2008 WNBA Model Performance
Season
Every ranked model predicted the same 208 games (2008-06-09 to 2008-10-05, 17 weekly windows) out of sample: each week's games were predicted by models fit only on the season's earlier games.
Model Performance Leaderboard
Ranked by full-season AUC. The final week is the season's last 7 played days (4 games, 2008-09-29 to 2008-10-05), 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 Margin Linear team-strength model fit on point differential instead of binary wins. More โ | 0.711 | 67.3% | 0.222 | 0.637 | 1.000 | 50.0% |
| 2 | Margin Recency Margin Recency Margin regression with exponential recency weights on newer games. More โ | 0.708 | 66.8% | 0.223 | 0.638 | 1.000 | 50.0% |
| 3 | Points Off/Def Points Off/Def Raw points regression with separate offensive and defensive team parameters. More โ | 0.705 | 67.8% | 0.222 | 0.636 | 1.000 | 50.0% |
| 4 | Points Off/Def Recency Points Off/Def Recency Off/def points regression with exponential recency weights. More โ | 0.704 | 68.8% | 0.223 | 0.637 | 1.000 | 50.0% |
| 5 | Adjusted Context Blend Adjusted Context Blend Chronologically trained component blend with rest and signed venue context. More โ | 0.687 | 67.3% | 0.230 | 0.664 | 0.333 | 50.0% |
| 6 | Elo Elo Streaming paired-comparison rating with recency baked into sequential updates. More โ | 0.679 | 63.0% | 0.230 | 0.651 | 0.333 | 25.0% |
| 7 | Core Ensemble Core Ensemble Equal-logit blend of Elo, recency BT, recency margin, log-adjusted pyth, and points off/def. More โ | 0.644 | 61.1% | 0.233 | 0.657 | 0.333 | 25.0% |
| 8 | Recency Ensemble Recency Ensemble Equal-logit blend of Elo, recency BT, recency margin, log-adjusted pyth, and recency points off/def. More โ | 0.643 | 62.0% | 0.233 | 0.657 | 0.333 | 25.0% |
| 9 | Home Team Baseline Home Team Baseline Always favor the home team with a fixed prior. More โ | 0.630 | 63.0% | 0.234 | 0.661 | 0.667 | 50.0% |
| 10 | Adjusted Efficiency Adjusted Efficiency Opponent-adjusted efficiency model with separate offensive and defensive components. More โ | 0.618 | 58.2% | 0.239 | 0.670 | 1.000 | 100.0% |
| 11 | Log Adjusted Log Adjusted Log-scale adjusted efficiency model that downweights blowout leverage. More โ | 0.617 | 58.2% | 0.239 | 0.670 | 1.000 | 100.0% |
| 12 | Bradley-Terry Bradley-Terry Static logistic paired-comparison model with one team strength parameter. More โ | 0.615 | 58.2% | 0.241 | 0.673 | 0.000 | 25.0% |
| 13 | Dynamic Bradley-Terry Dynamic Bradley-Terry Time-evolving paired-comparison model with latent team strength drift. More โ | 0.611 | 57.7% | 0.253 | 0.705 | 0.000 | 25.0% |
| 14 | Pythagorean Pythagorean Pythagorean win expectation from raw points scored and allowed. More โ | 0.600 | 57.7% | 0.237 | 0.666 | 1.000 | 100.0% |
| 15 | Avg Margin Baseline Avg Margin Baseline Predict from simple average scoring margin in the training window. More โ | 0.596 | 55.3% | 0.242 | 0.676 | 1.000 | 100.0% |
| 16 | Bradley-Terry Recency Bradley-Terry Recency Static Bradley-Terry with exponential recency weights on newer games. More โ | 0.592 | 55.3% | 0.243 | 0.678 | 0.000 | 25.0% |
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 120-day half-life in NCAA leagues (keeping the non-conference games that link conferences) and a 30-day half-life in pro leagues
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