Who Will Hate Robot Umps the Most?

Ever since Eric Byrnes used a computer to help umpire an independent-league baseball game last year, and then Brian Kenny took up the mantle of #RobotUmpsNow on the MLB Network, I’ve been fascinated with the idea that robot umpires will soon call strike zones in baseball. The more I talk to players about it, though, the more I doubt that it’s an eventuality. Because the players, well, the players are going to hate it.

I can’t speak for all players, obviously. I haven’t talked to all of them. But I’ve talked to plenty on both sides, even ones I can’t quote here, and the biggest endorsement I could get was a tepid version of “It’s going to happen.”

So instead of asking each player what they thought about robot umpires, I changed the question a bit. Instead, I asked pitchers, catchers, and hitters, “Who will hate robot umps the most?”

The short answer? Everyone. The long answer? Much more interesting.

The Catchers
Okay, maybe this one was obvious. Catchers get to frame the pitch and add value to every play right now. They’re more plugged in than anyone. They’ll lose their privileged position.

“Catchers will just be, no pun intended, robots themselves,” said A.J. Ellis. “They’ll just be blocking and catching the ball and maybe throwing a guy out. It’ll take some of the art away.”

Former Cubs (and Marlins and Padres) catcher John Baker tried to imagine what the game would look like for the catcher, as well. “Guys wouldn’t even have to catch balls unless there was two strikes or someone on base,” he thought. “Every game could start with the catcher leaning up against the back stop.”

He didn’t stop there. “Pitch clocks was the first step, next they want our umpires, then they’ll say we can have robot first-base coaches. This line of logic clearly leads to Baseball Simulator 1000 for the NES. RoboBall.”

A few other catchers agreed, but wouldn’t put their name on it. It’s political back there behind the plate. Let’s remove catchers from the pool, because they would hate it the most. Obviously.

The Pitchers
Speaking of political answers, this one from Zack Greinke should have come with an all-caps caveat. “It would be a bad idea,” he said. “I am very pro umpire, and everything they bring to the game.” Love you, Zack, but was that for me or the umpires possibly reading this post?

Some felt like they’d lose the catcher as an ally, stealing them strikes. “The zone will get smaller,” thought Giants reliever George Kontos. “My two-seam away, I get that call because it’s moving and it looks like a strike.” And his catcher, Buster Posey, agreed: “Pitchers will hate it more. From my perspective, more calls are flipped are towards the pitchers.” But that’s just because Posey is a good framer.

Giants reliever Hunter Strickland thought there was no need. “If it ain’t broke, don’t fix it,” he exclaimed. “Why change the game?” There’s been an undercurrent against change in baseball, and it hasn’t necessarily served the game wrong. Jake Peavy imagined ahead to a dystopian future: “Is there going to be a point, 15 years from now, when our kids are not able to be attentive for two or three hours so we’re suddenly playing seven-inning games?”

It’s a defensible stance, that the game doesn’t need a fix, depending on how you feel about error. “Honestly, home-plate umpiring is really hard, those guys do a pretty good job,” admitted Baker. Examining the matter of strike-calling, Scott Lindholm of Beyond the Box Score estimated that umpires got it right about 85% of the time. Is that acceptable?

Maybe it’s not when you consider bias. Brayden King of Northwestern University and Jerry Kim of Columbia Business School worked together to find that umpire bias was prevalent. The All-Star status of the pitcher, the skin color of the pitcher, the stadium, and the count all changed the likelihood of a strike call according to their 2014 study.

“I think if it were done right, all players would prefer it,” said Dan Haren. “Players get frustrated with inconsistency, bias, and the way zones change from umpire to umpire.” So certain subsets of pitchers wouldn’t mind getting a fairer shake.

Still, there’s some reticence because of the removal of the human element, a common refrain. Strickland: “I’m not trying to play a video game, I’m trying to play baseball. Anyone can sit in front of a computer.” Peavy: “I think the human element is part of the game.”

A slightly more nuanced version, from Ellis: “The back and forth between the catcher and the umpire will go away. The game takes its queues from the umpire, and you can tell when an umpire has control of the pace of a game.” And Brian Dozier said the same, highlighting the interaction between hitter, pitcher and umpire as unique to the game: “There’s a thing about baseball that sets it apart from other sports, and that’s the interaction you can have with the umpire. In football you can yell, but in basketball you get T’ed up pretty quick.”

The Hitters
Josh Donaldson had one of the more intriguing perspectives on the issue. He thought hitters would hate it more at first, and then would change their tune. “The first year or so hitters aren’t going to like it,” he said earlier this year. “You’re going to get those breaking balls that are in the dirt — big 12-to-6 curveballs that hit the plate — they’re going to be strikes. You’re going to get the high fastballs up there, they’re up, but they’ll be strikes.”

Everything he says is demonstrably true. At least it has been in previous years. Look at the map that Jon Roegele developed to show how big the strike zone was in 2014.

RHH-2009-2014-w-title

Yes, the strike zone has gotten really deep at the bottom. There’s still undiscovered strikes at the top of the zone that the robot would pick up if this is to be the rulebook strike zone. “You’ll be able to throw your high-90s stuff up in the zone all day!” exclaimed high-90s Giants reliever Derek Law. Fellow reliever Javier Lopez agreed, saying “Hitters aren’t used to hitting up in the zone. There will be a lot of strikes to get up in the zone.”

And the curves, nipping the zone on the way to bouncing on the plate? Yeah, that’s going to happen too. “As a guy who’s now more attentive to framing stats, I want to know how many times over my career I’ve lost a curveball because Clayton Kershaw’s 12-6 curveball is so nasty,” pointed out Ellis. “I had one the other day where I went to block it and I checked later and it was a strike. And I lost points for that.” Oakland starter Rich Hill agreed: “Pitches that enter the zone as strikes and leave as balls, and are caught past the plate will present as strikes. It would especially benefit myself a lot more than it would benefit a hitter.”

Here’s a 2012 plot from one of Jon Roegele’s early pieces on the subject. It shows curveball called strikes (left), and a marked difference down in the zone (and on the corners low) from other pitches like fastballs (right).

2012CurvesFastballs

So we’d see more curveball strikes.

It gets worse for hitters. This great Hardball Times post from Eric Lang, which imagined the strike zone as a three-dimensional volume, showed that back-door and front-door strikes — balls that aren’t necessarily strikes at the front of the strike zone, but nip the back ends of the strike zone — are called strikes at a 30% rate compared to the 80-85% rate on other pitches.

CORRECT CALL PERCENTAGE
All Pitches Strikes At Front Edge Back-door Strikes
80.4 81.8 31.8
SOURCE: http://www.hardballtimes.com/analyzing-the-strike-zone-as-a-three-dimensional-volume/

It looks like there would be three added sources of strikes from the robot strike zone. When provided this information, some pitchers changed their tune. “I could be convinced,” said Peavy with a smile. (He throws front-door sinkers to lefties, you see.)

Of course, there could be a tweak. “I think the strike zone needs to be defined better before it’s implemented. It’s hard to describe, but I think the edges need to be expanded towards the middle of the plate, and it’s a little too high,” thought Haren. Others agreed that tweaking the amount of the ball that needed to be in the zone could help mitigate the impact of the extra strikes from bouncing breaking balls and the sweeping sinkers.

Donaldson did suggest it would come back around to the hitters’ favor eventually: “Day in and day out, our strike zones are different. I’ve been punched out two times this year at balls in my neck. I would be into it. If you could have the strike zone the same every day, you’d get used to it and have better awareness.” Ellis agreed, to an extent. “One of the biggest gripes a hitter will have is when he asks an umpire, ‘Is that as far as you will go?’ and then he goes further,” the Dodgers catcher said.

There’s not going to be a good way to tie this all up in a bow, and you’re always going to get very subjective answers when you ask players. “If they blow a 3-2 call, you’re suddenly into robot umps,” joked Law. “Let’s say I go out there and pitch and the umpire has a bad night and blows eight to ten calls, yeah I’m not happy and that might be a night where I say, yeah let’s do this. But I know there will be another night when four to five calls go my way,” said Hill.

Personally — and this surprises me a bit because I am a fan of accuracy and did like that replay was implemented — I’m not sure I’m a fan. It may sound ridiculous to talk about human elements and video games, but there’s some truth in there. “I love the fact that we can sit here and argue he was safe, he was out, that was a strike. That’s baseball,” said Hill, and that’s part of it for me.

But the strike zone is also currently a negotiated entity. The pitcher, the catcher, the hitter and the umpire (and sometimes even the public analyst on a free website) all come together to produce a living, breathing document. No, it’s not consistent every night, and it has holes. But that negotiated zone is a unique thing in sports, and losing it would lose something real, call it whatever you will.

If we actually push that strike zone to the rulebook edges, all the way to every border in all directions, we’ll probably see offense plummet. Hitters may find a way to adjust to plate-hitting, side-winding called strikes. They may not, though — it’s already hard enough for hitters to cover both sides of the plate. Even a plus-plus eye like the one on Joey Votto has trouble with the inside pitch. Even Bryce Harper looks past that inside pitch.

What will happen when our best hitters have to swing at curveballs headed for the plate? They can’t all be Mike Trout on those pitches. There’s no way this doesn’t lead to reduced offense and a huge spike in strikeouts… unless we change the rulebook definition of the strike zone fairly drastically. And then we’ll have changed baseball in three different ways — the rule book, the implementation, and the feel on the field — and along the way we will upset the catchers, the pitchers, the hitters, the umpires, and the traditionalists.

Is it worth it?

We hoped you liked reading Who Will Hate Robot Umps the Most? by Eno Sarris!

Please support FanGraphs by becoming a member. We publish thousands of articles a year, host multiple podcasts, and have an ever growing database of baseball stats.

FanGraphs does not have a paywall. With your membership, we can continue to offer the content you've come to rely on and add to our unique baseball coverage.

Support FanGraphs




With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.

newest oldest most voted
Damaso
Member
Damaso

red sox fans.

Damaso
Member
Damaso

when you think about it, it’s pretty hilarious that we expect a human to consistently judge whether a wickedly swerving 95mph object has passed through an invisible 3 dimensional box which changes dimensions with every batter, all while every player involved is doing their best to fool you into calling it their way.

374285942768
Member
374285942768

exactly and that’s just one human, let a alone a different human set by their own set of inconsistent parameters each day. how can we determine what is statistical ‘noise’ when there is so much inconsistency in the tools that measure performance?