🐻⬇️🏀

2013 WNBA Model Performance

Season
2026 2025 2024 2023 2022 2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 1998 1997

Every ranked model predicted the same 194 games (2013-06-14 to 2013-10-10, 17 weekly windows) out of sample: each week's games were predicted by models fit only on the season's earlier games.

Model Catalog · Player models

Model Performance Leaderboard

Ranked by full-season AUC. The final week is the season's last 7 played days (3 games, 2013-10-04 to 2013-10-10), 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 Points Off/Def Points Off/Def Raw points regression with separate offensive and defensive team parameters. More → 0.700 64.4% 0.221 0.632 - 100.0%
2 Margin Margin Linear team-strength model fit on point differential instead of binary wins. More → 0.694 63.9% 0.222 0.636 - 100.0%
3 Adjusted Context Blend Adjusted Context Blend Chronologically trained component blend with rest and signed venue context. More → 0.691 61.9% 0.223 0.636 - 100.0%
4 Points Off/Def Recency Points Off/Def Recency Off/def points regression with exponential recency weights. More → 0.685 63.9% 0.224 0.639 - 100.0%
5 Margin Recency Margin Recency Margin regression with exponential recency weights on newer games. More → 0.681 61.3% 0.226 0.643 - 100.0%
6 Core Ensemble Core Ensemble Equal-logit blend of Elo, recency BT, recency margin, log-adjusted pyth, and points off/def. More → 0.677 61.3% 0.225 0.640 - 100.0%
7 Recency Ensemble Recency Ensemble Equal-logit blend of Elo, recency BT, recency margin, log-adjusted pyth, and recency points off/def. More → 0.672 60.8% 0.226 0.642 - 100.0%
8 Elo Elo Streaming paired-comparison rating with recency baked into sequential updates. More → 0.665 62.4% 0.232 0.659 - 100.0%
9 Log Adjusted Log Adjusted Log-scale adjusted efficiency model that downweights blowout leverage. More → 0.662 59.8% 0.232 0.655 - 100.0%
10 Avg Margin Baseline Avg Margin Baseline Predict from simple average scoring margin in the training window. More → 0.662 61.3% 0.230 0.650 - 100.0%
11 Adjusted Efficiency Adjusted Efficiency Opponent-adjusted efficiency model with separate offensive and defensive components. More → 0.662 60.3% 0.232 0.655 - 100.0%
12 Bradley-Terry Bradley-Terry Static logistic paired-comparison model with one team strength parameter. More → 0.658 59.3% 0.230 0.652 - 100.0%
13 Pythagorean Pythagorean Pythagorean win expectation from raw points scored and allowed. More → 0.653 60.3% 0.231 0.654 - 100.0%
14 Dynamic Bradley-Terry Dynamic Bradley-Terry Time-evolving paired-comparison model with latent team strength drift. More → 0.641 61.9% 0.240 0.673 - 100.0%
15 Bradley-Terry Recency Bradley-Terry Recency Static Bradley-Terry with exponential recency weights on newer games. More → 0.640 57.2% 0.233 0.657 - 100.0%
16 Home Team Baseline Home Team Baseline Always favor the home team with a fixed prior. More → 0.584 58.2% 0.244 0.680 - 66.7%

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