Boston,,Massachusetts,,June,20,,2010:,A,Look,At,Boston,The

Credit: Gerald Peplow on Shutterstock

What makes a genuine ‘robot’? MLB’s strike zone is at the center of cutting edge technology and philosophy.

In A Nutshell

  • MLB’s “robot umpire” isn’t a robot. It’s a system of cameras, software, and human decision-making.
  • Researchers argue it’s a “fluid technology,” shaped by players, fans, engineers, and umpires.
  • The system does more than call strikes: it influences strategy, entertainment, and trust in the game.
  • The study suggests robots aren’t fixed machines, but evolving systems shaped by people and context.

When the New York Times declared “Robot umps are here” ahead of Major League Baseball’s rollout of its Automated Ball-Strike System, fans might have imagined a gleaming humanoid crouching behind home plate, calling balls and strikes with cold mechanical precision. Instead, what showed up at spring training was a bunch of cameras, a graphic on the scoreboard, and the same human umpire who’s always been there, still making every call. MLB’s Commissioner has been blunt about it: “There’s no robot.” Yet players, reporters, fans, and even academics can’t stop calling it one. One fan interviewed for a new study went so far as to call the system “a robot god.”

That stubborn gap between what people see and what’s actually there caught the attention of researchers at Cornell University and Indiana University Bloomington. In a paper presented at the 2026 International Conference on Human-Robot Interaction in Edinburgh, Scotland, they argue that the so-called robot umpire is a perfect case study for rethinking a deceptively simple question: What actually counts as a robot?

Their answer challenges the tidy image of a robot as a single machine built by a single team to do a single job. The robot umpire, they say, is something far messier and more interesting: a constantly shifting creation shaped by engineers, league officials, umpires, players, fans, and even decades-old dreams of an “electronic umpire.”

The study draws on a concept from social science called “fluid technology,” borrowed from a famous analysis of a hand-operated water pump used across rural Zimbabwe. Just as that pump turned out to be far more than a simple piece of hardware, with its identity and purpose reshaped by every community that used it, the researchers argue that baseball’s automated strike system resists any fixed definition.

How Researchers Studied Baseball’s Robot Umpire

The research team spent roughly seven months, from December 2024 through July 2025, embedding themselves in the world of baseball to watch, listen, and ask questions. Lead author Waki Kamino and co-author Andrea W. Wen-Yi attended four MLB Spring Training games (including one without the automated system for comparison), one Triple-A minor league game, and watched the All-Star Game online. Across those visits, the pair logged about 18 hours of on-the-ground observation, snapping photos and videos, capturing fan reactions, and taking detailed notes on how challenges played out in the stadium.

They also conducted nine in-depth interviews, each lasting about an hour, over video calls. Six were with MLB personnel, including a Vice President of Umpire Operations, a Director of On-Field Strategy, a Senior Vice President of Engineering, and others involved in building and running the system. Three interviews were with fans recruited through the researchers’ social network, all American men ranging in age from their late 20s to their 60s, all lifelong baseball followers who regularly attended games. The interviews totaled five hours and 40 minutes of recorded conversation. No one was paid to participate.

Kamino also went deeper into the umpiring world, attending two umpire training camps: a four-day program in Atlanta taught by former and current major and minor league umpires, and an official one-day MLB Umpire Camp held in Dallas. Shadowing a former MLB umpire coach, participating in on-field drills, and sitting in on classroom lectures gave the researcher firsthand insight into the skills, decision-making, and culture that define umpiring, the very world the automated system is entering.

“Ambiguity is core to MLB and to the idea of what makes a good game and a good experience,” Kamino said in a statement. “The strike zone is ambiguous. It’s a social construct. How do you even automate that?”

Shohei Ohtani takes a practice swing during a game against the Oakland Athletics at the Oakland Coliseum.
MLB’s “robot umpire” isn’t a robot at all. A new study shows how cameras, humans, and fans together are redefining what a robot really is. (Photo by Conor P. Fitzgerald on Shutterstock)

What Exactly Makes A Robot Ump?

One of the study’s most revealing discoveries is that nobody can agree on what the robot umpire actually is in physical terms. At its core, the system consists of high-speed cameras made by a company called Hawk-Eye Innovations, software that defines the strike zone, and a server that processes pitch paths and generates calls. Simple enough. But those neat boundaries fall apart quickly.

Does the stadium scoreboard count as part of the robot? That’s where fans actually see the system’s decisions displayed. What about the Wi-Fi networks carrying the data, or the electrical systems powering the cameras? And then there are the people: MLB engineers maintaining the code, press box operators triggering the graphics, control room staff routing visuals to the scoreboard, and umpires wearing earpieces and belt packs to relay calls. An MLB official told the researchers that technicians arrive hours before each game to confirm “data is flowing properly,” clean equipment, and run tests. Strip away any of these human and technical layers, and the system stops working. Yet none of them look like what most people picture when they hear the word “robot.”

The label itself has a life of its own. Reporters ironically started calling the human umpires “robo-umps,” and the nickname stuck, spreading to describe the entire system even as MLB officially rejected the term. A Rolling Stone headline captured the tension perfectly: “Technology Is Coming for Baseball’s Strike Zone. Just Don’t Call It a Robo-Ump.

More Than Just Calling Balls And Strikes

On paper, the system decides whether a pitch lands inside the strike zone. In practice, the researchers found at least eight distinct functions it serves, most of which have little to do with accuracy.

Cultural preservation turned out to be a major concern. MLB tested two versions of the system: full automation, where every call is made by the machine and simply announced by the umpire, and a challenge system, where umpires still make calls but players can request a review. The challenge system won overwhelmingly. As one MLB official told the researchers, “One of the most surprising takeaways has been how many people have disliked the fully automated system.” People preferred keeping the human element, including subtle skills like “pitch framing,” where catchers try to make borderline pitches look like strikes through glove positioning. MLB officials called these strategic elements “the beauty of baseball” and “the romantic aspects” of the game.

Game flow mattered enormously too. MLB worked to keep each challenge sequence under 14 seconds from start to finish. One official noted that “a lot of people are involved, and a lot has to go right, just for that 14 seconds to be successful.” Even fans who liked the system expressed worry about its use in the highest-stakes moments. One cautioned that adding 12 to 14 seconds of challenge time during the World Series “could ruin a moment.”

The system also had to earn trust by matching real-world umpiring norms rather than rigidly enforcing the rulebook. Officials described constantly asking umpires for feedback with questions like, “Is this a pitch you’d never call a strike?” When umpires warned that certain calls would make dugouts “go crazy,” that feedback actually reshaped the strike zone the system uses.

Then there’s the entertainment factor. At spring training, researchers watched fans gasp with eyes locked on the scoreboard during challenges, followed by sighs or cheers. One official called it “good theater.” In the minor leagues, promotions tied to the system, like free chicken sandwiches if the home team recorded ten strikeouts, sometimes hinged on an automated challenge that resulted in a strikeout, sending entire stadiums into a frenzy.

For umpires themselves, the system became a professional development tool. Officials described umpires craving the real-time performance feedback the system provides. One called it “another tool for them to get it right.” It also served as a conflict diffuser. A minor league umpire said he enjoyed the challenge system because he could defuse arguments by simply telling players, “Challenge me.” And perhaps most surprisingly, it became a way to validate umpire skill. Spring Training in 2025 showed an overturn rate of about 50 percent on challenged calls, meaning umpires were getting it right roughly half the time players disagreed. An MLB official explained that keeping the rate near 50 percent meant “no one’s really getting shown up on either side.”

At its most ambitious, the system serves as a platform for building agreement. One MLB official, himself a former umpire, offered a memorable observation: “The strike zone is the mass delusion that we all agree on.” He described the system as “a final arbiter for those arguments that have gone on for 100 or 150 years.”

mlb umpires
Human umpires aren’t going anywhere yet. (Credit: Eric Broder Van Dyke on Shutterstock)

It Takes A Village to Build A ‘Robot’

The question of who made the robot umpire proved just as slippery as defining what it is. Traditional thinking might point to Hawk-Eye Innovations, which supplies the camera technology, or to MLB’s engineering team. But the researchers found the system’s creation spread across an almost comically wide cast of contributors.

MLB’s On-Field Strategy team coordinates the project. Hawk-Eye provides the tracking technology. Engineers continuously modify code based on field feedback. Stadium technicians maintain hardware. Players reshape the system by adjusting their swings and plate approaches based on its tendencies. Umpires influence the strike zone through their practiced judgment and feedback. Fans shape it through post-game surveys that inform decisions like how many challenges are allowed per game. Media coverage molds public perception and acceptance. Even history plays a role. One MLB official pointed to a photo from 1950 showing Jackie Robinson and Duke Snider testing “a prototype for the electronic umpire,” making contributors from more than 75 years ago, in a sense, co-authors of today’s system.

As one official put it simply: “It takes a village to put ABS together.”

Since its initial trial in 2019, the system has been tested in thousands of games. By 2023, it was running in every game at the highest minor league level, roughly 2,000 contests per year. In 2025, it appeared in 13 spring training stadiums and at the All-Star Game, exposing every major league team to it directly. In September 2025, MLB announced the system would officially debut in the 2026 regular season.

The researchers argue that this sprawling, years-long rollout, scrutinized by millions of passionate fans in real time, gives researchers a rare real-world window into how a “robot” gets born, contested, reshaped, and gradually accepted out in the open rather than in a controlled lab. Their central claim is that robots aren’t fixed objects but fluid creations whose identities, purposes, and boundaries are continuously negotiated by everyone who touches them. That framework, they say, applies far beyond baseball to everyday machines like home cleaning robots, delivery bots, and surgical systems whose meaning and function shift depending on who’s using them and where.

For a sport built on 150 years of tradition, arguments, and shared illusions about where the strike zone really is, the “robot umpire” turns out to be a surprisingly human creation, assembled not from circuits alone but from compromises, chicken giveaways, and a collective willingness to keep negotiating what the machine should be.


Paper Notes

Limitations

The study’s fan sample was limited to three American males, all recruited through the researchers’ social networks, which restricts the diversity of fan perspectives represented. The researchers acknowledged this narrow demographic scope. Additionally, the approach, while rich in qualitative detail, focused on a single technology case in a specific cultural domain, professional baseball, and the authors’ analytical lens was shaped from the outset by their interest in the contested nature of the “robot umpire” label, which they acknowledged influenced their approach. The study did not include direct interviews with active players or coaches, whose perspectives were represented only indirectly through MLB officials’ accounts.

Funding and Disclosures

The paper states that open access support was provided by Cornell University and Indiana University Bloomington. The research was also funded by the National Science Foundation (CAREER #1942085) and the Office of Naval Research (N00014-19-1-2299). The authors note that any opinions, findings, conclusions, or recommendations are their own and do not reflect the views of the funding agencies. No interview participants received compensation.

Publication Details

Title: What Is a Robot? Understanding Baseball’s “Robot Umpire” through the Lens of Fluid Technology | Authors: Waki Kamino (Cornell University), Andrea W. Wen-Yi (Cornell University), Guy Hoffman (Cornell University), Selma Šabanović (Indiana University Bloomington), Malte F. Jung (Cornell University) | Published in: HRI ’26: Proceedings of the 21st ACM/IEEE International Conference on Human-Robot Interaction, March 16–19, 2026, Edinburgh, Scotland, UK | DOI: 10.1145/3757279.3785604 | ISBN: 9798400721281 | License: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License

About StudyFinds Analysis

Called "brilliant," "fantastic," and "spot on" by scientists and researchers, our acclaimed StudyFinds Analysis articles are created using an exclusive AI-based model with complete human oversight by the StudyFinds Editorial Team. For these articles, we use an unparalleled LLM process across multiple systems to analyze entire journal papers, extract data, and create accurate, accessible content. Our writing and editing team proofreads and polishes each and every article before publishing. With recent studies showing that artificial intelligence can interpret scientific research as well as (or even better) than field experts and specialists, StudyFinds was among the earliest to adopt and test this technology before approving its widespread use on our site. We stand by our practice and continuously update our processes to ensure the very highest level of accuracy. Read our AI Policy (link below) for more information.

Our Editorial Process

StudyFinds publishes digestible, agenda-free, transparent research summaries that are intended to inform the reader as well as stir civil, educated debate. We do not agree nor disagree with any of the studies we post, rather, we encourage our readers to debate the veracity of the findings themselves. All articles published on StudyFinds are vetted by our editors prior to publication and include links back to the source or corresponding journal article, if possible.

Our Editorial Team

Steve Fink

Editor-in-Chief

John Anderer

Associate Editor

Leave a Comment