Officials
Accuracy doesn't differ beyond chance. Call volume does.
This site does not name or rank officials. The tests below found no official whose accuracy stands out. What does vary is how many of the crew's late fouls each official calls, and that is shaped by court position and crew role, which the reports don't record.
Each official's share of the crew's late foul calls
Shares run from 23.5% to 41.5%; 15 of 82 differ from an even split after correction. Listed crew position doesn't explain it, but on-court positions (lead, slot, trail) rotate during a game and aren't in the data, so volume partly reflects where an official stood.
Tests
| Question | Officials | p | Differ after correction |
|---|---|---|---|
| Incorrect-call rate differs by official (adj. call type) | 66 | 0.33 | 0 |
| Share of fouls called on the home team differs by official | 57 | 0.086 | 0 |
| Share of the crew's late foul calls differs by official | 82 | 4.1e-26 | 15 |
| Error rate in games worked differs by official (adj. season) | 86 | 0.025 | 0 |
Foul calls matched to the calling official with high confidence (clock and player), 2016-17 on. Officials with too few calls for a test are left out of it. The error-rate-in-games-worked test is judged against random crew assignment: the observed statistic is within the range random crews produce.