2025-2026 NCAAWD2 Model Performance
Season
2026-2027 has not had enough games yet: results start once three weeks of games are in (the first holdout week is predicted from 21 days of earlier games). Showing 2025-2026, the newest season with results.
Every ranked model predicted the same 3312 games (2025-11-23 to 2026-03-28, 18 weekly windows) out of sample: each week's games were predicted by models fit only on the season's earlier games. 51 of 3363 games are left out because a model had no rating for a team yet.
Model Performance Leaderboard
Ranked by full-season AUC. The final week is the season's last 7 played days (10 games, 2026-03-22 to 2026-03-28), predicted from every earlier game: a small sample that shows the latest form.
| # | Model | AUC | Acc | Brier | LogLoss | AUC final wk | Acc final wk |
|---|---|---|---|---|---|---|---|
| 1 | Margin Recency Margin Recency Margin regression with exponential recency weights on newer games. More → | 0.857 | 76.7% | 0.156 | 0.477 | 0.708 | 70.0% |
| 2 | Points Off/Def Points Off/Def Raw points regression with separate offensive and defensive team parameters. More → | 0.857 | 77.3% | 0.159 | 0.485 | 0.875 | 70.0% |
| 3 | Margin Margin Linear team-strength model fit on point differential instead of binary wins. More → | 0.857 | 76.8% | 0.156 | 0.477 | 0.833 | 60.0% |
| 4 | Points Off/Def Recency Points Off/Def Recency Off/def points regression with exponential recency weights. More → | 0.855 | 77.0% | 0.161 | 0.491 | 0.833 | 80.0% |
| 5 | Core Ensemble Core Ensemble Equal-logit blend of Elo, recency BT, recency margin, log-adjusted pyth, and points off/def. More → | 0.851 | 76.4% | 0.158 | 0.480 | 0.875 | 80.0% |
| 6 | Recency Ensemble Recency Ensemble Equal-logit blend of Elo, recency BT, recency margin, log-adjusted pyth, and recency points off/def. More → | 0.851 | 76.4% | 0.158 | 0.480 | 0.833 | 80.0% |
| 7 | Adjusted Context Blend Adjusted Context Blend Experimental context-heavy win model blending strong team components with rest and venue context. More → | 0.844 | 76.1% | 0.166 | 0.507 | 0.875 | 60.0% |
| 8 | Adjusted Efficiency Adjusted Efficiency Opponent-adjusted efficiency model with separate offensive and defensive components. More → | 0.844 | 75.5% | 0.169 | 0.527 | 0.917 | 80.0% |
| 9 | Log Adjusted Log Adjusted Log-scale adjusted efficiency model that downweights blowout leverage. More → | 0.844 | 75.4% | 0.170 | 0.529 | 0.917 | 80.0% |
| 10 | Dynamic Bradley-Terry Dynamic Bradley-Terry Time-evolving paired-comparison model with latent team strength drift. More → | 0.818 | 73.9% | 0.175 | 0.521 | 0.875 | 80.0% |
| 11 | Avg Margin Baseline Avg Margin Baseline Predict from simple average scoring margin in the training window. More → | 0.817 | 73.4% | 0.174 | 0.523 | 0.833 | 80.0% |
| 12 | Bradley-Terry Bradley-Terry Static logistic paired-comparison model with one team strength parameter. More → | 0.813 | 73.3% | 0.177 | 0.528 | 0.833 | 70.0% |
| 13 | Bradley-Terry Recency Bradley-Terry Recency Static Bradley-Terry with exponential recency weights on newer games. More → | 0.812 | 73.3% | 0.181 | 0.541 | 0.792 | 80.0% |
| 14 | Pythagorean Pythagorean Pythagorean win expectation from raw points scored and allowed. More → | 0.808 | 73.3% | 0.181 | 0.538 | 0.750 | 70.0% |
| 15 | Elo Elo Streaming paired-comparison rating with recency baked into sequential updates. More → | 0.788 | 70.2% | 0.188 | 0.551 | 0.833 | 70.0% |
| 16 | Home Team Baseline Home Team Baseline Always favor the home team with a fixed prior. More → | 0.579 | 57.9% | 0.244 | 0.682 | 0.667 | 70.0% |
| - |
Efficiency
Efficiency
Tempo-adjusted efficiency version of Pythagorean ratings.
More →
not compared: predicted 0 of 3363 games
|
- | - | - | - | - | - |
Methodology
Out-of-sample windows
- The season is split into 7-day windows ending on its last played day
- Each window is predicted by models refit on all earlier games of the season
- The first window starts after 21 days of games
- Every model is scored on the same games; a model that cannot predict most of them is listed as not compared
Elo / Bradley-Terry
- Elo: the published league Elo (preseason priors and carryover where configured), frozen at the window start
- Bradley-Terry: static logistic fit on the earlier games
- Recency variants weight games with a 30-day half-life
Pythagorean and margin models
- Raw, adjusted and log-adjusted Pythagorean from scores; efficiency from box-score possessions
- Margin: ridge regression on point margin with a learned home advantage
- Ensembles average the log-odds of five families
Baselines
- Home Team: the home team wins 60% (50% at neutral sites)
- Avg Margin: the higher average margin so far wins
- Models should beat these to add value