🐻⬇️🏀

Team model

Points Off/Def Recency

Off/def points regression with exponential recency weights.

Status implemented Short Name PtsR
Rating Style probability Where Shown teams rankings, team pages, model performance

Model Card

Same two-parameter points model as Points Off/Def, but with recency weighting. Useful for testing whether newer offensive and defensive form should move faster than season-long estimates.

Validation on this page uses stored season-level full-fit rows plus strict weekly rolling holdouts. Rolling ranks are based on average weekly log loss in the selected league/season. Current scope: All Games. All scored games in the selected league and season. AP Poll is excluded here.

Selected League / Season

Use the selected scope to inspect this model against peers and across rolling weekly holdouts.

Full Fit AUC - · 0 games
7-Day Holdout Log Loss 0.781 (#8) · 4 games
Rolling Avg Log Loss 0.660 (#1) · 22 windows
Window Wins 1 · 61.1% rolling accuracy

Full-fit AUC is only available where the older stored-rating batch produced that model for the selected scope. Recency and ensemble families can still have valid 7-day and rolling holdout results even when the full-fit columns are blank.

Rolling Holdout Curves

Selected model plus comparison peers for GLEAGUE 2025-2026. The focus line is Points Off/Def Recency. Click legend items to hide/show peers.

Log Loss Brier AUC Accuracy

Weekly strict holdout log loss. Lower is better. Showing 6 models across 22 windows.

Recent Windows

Holdout Log Loss Rank Winner
Apr 8 - Apr 10 0.639 #6/16 Log Adjusted (0.567)
Apr 1 - Apr 7 0.619 #11/16 Log Adjusted (0.563)
Mar 25 - Mar 31 0.669 #2/16 Home Team Baseline (0.650)
Mar 18 - Mar 24 0.685 #7/16 Dynamic Bradley-Terry (0.674)
Mar 11 - Mar 17 0.727 #5/16 Home Team Baseline (0.601)
Mar 4 - Mar 10 0.644 #6/16 Pythagorean (0.624)
Feb 25 - Mar 3 0.619 #8/16 Dynamic Bradley-Terry (0.591)
Feb 18 - Feb 24 0.692 #1 Points Off/Def Recency

Selected Scope Peer Table

Comparison snapshot for GLEAGUE 2025-2026. Rows are ordered by rolling log-loss rank; full-fit AUC may be blank for models that do not have legacy full-season rows stored for this scope.

Model Full AUC AUC Rank 7d LogLoss 7d Rank Rolling LL Rolling Rank Windows
Points Off/Def Recency - - 0.781 #8 0.660 #1 22
Pythagorean 0.532 #6 0.680 #4 0.661 #2 22
Points Off/Def 0.521 #7 0.692 #6 0.664 #3 22
Avg Margin Baseline 0.686 #1 0.655 #3 0.664 #4 22
Recency Ensemble - - 0.811 #12 0.668 #5 22
Core Ensemble - - 0.793 #11 0.668 #6 22
Margin Recency - - 0.787 #9 0.671 #7 22
Margin 0.474 #10 0.692 #5 0.673 #8 22

League × Season Summary

Cross-scope validation history for Points Off/Def Recency. Rolling ranks are by average weekly log loss.

League Season Full AUC AUC Rank 7d LL 7d Rank Rolling LL Rolling Rank Windows Wins
WNBA 2025 - - 0.615 #9 0.633 #7 21 0
NCAAWD2 2025-2026 - - 0.498 #6 0.687 #15 14 1
NCAAW 2025-2026 - - 0.539 #2 0.551 #7 22 0
NCAAMD2 2025-2026 - - 0.533 #1 0.573 #7 14 0
NCAAM 2025-2026 - - 0.645 #9 0.576 #6 22 2
NBA 2025-2026 - - 0.581 #6 0.616 #4 26 0
GLEAGUE 2025-2026 - - 0.781 #8 0.660 #1 22 1

Output Coverage

Stored site-output rows for this model.

League Season Rows
WNBA 2025 17
NCAAWD3 2025-2026 730
NCAAWD2 2025-2026 697
NCAAW 2025-2026 592
NCAAMD3 2025-2026 781
NCAAMD2 2025-2026 603
NCAAM 2025-2026 704
NBA 2025-2026 37
GLEAGUE 2025-2026 31