Can Umpires and Pitch-Tracking Coexist?
Rob Manfred might be distancing himself from robot umps.
Last October, Manfred addressed and seemed open to the idea that tracking systems could be used to call balls and strikes:
“As technology continues to improve and those sorts of adjustments can be made [in] real time, that technology will become more feasible for use on the field. I don’t believe we are there yet.”
But last week, while speaking with USA Today’s Bob Nightengale, the MLB commissioner seemed to back away from the idea of turning over ball and strike calls to machines.
“I don’t believe the current technology is sufficient to call balls and strikes on a real-time basis,’’ Manfred says. “If and when we get to that technology — and sooner or later we’re going to get there — there’s still a fundamental question about whether or not we want to remove that human element from the game.
“There is a human aspect to that, a work aspect to it, that’s always been an important part of our game… I don’t think you can just jump to the conclusion that if you have the technology to do it that’s the right thing for your product.”
Added Nightengale: “Bravo.”
As indicated by Nightengale’s response, there’s still plenty of resistance to technology in the game — or anything that challenges tradition.
Maybe we’re further away from seeing automated calls than some thought, a fact that might be frustrating to Ian Kinsler and Joe Maddon, both of whom have voiced frustrations with calls recently.
Still, it seems the technology will eventually get there. The systems in place today can call the east-west zones more accurately than umpires, though determining the north-south zones have yet to be solved. Given the relatively short window within which sophisticated pitch-tracking has become available, however, the solution is likely to arrive sooner than later.
When it does, the question is whether baseball will be willing to use it.
While a perfect strike zone might never exist even with the introduction of radar and computers, it seems a more consistent zone could be created by turning ball and strike calls over to robot umpires. The strike zone continues to evolve year by year, according to Jon Roegele’s analysis, and I wrote earlier this year how a more consistent strike zone could help teams and players with in-game strategy and big-picture planning.
For instance, if a club had built a team designed to take advantage of the lower part of the strike zone — an area which had expanded every year from 2008 to -15 — they would have been disappointed to learn that the bottom of the zone has actually contracted a bit over the last two years.
Of course, doing away with umpires could have other unintended consequences, like eroding home-field advantage, for example.
The debate seems to center around whether consistency and integrity of the zone is worth sacrificing the human element. Umpires might soon have another issue around which to express solidarity: a fight against technology.
Umpire Joe West before taking the field today wearing white wrist band to protest escalating verbal attacks on MLB umpires! pic.twitter.com/a8TkW2JCip
— Major League Baseball Umpires Association (@MLBUA) August 19, 2017
Perhaps what’s being missed, though, is that this doesn’t have to be debate about whether technology should render human umpires redundant. Home-plate umpires would still likely exist even with automated ball and strike calls, and technology could free them to focus on other tasks.
In some ways, the debate has already been settled: as I noted on Monday, Statcast (and PITCHf/x before it) is collecting the same data for which scouts were once responsible. We no longer need scouts to calculate pop times and lead lengths. Not at the major-league level, certainly, and — depending on the proliferation of pitch-tracking systems — probably soon not at the minor-league level, either. That doesn’t mean the scout is incapable of adding value. The introduction of ever more sophisticated technology don’t mean the computer and the human umpire cannot coexist.
There’s been this idea that the human umpire could be responsible for north-south calls while tracking systems call the east-west zone. Beyond the prospect of that somewhat intricate hybrid solution, however, there’s this: with the home-plate umpire freed of ball-strike calls, he could better serve another important function — that is, on-field traffic cop.
Consider the present concerns about pace of play. The commissioner has expressed his intentions to help make game move more quickly — and rightfully so, as the amount of time between pitches continues to increase. This is at a moment in the game’s history when balls in play are at an all-time low.
Pace isn’t just about pitchers taking forever to throw, it’s about batters stepping out of the box. Since MLB and the union agreed to drop fines for batters who violate pace rules, I have doubts about how well batters are adhering to the rule and about how interested umpires are in enforcing it. But without ball-strike calls, the umpire could focus on seconds between pitch and keeping batters in the box. The home-plate umpire could focus on keeping the game moving. It’s not as interesting a job title or responsibility — and there would be other duties that remain — but the role would still serve a purpose.
Some have wondered what a game might look like without umpires involved with ball-strike calls.
Wrote ESPN’s Buster Olney recently:
Last weekend, ESPN’s Will Dorney — in St. Louis to do graphics for Sunday Night Baseball — mentioned an idea that you could envision the moment he said it out loud: Strike and strikeout music for the home pitchers, activated automatically by the strike zone electronics.
In other words: walk-off music for pitchers.
Just imagine: Craig Kimbrel is trying to work his way through a tense situation. Bases loaded and two outs, Aaron Judge at the dish, and Kimbrel tries to do what the Red Sox pitchers tried to do repeatedly against Judge on Friday night — attacking the top of the strike zone.
And with a full count, Kimbrel nicks the strike zone, and in an instant, there is a flash of red on the scoreboard, and Kimbrel’s strikeout music is activated.
The umpires have understood for decades the importance of letting the fans and players know the outcome of important pitches, which is how and why so many came to develop their own personal style for signaling a strike three.
But even if they are not calling balls and strikes, a home-plate umpire could still serve roles, and new ones, and peacefully coexist with a computer. Computers can give us a more accurate, consistent game and they could help umpires give us a quicker, more satisfying one.
Maybe this shouldn’t be an either/or debate. We should instead focus on what the computer and human do well and build around those strengths.
A Cleveland native, FanGraphs writer Travis Sawchik is the author of the New York Times bestselling book, Big Data Baseball. He also contributes to The Athletic Cleveland, and has written for the Pittsburgh Tribune-Review, among other outlets. Follow him on Twitter @Travis_Sawchik.
There is a middle ground.
Instead of using the tech to replace the umpire, you give the tech to the ump to help them do the job better.
The tech already exists to give umps a visor that overlays a player’s strike zone in their field of view and that can freeze a marker showing where the ball crossed the strike zone volume. (It’s pricey, ~$3000 per visor, but it exists.) The strike zone presented would be independent of the ump’s position so the ump’s posture and angle and catcher framing would have no impact.
The ump still makes the call.
But he doesn’t rely just on his eyeballs.
Instead of replacing the ump, you augment him.
(And since the visor feed can be recorded and shared it’ll cut down on whining and arguing. It’ll make for faster games, too.)
Yeah, the augmented reality technology angle is interesting, and more feasible than truly automated strike zones IMO. Frankly, $3000 is a fairly trivial expense compared even to umpire salaries.
This is almost certainly the way to go, but first we’ll have to get the strikezone readings right.
I watched a game last week in Detroit where the strike zone was just about 2 inches off left to right, to say nothing of the challenges with the top and bottom of the zone changing from batter to batter and pitch to pitch.
That’s the part of the technology that isn’t there yet. While in aggregate the readings are accurate, we’re not interested in in aggregate results on a single pitch in a playoff game.
I was going to say pretty much tha same thing. Your response beat me by an hour.
It would seem to reason that if we all have StatCast in near-live time, that could be incorporated into a pair of Oakleys and an overlay to what they see live. Reviewing the calls can help tweak the systems. It could also include a timer to assist with pace of play.
Couldn’t agree more. I envision, basically the home plate ump wearing Google glass. I’m all about change, but I can’t imagine having to reach to a scoreboard to know ball/strike calls. Part of the entertainment for fans (and emotion for players) is built into the immediacy and presentation of the calls. With augmented umps, you keep that, but give the ump another tool aside from his eyes (well, he’d being using his eyes on the tool, but I digress…) to make the call.
what the hell is the difference….the umps standing their? Thats just optics. Your still using the computer. It still deletes the catchers out of the game, why is everyone trying to change the game all of the sudden? why play with 9, lets just play with 8. Ridiculous
Well-reasoned and thoughtful argument.
There is definitely going to need to be a combined approach – while the tech will clearly be there soon, umpires will probably still need to be on the field. It will be interesting to see how the league/players union/umpires union deals with this going forward. Good closing statement about focusing on the strengths of both.
This always seems like a silly conversation to me. The con of the electronic strike zone is “don’t remove the human element,” which translates to “continue to allow human error to impact game outcomes.”
The pro of the electronic strike zone is IT WILL ALWAYS GET THE CALL RIGHT.
Hawkeye has been in use since 2001, it’s absurd to say we don’t have the technology.
MLB Baseball is an entertainment business. And one that relies a lot on continuity and tradition so the powers that be are by necessity conservative in making changes. Nobody wants to annoy the paying customers.
the con is that offense *could* plummet with some of the borderline/strange angle/late breaking pitches that are almost never called strikes by a human ump that would be strikes electronically.
Of course, that’s a could, and we’ll never know if they experiment with it, but saying the con is only “don’t remove the human element” is disingenuous.
FWIW, I personally would really like to see them at least examine it further.
no. any program can be manipulated to call the low strike or give the corners to the pitcher, etc.
this is a false solution to a human problem
if we go to mlb robots calling balls and strikes, we might as well just delete the catcher position. put a net behind home plate and put a bucket of balls on the mound. good luck finding someone that wants to squat all day just to catch a ball. The art of catching is why catchers play the position…..some are just much better at it than others.
Does the “art of catching” not involve other factors like pitch-calling, observing hitters’ movement in the box, building relationships with pitchers and shepherding them through perilous situations, combating the running game, and being involved in virtually every single defensive play?
Pitchers need a visible target in the form of a catcher’s glove to throw to.
you honestly think the catchers will be calling pitches in 2 years if we start letting computers and earphones on them?
Umpires are actually great at what they do 99% of time. That 1% does not justify changing the entire game over.
Hypothetical: Every instance of that 1% was a game-altering situation.
Does that 1% still not warrant investigating change?
At 300 pitches per game, 1% in my hypothetical is 3 pitches, and 3 game-changing instances…
They are not great at what they do 99% of the time. Even the MLB only claims 97%. Independent analysis has placed the number closer to 81% overall and 61% on the edges of the strikezone. You can make various arguments justifying the continued existence of Umpires. Their performance is not among them.
its the same ump for both teams. both teams get same calls. bipartisan. 99% of the time the calls are the same all game. 1 teams does not benefit over the other, unless they are wise and hired a great framing catcher. In that case, the catcher may only get you 2 calls a game at best.
“both teams get same calls” – sounds like a comment from someone who’s never played an officiated sport.
Also sounds like you’re arguing that they make so many mistakes it makes it ok. You a politician? The team that hires the guy who’s better at tricking the umpires wins, that’s the baseball you want to see?
Your made up numbers are far from accurate. All the studies done have shown that umpires are missing roughly 1/3rd of the calls on the edges of the plate (the calls that they are “paid” to make), and they are missing on average 8 calls per game more in favor of the home team than the visiting team. It’s not bipartisan. It’s not 99% accurate. These guys are horrible at their job and the entire game suffers.
there you are… well maybe more like 80% of the time. But still.
I’ll keep beating the drum on my, probably inane, middle ground idea. I think the issue has been mis-framed from the start. Rather than have the goal be ‘getting it right’ 100% of the time, the goal should be *consistency*. From a fan & player perspective, isn’t the real problem not that a borderline ball was called a strike, but that it was called a strike in one instance but a ball in another?
So let the umps call their game, and map out their strike zone tendencies, then use that map to enforce that strike zone against them. Technology would only step in when an umpire calls a ball where he usually tends to call a strike (regardless of if it’s *actually* a ball or strike), or vice versa. You retain the (in my opinion) entertaining gamesmanship that comes with a somewhat amorphous strike zone – teams would presumably have access to this data and you’d reward pitchers/hitters who can action it. You could even expand this a bit to say the umpire gets a +- 5% ‘close enough’ buffer to their personal strike zones, so umpires would have less concern that they’d be constantly contradicted on the borderline calls.
I guess fans who aren’t comfortable with ambiguity would hate this, but I think it’d be a great compromise.
To marry this to some if the ideas thrown around here, I could envision an ump with a Google glass type fixture that has a plot similar to the strike-zone plot on the FG game pages. Except it would would only show pitches on the edges of the zone, colored for balls/strikes showing the umpire’s historical data.
North/south robotics can’t be that far off. For each level, you only need about 500 inputs, and those can be set at the beginning of the game (18) and then at each substitution. Maybe that’s a lot of data…for an abacus. A bad abacus.
What might be more interesting than this incessant whining about The Zone, is to construct the zone an umpire is pitching to as the pitch count accrues in a game. That might really be interesting, this other nonsense just isn’t.
one of the best part about the game is arguing with those umps. want to make it more boring? go to computers. The game has been played for over 100 years, now we have these millennials in charge and writing about it trying to change it. So sad
Two random thoughts:
–This has been said, but it’s worth repeating. Suppose baseball never existed, and you were creating it in 2017. You have two options for balls and strikes: a computer with 99% accuracy (assuming we get to that point) or a human. Would it even be a conversation?
–If gambling on baseball becomes legal – and Manfred seems to support it – I think that could be a catalyst for moving to an automated zone.
Suppose you or I never existed?… absurd supposition. but that’s where your logic is going. Watch “2001 A Space Oddity”.
Ok. Focus. On what computers do best… record and store information, formulate outcomes based on the information. what else?
I for one would not want some computer doing that for what tetermines my paycheck/compensation. Who’s to say that a program is not adjusted to predict a certain outcome more often? then who do I have a beef with?
Now let’s once more Focus. On what humans do best… bring personality and life to our existance. We are social beings and need the input/judgements of our peers to guide our lives.
Do you suggest robo judges in the court room…of course not (i hope).
Sorry, computers are no certain solution to your troubles with the game… people will find ways to manipulate that system and then where are we? Can’t tell you all the times I go to the bank, utility company, or hardware store… “sorry you can’t have your money/favorite grass seed Sir, our computer went down.”
Well, you’d certainly see more hackers at the game
I think the most obvious solution could be implemented almost immediately: provide the HP umpire with a device that vibrates when the pitch is within the robot strike zone, to give them real-time info. On the close pitches (the ones they most frequently get wrong), they can wait half a second to help them make the right call. And if a player bitches about a call, they can always say “robe ump said it was a strike,” and that should shut them up.