How the NBA model is tested
The method itself isn't published here. What is published: every pick is locked before tipoff and graded after, the model is evaluated on games it never saw, and a change only ships if it holds up there. Winners are the record — there is no spread or total for NBA, only a probability the favored side wins. See the WNBA modelfor the other league's numbers.
Backtest — held out 2020–25
Walk-forward: every game priced from ratings that have seen only the games before it. The model was fixed before this window was ever scored.
Calibration
The tiers below are the live record — thin so far, since basketball picks only started grading this season — next to the backtest's own headline as a reference. As more games grade, this section becomes the more meaningful check.
No graded picks yet — the live calibration tiers fill in as NBA games grade. Backtest overall: 65.0% on 7,233 held-out games.
What we don't publish
The method, its parameters, and what was tried and rejected are not published — that's the part of the work we keep to ourselves. What's always public is the result: every pick locked pre-game, graded after, calibration shown honestly, and nothing about the record edited after the fact.
No player-prop lines are ingested yet for NBA (TheRundown has no NBA events posted this early in the build).
Informational only. 21+, 1-800-GAMBLER.