Refs don't miss calls at random. The NBA's Last Two Minute report is the league's own accounting of what they got wrong in clutch time. 51,130 events across four seasons of holdout data. I used it to figure out which situations break officiating and which ones don't.
Errors aren't spread evenly. Ordinary contact fouls, the thing everyone argues about, are missed least often. The situations that produce the most errors are the ones with the most moving parts: traveling, out-of-bounds, defensive three-seconds, off-ball monitoring. In the highest-risk contexts they hit roughly 3.5× the baseline miss rate.
The lift holds across all four seasons. It's structural, not a single-year artifact.
| Season | Baseline error rate | Top 10% risk contexts |
|---|---|---|
| 2021-22 | 7.7% | 3.45× baseline |
| 2022-23 | 6.2% | 3.93× baseline |
| 2023-24 | 6.0% | 3.92× baseline |
| 2024-25 | 4.5% | 4.00× baseline |
You can predict which games will have more missed calls before tip-off. Not because refs are biased, but because certain situations are structurally harder to officiate. Multiple moving parts, off-ball action, judgment calls on traveling and defensive positioning. That's where the errors concentrate. The refs-are-inconsistent feeling late in games isn't a vibe. It's measurable, and it's situation-driven.