🐻⬇️🏀

Player model

Adjusted Plus/Minus

Unregularized adjusted plus-minus per 70 possessions.

Status implemented Short Name APM
Rating Style impact Where Shown player pages, players leaderboard

Model Card

Unregularized weighted lineup regression can be unstable, especially for players who share lineups. The fit uses a minimum-norm solution, then centers on published players by minutes. Segment outcomes are per 100 possessions, weighted by the square root of possessions, and scaled by 0.70 to per 70 possessions (about 40 college minutes, not a per-minute rate).

Validation here is out of sample. From two weeks into each season, every week's lineups are predicted by a fit on the season's earlier play, with the served settings; each player's point is the predicted and actual margin per 100 possessions of the held-out play the player was on the floor for.

Selected League / Season

Change league with the League selector at the top of the page.

Held-out player predictions

Plot Adjusted Plus/Minus against any counting stat (one point per player; choose any season with a served list).

Not enough held-out play to show a validation plot. NCAAM 2026-2027: 0 eligible players. A player needs at least 200 held-out possessions. Held-out windows: 1. Held-out possessions: 2238.1. Computed 2026-11-17 13:18:03.777861+00:00.

NCAAM 2025-2026: 3791 players with at least 200 held-out possessions. Across players, predicted and actual correlate at r = 0.67 (weighted by held-out possessions). Across these player averages, weighted R² = -0.16 (relative to the weighted mean of actual held-out outcomes). Per held-out segment, squared error is 20.0% above predicting the league average. Each dot is a player (500 with the most held-out possessions are plotted; larger dots had more); on the dashed line, actual equals predicted. Select a dot for the player.

Select a dot to see the player.

Held-out miss against playing time

Each dot is a player's actual minus predicted margin per 100 possessions with the player on the floor against their held-out possessions (log scale). Orange points are the possessions-weighted mean miss of eight equal groups of players by playing time: a model that shrinks its low-possessions players too far, or not far enough, shows a slope.

Above prediction

PlayerHeld-out possessionsPredictedActual
Chase Ross2871 -10.9-2.0
Amare Bynum2935 -3.7+4.9
Jaron Pierre Jr.2976 -6.2+2.0
Larry Hughes II3022 -6.4+1.3
Josiah Davis3307 -8.3-0.7
Jayme Mitchell Jr.3207 -14.4-7.4
Derek Simpson3076 -4.8+1.8
Brant Byers3165 -0.4+5.9
Joshua O'Garro3189 -5.1+1.0
Josiah Lake II2874 -5.6-0.0

Below prediction

PlayerHeld-out possessionsPredictedActual
Dug McDaniel3061 +14.6-2.1
Baraka Okojie2892 +13.0+2.5
Lazar Djokovic2919 +15.9+6.6
Micah Simpson3010 +7.6-1.6
Marquel Sutton2881 +6.8-1.8
Maleek Arington2916 +8.8+0.2
Tylik Weeks2921 +9.7+1.3
Elyjah Freeman2978 +9.8+1.8
Jakobi Heady3086 +9.3+1.4
Cory Wells3392 +7.2-0.1

Output Coverage

Rows in each model's current published run (or legacy rows where no run is selected).

League Season Rows
WNBA 2026 238
NCAAW 2026-2027 4403
NCAAW 2025-2026 4656
NCAAW 2024-2025 4718
NCAAMD2 2025-2026 3953
NCAAM 2026-2027 4462
NCAAM 2025-2026 4910
NCAAM 2024-2025 5088
NCAAM 2015-2016 4619
NCAAM 2014-2015 4677
NCAAM 2013-2014 4682
NCAAM 2012-2013 4534
NCAAM 2011-2012 4516
NCAAM 2010-2011 4478
NCAAM 2009-2010 4627
NBA 2026-2027 470
NBA 2025-2026 579