Methodology

What is APEX?

APEX (Adjusted Player Experience) is a player impact metric that measures how many points a player contributes per 36 minutes of play. It starts from raw plus-minus — the actual point differential when a player is on the court — then adjusts for game context, recency, and the quality of teammates and opponents to isolate individual contribution.

The Bridge

APEX is built in four layers, each adding an adjustment to the raw signal:

Example: Nikola Jokic — how Raw +/- becomes APEX

  1. Raw Plus/Minus — What actually happened on court. The point differential per 36 minutes during every stint (continuous period with the same 10 players).
  2. +Leverage — Close games are weighted more heavily than blowouts. A +5 in a 2-point game tells us more than a +5 in a 30-point game.
  3. +Recency — Recent performance is weighted more via a per-player statistical test (t-test). Players improving or declining have their recency adjusted accordingly.
  4. +Teammate/Opponent — An iterative solve that separates individual contribution from lineup context. Playing with great teammates inflates raw numbers; this adjustment accounts for that.
  5. = APEX per 36 — The final metric: a player's isolated impact per 36 minutes, adjusted for all of the above.

Data Source

APEX is built on stint-level data from the NBA API — every continuous period with the same 10 players on court. This is the most granular play-by-play-adjacent data available, capturing exactly who was on the floor and what happened.

FAQ

How often is it updated?

Daily. The pipeline runs every morning, collecting new game data and recalculating all ratings.

How does it compare to EPM/LEBRON/RAPTOR?

APEX uses a different methodology. Unlike metrics that incorporate box-score priors, APEX focuses purely on stint-level plus-minus with leverage weighting and per-player recency adjustments. This makes it more reactive to current performance while still accounting for sample size.