The Question Nobody Asks
Jorge Soler charged the mound. Reynaldo Lopez threw punches with a baseball in his hand. Walt Weiss tackled a baserunner. Seven-game suspensions came down from the league office.
Everyone debates the suspensions, the provocation, whether the unwritten rules justified it. But nobody asks the question that matters for the next game: does the umpire's zone change after a bench-clearing incident?
The conventional assumption goes one of two ways: either the umpire expands the zone to "keep control" of the next game, or they tighten up to avoid anything borderline. Both stories make intuitive sense. Neither has been tested with data.
We tested it. Seven Statcast-era bench-clearing brawls (2019–2026). Every incident that resulted in suspensions. For each one, we tracked the home plate umpire's zone in the incident game, then in their next game behind the plate, and compared both to their season baseline.
The Zone Doesn't Move
The first finding is the myth-bust. Called-strike rate — the simplest measure of zone size — doesn't significantly change after a brawl.
Each line traces one umpire's called-strike rate from season baseline through the incident game to their next HP assignment. Lines scatter in both directions — no consistent pattern.
The mean shift is -1.3 percentage points from baseline (p=0.302, 95% CI: [-4.1pp, +1.5pp]). Four umpires were slightly below their baseline in the next game; two were above. The confidence interval comfortably includes zero.
If umpires were systematically expanding or shrinking the zone after fights, we'd see a directional cluster. We don't. The zone stays the same size.
But the Zone Gets Cleaner
The second finding is the surprise. While the zone doesn't move, it gets more precise. Every metric related to call quality improves:
Forest plot showing the shift from baseline for each zone metric in post-brawl games. Metrics where the CI excludes zero are starred. Accuracy and out-of-zone strike rate both show significant improvement.
Three findings are statistically significant:
- Accuracy improves by +2.0pp (p=0.001). All 6 of 6 umpires were more accurate than their season average. Unanimous direction.
- Out-of-zone called strikes drop by -2.6pp (p=0.004). All 6 of 6 umpires called fewer strikes on pitches off the plate. This is the specific mechanism.
- Miss rate drops by -2.0pp (p=0.001). Fewer total wrong calls of any kind.
The zone doesn't get bigger or smaller. It gets sharper. Pitches off the plate that might normally get a generous call — the "freebies" — stop getting called strikes.
Where the Calls Change
The edge-distance profile shows exactly where the behavior shifts. Compare the called-strike probability at each distance from the zone edge: season baseline (gray) vs post-brawl next game (red).
Called-strike probability by distance from the rulebook zone edge. The gap between curves is largest in the -4" to 0" range — pitches just off the plate. Inside the zone, the curves converge.
The curves are nearly identical for pitches clearly inside the zone (+2" and beyond). The divergence happens in the -4" to 0" band — pitches that are technically off the plate but close enough that umpires sometimes give them. After a brawl, they stop giving them.
This is consistent with a "focus up" explanation: the umpire is more disciplined at the margins, not fundamentally calling a different zone. The pitch 3 inches inside the zone still gets called a strike. The pitch half an inch off the plate no longer does.
The Full Dataset
Here are all seven incidents with their umpire assignments and key metrics:
| Incident | Umpire | Inc CSR | Next CSR | Baseline | Acc +/- | OOZ +/- |
|---|---|---|---|---|---|---|
| Soler-LopezApr 7, 2026 | Moscoso | 26.2% | pending | 31.6% | — | — |
| Peralta-SiriMay 1, 2024 | MacKay | 36.7% | 34.3% | 33.3% | +1.1pp | -2.6pp |
| Anderson-RamirezAug 5, 2023 | Wegner | 32.4% | 27.5% | 32.8% | +2.7pp | -4.1pp |
| Winker-RendonJun 26, 2022 | Bacon | 27.7% | 32.5% | 33.1% | +1.8pp | -2.1pp |
| Arenado-MetsApr 27, 2022 | Rehak | 34.1% | 34.9% | 33.0% | +1.4pp | -0.7pp |
| LaureanoAug 9, 2020 | Mahrley | 38.3% | 31.4% | 33.0% | +3.0pp | -2.5pp |
| Reds-PiratesJul 30, 2019 | Vanover | 30.5% | 29.6% | 32.6% | +1.7pp | -3.8pp |
The Soler-Lopez incident (April 7, 2026) is the freshest — Edwin Moscoso hasn't returned to home plate yet. His incident game was already unusual: a 26.2% called-strike rate (6th percentile among his own games), with an out-of-zone strike rate at the 3rd percentile. We're tracking his next HP assignment and will update this analysis when it happens.
What This Means
The myth is partly right and mostly wrong. Something does change after a bench-clearing incident — but it's not the zone size. It's the zone quality.
The most likely explanation is simple: umpires focus up. A bench-clearing brawl is a high-profile event. The umpire knows their next game will be scrutinized. They come in sharp, they stop giving borderline calls, and the result is a cleaner, more literal rulebook zone.
This is good news. It means the umpiring system is self-correcting in at least one way: when the stakes rise, precision improves. The question is why this level of focus isn't the default — but that's a question for the ABS era to answer.
Methodology
Incidents: 7 Statcast-era bench-clearing brawls (2019–2026) that resulted in MLB suspensions. We identified each game's home plate umpire, then found their next HP assignment after the incident (typically 4–12 days later due to crew rotation).
Data: Statcast pitch-level data via pybaseball. Called pitches (called_strike + ball) extracted for incident games, next games, and a 60-day season window for each umpire's baseline. 6 of 7 incidents have complete incident/next-game pairs; Soler-Lopez (2026) is pending.
Zone model: Rulebook zone expanded by one baseball radius (1.45"/12 = 0.121 ft) on all sides, using per-batter sz_top and sz_bot from Statcast. This matches the MLB rule that a pitch only needs to clip the zone.
Statistical tests: One-sample t-tests on (next game − umpire baseline) deltas, testing whether the mean shift differs from zero. Bootstrap 95% CIs (5,000 resamples). n=6 complete pairs for all significance tests.
Limitations: Small sample (n=6). No control for opponent, park, or pitcher mix. The OOZ finding is exploratory (multiple metrics tested). Season baseline is a 60-day window, not full career. Analysis uses Statcast geometry, not CalledThird's D1 grading table.
Dual analysis: Two independent research agents analyzed the data separately. Both converged on the same conclusion: no zone size change, significant accuracy improvement. The out-of-zone finding was identified by the Codex analysis and confirmed by the independent analysis. Full code on GitHub.
Cite this analysis
CalledThird. "After a Fight, the Zone Gets Cleaner." CalledThird.com, April 10, 2026. https://calledthird.com/analysis/the-zone-after-a-fight
All CalledThird analysis is original research. If you reference our findings, data, or charts in your work, please link back to the original article. For data inquiries: hello@calledthird.com