Methodology

Z-scores use the top 156 players by total minutes as the reference group. This group sets the average and standard deviation for each category. It does not limit the player list.

Every player with at least one game gets a z-score and a score against that reference group. Players outside the group are included and can have negative z-scores.

The reference size is configurable. A 10-team league is 130; a 14-team league is 182.

A z-score is the number of standard deviations a player is from the reference group average: zero is average, +1 is one standard deviation above, and negative is below. Adding z-scores lets points, rebounds, steals, and turnovers share one scale.

FG% and FT% are not z-scored as raw percentages. We use made-above-expected impact, which rewards volume:

FG impact = (FGM - FGA × pool FG%) / GP
FT impact = (FTM - FTA × pool FT%) / GP

A player who is 1-for-1 from the line does not outrank a high-volume shooter. Turnovers are inverted, so a lower turnover rate is better.

My Score is the sum of the z-scores for your selected categories, excluding any categories you punt. 9-Cat is the sum across all nine standard categories. Both use the same reference group.

Ranks use all players who played at least one game. Search and sort do not change a rank.

Per-game mode scores rate stats and is the default for “who is better right now.” Totals mode scores season totals and rewards availability. Percentage categories still use impact, not raw percentages, in both modes.

Data comes from ESPN’s public scoreboard and box-score endpoints. The nightly pipeline ingests completed regular-season games, reconciles player totals, rebuilds the season aggregates, and publishes the marts atomically. The board shows the last successful publish time.

9Cat is an independent project. It is not affiliated with the NBA or ESPN. Team and league marks belong to their owners.