Robo-Umps and Strike Zone challenges: the future of Major League Baseball

Of all the rule changes MLB has implemented these past several seasons, the most important is one they’ve shied away from. But that may be changing.

Yahoo Sports’ Jack Baer reported last week that MLB will test a robo-ump ball-strike challenge system during 2025 spring training games. If the test goes well, such a challenge system could be implemented for the regular season as early as 2026.

Through all the tinkering with rules that has preoccupied baseball since 2020—ghost runners, banning shifts, larger bases, limits on throwing to first, pitch clocks, three-batter minimums—the plate umpire’s hegemony over ball-strike decisions has been the one ‘no-go’ area. That’s been true despite technological advancements which have made it clear how frequently and how profoundly missed ball-strike counts can influence game outcomes.

The need for change has only been underscored by the profusion of electronic strike zone boxes that make it instantaneously clear when an ump blows a call.

I was one of the first people to reduce this problem to numbers. In 2004, Thomas Dunne Books published my work, “The Book On The Book,” including a chapter titled “The Most Under-appreciated Words in Baseball.”

In that chapter, I published calculations illustrating the profound impact on the outcome of an at-bat simply by whether an umpire got a ball-strike call correct. Based on thousands of real-life samples and listed in descending order of batter performance, this was the MLB batting average I found for each possible ball-strike count:

That research clearly demonstrates that the key point of demarcation is strike two. Literally every pitch count with fewer than two strikes results in an outcome that is significantly better (from a hitter’s standpoint) than the MLB average, which, at the time, was .267. Literally every result in a two-strike count is significantly worse… and the differences can amount to 150 points in batting average.

Since 2004, there have been several similar studies, and although the data has changed at the  margins, they’ve all come to pretty much the same conclusion. A few years ago, a research group called Spiders Elite conducted a similar study, except theirs focused on OPS, which by then had superseded batting average as the sine qua non of offensive stats.

Spiders Elite found that if on a 1-1 count the umpire were to call the pitch a ball, the eventual outcome of the at-bat resulted in a .927 OPS. If, however, the umpire called the 1-1 pitch a strike, the eventual OPS outcome of the at-bat fell to .428. That is a profound difference, indeed, and it illustrates the impact of an umpire's calls on the game.

Of course, there are dissenters; more "old-fashioned" fans who oppose the implementation of robots in baseball. Those resistant to removing dictatorial control of the strike zone from the plate umpire generally cite three arguments:

1. The first is that the robot umpires have not been proven to be sufficiently precise. But, if that is true, why does MLB furnish literally even telecasting outlet with the electronically-tracked results of each pitch? Does it not stand behind its own data?

2. The second criticism is that using robot umps would destroy the ‘skill’ of pitch framing, as if the MLB has any legitimate interest in protecting cheating catchers.

3. The third criticism is that MLB umpires already do a good enough job calling balls and strikes that the technology is not needed. In support of this argument, they cite data establishing that umpires already get, on average, about 93 percent of ball and strike calls right.

But that argument is fallacious given the high number of ‘easy’ ball-strike calls. A pitch in the dirt, one that brushes a batter back, or one that misses the strike zone a foot outside all are obvious balls. A 3-0 or 3-1 pitch fired right down the middle and taken are obvious strikes.

What umpires are really paid to do is get the marginal calls correct. And if you subtract those ‘easy’ calls from the supposed 93 percent ‘correct’ figure touted by the MLB, it turns out that major league umps miss about one-fourth of the missable ball-strike calls.

Seen in that context, praising umpires for getting 93 percent of calls right is a lot like praising a student for scoring 93 on a math test in which three-quarters of the questions are ‘"how much is two plus two?”

Batters and pitchers both know all of this, which is why nothing infuriates a hitter more to be thrown into a two-strike count by an umpire’s erroneous call of ‘strike’ on a pitch just off the zone. As an aside, it’s also why a lot of hitters ‘fish’ for pitches off the plate: they don’t trust the umpire to get the call right.

Pitchers have an equal complaint when their best one-strike pitch clips the corner or the knees only to have an umpire mistakenly call it a ball, and in the process swing the count leverage decidedly toward the hitter.

Personally, I’d be all-in for letting the robot umps take over ball-strike calls entirely, but emotionally, MLB isn’t ready to go that far yet. It is at least willing to try out a challenge system, which would give each team two appeals of bad calls.

That may not be a perfect solution, but it’s a darn sight better than the status quo.

More From Call To The Pen: