Umpires arguably hold the most power of any single person on the field during a baseball game. Any pitch that doesn’t draw a swing is determined by the home plate ump alone, the same way they did it a century ago. They stand there behind the catcher and make decisions based on what they see in front of them. Today’s umps don’t use supportive technology to back up their calls, and pitch calls aren’t eligible for video review. It’s their call above all else.
Meanwhile, Major League Baseball has long been secretive about their umpiring corps. The schedules for the 91 active umpires are kept a secret until mere minutes before game time to reduce the chance of outside influence. Thanks to the strength of the MLB Umpires Association, umps with poor performance outcomes remain active in the league until they choose to retire. MLB and its data partners also don’t release statistics on the accuracy of umpires’ calls, nor on their trends. Any aggregated data on individual umpires has to be collected independently.
What data has reached the public is not favourable to toward umps. In fact, the stats indicate that the longer umpires are tenured, the worse they perform. Umpire performance behind the plate has long been a source of anecdotal ire, but with statistics uncovering thousands of examples of botched calls it’s not surprising to see their reputations souring even further among players, managers, and fans.
But how bad is it really? What kind of technology is possible? And, crucially, will umpires accept changes that would diminish their role in the sport?
Boston University lecturer Mark T. Williams and his research team analysed over 4 million pitches thrown between 2008 and 2018. What they found makes a strong argument for helping plate umps evaluate pitch location, or even implementing a fully automated system for calling balls and strikes.
In 2018, MLB umpires made incorrect ball and strike calls an average of 14 times per game. That’s 1.6 incorrect calls per inning. When the count got to two strikes, umpires got even worse with 29% of calls being incorrect – almost double the rate of error than that in counts with fewer than two strikes on the batter.
To see the numbers from 2018 to 2020 I ran some data using MLB’s BaseballSavant illustrator tool. I looked at the players with the most plate appearances in each league since 2018. From the AL it was Francisco Lindor (‘18) and Whit Merrifield (’19 and ‘20). The NL leaders were Trea Turner (’18), Ozzie Albies (’19), and Dansby Swanson (’20). I took those players and narrowed their stats down to pitches that were called strikes. The results were…striking.
Albies’ numbers are the most concerning of the group. Over the last three seasons, Atlanta’s young star took significantly fewer called strikes than the other four players, but that has only magnified the impact of the ones he did take. Of 525 called strikes, 137 of those were outside the zone and should have been called balls – good for 26% of his total called strikes.
Lindor, Merrifield, Swanson, and Turner were all subject to incorrect strike calls 16% to 18% of the time. Washington’s Turner had the worst time in terms of volume. Umpires around the league tagged him with 200 extra strike calls over that same time span, bringing his incorrect strike percentage up to 18%.
When the numbers are displayed so clearly, it begs the question whether player performance is being impacted by the rate of error behind the plate. The thinking behind implementing replay review for calls in the field came from a similar place; the role of human error can and should be mitigated by technology.
The argument in favour of plate umps having technological backup isn’t necessarily about working them out of a job. Aligning their experience with more precise calls can ensure players can perform at their best – consider the amount of time major league hitters spend on building pitch recognition and strike zone discipline. There’s an argument to be made that their skill level should be met by the skill level of the umpiring crew.
There isn’t much standing in the way of accessing instantaneous pitch data. The systems are readily available and, in some cases, already in use in some leagues. The only thing standing in the way is likely the willingness of the umpires’ union to agree. But let’s look at what kind of technology is out there that could improve called strike accuracy.
Major League Baseball doesn’t just have access to the technology that would make it possible to track balls and strikes; they own it. Statcast, a subsidiary of MLB, uses dozens of cameras installed in every major league ballpark to triangulate stats like launch angle and sprint speed. It’s also the basis for on-screen strike zone graphics during broadcasts. Statcast is the primary technology for statistical data gathering during games and is understood to be accurate for pitch tracking within an inch.
Given its existing use by the league, Statcast seems to be the robo-ump heir apparent. It wouldn’t be surprising to see the league use their own tech along with existing infrastructure if (or when) Commissioner Manfred et al. decide to pursue electronic strike zone systems.
There is, however, another version of pitch tracking technology already in play. The independent Atlantic League has been using a system called TrackMan since 2019. The trial was actually instigated by Major League Baseball to get a clearer picture of how it would look to implement a robo-ump at the pro level.
TrackMan is similar to Statcast in that it determines pitch information in real time. Instead of using cameras, though, it uses Doppler radar to measure 3D objects in space. It takes into account the stature of the batter among other information, which ensures the size of the strike zone is relative to the height of the batter. It works simply: the ump wears an earpiece connected to the main system that pipes in “strike” or “ball” as the pitch comes through the zone. The ump physically makes the call as they usually would.
High-profile proponents of strike zone technology include retired super-utility man Ben Zobrist and Miami Marlins manager Don Mattingly who have both expressed support for bringing it to the majors. For what it’s worth, their respective comments on the topic were the result of being negatively impacted by game-ending umpiring mistakes. And now, a few years into the experimental phase, reaction to the TrackMan experiment has been in line with their original opinion. It’s been largely positive from umpires and players alike.
In 2019 CBS Sports’ Katherine Acquavella reported on how it was going a few months into TrackMan’s inaugural season in the Atlantic League. Traditionalists, she says, are having a tough time balancing the so-called “human element” of umpire-player relationships. Reports out of the Arizona Fall League have a similar bent:
“[Mets catching prospect Ali] Sanchez has learned to vary his conversations, chatting with the home plate umpire about everything from the weather to family to their upcoming vacations over the winter holidays. No more negotiating about the strike zone. Nothing can be done.”
One can argue that relationships shouldn’t have any effect on the size or shape of the strike zone at all. A computer’s impartiality can make the game more predictable on both sides of the ball, which is what some players are coming around to as well.
“It gives me confidence knowing that every single pitch I throw is going to be called exactly what it is,” [Giants pitching prospect Tristan] Beck said. “I throw a close pitch and it’s a ball? No argument. No leaning on the umpire that it should have been a strike. I know for a fact that it’s a ball or a strike.”
The system tends to be well received when it comes to pitches on the inside and outside corners, but there’s some chatter that pitches at the top and bottom of the zone tend to be less consistent. Pitches that break at or under the bottom of the zone have a tendency to trip up TrackMan, which mistakenly calls some of them strikes even if they skip into the dirt.
That’s where the human element does play a role. The umpire still has the authority to call a pitch one way or the other if the system is clearly misreading the outcome. Atlantic League umpire Freddie DeJesus told Acquavella he thinks the future is bright for umpires who embrace the new tech.
“As I’ve had the opportunity to do it now, it’s great,” DeJesus said. “It’s a great opportunity and it’s good for the game. I can see it down the line getting to the next level. It’s just an opportunity for bigger things to happen within baseball.”
Following the 2019 season, the MLBUA’s labour agreement was up for negotiation with the league. The new five-year labour agreement includes agreement with the development and testing of electronic strike zone technology. The umps have said they’ll use the systems if Manfred chooses to bring them into MLB games.
This from a union who, in 1999, used mass resignations as a bargaining tool against having a taller strike zone. Perhaps umpires are beginning to see what everyone else can – that the game needs consistency to get the best out of its players.
While baseball might be leaning toward optimism, some fans aren’t so sure. One game in the 2019 Arizona Fall League stoked so much fan frustration after a questionable TrackMan strike call that the league began making an announcement before every game reminding attendees that the umpires weren’t responsible for ball-strike calls. The fans’ feelings about that?
“Get rid of it,” one yelled [at the PA announcer’s echo]. “Let the umps do their jobs!”