The ABS Is Catching What Umpires Miss: 12 Days of Strike Zone Data

Two weeks into the 2026 season, baseball has a problem: human umpires are calling too many pitches wrong. Well, that’s what the data says, anyway. After 542 challenges in the ABS system’s first 12 days, we’re looking at a 55.17% overturn rate, which means more than half of all challenged calls are getting reversed. That’s not margin for error. That’s a pattern.

I wrote about the geometry of the strike zone back in spring, breaking down how the new zone fundamentally changed what constitutes a strike. The math was clean: the zone has measurable boundaries, defined by inches and angles. The system tracking it can detect violations to within 0.16 inches median accuracy. But math on paper and math in practice are different animals. So let’s talk about what’s actually happening out there.

The Overall Picture

The numbers are striking. In 12 days, we’ve seen 542 challenges across 20,976 total pitches thrown. That’s a 2.58% challenge rate, which feels right; not every borderline pitch gets contested. Of those 542 challenges, 299 were overturned. That’s roughly 6 per game, and it tells us the system is catching real errors.

For context, compare that to spring training: 53% of 1,844 challenges were successful, 4.32 challenges per game. We’re at 55.17% now, with more challenges overall. The system is settling in, and teams are getting smarter about when to use their challenges. The accuracy specs on the Automated Ball-Strike system itself are tight: 0.39 inches at 95% confidence, 0.48 inches at 99% confidence. That’s the margin the old human umpiring system couldn’t touch.

Who’s Challenging, and Who’s Winning?

Not everyone challenges equally, and that’s where it gets interesting. The data from opening weekend tells a clear story:

Catchers know the zone better than anyone else on the field. They see every pitch from directly behind the batter. Batters are guessing, sometimes successfully, but with less information. Pitchers? They lose more challenges than they win, which probably tells us that pitchers are worse at accepting they missed, not that the system is wrong.

Timing Is Everything

When you challenge matters:

The pattern here isn’t what you’d expect. Umpires don’t simply get worse as the game goes on. They start reasonably sharp (50% in the first, just 35.3% in the second), then lose focus in the middle innings. The 5th and 6th are the worst, with overturn rates hitting 59.4% and 60%. Then something clicks. The 8th inning has the lowest overturn rate of any full inning at 37.8%, meaning umpires are at their most accurate when the game is tightest. The 9th ticks back up to 54.5%, possibly because closers throw nastier stuff and the zone gets harder to read.

This isn’t a fatigue story. It’s a concentration curve. Umpires coast through the middle of the game, then lock in when it matters most. Full count situations tell a similar story: when the count reaches 3-2, only 38.98% of challenges succeed, suggesting umpires bear down on the pitches with the highest consequences.

The Pitch Type Problem

Some pitches get called right more often than others:

Four-seam fastballs are getting overturned at nearly 2 to 1 compared to two-seamers. That’s weird. Both are fastballs, both should be easier to track than breaking balls. My guess: umpires are more confident calling the seam variation correctly because it’s the more "expected" pitch, so they’re overconfident. Or maybe the movement on a two-seamer genuinely confuses even the best umpires. The sample size is small for spring, but this is a signal worth watching as April unfolds.

The Umpire Question

Some men are just better at calling strikes:

Fletcher and Estabrook have been rough. A 91.7% overturn rate is indefensible. Either they don’t understand the new zone, or they’re not paying close enough attention to borderline pitches. Chad Whitson’s perfect 7-for-7 in one game could be noise, but Fletcher and Estabrook have enough volume to suggest a real problem.

On the flip side, Bacchus and Wolcott haven’t had a single challenge succeed. That’s either excellence or luck, but the consistency is notable. CB Bucknor, who has historically been below expected in his zone management since 2020 (253 calls below expectation), is at 75% here. Maybe the system is forcing a reset, or maybe he’s consciously adjusting. Time will tell.

Teams and Strategy

Detroit leads the league at 75% challenge success, while Cleveland sits at the basement at 31.6%. That’s not random. Better teams recognize better pitches. They have more skilled hitters who understand the zone. They have veteran catchers calling better challenges. The Tigers are doing something right, and the Guardians need to think harder about when to use their opportunities.

The Unforgettable Moments

Then there are the individual moments that make you remember why we watch. Eugenio Suarez got back-to-back overturned strikeouts with the bases loaded, the kind of sequence that changes a game. Randy Arozarena started jogging to first while the challenge was being reviewed and won by 0.2 inches, the smallest margin I’ve seen so far. Salvador Perez went 4 for 4 on challenges in one game, which is just dominant. And Dillon Dingler, the Tigers catcher, went 7 for 7 on challenges. That’s not skill. That’s knowledge.

There were also the brutal ones. Tripp Gibson called George Springer out on a pitch 4.3 inches outside the zone. Matt Wallner challenged a pitch 4.8 inches inside the zone, maybe the worst challenge I’ve seen, a swing and a miss turned into a play at the plate that didn’t need to happen.

What It All Means

The data is clear: the strike zone is being called differently now, and more accurately overall. A 55% overturn rate feels high, but remember, these are the pitches people thought were wrong enough to challenge. The system isn’t fixing every call; it’s catching the worst ones, the ones that matter, the ones that move runners.

But the inning-by-inning data reveals something deeper about how umpires actually work behind the plate. They aren’t grinding down from fatigue over nine innings. They’re losing concentration in the middle of games, then snapping back into focus when the stakes rise. A 60% overturn rate in the 6th dropping to 37.8% in the 8th is a 22-point swing. That’s not noise. That’s human psychology showing up in the data: when the pressure is low, attention drifts. When it’s high, it sharpens.

If MLB is paying attention (and Baseball Savant’s ABS dashboard suggests they are), this is actionable. Maybe umpires need a mid-game reset routine. Maybe the 5th-inning stretch should be for the crew, not just the crowd. Or maybe the answer is simpler: the ABS challenge system itself is the correction mechanism. Teams know the middle innings are soft, so they challenge more aggressively there, and the system catches what the human eye misses.

We’re only 12 days in. The season is a long story, and this is just the first chapter. But the numbers are telling us something important: baseball’s oldest tradition is changing, and you can measure that change down to a fifth of an inch.