2025-2026 NBA Model Performance
Season
2026-2027
2025-2026
2024-2025
2023-2024
2022-2023
2021-2022
2020-2021
2019-2020
2018-2019
2017-2018
2016-2017
2015-2016
2014-2015
2013-2014
2012-2013
2011-2012
2010-2011
2009-2010
2008-2009
2007-2008
2006-2007
2005-2006
2004-2005
2003-2004
2002-2003
2001-2002
2000-2001
1999-2000
1998-1999
1997-1998
1996-1997
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 1134 games (2025-11-16 to 2026-06-13, 30 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 (3 games, 2026-06-07 to 2026-06-13), 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 Recency Points Off/Def Recency Off/def points regression with exponential recency weights. More → | 0.741 | 68.3% | 0.209 | 0.606 | 0.250 | 66.7% |
| 2 | Margin Recency Margin Recency Margin regression with exponential recency weights on newer games. More → | 0.740 | 68.1% | 0.208 | 0.605 | 0.250 | 66.7% |
| 3 | Points Off/Def Points Off/Def Raw points regression with separate offensive and defensive team parameters. More → | 0.739 | 68.7% | 0.210 | 0.609 | 0.250 | 33.3% |
| 4 | Margin Margin Linear team-strength model fit on point differential instead of binary wins. More → | 0.739 | 68.7% | 0.210 | 0.609 | 0.250 | 33.3% |
| 5 | Core Ensemble Core Ensemble Equal-logit blend of Elo, recency BT, recency margin, log-adjusted pyth, and points off/def. More → | 0.734 | 67.9% | 0.209 | 0.606 | 0.250 | 66.7% |
| 6 | Recency Ensemble Recency Ensemble Equal-logit blend of Elo, recency BT, recency margin, log-adjusted pyth, and recency points off/def. More → | 0.733 | 67.5% | 0.209 | 0.606 | 0.250 | 66.7% |
| 7 | Pythagorean Pythagorean Pythagorean win expectation from raw points scored and allowed. More → | 0.733 | 66.4% | 0.220 | 0.632 | 0.500 | 33.3% |
| 8 | Avg Margin Baseline Avg Margin Baseline Predict from simple average scoring margin in the training window. More → | 0.733 | 66.4% | 0.211 | 0.612 | 0.500 | 33.3% |
| 9 | Adjusted Efficiency Adjusted Efficiency Opponent-adjusted efficiency model with separate offensive and defensive components. More → | 0.731 | 66.8% | 0.210 | 0.610 | 0.500 | 33.3% |
| 10 | Log Adjusted Log Adjusted Log-scale adjusted efficiency model that downweights blowout leverage. More → | 0.731 | 66.8% | 0.210 | 0.610 | 0.500 | 33.3% |
| 11 | Adjusted Context Blend Adjusted Context Blend Experimental context-heavy win model blending strong team components with rest and venue context. More → | 0.729 | 67.9% | 0.212 | 0.616 | 0.250 | 33.3% |
| 12 | Dynamic Bradley-Terry Dynamic Bradley-Terry Time-evolving paired-comparison model with latent team strength drift. More → | 0.721 | 66.9% | 0.214 | 0.618 | 0.000 | 66.7% |
| 13 | Bradley-Terry Recency Bradley-Terry Recency Static Bradley-Terry with exponential recency weights on newer games. More → | 0.720 | 66.5% | 0.214 | 0.616 | 0.500 | 66.7% |
| 14 | Bradley-Terry Bradley-Terry Static logistic paired-comparison model with one team strength parameter. More → | 0.720 | 66.8% | 0.214 | 0.620 | 0.500 | 33.3% |
| 15 | Elo Elo Streaming paired-comparison rating with recency baked into sequential updates. More → | 0.713 | 65.4% | 0.220 | 0.635 | 0.250 | 33.3% |
| 16 | Home Team Baseline Home Team Baseline Always favor the home team with a fixed prior. More → | 0.550 | 55.0% | 0.250 | 0.693 | 0.250 | 33.3% |
| - |
Efficiency
Efficiency
Tempo-adjusted efficiency version of Pythagorean ratings.
More →
not compared: predicted 0 of 1134 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