🐻⬇️🏀

2025-2026 NCAAMD2 Model Performance Analysis

Comparing prediction accuracy across 3525 games using multiple rating models.

Model Performance Leaderboard

Models ranked by AUC on 2025-2026 games. Hover over column headers for explanations.

# Model AUC Acc Brier LogLoss n AUC 7d Acc 7d Brier 7d n 7d
1 Avg Margin (Baseline) 0.874 77.9% 0.150 0.458 262 0.780 69.5% 0.191 285
2 Pure ELO 0.844 79.0% 0.179 0.540 262 0.760 70.2% 0.197 285
3 Margin Regression 0.802 71.8% 0.184 0.544 262 0.813 73.0% 0.179 285
4 Pythagorean Adjusted 0.799 71.4% 0.184 0.541 262 0.779 73.0% 0.196 285
5 Bradley-Terry 0.796 70.2% 0.185 0.547 262 0.777 69.8% 0.192 285
6 Points O/D 0.796 72.9% 0.191 0.562 262 0.816 71.6% 0.178 285
7 Pythagorean Log 0.794 72.5% 0.187 0.548 262 0.779 73.3% 0.196 285
8 Pythagorean Raw 0.787 70.6% 0.195 0.573 262 0.772 72.3% 0.199 285
9 Pythagorean Efficiency 0.638 57.6% 0.331 1.499 224 - - - 0
10 Home Team (Baseline) 0.592 59.2% 0.242 0.676 262 0.635 63.2% 0.234 285
11 Bradley-Terry Recency - - - - 0 0.760 68.4% 0.198 285
12 Margin Recency - - - - 0 0.810 72.6% 0.180 285
13 Points O/D Recency - - - - 0 0.820 72.6% 0.178 285
14 Core Ensemble - - - - 0 0.798 72.6% 0.183 285
15 Recency Ensemble - - - - 0 0.798 73.0% 0.183 285
16 Adjusted Context Blend - - - - 0 0.810 72.3% 0.180 285
17 Dynamic Bradley-Terry - - - - 0 0.778 71.9% 0.191 285

Methodology

ELO / Bradley-Terry

  • ELO: Iterative updates, K=64, HCA=100
  • BT: Static logistic regression on all games
  • Both model win probability, not margin
  • ELO updates after each game; BT fits once

Pythagorean Models

  • Raw: Bill James formula (pts scored/allowed)
  • Efficiency: Pace-adjusted (pts per possession)
  • Adjusted: KenPom-style opponent adjustment
  • Log: Log-linear multiplicative scale

Margin Regression

  • Team-level ridge regression on point margin
  • Linear Bradley-Terry (margin target)
  • Alpha=0.05 (CV-tuned)
  • Learns home advantage from data (~6 pts)

Baselines

  • Home Team: Always predict home wins (60%)
  • Avg Margin: Higher average margin wins
  • Models should beat these to add value