higher? - Consider: Match or slightly exceed likely competitor promises for Tier 1 targets - Maintain "underpromise" for zero-competition targets (Iruafemi model)

Budget Allocation

Problem: (fill in: did we under/over-spend in any phase?) Fix: Proposed phase budget allocation for 2051: - Phase 0: Spend full $100K (no rollover) - Phase 1: Heavy L5 scouting (60%+ of Phase 1 budget on scouting) - Phase 2: Confirm character on all potential scholarship targets BEFORE campus visits - Phase 3+: Concentrated investment on confirmed-clean targets only

Early Analytics Integration

Problem: Cross-project data review (Ridge scores, pro projections) happened in Phase 3 prep, not Phase 1. Fix: Run Ridge scores and Morey Ball fit analysis in Phase 1 alongside initial scouting. Use analytics to triage the target list before spending calls/visits.


7. Data to Collect Before Next Cycle

These questions should be answered before the 2051 recruiting cycle begins:

From This Cycle's Results

  • [ ] What promise did Western Illinois make Haug? (Check if visible in commit data)
  • [ ] Did Delaney's interest bar change after our pitch? (Compare Phase 2 vs Phase 4 if data available)
  • [ ] Final interest rankings for all targets at time of commit/loss
  • [ ] Total budget spent vs available -- was there waste?
  • [ ] Did any Tier 3 (mixed character) players commit to SIUE without scholarship? Walk-ons?

From the 2050 Season (After It Plays Out)

  • [ ] Iruafemi's actual production vs Ridge score prediction
  • [ ] Did any dropped targets (Seumalo, Cannon) cause problems at their schools? (Validates character screening)
  • [ ] SIUE's final prestige after 2050 season -- did it move up from Very Low?
  • [ ] Which Morey Ball skills actually correlated with SIUE wins this season?

For 2051 Planning

  • [ ] How many scholarships will be available? (Check graduation/departures)
  • [ ] What positions will need replacing?
  • [ ] Will SIUE's prestige have changed? (Affects targeting thresholds)
  • [ ] Budget projections for 2051 recruiting
  • [ ] Any coaching changes that affect philosophy or recruiting appeal?

Appendix: Full Target List Outcomes

# Name Pos Tier Character Outcome Resources Spent ROI
1 Emanuel Delaney PG 1 CLEAN LOST Max (scholarship, pitch, all visits/calls) Negative
2 Jon Haug SG 1 BORDERLINE LOST (Western Illinois) Max (scholarship, pitch, all visits/calls) Negative
3 Clent Seumalo G 1 (dropped) FAIL (Disc 8, Prof 9) DROPPED Phase 2 High (campus visit before L5) Wasted
4 Sai Iruafemi SG/SF 2 (elevated) CLEAN PERFECT COMMITTED Full Positive
5 Tshiefu Cannon SF 1 (dropped) HARD FAIL (Disc 6) DROPPED Phase 2 Medium (HC visits, calls) Wasted
6 Jules Toone PG/SG 1.5 CLEAN Workhorse (fill in) Medium (fill in)
7 Ikenna Ranson SF 2 CLEAN (fill in) Medium (fill in)
8 Sarp Anderton PG/SG 3 BORDERLINE (Disc 11) (targeting -- fill in) Low-Medium (fill in)
9 Nolen Ratcliff SF 3 Mixed (WE 10) (fill in) Low (fill in)
10 Dima Bingaya SF 3 Mixed (Disc 11) (fill in) Low (fill in)
11 Matiss Glasford SG 3 Mixed (Disc 8) (fill in) Low (fill in)
12+ (others)

This document should be completed at the end of the 2050 recruiting cycle and referenced at the start of the 2051 cycle to inform Phase 0/1 planning.

SIUE Roster Data Complete

===============================================

RAMEL BRINK - PLAYER PROFILE DATA

===============================================

BASIC INFORMATION


Name: Ramel Brink

Team: Southern Illinois-Edwardsville

Position: PG

Height: 5'11

Weight: 197 lbs

Age: 23 y.o. (Senior)

Shoots: Right

Wingspan: 6'2

Hometown: Villa Park, Illinois

High School: Willowbrook

Current Stats: 0.0 PPG, 0.0 RPG, 0.0 APG, 0.00 SPG, 0.00 BPG, 0.0 MPG, 0.0 PER

===============================================

TAB 1: STATISTICS

===============================================

PER GAME AVERAGES


Season Age Team Tm G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 22 Southern Illinois-Edwardsville 0 0 0.0 0 0 0 0 0 0 0 0 0 0.000 0 0 0.840 0.0 0.0 0.0 0.0 0.00 0.00 0.00 0.0 0.0

2048 21 Southern Illinois-Edwardsville 37 0 16.7 1.5 3.9 0.384 0.8 2.3 0.326 0.8 1.6 0.467 0.479 0.6 0.7 0.840 0.2 0.6 0.8 2.8 1.03 0.03 0.89 1.0 4.4

2047 20 Southern Illinois-Edwardsville 1 0 1.9 0.0 0.0 0 0.0 0.0 0 0.0 0.0 0 0.000 0.0 0.0 0 0.0 0.0 0.0 1.0 0.00 0.00 0.00 0.0 0.0

Career 38 0 16.3 1.5 3.8 0.384 0.7 2.3 0.326 0.7 1.6 0.467 0.479 0.6 0.7 0.840 0.2 0.6 0.8 2.7 1.00 0.03 0.87 1.0 4.2

TOTALS


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 22 Southern Illinois-Edwardsville 0 0 0 0 0 0 0 0 0 0 0 0 0.479 0 0 0 0 0 0 0 0 0 0 0 0

2048 21 Southern Illinois-Edwardsville 37 0 617 56 146 0.384 28 86 0.326 28 60 0.467 0.479 21 25 0.840 9 22 31 102 38 1 33 38 161

2047 20 Southern Illinois-Edwardsville 1 0 2 0 0 0 0 0 0.467 0 0 0.479 0 0 0 0.840 0 0 0 1 0 0 0 0 0

Career 38 0 619 56 146 0.384 28 86 0.326 28 60 0.467 0.479 21 25 0.840 9 22 31 103 38 1 33 38 161

PER 36 MINUTES


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 22 Southern Illinois-Edwardsville 0 0 0 0.0 0.0 0 0.0 0.0 0 0.0 0.0 0.467 0.479 0.0 0.0 0.840 0 0 0 0 0 0 0 0 0

2048 21 Southern Illinois-Edwardsville 37 0 617 3.3 8.5 0.384 1.6 5.0 0.326 1.6 3.5 0.467 0.479 1.2 1.5 0.840 0.5 1.3 1.8 6.0 2.22 0.06 1.93 2.2 9.4

2047 20 Southern Illinois-Edwardsville 1 0 2 0.0 0.0 0.384 0.0 0.0 0.326 0.0 0.0 0.467 0.479 0.0 0.0 0.840 0.0 0.0 0.0 18.0 0.00 0.00 0.00 0.0 0.0

Career 38 0 36.0 3.3 8.5 0.384 1.6 5.0 0.326 1.6 3.5 0.467 0.479 1.2 1.5 0.840 0.5 1.3 1.8 6.0 2.2 0.1 1.9 2.2 9.4

SHOOTING SPLITS


Season Age Team G GS MP AtRim AtRimA AtRim% Close CloseA Close% Mid MidA Mid% Long LongA Long% 3P 3PA 3P%

2049 22 Southern Illinois-Edwardsville 0 0 2 0.0 0.0 .000 0.0 0.0 .000 0.0 0.0 .000 0.0 0.0 .000 0.0 0.0 .000

2048 21 Southern Illinois-Edwardsville 37 0 2 0.4 0.8 .552 0.2 0.4 .467 0.1 0.4 .286 0.0 0.1 .500 0.8 2.3 .326

CLUTCH STATS


Season Age Team G MIN FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS +/-

2049 22 Southern Illinois-Edwardsville 0 0 0 0 .000 0 0 0 0 0 .000 .000 0 0 .000 0 0 0 0 0 0 0 0 0 0

2048 21 Southern Illinois-Edwardsville 1 0 0 0 .000 0 0 0 0 0 .000 .000 0 0 .000 0 0 0 0 0 0 0 0 0 -1

2047 20 Southern Illinois-Edwardsville 0 0 0 0 .000 0 0 0 0 0 .000 .000 0 0 .000 0 0 0 0 0 0 0 0 0 0

POSTSEASON PER GAME


Career: G=0, GS=0, MP=16.3, FG=1.5, FGA=3.8, FG%=0.7, 3P=2.3, 3PA=0, 3P%=0.7, 2P=1.6, 2P%=0.467, eFG%=0.000, FT=0.6, FTA=0.7, FT%=0, ORB=0.2, DRB=0.6, TRB=0.8, AST=2.7, STL=1.00, BLK=0.03, TOV=0.87, PF=1.0, PTS=4.2

HIGH SCHOOL PER GAME


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2046 19 Willowbrook 32 32 33.8 6.8 13.8 0.492 1.9 5.0 0.377 4.9 8.8 0.557 0.560 2.3 3.3 0.695 0.2 1.4 1.7 5.4 2.67 0.06 2.73 0.0 17.5

Career 32 32 34.9 6.8 13.8 0.492 1.9 5.0 0.377 4.9 8.8 0.557 0.560 2.3 3.3 0.695 0.2 1.4 1.6 5.4 2.72 0.06 2.59 1.6 17.7

===============================================

TAB 2: GAME LOG (2048 Season - 37 Games)

===============================================

Day Team Opp Result G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS +/- GmSc ORtg DRtg

144 Southern Illinois-Edwardsville Morehead State L 78-80 1 0 17 3 4 0.750 2 2 1.000 1 2 0.500 1.000 0 0 0.000 0 0 2 0 0 2 0 8 -10 5.8 121.8 132.5

133 Southern Illinois-Edwardsville Eastern Illinois W 81-80 1 0 15 2 3 0.667 1 2 0.500 1 1 1.000 0.833 0 0 0.000 0 2 2 2 1 0 0 5 3 6.7 169.3 112.9

127 Southern Illinois-Edwardsville Lindenwood W 70-59 1 0 15 2 4 0.500 2 3 0.667 0 1 0.000 0.750 0 0 0.000 0 0 2 0 0 2 2 6 7 2.6 98.7 108.4

123 Southern Illinois-Edwardsville Tennessee Tech L 56-79 1 0 16 0 6 0.000 0 1 0.000 0 5 0.000 0.000 0 0 0.000 0 0 3 0 0 2 0 0 -6 -4.9 15.9 126.6

119 Southern Illinois-Edwardsville Southern Indiana W 92-90 1 0 16 1 3 0.333 0 1 0.000 1 2 0.500 0.333 2 2 1.000 0 0 4 0 0 1 4 4 4.7 153.5 145.2

115 Southern Illinois-Edwardsville Tennessee-Martin W 91-81 1 0 15 0 3 0.000 0 2 0.000 0 1 0.000 0.000 0 0 0.000 0 0 1 1 0 0 1 0 -2 -1.8 18.7 122.4

111 Southern Illinois-Edwardsville Little Rock L 70-72 1 0 15 1 4 0.250 0 2 0.000 1 2 0.500 0.250 4 4 1.000 1 1 2 4 0 0 1 9 6.4 117.4 115.9

107 Southern Illinois-Edwardsville Southeast Missouri State W 86-76 1 0 16 2 3 0.667 1 1 1.000 0 1 0.500 0.833 3 3 1.000 0 3 3 3 4 0 0 3 8 29 12.5 194.3 89.6

103 Southern Illinois-Edwardsville Tennessee State W 84-81 1 0 15 2 6 0.333 2 5 0.400 0 1 0.000 0.500 0 0 0.000 0 0 1 3 1 0 1 1 6 1 4.6 108.2 104.2

99 Southern Illinois-Edwardsville Eastern Illinois L 71-80 1 0 15 1 2 0.500 1 2 0.500 0 0 0.000 0.750 0 0 0.000 0 0 4 1 0 0 2 3 -2 5.0 172.4 111.8

95 Southern Illinois-Edwardsville Morehead State L 76-89 1 0 15 1 3 0.333 1 2 0.500 0 1 0.000 0.500 0 0 0.000 0 0 4 0 1 0 1 0 3 -1 1.3 68.7 119.8

91 Southern Illinois-Edwardsville Lindenwood W 71-69 1 0 17 2 3 0.667 2 3 0.667 0 0 0.000 1.000 0 0 0.000 0 0 4 0 0 0 0 6 8 7.5 214.8 109.6

87 Southern Illinois-Edwardsville Tennessee Tech L 83-89 1 0 17 3 5 0.600 1 2 0.500 2 3 0.667 0.700 0 0 0.000 0 1 1 3 0 1 1 2 7 -5 6.0 125.9 132.0

83 Southern Illinois-Edwardsville Southern Indiana W 76-65 1 0 16 6 8 0.750 3 5 0.600 3 3 1.000 0.938 2 2 1.000 0 5 3 3 2 0 0 1 17 4 18.4 198.5 90.2

79 Southern Illinois-Edwardsville Tennessee-Martin L 73-87 1 0 17 1 2 0.500 0 0 0.000 1 2 0.500 0.500 0 0 0.000 0 0 3 0 0 2 2 1 2 -22 0.7 76.0 136.3

75 Southern Illinois-Edwardsville Little Rock L 64-85 1 0 16 2 3 0.667 0 1 0.000 2 2 1.000 0.667 1 1 1.000 0 1 1 2 0 0 0 5 5 -13 5.0 155.8 132.9

71 Southern Illinois-Edwardsville Southeast Missouri State L 74-80 1 0 17 1 3 0.333 1 2 0.500 0 1 0.000 0.500 0 0 0.000 1 0 1 3 1 0 1 0 3 -1 4.7 138.6 116.9

67 Southern Illinois-Edwardsville Tennessee State W 86-78 1 0 16 2 4 0.500 1 2 0.500 1 2 0.500 0.625 0 0 0.000 1 0 1 4 1 0 1 0 5 7 6.5 135.4 112.2

63 Southern Illinois-Edwardsville Morehead State W 83-81 1 0 16 0 4 0.000 0 4 0.000 0 0 0.000 0.000 0 0 0.000 0 0 4 4 0 1 0 0 -3 3.0 48.5 95.4

59 Southern Illinois-Edwardsville Jacksonville State W 81-76 1 0 15 2 2 1.000 0 0 0.000 2 2 1.000 1.000 0 0 0.000 2 1 3 3 1 0 1 1 4 5 6.8 154.4 106.7

54 Southern Illinois-Edwardsville Nebraska L 78-81 1 0 14 3 5 0.600 1 1 1.000 2 4 0.500 0.700 2 2 1.000 0 0 2 2 0 0 1 0 9 1 9.1 136.0 112.0

50 Southern Illinois-Edwardsville Illinois State L 69-74 1 0 16 2 7 0.286 1 3 0.333 1 4 0.250 0.357 0 0 0.000 0 0 2 0 0 2 0 5 -6 1.5 85.6 121.5

48 Southern Illinois-Edwardsville California Baptist L 74-84 1 0 16 1 2 0.500 0 1 0.000 1 1 1.000 0.500 0 0 0.000 0 0 5 0 0 0 1 2 5 4.1 159.2 135.3

44 Southern Illinois-Edwardsville Alabama A&M W 73-66 1 0 16 3 4 0.750 1 2 0.500 2 2 1.000 0.875 0 0 0.000 0 0 5 2 0 3 2 7 15 7.1 110.1 97.6

41 Southern Illinois-Edwardsville Pepperdine L 59-82 1 0 16 0 1 0.000 0 0 0.000 0 1 0.000 0.000 0 0 0.000 1 1 2 1 0 0 1 1 0 -4 -0.4 53.1 136.8

37 Southern Illinois-Edwardsville Colgate L 70-72 1 0 17 3 7 0.429 2 6 0.333 1 1 1.000 0.571 1 2 0.500 1 0 1 1 3 0 1 1 9 4 7.9 118.2 101.8

34 Southern Illinois-Edwardsville Bradley L 71-96 1 0 22 0 2 0.000 0 1 0.000 0 1 0.000 0.000 1 4 0.250 0 1 1 3 2 0 1 1 1 -18 1.4 51.7 128.4

31 Southern Illinois-Edwardsville Utah Valley L 53-84 1 0 17 1 5 0.200 0 2 0.000 1 3 0.333 0.200 0 0 0.000 0 0 2 3 0 0 1 1 2 -15 1.9 51.2 114.0

28 Southern Illinois-Edwardsville Western Kentucky L 87-99 1 0 18 3 4 0.750 2 3 0.667 1 1 1.000 1.000 2 2 1.000 0 0 4 0 0 2 0 10 -12 9.2 148.1 141.0

25 Southern Illinois-Edwardsville McNeese State L 69-88 1 0 19 3 7 0.429 1 3 0.333 2 4 0.500 0.500 0 0 0.000 1 0 1 6 2 0 1 3 7 0 8.0 114.6 117.8

23 Southern Illinois-Edwardsville Monmouth L 68-81 1 0 19 2 5 0.400 1 2 0.500 1 3 0.333 0.500 1 1 1.000 0 3 3 3 3 0 3 0 6 -12 6.3 81.5 102.1

21 Southern Illinois-Edwardsville Tulane L 66-85 1 0 15 1 6 0.167 1 6 0.167 0 0 0.000 0.250 0 0 0.000 0 0 1 1 0 0 1 1 3 -6 -0.1 49.7 112.8

17 Southern Illinois-Edwardsville Youngstown State L 66-83 1 0 15 0 4 0.000 0 4 0.000 0 0 0.000 0.000 0 0 0.000 0 0 0 0 0 2 0 0 -20 -3.6 0.0 128.1

15 Southern Illinois-Edwardsville Prairie View W 86-84 1 0 15 0 1 0.000 0 1 0.000 0 0 0.000 0.000 0 0 0.000 0 1 1 4 0 0 0 1 0 4 2.0 144.8 119.9

12 Southern Illinois-Edwardsville Evansville L 61-89 1 0 15 0 4 0.000 0 4 0.000 0 0 0.000 0.000 0 0 0.000 0 3 2 0 1 2 0 0 -14 -0.5 35.0 124.7

9 Southern Illinois-Edwardsville Wyoming L 52-68 1 0 16 0 4 0.000 0 2 0.000 0 2 0.000 0.000 2 2 1.000 0 2 2 2 0 0 1 1 2 -10 0.8 66.5 110.9

1 Southern Illinois-Edwardsville McNeese State W 83-75 1 1 2 1 0 3 0.000 0 3 0.000 0 0 0.000 1 1 2 1 0 0 1 0 -12 -1.8 37.6 114.1

2048 SEASON TOTALS:

Total: 37 G, 0 GS, 599 MP, 56 FG, 146 FGA, 0.384 FG%, 28 3P, 86 3PA, 0.326 3P%, 28 2P, 60 2PA, 0.467 2P%, 0.840 FT%, 21 FT, 25 FTA, 9 ORB, 22 DRB, 31 TRB, 102 AST, 38 STL, 1 BLK, 33 TOV, 38 PF, 161 PTS, -96 +/-

Averages: 16.2 MP, 1.5 FG, 3.9 FGA, 0.384 FG%, 0.8 3P, 2.3 3PA, 0.326 3P%, 0.8 2P, 1.6 2PA, 0.467 2P%, 0.479 eFG%, 0.6 FT, 0.7 FTA, 0.840 FT%, 0.2 ORB, 0.6 DRB, 0.8 TRB, 2.8 AST, 1.03 STL, 0.03 BLK, 0.89 TOV, 1.0 PF, 4.4 PTS, -2.6 +/-, 4.2 GmSc

===============================================

TAB 3: ADVANCED STATS

===============================================

ADVANCED METRICS


Season Age Team G GS MP AvgGS PER EWA OWS DWS WS WS/48 BPM VORP ORtg DRtg TS% 3PAr FTr ORB% DRB% TRB% AST% STL% BLK% TOV% USG%

2049 22 Southern Illinois-Edwardsville 0 0 0 0.0 0.0 0.0 0.0 0.0 0.000 0 0 0.0 0.000 0.000 0.000 0.0 0.0 0.0 0.0 0.0 0.0

2048 21 Southern Illinois-Edwardsville 37 0 617 4.2 11.7 -0.9 0.2 0.4 0.6 0.043 -2.1 0.0 106.3 117.4 0.513 0.589 0.171 1.8 4.9 3.2 28.9 3.6 0.2 17.4 16.1

2047 20 Southern Illinois-Edwardsville 1 0 2 0.7 19.1 0.0 0.0 0.0 0.0 0.332 -24.6 0.0 238.3 122.7 0.000 0.000 0.000 0.0 0.0 0.0 70.7 0.0 0.0 0.0 0.0

COURT STATISTICS


Season Age Team G GS MP +/- +/- G On Off PG% SG% SF% PF% C%

2049 22 Southern Illinois-Edwardsville 0 0 0 0 0.0 0 -95

2048 21 Southern Illinois-Edwardsville 37 0 617 -96 -2.6 -94 -143 48.2 51.8

2047 20 Southern Illinois-Edwardsville 1 0 2 4 4.0 4 -7 100.0

SHOT LOCATIONS


                                                Shot Preference %                           At Rim              Close               Mid                 Long                3PT

Season Age Team G GS MP AtRim Close Mid Long 3P FGM FGA FG% FGM FGA FG% FGM FGA FG% FGM FGA FG% 3P 3PA 3P%

2049 22 Southern Illinois-Edwardsville 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 .000 0.0 0.0 .000 0.0 0.0 .000 0.0 0.0 .000 0.0 0.0 .000

2048 21 Southern Illinois-Edwardsville 37 0 617 19.9 10.3 9.6 1.4 58.9 0.4 0.8 .552 0.2 0.4 .467 0.1 0.4 .286 0.0 0.1 .500 0.8 2.3 .326

2047 20 Southern Illinois-Edwardsville 1 0 2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 .000 0.0 0.0 .000 0.0 0.0 .000 0.0 0.0 .000 0.0 0.0 .326

PLAY TYPES


                                                PnR Ball Handler        PnR Roll Man           Isolation               Transition              Spot Up                 Off Screen              Hand Off                Post Up

Season Age Team G GS MP Role Plays Pos PTS PPP eFG Pos PTS PPP eFG Pos PTS PPP eFG Pos PTS PPP eFG Pos PTS PPP eFG Pos PTS PPP eFG Pos PTS PPP eFG Pos PT

2049 22 Southern Illinois-Edwardsville 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

2048 21 Southern Illinois-Edwardsville 37 0 617 188 56 38 0.68 0.337 4 6 1.50 1.000 8 7 0.88 0.438 21 29 1.38 0.722 57 40 0.70 0.452 11 15 1.36 0.833 17 11 0.65 0.231 3 2 0.67 0.500 7

2047 20 Southern Illinois-Edwardsville 1 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

PLAY TENDENCIES


Season Age Team G GS MP Plays PnR BH % PnR RM % Isolation % Transition % Spot Up % Off Screen % Hand Off % Post Up % Putback %

2049 22 Southern Illinois-Edwardsville 0 0 0 0 0 0 0 0 0 0 0 0 0

2048 21 Southern Illinois-Edwardsville 37 0 617 188 29.8 2.1 4.3 11.2 30.3 5.9 9.0 1.6 3.7 2.1

2047 20 Southern Illinois-Edwardsville 1 0 2 0 0 0 0 0 0 0 0 0 0

HUSTLE STATS


Season Age Team G GS MP Screen Assists SAPG Deflections DFPG Contested Shots CSPG Loose Balls Rcvd LBRPG Charges Drawn CDPG

2049 22 Southern Illinois-Edwardsville 0 0 0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0

2048 21 Southern Illinois-Edwardsville 37 0 617 1 0.0 4 0.1 62 1.7 7 0.2 0 0.0

2047 20 Southern Illinois-Edwardsville 1 0 2 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0

===============================================

TAB 4: SCOUTING REPORT

===============================================

PLAYER RATINGS (Scale: 1-20)


SHOOTING:

Inside Shooting: 13

  • Mid-Range Shooting: 17

  • 3PT Shooting: 17

  • Free Throws: 15

CREATION:

Self-Creation: 19

  • Shoot Off Dribble: 18

  • Ball Dominance: 7

  • Foul Drawing: 8

  • Finishing: 9

EXECUTION:

Isolation Execution: 13

  • Post Execution: 5

  • Pick & Roll Execution: 12

  • Basketball IQ: 7

PLAYMAKING:

Playmaking: 10

  • Passing: 17

  • Gravity: 10

  • Spacing: 9

  • Ball Handling: 13

DEFENSE:

On-Ball Perimeter Defense: 19

  • Off-Ball Defense: 17

  • Help Defense: 11

  • Stealing: 16

  • Post Defense: 3

  • Rim Protection: 3

REBOUNDING:

Offensive Rebounding: 5

  • Defensive Rebounding: 7

PHYSICAL:

Athleticism: 17

  • Endurance: 15

  • First Step: 17

  • Quickness: 13

  • Speed: 15

  • Strength: 4

TENDENCIES:

Driving: 11

  • Roll : Pop: 25:75

  • Catch & Shoot : Pull-Up: 39:61

  • Corner 3 : ATB: 22:78

POSITIONAL FIT:

Guard: 3 stars

  • Wing: 2 stars

  • Big: 1 star

INTANGIBLES:

Work Ethic: 20

  • Leadership: 6

  • Discipline: 14

  • Professionalism: 19

  • Aggression: 9

  • Toughness: 16

  • Hustle: 19

PERSONALITY:

Personality: Driven

SCOUTING INFORMATION:


Player Description: Athletic guard

Offensive Archetype: Primary Ballhandler

Defensive Archetype: Point Of Attack

College Ability: 3 stars

Pro Potential: Longshot prospect with limited JBL upside

Declaration Chance: Unknown

Media Hype: None

Rankings: JBLDraft.Net: Unranked, HoopProspects.com: Unranked, JSPN: Unranked

OVERVIEW:


Ramel Brink is a 22-year-old guard from Villa Park who arrived at Southern Illinois-Edwardsville as an under-recruited spark and has settled into a limited backcourt role in the OVC... He logged 4.4 points and 2.8 assists in 16-17 minutes per game as a sophomore, coming off the bench as a matchup piece rather than a primary creator... He operates as a distributor on offense and the primary point-of-attack defender on the other end, but his minutes and usage remain tightly controlled by staff decisions...

STRENGTHS:


Brink defends the ball with intensity and discipline on straight-line matchups, staying in the pocket and bothering ball-handlers enough to generate steals and deflections... He shows a compact first step and can blow by closeouts to create midrange separation or pull-up triples rather than attacking the rim... He comfortably strokes pull-up jumpers and midrange looks off the dribble and connects from deep when given space, turning pressure into immediate shooting opportunities off single coverage... He sees the floor in short windows and delivers catchable passes in traffic, producing a positive assist rate for his minutes... He brings relentless effort to drills and games alike and shows textbook work habits in the gym, which makes him coachable and trustworthy in situational defensive assignments...

WEAKNESSES:


Brink struggles to turn creation into efficient finishes because he cannot convert through contact or finish in traffic, so drives often end in missed layups or turnovers rather than free throws... He gets pulled out of help rotations and frequently ends up out of position when the possession stretches beyond his man, which undermines team defense despite his strong on-ball instincts... His playmaking reads are limited - he will find the open man in clear looks but repeatedly misses secondary options and timing on complex reads... He lacks functional upper-body strength and is easily bumped off drives or boxed out on the glass, and he offers almost nothing as a rebounder or rim protector... He rarely draws fouls and can be turnover-prone when forced to create against pressure, and his fundamentals on finishing and post work are underdeveloped, leaving questions about his ability to handle heavier workloads in late-game lineups... His consistency is a problem; he can flash impact and then disappear for stretches, and his leadership on the floor is minimal, which makes his long-term role fragile... One clear metaphor: he plays like a hunting dog with a short memory - ferocious in bursts but prone to losing the scent...

PROJECTION:


At the small-conference level Brink projects as a situational rotation guard who can help his college team by guarding ball-screens, hitting pull-up shots and making quick reads off drive-and-kick actions if he remains in a limited role... Given his current toolset and polish, he is unlikely to carve out a JBL roster spot without major changes to strength, finishing and decision-making under pressure, so expect his ceiling to be a dependable bench piece in the OVC rather than a pro prospect... Development priorities that would actually move the needle: add functional strength to finish through contact and fight for boards... refine footwork and basic finishing mechanics so drives produce shots or free throws instead of turnovers... and expand pick-and-roll reads and spacing awareness so his passing ability can translate into consistent playmaking for a second unit role... If Brink leverages his very high work ethic and professional habits to address those three items, he becomes a safer, more reliable college rotation guard; if not, his role will remain marginal and his projection will be limited to the conference level...

Scouting Accuracy: High

Scouting from: Southern Illinois-Edwardsville

All CJBL ratings are scaled to college relative levels.

===============================================

TAB 5: HEATMAP

===============================================

SHOT HEATMAP (2049 Season - No data, player has not played)


All zones showing 0/0 (0.0%) - Did Not Play - Coach's Decision

Court Zones (from baseline):

Left Corner 3: 0/0 (0.0%)

  • Left Wing 3: 0/0 (0.0%)

  • Top of Key 3: 0/0 (0.0%)

  • Right Wing 3: 0/0 (0.0%)

  • Right Corner 3: 0/0 (0.0%)

  • Left Short Corner: 0/0 (0.0%)

  • Left Elbow: 0/0 (0.0%)

  • Paint/Restricted Area: 0/0 (0.0%)

  • Right Elbow: 0/0 (0.0%)

  • Right Short Corner: 0/0 (0.0%)

  • Left Mid-Range: 0/0 (0.0%)

  • Free Throw Line Area: 0/0 (0.0%)

  • Right Mid-Range: 0/0 (0.0%)

Note: Player did not play in 2049 season (Coach's Decision for all games)

For 2048 season shot data, see Statistics tab - Shooting Splits section

===============================================

END OF PLAYER PROFILE DATA

===============================================

Data extracted from: https://jblfl.com/cjbl/players.php?pid=462992

Player ID: 462992

Team: Southern Illinois-Edwardsville (SIUE)

Conference: OVC

PLAYER 2 OF 12: IZAYAH CODDINGTON

===============================================

IZAYAH CODDINGTON - PLAYER PROFILE DATA

===============================================

BASIC INFORMATION


Name: Izayah Coddington

Team: Southern Illinois-Edwardsville

Position: SG

Height: 6'7

Weight: 195 lbs

Age: 20 y.o. (Sophomore)

Shoots: Right

Wingspan: 7'3

Hometown: Skokie, Illinois

High School: Niles North

Current Stats: 15.5 PPG, 3.2 RPG, 2.4 APG, 1.21 SPG, 0.21 BPG, 29.0 MPG, 16.4 PER

===============================================

TAB 1: STATISTICS

===============================================

PER GAME AVERAGES


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 19 Southern Illinois-Edwardsville 38 38 29.0 5.6 13.5 0.414 1.8 5.3 0.335 3.8 8.1 0.466 0.480 2.5 3.1 0.814 1.4 1.9 3.2 2.4 1.21 0.21 2.47 1.8 15.5

Career 38 38 29.0 5.6 13.5 0.414 1.8 5.3 0.335 3.8 8.1 0.466 0.480 2.5 3.1 0.814 1.4 1.9 3.2 2.4 1.21 0.21 2.47 1.8 15.5

TOTALS


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 19 Southern Illinois-Edwardsville 38 38 1102 212 512 0.414 68 203 0.335 144 309 0.466 0.480 96 118 0.814 52 71 123 93 46 8 94 68 588

Career 38 38 1102 212 512 0.414 68 203 0.335 144 309 0.466 0.480 96 118 0.814 52 71 123 93 46 8 94 68 588

PER 36 MINUTES


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 19 Southern Illinois-Edwardsville 38 38 1102 6.9 16.7 0.414 2.2 6.6 0.335 4.7 10.1 0.466 0.480 3.1 3.9 0.814 1.7 2.3 4.0 3.0 1.50 0.26 3.07 2.2 19.2

Career 38 38 36.0 6.9 16.7 0.414 2.2 6.6 0.335 4.7 10.1 0.466 0.480 3.1 3.9 0.814 1.7 2.3 4.0 3.0 1.5 0.3 3.1 2.2 19.2

SHOOTING SPLITS


Season Age Team G GS MP AtRim AtRimA AtRim% Close CloseA Close% Mid MidA Mid% Long LongA Long% 3P 3PA 3P%

2049 19 Southern Illinois-Edwardsville 38 38 1102 2.1 3.1 .667 0.5 1.3 .373 0.6 1.8 .328 0.7 1.9 .338 1.8 5.3 .335

CLUTCH STATS


Season Age Team G MIN FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS +/-

2049 19 Southern Illinois-Edwardsville 21 86 18 43 .419 7 17 .412 11 26 .423 .500 16 19 .842 1 4 0 8 3 0 0 10 59 13

POSTSEASON PER GAME


Career: G=0, GS=0, MP=29.0, FG=5.6, FGA=13.5, FG%=0, 3P=1.8, 3PA=5.3, 3P%=0, 2P=3.8, 2PA=8.1, 2P%=0.466, eFG%=0.000, FT=2.5, FTA=3.1, FT%=0, ORB=1.4, DRB=1.9, TRB=3.2, AST=2.4, STL=1.21, BLK=0.21, TOV=2.47, PF=1.8, PTS=15.5

HIGH SCHOOL PER GAME


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2048 18 Niles North 32 32 35.6 6.1 13.5 0.450 2.1 4.9 0.424 4.0 8.6 0.465 0.528 2.6 3.3 0.800 0.0 0.0 4.5 3.6 1.36 0.15 1.73 0.0 16.7

2047 17 Niles North 24 23 30.7 6.5 12.3 0.530 2.2 5.0 0.445 4.3 7.4 0.588 0.620 1.4 1.8 0.773 1.0 3.8 4.8 3.7 1.88 0.04 1.80 0.0 17.0

Career 56 55 34.7 6.3 13.0 0.483 2.1 4.9 0.433 4.1 8.1 0.513 0.565 2.1 2.7 0.792 1.3 3.3 4.6 3.6 1.63 0.11 1.79 1.4 16.8

===============================================

TAB 4: SCOUTING REPORT

===============================================

PLAYER RATINGS (Scale: 1-20)


SHOOTING:

Inside Shooting: 15

  • Mid-Range Shooting: 11

  • 3PT Shooting: 18

  • Free Throws: 17

CREATION:

Self-Creation: 13

  • Shoot Off Dribble: 13

  • Ball Dominance: 18

  • Foul Drawing: 10

  • Finishing: 10

EXECUTION:

Isolation Execution: 17

  • Post Execution: 3

  • Pick & Roll Execution: 18

  • Basketball IQ: 8

PLAYMAKING:

Playmaking: 14

  • Passing: 14

  • Gravity: 18

  • Spacing: 10

  • Ball Handling: 16

DEFENSE:

On-Ball Perimeter Defense: 10

  • Off-Ball Defense: 10

  • Help Defense: 8

  • Stealing: 6

  • Post Defense: 3

  • Rim Protection: 3

REBOUNDING:

Offensive Rebounding: 14

  • Defensive Rebounding: 14

PHYSICAL:

Athleticism: 10

  • Endurance: 17

  • First Step: 10

  • Quickness: 15

  • Speed: 14

  • Strength: 5

TENDENCIES:

Driving: 3

  • Roll : Pop: 27:73

  • Catch & Shoot : Pull-Up: 58:42

  • Corner 3 : ATB: 23:77

POSITIONAL FIT:

Guard: 2 stars

  • Wing: 3 stars

  • Big: 1 star

INTANGIBLES:

Work Ethic: 16

  • Leadership: 9

  • Discipline: 8

  • Professionalism: 7

  • Aggression: 8

  • Toughness: 10

  • Hustle: 7

PERSONALITY:

Personality: Friendly

SCOUTING INFORMATION:


Player Description: Versatile three-point catch and shoot offensive wing

Offensive Archetype: Secondary Creator

Defensive Archetype: Chaser

College Ability: 3 stars

Pro Potential: Fringe prospect with development needed, possible rotation upside

Declaration Chance: Unknown

Media Hype: None

Rankings: JBLDraft.Net: Unranked, HoopProspects.com: Unranked, JSPN: Unranked

OVERVIEW:


Skokie native who flew under the radar in high school and lands at an OVC program after a subdued recruiting run... He projects as a secondary creator and perimeter chaser in a balanced offense with frequent isolation looks and regular spot-up duties off the ball... Coaches will likely lean on him to stretch the floor and run pick-and-pop actions while he develops as a primary driver and play initiator...

STRENGTHS:


Shoots from distance with a repeatable release that forces closeouts and changes defensive reads around the perimeter... Prefers catch-and-shoot opportunities yet will sporadically step into pull-ups when defenders stay home, creating gravity that opens lanes for teammates on kick-outs... Handles the ball enough to start actions and is comfortable running pick-and-roll sets where he can pop and space the floor effectively... Shows the conditioning to stay on the court for extended minutes and sustain effort across games... Free-throw mechanics are steady enough to trust late free-throw opportunities and to convert when he gets to the line...

WEAKNESSES:


Struggles to finish through contact and offers limited craft around the rim which compounds a thin frame and very low leg drive that opponents exploit on drives and post touches... Demonstrates poor defensive positioning, misses rotations, and is slow to recover in help scenarios which makes him a liability on the ball and off it... Plays with low on-court discipline and fundamentals are inconsistent which shows up as weak spacing off the ball and sloppy decision-making in secondary playmaking duties... Passing and court awareness are still raw so possessions often end without a quality kick or timely read... Shows minimal aggression and hustle on the glass and in loose-ball moments which further reduces his two-way impact... Like a nice lamp that doesn't survive a move, his offensive shine risks breaking under physical pressure...

PROJECTION:


Ceiling domestic to college is as a floor-spacing wing who can carry a role in a small-conference rotation if he accepts a narrow, well-defined job description and puts work into strength and finishing... To matter in conference play he must add upper-body strength, refine finishing at the rim, clean up defensive positioning, and commit to consistent fundamentals and film study or the minutes will plateau... Low discipline and professionalism are the main growth brakes and without marked behavioral buy-in his upside evaporates quickly... Realistic path this season is a rotational spot as a catch-and-shoot specialist and situational isolation option with upside to be a steady conference-level contributor if he chooses to grind in the weight room and on the practice floor...

Scouting Accuracy: High

Scouting from: Southern Illinois-Edwardsville

PLAYER 3 OF 12: CANYON JIHAD

===============================================

CANYON JIHAD - PLAYER PROFILE DATA

===============================================

BASIC INFORMATION


Name: Canyon Jihad

Team: Southern Illinois-Edwardsville

Position: C

Height: 7'1

Weight: 239 lbs

Age: 21 y.o. (Junior)

Shoots: Right

Wingspan: 7'6

Hometown: New Concord, Ohio

High School: John Glenn

Current Stats: 5.8 PPG, 8.6 RPG, 0.6 APG, 0.42 SPG, 2.13 BPG, 25.5 MPG, 16.6 PER

===============================================

TAB 1: STATISTICS

===============================================

PER GAME AVERAGES


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 20 Southern Illinois-Edwardsville 38 38 25.5 2.0 3.7 0.539 0.0 0.0 0.000 2.0 3.7 0.543 0.539 1.8 2.8 0.630 1.4 7.2 8.6 0.6 0.42 2.13 0.32 2.5 5.8

2048 19 Southern Illinois-Edwardsville 37 37 26.3 2.2 4.4 0.506 0.0 0.1 0.000 2.2 4.3 0.519 0.506 1.4 2.5 0.565 1.6 6.2 7.8 0.5 0.41 0.92 0.95 3.2 5.8

Career 75 75 25.9 2.1 4.0 0.521 0.0 0.1 0.000 2.1 4.0 0.530 0.521 1.6 2.7 0.600 1.5 6.7 8.2 0.5 0.41 1.53 0.63 2.8 5.8

TOTALS


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 20 Southern Illinois-Edwardsville 38 38 967 76 141 0.539 0 1 0.000 76 140 0.543 0.539 68 108 0.630 54 274 328 21 16 81 12 94 220

2048 19 Southern Illinois-Edwardsville 37 37 972 82 162 0.506 0 4 0.000 82 158 0.519 0.506 52 92 0.565 59 231 290 18 15 34 35 119 216

Career 75 75 1939 158 303 0.521 0 5 0.000 158 298 0.530 0.521 120 200 0.600 113 505 618 39 31 115 47 213 436

PER 36 MINUTES


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 20 Southern Illinois-Edwardsville 38 38 967 2.8 5.2 0.539 0.0 0.0 0.000 2.8 5.2 0.543 0.539 2.5 4.0 0.630 2.0 10.2 12.2 0.8 0.60 3.02 0.45 3.5 8.2

2048 19 Southern Illinois-Edwardsville 37 37 972 3.0 6.0 0.506 0.0 0.1 0.000 3.0 5.9 0.519 0.506 1.9 3.4 0.565 2.2 8.6 10.7 0.7 0.56 1.26 1.30 4.4 8.0

Career 75 75 36.0 2.9 5.6 0.521 0.0 0.1 0.000 2.9 5.5 0.530 0.521 2.2 3.7 0.600 2.1 9.4 11.5 0.7 0.6 2.1 0.9 4.0 8.1

SHOOTING SPLITS


Season Age Team G GS MP AtRim AtRimA AtRim% Close CloseA Close% Mid MidA Mid% Long LongA Long% 3P 3PA 3P%

2049 20 Southern Illinois-Edwardsville 38 38 972 1.4 2.3 .632 0.5 1.1 .439 0.0 0.2 .000 0.1 0.2 .500 0.0 0.0 .000

2048 19 Southern Illinois-Edwardsville 37 37 972 1.7 3.2 .547 0.4 0.9 .485 0.0 0.2 .167 0.0 0.1 .500 0.0 0.1 .000

CLUTCH STATS


Season Age Team G MIN FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS +/-

2049 20 Southern Illinois-Edwardsville 20 64 6 15 .400 0 1 .000 6 14 .429 .400 14 18 .778 5 21 0 1 5 1 7 26 -3

===============================================

TAB 4: SCOUTING REPORT

===============================================

PLAYER RATINGS (Scale: 1-20)


SHOOTING:

Inside Shooting: 13

  • Mid-Range Shooting: 5

  • 3PT Shooting: 6

  • Free Throws: 10

CREATION:

Self-Creation: 9

  • Shoot Off Dribble: 3

  • Ball Dominance: 5

  • Foul Drawing: 19

  • Finishing: 8

EXECUTION:

Isolation Execution: 5

  • Post Execution: 11

  • Pick & Roll Execution: 6

  • Basketball IQ: 8

PLAYMAKING:

Playmaking: 1

  • Passing: 3

  • Gravity: 4

  • Spacing: 3

  • Ball Handling: 4

DEFENSE:

On-Ball Perimeter Defense: 2

  • Off-Ball Defense: 5

  • Help Defense: 15

  • Stealing: 8

  • Post Defense: 19

  • Rim Protection: 16

REBOUNDING:

Offensive Rebounding: 14

  • Defensive Rebounding: 18

PHYSICAL:

Athleticism: 13

  • Endurance: 19

  • First Step: 9

  • Quickness: 9

  • Speed: 6

  • Strength: 12

TENDENCIES:

Driving: 19

  • Roll : Pop: 80:20

  • Catch & Shoot : Pull-Up: 87:13

  • Corner 3 : ATB: 24:76

POSITIONAL FIT:

Guard: 1 star

  • Wing: 1 star

  • Big: 3 stars

INTANGIBLES:

Work Ethic: 8

  • Leadership: 17

  • Discipline: 18

  • Professionalism: 20

  • Aggression: 7

  • Toughness: 8

  • Hustle: 8

PERSONALITY:

Personality: Professional

SCOUTING INFORMATION:


Player Description: Rim-protecting five

Offensive Archetype: Roll-and-Cut Big

Defensive Archetype: Anchor Big

College Ability: 3 stars

Pro Potential: Average prospect with solid bench contributor potential

Declaration Chance: Unknown

Media Hype: None

Rankings: JBLDraft.Net: Unranked, HoopProspects.com: Unranked, JSPN: Unranked

OVERVIEW:


A 7'1", 239 lb sophomore from New Concord, Ohio who landed at a small-conference scholarship program after a low-profile recruiting run and has settled into a roll-and-cut big role at Southern Illinois-Edwardsville... And he plays heavy minutes as a defensive anchor and rebounder in the OVC while featuring sparing usage on offense and a lot of rim attempts rather than perimeter work...

STRENGTHS:


Uses length and size to finish a large share of looks at the rim, converting the majority of his attempts in close traffic which drives his efficient field-goal share... And he consistently cleans the defensive glass with physical positioning and timing rather than leaping, giving his team extra possessions and steady defensive rebounding numbers... He deters interior shots and produces swats and alterations when stationed in the paint, helping the team limit opponent looks in his area... And he stays available and plays through a full minute load thanks to clear endurance, which lets coaches count on him late into games... He is professional in approach around staff and practice structure and shows the discipline to run set plays and stay within his role...

WEAKNESSES:


Offers almost no shooting threat from midrange or three, which collapses spacing and funnels everything into the paint where he struggles to create or finish through contact... And his handle and playmaking are unreliable with a turnover problem that outpaces his assist output, so possessions often die when the ball reaches him in traffic or when he's asked to make reads out of the post or pick-and-roll... He lacks first step, quickness and on-court burst, which limits his ability to chase perimeter actions or threaten defenders off the catch, and his footwork and post execution are underdeveloped so touches near the basket are low-value more often than not... He draws fouls effectively but misses enough free throws to blunt that edge and leaves points at the line... And while he positions well and sees rotations, his low aggression and minimal hustle mean he rarely convert that awareness into help plays or loose-ball activity - a tame rim-runner rather than a wrecking ball...

PROJECTION:


On current trajectory he profiles as a high-level mid-major starter who can control the paint defensively and on the glass if he accepts a limited offensive script, with a possible JBL workout invite only if he adds consistent shooting, drastically tightens decision-making and develops strength to handle bigger pros... And his professional manner and discipline make him coachable, but low drive and weak fundamentals cap the ceiling unless work ethic changes and he commits to repetitive guardable skills like free-throw mechanics and short-area quickness drills... He carries a clear boom/bust risk - length and rim defense give upside but limited athleticism and offensive polish make a JBL draft ticket unlikely without major strides...

Scouting Accuracy: High

Scouting from: Southern Illinois-Edwardsville

PLAYER 4 OF 12: STANISLAV KUNCHIK

===============================================

STANISLAV KUNCHIK - PLAYER PROFILE DATA

===============================================

BASIC INFORMATION


Name: Stanislav Kunchik

Team: Southern Illinois-Edwardsville

Position: SG/PG

Height: 6'3

Weight: 188 lbs

Age: 21 y.o. (Sophomore)

Shoots: Right

Wingspan: 6'6

Hometown: Berezan, Ukraine

High School: Berezan (HS)

Current Stats: 2.2 PPG, 0.8 RPG, 1.9 APG, 0.55 SPG, 0.05 BPG, 11.7 MPG, 11.5 PER

===============================================

TAB 1: STATISTICS

===============================================

PER GAME AVERAGES


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 20 Southern Illinois-Edwardsville 38 0 11.7 0.8 2.0 0.390 0.2 0.7 0.231 0.6 1.3 0.471 0.429 0.5 0.5 0.947 0.3 0.5 0.8 1.9 0.55 0.05 0.53 0.6 2.2

2048 19 Southern Illinois-Edwardsville 6 0 1.2 0.0 0.2 0.000 0.0 0.2 0.000 0.0 0.0 0 0.000 0.0 0.0 0 0 0.0 0.0 0.7 0.17 0.00 0.00 0.2 0.0

Career 44 0 10.2 0.7 1.8 0.385 0.1 0.6 0.222 0.5 1.2 0.471 0.423 0.4 0.4 0.947 0.2 0.5 0.7 1.7 0.50 0.05 0.45 0.5 1.9

TOTALS


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 20 Southern Illinois-Edwardsville 38 0 443 30 77 0.390 6 26 0.231 24 51 0.471 0.429 18 19 0.947 10 20 30 71 21 2 20 23 84

2048 19 Southern Illinois-Edwardsville 6 0 7 0 1 0.000 0 1 0.000 0 0 0 0.000 0 0 0.947 0 0 0 4 1 0 0 1 0

Career 44 0 450 30 78 0.385 6 27 0.222 24 51 0.471 0.423 18 19 0.947 10 20 30 75 22 2 20 24 84

===============================================

TAB 4: SCOUTING REPORT

===============================================

PLAYER RATINGS (Scale: 1-20)


SHOOTING:

Inside Shooting: 12

  • Mid-Range Shooting: 17

  • 3PT Shooting: 12

  • Free Throws: 17

CREATION:

Self-Creation: 18

  • Shoot Off Dribble: 11

  • Ball Dominance: 11

  • Foul Drawing: 19

  • Finishing: 7

EXECUTION:

Isolation Execution: 9

  • Post Execution: 5

  • Pick & Roll Execution: 6

  • Basketball IQ: 17

PLAYMAKING:

Playmaking: 14

  • Passing: 18

  • Gravity: 17

  • Spacing: 11

  • Ball Handling: 17

DEFENSE:

On-Ball Perimeter Defense: 10

  • Off-Ball Defense: 13

  • Help Defense: 11

  • Stealing: 6

  • Post Defense: 10

  • Rim Protection: 2

REBOUNDING:

Offensive Rebounding: 7

  • Defensive Rebounding: 8

PHYSICAL:

Athleticism: 16

  • Endurance: 14

  • First Step: 18

  • Quickness: 14

  • Speed: 14

  • Strength: 2

TENDENCIES:

Driving: 13

  • Roll : Pop: 15:85

  • Catch & Shoot : Pull-Up: 53:47

  • Corner 3 : ATB: 13:87

POSITIONAL FIT:

Guard: 2 stars

  • Wing: 2 stars

  • Big: 1 star

INTANGIBLES:

Work Ethic: 19

  • Leadership: 10

  • Discipline: 15

  • Professionalism: 13

  • Aggression: 19

  • Toughness: 13

  • Hustle: 17

PERSONALITY:

Personality: Confident

SCOUTING INFORMATION:


Player Description: Explosive three-point shot creating floor general finishing guard

Offensive Archetype: Shot Creator

Defensive Archetype: Point Of Attack

College Ability: 2 stars

Pro Potential: Fringe prospect with development needed, possible rotation upside

Declaration Chance: Unknown

Media Hype: None

Rankings: JBLDraft.Net: Unranked, HoopProspects.com: Unranked, JSPN: Unranked

Scouting Accuracy: High

Scouting from: Southern Illinois-Edwardsville

PLAYER 5 OF 12: CYNCIER LEVETT

===============================================

CYNCIER LEVETT - PLAYER PROFILE DATA

===============================================

BASIC INFORMATION


Name: Cyncier Levett

Team: Southern Illinois-Edwardsville

Position: SF

Height: 6'7

Weight: 216 lbs

Age: 22 y.o. (Senior)

Shoots: Right

Wingspan: 6'11

Hometown: Flossmoor, Illinois

High School: Homewood-Flossmoor

Current Stats: 9.5 PPG, 1.3 RPG, 0.4 APG, 0.64 SPG, 0.05 BPG, 13.3 MPG, 22.4 PER

===============================================

TAB 1: STATISTICS

===============================================

PER GAME AVERAGES


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 21 Southern Illinois-Edwardsville 22 0 13.3 3.4 7.5 0.457 1.2 2.9 0.429 2.2 4.6 0.475 0.540 1.4 1.7 0.838 0.5 0.9 1.3 0.4 0.64 0.05 1.00 0.4 9.5

2048 20 Southern Illinois-Edwardsville 37 0 12.1 2.3 5.0 0.465 0.6 1.9 0.333 1.7 3.1 0.543 0.527 1.5 1.8 0.821 0.6 0.5 1.1 0.8 0.35 0.16 0.78 1.1 6.8

2047 19 Southern Illinois-Edwardsville 2 0 3.1 0.5 0.5 1.000 0.5 0.5 1.000 0.0 0.0 0 1.500 0.0 0.0 0 0.0 0.5 0.5 0.0 0.00 0.00 0.50 0.5 1.5

Career 61 0 12.2 2.7 5.7 0.463 0.8 2.2 0.383 1.8 3.6 0.512 0.536 1.4 1.7 0.827 0.5 0.6 1.1 0.6 0.44 0.11 0.85 0.8 7.6

===============================================

TAB 4: SCOUTING REPORT

===============================================

PLAYER RATINGS (Scale: 1-20)


SHOOTING:

Inside Shooting: 18

  • Mid-Range Shooting: 12

  • 3PT Shooting: 18

  • Free Throws: 18

CREATION:

Self-Creation: 18

  • Shoot Off Dribble: 12

  • Ball Dominance: 17

  • Foul Drawing: 18

  • Finishing: 16

EXECUTION:

Isolation Execution: 12

  • Post Execution: 3

  • Pick & Roll Execution: 19

  • Basketball IQ: 10

PLAYMAKING:

Playmaking: 7

  • Passing: 10

  • Gravity: 15

  • Spacing: 10

  • Ball Handling: 20

DEFENSE:

On-Ball Perimeter Defense: 9

  • Off-Ball Defense: 10

  • Help Defense: 9

  • Stealing: 5

  • Post Defense: 9

  • Rim Protection: 4

REBOUNDING:

Offensive Rebounding: 8

  • Defensive Rebounding: 6

PHYSICAL:

Athleticism: 14

  • Endurance: 11

  • First Step: 18

  • Quickness: 18

  • Speed: 16

  • Strength: 3

INTANGIBLES:

Work Ethic: 16

  • Leadership: 13

  • Discipline: 8

  • Professionalism: 9

  • Aggression: 11

  • Toughness: 10

  • Hustle: 10

PERSONALITY:

Personality: Friendly

SCOUTING INFORMATION:


Player Description: Penetrating finishing three-point shot creating offensive finishing wing

Offensive Archetype: Secondary Creator

Defensive Archetype: Chaser

College Ability: 3 stars

Pro Potential: Longshot prospect with limited JBL upside

Scouting Accuracy: High

Scouting from: Southern Illinois-Edwardsville

PLAYER 6 OF 12: MARK MIMMS

===============================================

MARK MIMMS - PLAYER PROFILE DATA

===============================================

BASIC INFORMATION


Name: Mark Mimms

Team: Southern Illinois-Edwardsville

Position: PG

Height: 6'0

Weight: 202 lbs

Age: 21 y.o. (Junior)

Shoots: Right

Wingspan: 6'3

Hometown: Woodmere, New York

High School: Woodmere Academy

Current Stats: 3.6 PPG, 1.0 RPG, 2.4 APG, 0.58 SPG, 0.00 BPG, 13.5 MPG, 16.3 PER

===============================================

TAB 1: STATISTICS

===============================================

PER GAME AVERAGES


Season Age Team G GS MP FG FGA FG% 3P 3PA 3P% 2P 2PA 2P% eFG% FT FTA FT% ORB DRB TRB AST STL BLK TOV PF PTS

2049 20 Southern Illinois-Edwardsville 38 0 13.5 1.3 2.5 0.500 0.7 1.4 0.455 0.6 1.1 0.561 0.630 0.4 0.5 0.833 0.4 0.6 1.0 2.4 0.58 0.00 0.61 0.6 3.6

2048 19 Hofstra 38 0 17.0 2.1 4.7 0.436 0.8 2.0 0.400 1.3 2.7 0.462 0.520 0.7 0.9 0.818 0.4 0.4 0.9 2.4 0.74 0.03 0.58 1.1 5.6

Career 76 0 15.3 1.7 3.6 0.458 0.7 1.7 0.423 0.9 1.9 0.490 0.558 0.6 0.7 0.824 0.4 0.5 0.9 2.4 0.66 0.01 0.59 0.8 4.6

===============================================

TAB 4: SCOUTING REPORT

===============================================

PLAYER RATINGS (Scale: 1-20)


SHOOTING:

Inside Shooting: 16

  • Mid-Range Shooting: 18

  • 3PT Shooting: 18

  • Free Throws: 16

CREATION:

Self-Creation: 13

  • Shoot Off Dribble: 14

  • Ball Dominance: 8

  • Foul Drawing: 10

  • Finishing: 15

EXECUTION:

Isolation Execution: 16

  • Post Execution: 2

  • Pick & Roll Execution: 14

  • Basketball IQ: 8

PLAYMAKING:

Playmaking: 10

  • Passing: 17

  • Gravity: 13

  • Spacing: 16

  • Ball Handling: 19

DEFENSE:

On-Ball Perimeter Defense: 15

  • Off-Ball Defense: 12

  • Help Defense: 6

  • Stealing: 7

  • Post Defense: 1

  • Rim Protection: 2

REBOUNDING:

Offensive Rebounding: 6

  • Defensive Rebounding: 7

PHYSICAL:

Athleticism: 13

  • Endurance: 9

  • First Step: 8

  • Quickness: 19

  • Speed: 15

  • Strength: 7

INTANGIBLES:

Work Ethic: 9

  • Leadership: 7

  • Discipline: 9

  • Professionalism: 13

  • Aggression: 14

  • Toughness: 8

  • Hustle: 8

PERSONALITY:

Personality: Ambitious

SCOUTING INFORMATION:


Player Description: Quick three-point shot creating perimeter guard

Offensive Archetype: Primary Ballhandler

Defensive Archetype: Point Of Attack

College Ability: 2 stars

Pro Potential: Fringe prospect with development needed, possible rotation upside

Scouting Accuracy: High

Scouting from: Southern Illinois-Edwardsville

PLAYER 7 OF 12: L.A. MISCH

===============================================

L.A. MISCH - PLAYER PROFILE DATA

===============================================

BASIC INFORMATION


Name: L.A. Misch

Team: Southern Illinois-Edwardsville

Position: PG

Height: 6'6

Weight: 193 lbs

Age: 21 y.o. (Junior)

Shoots: Right

Wingspan: 6'10

Hometown: Bloomington, Illinois

High School: Bloomington Central Catholic

Current Stats: 0.0 PPG, 1.0 RPG, 0.0 APG, 0.00 SPG, 0.00 BPG, 1.6 MPG, 10.5 PER

===============================================

TAB 4: SCOUTING REPORT

===============================================

PLAYER RATINGS (Scale: 1-20)


SHOOTING:

Inside Shooting: 17

  • Mid-Range Shooting: 11

  • 3PT Shooting: 18

  • Free Throws: 19

CREATION:

Self-Creation: 15

  • Shoot Off Dribble: 14

  • Ball Dominance: 7

  • Foul Drawing: 10

  • Finishing: 16

EXECUTION:

Isolation Execution: 14

  • Post Execution: 4

  • Pick & Roll Execution: 11

  • Basketball IQ: 9

PLAYMAKING:

Playmaking: 9

  • Passing: 16

  • Gravity: 12

  • Spacing: 16

  • Ball Handling: 11

DEFENSE:

On-Ball Perimeter Defense: 10

  • Off-Ball Defense: 13

  • Help Defense: 8

  • Stealing: 9

  • Post Defense: 3

  • Rim Protection: 2

REBOUNDING:

Offensive Rebounding: 3

  • Defensive Rebounding: 5

PHYSICAL:

Athleticism: 9

  • Endurance: 17

  • First Step: 12

  • Quickness: 15

  • Speed: 17

  • Strength: 10

POSITIONAL FIT:

Guard: 2 stars

  • Wing: 2 stars

  • Big: 1 star

SCOUTING INFORMATION:


Player Description: Three-point shot creating perimeter guard

Offensive Archetype: Primary Ballhandler

Defensive Archetype: Chaser

College Ability: 2 stars

Pro Potential: Deep bench potential

Scouting Accuracy: High

Scouting from: Southern Illinois-Edwardsville

PLAYER 8 OF 12: DONCHEVELL ROSS

===============================================

DONCHEVELL ROSS - PLAYER PROFILE DATA

===============================================

BASIC INFORMATION


Name: Donchevell Ross

Team: Southern Illinois-Edwardsville

Position: C

Height: 6'10

Weight: 253 lbs

Age: 21 y.o. (Junior)

Shoots: Right

Wingspan: 7'2

Hometown: Mentor, Ohio

High School: Mentor

Current Stats: 3.5 PPG, 4.4 RPG, 0.3 APG, 0.21 SPG, 1.32 BPG, 14.9 MPG, 13.2 PER

===============================================

TAB 4: SCOUTING REPORT

===============================================

PLAYER RATINGS (Scale: 1-20)


SHOOTING:

Inside Shooting: 16

  • Mid-Range Shooting: 8

  • 3PT Shooting: 10

  • Free Throws: 11

CREATION:

Self-Creation: 10

  • Shoot Off Dribble: 3

  • Ball Dominance: 7

  • Foul Drawing: 12

  • Finishing: 15

EXECUTION:

Isolation Execution: 13

  • Post Execution: 8

  • Pick & Roll Execution: 7

  • Basketball IQ: 7

PLAYMAKING:

Playmaking: 5

  • Passing: 4

  • Gravity: 7

  • Spacing: 4

  • Ball Handling: 2

DEFENSE:

On-Ball Perimeter Defense: 4

  • Off-Ball Defense: 5

  • Help Defense: 14

  • Stealing: 10

  • Post Defense: 17

  • Rim Protection: 16

REBOUNDING:

Offensive Rebounding: 19

  • Defensive Rebounding: 17

PHYSICAL:

Athleticism: 11

  • Endurance: 5

  • First Step: 5

  • Quickness: 4

  • Speed: 4

  • Strength: 9

INTANGIBLES:

Work Ethic: 17

  • Leadership: 6

  • Discipline: 7

  • Professionalism: 10

  • Aggression: 11

  • Toughness: 10

  • Hustle: 7

PERSONALITY:

Personality: Friendly

SCOUTING INFORMATION:


Player Description: Shot-blocking five

Offensive Archetype: Roll-and-Cut Big

Defensive Archetype: Anchor Big

College Ability: 3 stars

Pro Potential: Fringe prospect with development needed, possible rotation upside

Scouting Accuracy: High

Scouting from: Southern Illinois-Edwardsville

PLAYER 9 OF 12: ARLON SANDERSON

===============================================

ARLON SANDERSON - PLAYER PROFILE DATA

===============================================

BASIC INFORMATION


Name: Arlon Sanderson

Team: Southern Illinois-Edwardsville

Position: SF

Height: 6'7

Weight: 199 lbs

Age: 22 y.o. (Junior)

Shoots: Right

Wingspan: 6'11

Hometown: Clarkston, Michigan

High School: Clarksto

Current Stats: 4.2 PPG, 2.2 RPG, 4.9 APG, 1.18 SPG, 0.08 BPG, 23.2 MPG, 16.3 PER

===============================================

TAB 4: SCOUTING REPORT

===============================================

PLAYER RATINGS (Scale: 1-20)


SHOOTING:

Inside Shooting: 14

  • Mid-Range Shooting: 12

  • 3PT Shooting: 9

  • Free Throws: 18

CREATION:

Self-Creation: 17

  • Shoot Off Dribble: 11

  • Ball Dominance: 15

  • Foul Drawing: 9

  • Finishing: 17

EXECUTION:

Isolation Execution: 15

  • Post Execution: 2

  • Pick & Roll Execution: 18

  • Basketball IQ: 10

PLAYMAKING:

Playmaking: 14

  • Passing: 17

  • Gravity: 17

  • Spacing: 9

  • Ball Handling: 18

DEFENSE:

On-Ball Perimeter Defense: 16

  • Off-Ball Defense: 15

  • Help Defense: 9

  • Stealing: 15

  • Post Defense: 8

  • Rim Protection: 2

REBOUNDING:

Offensive Rebounding: 10

  • Defensive Rebounding: 9

PHYSICAL:

Athleticism: 12

  • Endurance: 10

  • First Step: 10

  • Quickness: 15

  • Speed: 16

  • Strength: 2

INTANGIBLES:

Work Ethic: 19

  • Leadership: 10

  • Discipline: 7

  • Professionalism: 11

  • Aggression: 7

  • Toughness: 8

  • Hustle: 5

PERSONALITY:

Personality: Confident

SCOUTING INFORMATION:


Player Description: 3 & D perimeter wing

Offensive Archetype: Secondary Creator

Defensive Archetype: Wing Stopper

College Ability: 3 stars

Pro Potential: Fair prospect with potential to make a roster as a role player

Scouting Accuracy: High

Scouting from: Southern Illinois-Edwardsville

PLAYER 10 OF 12: QUION SLIWOSKI

===============================================

QUION SLIWOSKI - PLAYER PROFILE DATA

===============================================

BASIC INFORMATION


Name: Quion Sliwoski

Team: Southern Illinois-Edwardsville

Position: PF

Height: 6'10

Weight: 250 lbs

Age: 21 y.o. (Junior)

Shoots: Right

Wingspan: 7'1

Hometown: Sylvania, Ohio

High School: Southview

Current Stats: 6.5 PPG, 4.6 RPG, 0.6 APG, 0.55 SPG, 1.08 BPG, 24.3 MPG, 12.9 PER

===============================================

TAB 4: SCOUTING REPORT

===============================================

PLAYER RATINGS (Scale: 1-20)


SHOOTING:

Inside Shooting: 17

  • Mid-Range Shooting: 9

  • 3PT Shooting: 15

  • Free Throws: 14

CREATION:

Self-Creation: 9

  • Shoot Off Dribble: 9

  • Ball Dominance: 6

  • Foul Drawing: 20

  • Finishing: 13

EXECUTION:

Isolation Execution: 18

  • Post Execution: 8

  • Pick & Roll Execution: 9

  • Basketball IQ: 9

PLAYMAKING:

Playmaking: 8

  • Passing: 8

  • Gravity: 10

  • Spacing: 12

  • Ball Handling: 3

DEFENSE:

On-Ball Perimeter Defense: 7

  • Off-Ball Defense: 6

  • Help Defense: 8

  • Stealing: 7

  • Post Defense: 9

  • Rim Protection: 13

REBOUNDING:

Offensive Rebounding: 15

  • Defensive Rebounding: 12

PHYSICAL:

Athleticism: 7

  • Endurance: 11

  • First Step: 11

  • Quickness: 10

  • Speed: 4

  • Strength: 17

INTANGIBLES:

Work Ethic: 8

  • Leadership: 9

  • Discipline: 8

  • Professionalism: 5

  • Aggression: 9

  • Toughness: 8

  • Hustle: 8

PERSONALITY:

Personality: Eccentric

SCOUTING INFORMATION:


Player Description: Flashy three-point catch and shoot four

Offensive Archetype: Roll-and-Cut Big

Defensive Archetype: Mobile Big

College Ability: 3 stars

Pro Potential: Average prospect with solid bench contributor potential

Scouting Accuracy: High

Scouting from: Southern Illinois-Edwardsville

PLAYER 11 OF 12: ARKEL TOPERCER

===============================================

ARKEL TOPERCER - PLAYER PROFILE DATA

===============================================

BASIC INFORMATION


Name: Arkel Topercer

Team: Southern Illinois-Edwardsville

Position: SG

Height: 6'8

Weight: 208 lbs

Age: 22 y.o. (Senior)

Shoots: Right

Wingspan: 7'0

Hometown: Chicago, Illinois

High School: Simeon

Current Stats: 17.4 PPG, 2.4 RPG, 1.4 APG, 1.55 SPG, 0.13 BPG, 26.1 MPG, 22.4 PER

===============================================

TAB 4: SCOUTING REPORT

===============================================

PLAYER RATINGS (Scale: 1-20)


SHOOTING:

Inside Shooting: 14

  • Mid-Range Shooting: 18

  • 3PT Shooting: 19

  • Free Throws: 18

CREATION:

Self-Creation: 12

  • Shoot Off Dribble: 17

  • Ball Dominance: 18

  • Foul Drawing: 18

  • Finishing: 13

EXECUTION:

Isolation Execution: 9

  • Post Execution: 5

  • Pick & Roll Execution: 15

  • Basketball IQ: 13

PLAYMAKING:

Playmaking: 11

  • Passing: 6

  • Gravity: 16

  • Spacing: 17

  • Ball Handling: 15

DEFENSE:

On-Ball Perimeter Defense: 13

  • Off-Ball Defense: 13

  • Help Defense: 9

  • Stealing: 10

  • Post Defense: 4

  • Rim Protection: 5

REBOUNDING:

Offensive Rebounding: 11

  • Defensive Rebounding: 8

PHYSICAL:

Athleticism: 18

  • Endurance: 10

  • First Step: 9

  • Quickness: 15

  • Speed: 14

  • Strength: 16

INTANGIBLES:

Work Ethic: 4

  • Leadership: 6

  • Discipline: 8

  • Professionalism: 2

  • Aggression: 8

  • Toughness: 10

  • Hustle: 7

PERSONALITY:

Personality: Lazy

SCOUTING INFORMATION:


Player Description: Athletic three-point catch and shoot offensive finishing wing

Offensive Archetype: Secondary Creator

Defensive Archetype: Chaser

College Ability: 4 stars

Pro Potential: Decent prospect with rotation player upside

Scouting Accuracy: High

Scouting from: Southern Illinois-Edwardsville

================================================================================

PLAYER 12 OF 12: TABARI TOTTEN

================================================================================

BASIC INFORMATION:


Name: Tabari Totten

Team: Southern Illinois-Edwardsville

Position: PF

Height: 6'11"

Weight: 250lbs

Age: 21 y.o. (Junior)

Shoots: Right

Wingspan: 73

Hometown: Richmond, Illinois

High School: Richmond-Burton

SEASON AVERAGES:

4.7 PPG, 3.0 RPG, 0.8 APG, 0.34 SPG, 0.84 BPG, 16.7 MPG, 13.4 PER

STATISTICS:

===========

PER GAME:

Season | Age | Team | G | GS | MP | FG | FGA | FG% | 3P | 3PA | 3P% | 2P | 2PA | 2P% | eFG% | FT | FTA | FT% | ORB | DRB | TRB | AST | STL | BLK | TOV | PF | PTS

2049 | 20 | SIUE | 38 | 0 | 16.7 | 1.8 | 4.1 | .436 | 0.6 | 1.8 | .328 | 1.2 | 2.3 | .517 | .506 | 0.5 | 0.7 | .679 | 1.0 | 2.1 | 3.0 | 0.8 | 0.34 | 0.84 | 0.63 | 0.9 | 4.7

2048 | 19 | SIUE | 6 | 0 | 2.6 | 0.5 | 0.7 | .750 | 0.3 | 0.5 | .667 | 0.2 | 0.2 | 1.000 | 1.000 | 0.0 | 0.0 | - | 0.0 | 0.3 | 0.3 | 0.0 | 0.00 | 0.00 | 0.00 | 0.7 | 1.3

Career | - | - | 44 | 0 | 14.8 | 1.6 | 3.6 | .444 | 0.5 | 1.6 | .343 | 1.1 | 2.0 | .522 | .519 | 0.4 | 0.6 | .679 | 0.8 | 1.8 | 2.7 | 0.7 | 0.30 | 0.73 | 0.55 | 0.9 | 4.2

TOTALS:

Season | Age | Team | G | GS | MP | FG | FGA | FG% | 3P | 3PA | 3P% | 2P | 2PA | 2P% | eFG% | FT | FTA | FT% | ORB | DRB | TRB | AST | STL | BLK | TOV | PF | PTS

2049 | 20 | SIUE | 38 | 0 | 635 | 68 | 156 | .436 | 22 | 67 | .328 | 46 | 89 | .517 | .506 | 19 | 28 | .679 | 37 | 78 | 115 | 31 | 13 | 32 | 24 | 35 | 177

2048 | 19 | SIUE | 6 | 0 | 16 | 3 | 4 | .750 | 2 | 3 | .667 | 1 | 1 | 1.000 | 1.000 | 0 | 0 | - | 0 | 2 | 2 | 0 | 0 | 0 | 0 | 4 | 8

Career | - | - | 44 | 0 | 651 | 71 | 160 | .444 | 24 | 70 | .343 | 47 | 90 | .522 | .519 | 19 | 28 | .679 | 37 | 80 | 117 | 31 | 13 | 32 | 24 | 39 | 185

PER 36 MINUTES:

Season | Age | Team | G | GS | MP | FG | FGA | FG% | 3P | 3PA | 3P% | 2P | 2PA | 2P% | eFG% | FT | FTA | FT% | ORB | DRB | TRB | AST | STL | BLK | TOV | PF | PTS

2049 | 20 | SIUE | 38 | 0 | 635 | 3.9 | 8.8 | .436 | 1.2 | 3.8 | .328 | 2.6 | 5.0 | .517 | .506 | 1.1 | 1.6 | .679 | 2.1 | 4.4 | 6.5 | 1.8 | 0.74 | 1.81 | 1.36 | 2.0 | 10.0

2048 | 19 | SIUE | 6 | 0 | 16 | 6.8 | 9.0 | .750 | 4.5 | 6.8 | .667 | 2.3 | 2.3 | 1.000 | 1.000 | 0.0 | 0.0 | - | 0.0 | 4.5 | 4.5 | 0.0 | 0.00 | 0.00 | 0.00 | 9.0 | 18.0

Career | - | - | 44 | 0 | 36.0 | 3.9 | 8.8 | .444 | 1.3 | 3.9 | .343 | 2.6 | 5.0 | .522 | .519 | 1.1 | 1.5 | .679 | 2.0 | 4.4 | 6.5 | 1.7 | 0.7 | 1.8 | 1.3 | 2.2 | 10.2

SHOOTING SPLITS:

Season | Age | Team | G | GS | MP | AtRim | AtRimA | AtRim% | Close | CloseA | Close% | Mid | MidA | Mid% | Long | LongA | Long% | 3P | 3PA | 3P%

2049 | 20 | SIUE | 38 | 0 | 16 | 0.9 | 1.2 | .745 | 0.1 | 0.4 | .235 | 0.1 | 0.6 | .227 | 0.1 | 0.1 | .667 | 0.6 | 1.8 | .328

2048 | 19 | SIUE | 6 | 0 | 16 | 0.0 | 0.0 | .000 | 0.2 | 0.2 | 1.000 | 0.0 | 0.0 | .000 | 0.0 | 0.0 | .000 | 0.3 | 0.5 | .667

SCOUTING REPORT:

================

SHOOTING:

Inside Shooting: 16

  • Mid-Range Shooting: 16

  • 3PT Shooting: 16

  • Free Throws: 17

CREATION:

Self-Creation: 4

  • Shoot Off Dribble: 5

  • Ball Dominance: 9

  • Foul Drawing: 12

  • Finishing: 17

EXECUTION:

Isolation Execution: 9

  • Post Execution: 9

  • Pick & Roll Execution: 12

  • Basketball IQ: 12

PLAYMAKING:

Playmaking: 10

  • Passing: 7

  • Gravity: 9

  • Spacing: 13

  • Ball Handling: 3

DEFENSE:

On-Ball Perimeter Defense: 3

  • Off-Ball Defense: 5

  • Help Defense: 10

  • Stealing: 7

  • Post Defense: 12

  • Rim Protection: 11

REBOUNDING:

Offensive Rebounding: 15

  • Defensive Rebounding: 15

PHYSICAL:

Athleticism: 9

  • Endurance: 13

  • First Step: 8

  • Quickness: 5

  • Speed: 5

  • Strength: 9

INTANGIBLES:

Work Ethic: 16

  • Leadership: 19

  • Discipline: 6

  • Professionalism: 19

  • Aggression: 10

  • Toughness: 7

  • Hustle: 6

PERSONALITY:

Personality: Leader

SCOUTING INFORMATION:


Player Description: Three-point catch and shoot wing four

Offensive Archetype: Stretch Big

Defensive Archetype: Mobile Big

College Ability: 3 stars

Pro Potential: Fringe prospect with development needed, possible rotation upside

Scouting Accuracy: High

Scouting from: Southern Illinois-Edwardsville

================================================================================

END OF SOUTHERN ILLINOIS-EDWARDSVILLE ROSTER DATA

================================================================================

All 12 players compiled successfully.

Generated from: jblfl.com/cjbl

Unified Rating Guide

Unified Rating Guide — CJBL & JBL

Shared reference for both college (CJBL) and pro (JBL) projects. Both leagues use the same 0-100 raw scale converted to 1-20 display ratings.

Raw-to-Display Conversion Table

Both CJBL and JBL use identical conversion boundaries:

Raw Value Range Display Rating Tier
85-100 20 Elite
81-84 19 Elite
79-80 18 Excellent
75-78 17 Excellent
72-74 16 Very Good
69-71 15 Very Good
66-68 14 Good
63-65 13 Good
61-62 12 Above Average
58-60 11 Above Average
54-57 10 Average
50-53 9 Average
45-49 8 Below Average
42-44 7 Below Average
38-41 6 Poor
29-37 5 Poor
24-28 4 Very Poor
19-23 3 Very Poor
14-18 2 Minimal
0-13 1 Minimal

Complete Rating Mapping (CJBL <-> JBL)

29 ratings match between leagues. 3 are JBL-only, 6 are CJBL-only.

Matched Ratings (sorted by JBL win correlation)

CJBL Raw Field CJBL Display JBL Column JBL Win R²
InsideShooting Inside Shooting rating_shooting_Inside Shooting 0.522
DefRebounding Defensive Rebounding rating_rebounding_Defensive Rebounding 0.428
PnRExecution Pick & Roll Execution rating_execution_Pick & Roll Execution 0.418
HelpDefense Help Defense rating_defense_Help Defense 0.398
LongRangeShooting 3PT Shooting rating_shooting_3PT Shooting 0.373
OffRebounding Offensive Rebounding rating_rebounding_Offensive Rebounding 0.303
ShootOffDribble Shoot Off Dribble rating_creation_Shoot Off Dribble 0.232
PerimeterDefense On-Ball Perimeter Defense rating_defense_On-Ball Perimeter Defense 0.228
Endurance Endurance rating_physical_Endurance 0.220
BallDominance Ball Dominance rating_creation_Ball Dominance 0.190
Playmaking Playmaking rating_playmaking_Playmaking 0.186
BallHandling Ball Handling rating_playmaking_Ball Handling 0.178
SelfCreation Self-Creation rating_creation_Self-Creation 0.156
Stealing Stealing rating_defense_Stealing 0.148
Blocking Rim Protection rating_defense_Rim Protection 0.139
Spacing Spacing rating_playmaking_Spacing 0.134
Gravity Gravity rating_playmaking_Gravity 0.112
Speed Speed rating_physical_Speed 0.110
MidRangeShooting Mid-Range Shooting rating_shooting_Mid-Range Shooting
FTShooting Free Throw rating_shooting_Free Throws
FoulDrawing Foul Drawing rating_creation_Foul Drawing
Intelligence Basketball IQ rating_execution_Basketball IQ
Passing Passing rating_playmaking_Passing
OffBallDefense Off-Ball Defense rating_defense_Off-Ball Defense
PostDefense Post Defense rating_defense_Post Defense
Athletics Athleticism rating_physical_Athleticism
Quickness Quickness rating_physical_Quickness
Explosiveness First Step rating_physical_First Step
Strength Strength rating_physical_Strength

JBL-Only Ratings (no CJBL equivalent)

JBL Rating Category Notes
Finishing Creation Pro-level rim finishing. College uses InsideShooting broadly.
Isolation Execution Execution Not separately tracked in CJBL. R²=0.303 with wins.
Post Execution Execution Big-man post moves. Not tracked in CJBL.

CJBL-Only Ratings (no JBL equivalent)

CJBL Rating Category Notes
Offensive Awareness Execution College-specific awareness metric.
Scoring Execution Composite scoring instinct.
Defensive Awareness Defense College-specific defensive awareness.
Positioning Rebounding Rebounding positioning specifically.
Jumping Physical Vertical leap / aerial ability.
Durability Physical Injury resistance.

What "Good" Looks Like

College Context (CJBL — 1-star talent pool)

At SIUE's Very Low prestige level, targeting 1-star recruits:

Rating Level Display Meaning at College
Elite (18-20) Rare. Only 1-2 per recruit in 1-star pool. Franchise-caliber college skill.
Excellent (16-17) Strong. 3-5 per top recruit. Core rotation skill.
Very Good (14-15) Solid. Average for recruited targets. Meets minimum thresholds.
Below 12 Weak. Below Morey Ball minimums. Red flag for scheme fit.

Pro Context (JBL — League-wide)

Across 516 JBL1 players:

Rating Level Display Meaning at Pro Level
Elite (18-20) Star-caliber. Top 10% of league. Franchise skill — build around it.
Excellent (16-17) Above average. Top 25%. Starter quality in that skill.
Average (12-14) League median range. Rotation player level.
Below 10 Bottom quartile. Liability — avoid or scheme around.

Cross-League Evaluation Tips

  1. College 16 ≈ Pro 14-15: Development happens, but college standouts face stiffer pro competition.
  2. Endurance matters more in pro: 82-game pro season vs shorter college season. College End 13 may limit to 20 MPG in pros.
  3. PnR Execution translates directly: Top-3 win predictor in both leagues.
  4. BBIQ is universal: #1 college predictor, enables every pro skill to function better.
  5. Character predicts development speed: WE 20 players close the college-to-pro gap faster.
College Prospect Intel for Pro

College Prospect Intel for Pro Draft

Cross-project knowledge doc from CJBL SIUE AD (college scouting). Source: 368-team GB models, 22K recruit database, SIUE pipeline tracking.

What College Analytics Tell Pro GMs

BBIQ Is the #1 College Predictor

CJBL's gradient boosted models across 368 teams found that Basketball IQ has 2x the importance of any other skill in predicting college team success. This translates to pro value because:

  • High BBIQ players make better decisions under pressure
  • BBIQ correlates with scheme adaptability (fitting into any system)
  • BBIQ is hard to develop — it's largely innate or very slowly trained

Pro Application: When evaluating draft prospects, BBIQ 16+ is a strong positive signal even if other skills are moderate. BBIQ 12 or below is a yellow flag for pro adjustment speed.

Character Screening Framework

CJBL scouting uncovered that 87% of recruits fail character screening. The same framework applies to pro draft evaluation:

Attribute Threshold Impact
Work Ethic 15+ Development speed. WE below 12 = very slow growth.
Discipline 12+ Decision-making, foul trouble, consistency.
Professionalism 12+ Coachability, team chemistry, off-court risk.

Character Verdicts: - CLEAN (WE 15+, Disc 12+, Prof 12+): Safe investment. ~13% of prospects. - BORDERLINE (one attribute 10-11, rest pass): Acceptable risk if talent justifies. - FAIL (any below threshold): High-risk investment. Proceed only for elite talent. - AUTO-REJECT personalities: Troublemaker, Disruptive, Volatile, Lazy, Selfish. - Positive signals: Workhorse, Dedicated, Professional.

How to Use College Data in Pro Draft

  1. Map college ratings to pro equivalents using the shared rating_map.json
  2. Run the pro win model on college ratings (see pro_projection.py)
  3. Cross-reference character data — college scouting has deeper intangible visibility
  4. Check development trajectory — Work Ethic directly predicts growth rate
  5. Weight pro-predictive skills higher — PnR Execution, Inside Shooting, Def Rebounding

SIUE Pipeline — Draft-Eligible Prospects

This section is auto-updated by the nightly pipeline builder. Last updated: (awaiting first pipeline run)

Active SIUE Roster Players (Draft Eligible)

(To be populated by pipeline_builder.py with seniors and declaring juniors)

SIUE Recruiting Pipeline — Pro Potential Assessment

The SIUE 2050 recruiting class includes prospects with notable pro-relevant skills:

Prospect Pos College Rating Pro Win Model Character Key Pro Skills
(populated by pipeline builder)

College Scouting Advantages for Pro Teams

Pro teams often miss these college data points:

  1. Intangibles visibility — College scouting levels (L1-L5) reveal character attributes that pro scouts may not have
  2. Development projection — Work Ethic directly predicts growth; a WE 20 player at CA 2.5 will likely reach PA faster than WE 12
  3. Scheme fit data — College proficiency ratings (Morey Ball, Pace & Space, etc.) indicate system adaptability
  4. Competition context — College pipeline data shows who a prospect competed against and their scouting accuracy level