Opponent-adjusted efficiency
Per-possession production corrected for the quality of the defences and offences a team has faced.
Raw per-game numbers mix team quality with schedule quality and pace. Efficiency metrics first move to a per-possession basis so fast and slow teams can be compared, then adjust each result for the strength of the opponent that allowed it.
The adjustment is usually solved iteratively across the whole league: every team's rating depends on its opponents' ratings, which in turn depend on theirs, until the system settles.
Opponent-adjusted efficiency is one of the most predictive single inputs available in basketball, football and soccer models, and it is a core feature in AI4Gameday's ensembles.
Related terms
See it applied
Every published call on AI4Gameday carries a stated confidence and is graded in public once the game is final.