Improving baseball's replay not as easy as it looks

Better umpiring on the field is a must (2:53)

ESPN's David Schoenfield and Eric Karabell discuss the flaws found in MLB's replay system. (2:53)

The replay machines first started humming on the final day of March 2014, with a 105-second review of a call at first base in Pittsburgh. Can we all agree that baseball hasn't been the same since?

Once upon a time in this game, if an umpire called you out, you were out. If he called you safe, you were safe. About all you could do if you disagreed was kick dirt, heave the nearest base into the outfield or exhaust your supply of words you can't say on "Dora the Explorer."

Not anymore. It's now three seasons and nearly 3,000 replay reviews later. Here's how we'd describe life in a world with expanded replay:

More plays get called correctly than at any time in history. The right team wins pretty much every night. Also, we'd love to have 20 bucks for every minute we've spent watching umpires wearing their favorite noise-canceling headphones.

The bottom line? "We have done what we hoped to do, and I think most people in baseball are happy we have instant replay in place," said Atlanta Braves vice chairman John Schuerholz, the original chairman of baseball's first replay committee.

"Happy" is a relative term. Happy to be getting more calls right? You bet. Happy about 35 percent more calls being reviewed than at this time last year? Not so much. Happy to be spending 4 minutes, 53 seconds waiting for the replay umps to decide that they couldn't tell whether a guy was safe, as happened last week at Wrigley Field? Really, really not so happy.

Are we still hearing grumbling about replay? From players? From managers? From coaches? From GMs? Oh, yes. Do they think the replay operation can work better, faster and more smoothly? Oh, yes. But three seasons in, is it already too late to "fix" this replay system? Oh, no, Schuerholz said. MLB is very much open to suggestions.

"It's the third year, and the last year of our rollout," he said. "But that doesn't mean that in baseball we're going to stop doing what we can to make it pluperfect."

If baseball is open to suggestions, we have good news. We've been collecting suggestions from 16 people who work in all aspects of baseball, on and off the field.

Schuerholz is listening. He has been gracious enough to weigh in on many of the ideas we presented to him. Although he wants to make clear that he isn't speaking for Major League Baseball, he is at least willing to give us an idea of what's practical, what's reasonable, what's fair and, well, what isn't.

So here we go. It's time to reach into the replay suggestion box.


We heard this from players. We heard this from managers. We heard this from pretty much everyone: Speed. It. Up. Please.

"If it takes longer than two minutes [to review a play]," one player said, "then the call on the field stands."

Let's put this in perspective. Most reviews take less than two minutes. The average review time this season is 1:55. What's surprising is that replay experts predicted that by Year 3, that time would be shrinking. Instead, it's rising. From 1:46 in 2014 to 1:51 in 2015 to 1:55 in 2016. If we established a two-minute time limit, it would affect nearly half of all reviews. Is that a good idea?

"I don't know if it should be two minutes or two-and-a-half minutes," Schuerholz said. "But I understand the general feeling behind that sentiment. I don't think anybody would disagree that, if you asked people if they would prefer instant replay judgments to be rendered more quickly or less, they would all say more quickly."

That would be unanimous. But suppose we asked those same people a different question. Would they rather get more calls correct or incorrect? We're guessing they'd vote for "correct." Just a hunch. Which is more important? There is no doubt as to which way Schuerholz leans.

"We've had so many calls which have been called correctly and which have turned losses into wins and wins into losses because they've been called correctly," he said. "So would we like to have things transpire more quickly? Of course we would. But we prefer to focus on the accuracy of the calls. And if that sometimes takes a little longer than we'd like, then so be it."


This was another common theme. The managers might not be loitering in the infield while waiting for word from their video wizards on whether to challenge. But the general consensus was overwhelming: They. Take. Way. Too. Long.

We heard two executives -- and one current manager -- propose a time limit on how long managers have to challenge, and after that, they wouldn't be allowed to check with their replay experts. The time limits proposed ranged from 15 to 30 seconds.

"When that manager is on the top step, waiting, it's such a lull in the game," one exec complained. "We've got to find a way where that's timed."

Let's go to another fact check. On average this season, we're down to a 36-second "lull" between the end of a play and the manager's decision to challenge. That's reduced from 45 seconds in Year 1. How much would be gained by cutting 15 or even 20 seconds off that time, if the likely result would be more challenges that never would have been made?

Eliminating looks by the club replay wizards? Schuerholz isn't a fan of that idea.

"How many times have you seen a bang-bang play at first base where the runner said, 'I was safe. Challenge it.' And he's wrong?" he said. "So to do that might be faster. But that doesn't mean it's more accurate."

What about a cap on how much time a manager has to challenge? That, Schuerholz said, "might be something that should be talked about. ... I'd be OK with that, with whatever people decide is the right amount of time."

A 30-second challenge cap? Why not? It wouldn't save much time in a sport that still averages less than one challenge per game, but anything that encourages a faster pace is worth exploring.


Here's a thought: Seeing as all 30 teams go through New York every season, invite them to tour the replay center. We bet it would erase a lot of misconceptions.

We keep hearing people wonder why umpires in the replay center don't start looking at close calls the moment they're made, instead of (allegedly) waiting for a challenge. Schuerholz and others say that's already happening, and umpires in the replay center begin looking at different angles immediately. But execs across the sport find that hard to believe.

"That's supposed to happen, but it's not," one executive said.

Sources familiar with the behind-the-scenes operation in the replay center say that although the process starts within seconds, umpires feel obliged to check multiple angles, even on calls that appear obvious. That means it's the pursuit of accuracy that is eating up time. But again, some execs are dubious.

"They should be able to say 'safe' or 'out' as soon as the guys put on the headset," another executive said. "I love replay. I think it makes the game better. But it takes too long."

Of the first 313 calls reviewed this season, only five took less than 40 seconds, just one took less than 30 seconds and only 37 were decided in less than a minute. That, one of the execs quoted above said, "makes me crazy -- the time it takes when it's an obvious call."

No umpires agreed to be interviewed for this story. But informal conversations about replay with umpires over the years have led us to two conclusions: First, they have come to love it. Second, they're taking all that time because they've been encouraged to not rush if they think there's a better camera angle out there. As the venerable Joe West once said, "We don't want any umpire leaving a missed call on the field just because someone got impatient."


This is an idea managers have supported from the beginning. Add a fifth umpire to every crew. Rotate the extra umpire through the park's replay room every day. Then, when a controversial play goes down, that umpire already will be locked in on just that game. Boom, he's on it. He zips through the review, and back to the action we go -- at least in theory.

But Schuerholz said he is convinced that the current system, which involves assigning two crews of four umpires each to replay duty for a full week, works better. For one thing, the umpires in New York see the TV feed just as quickly. For another, each umpire in New York is assigned to oversee only two games a night, so he's watching closely. Also, Schuerholz is a fan of creating the opportunity for several umpires to collaborate on a call and use their cumulative expertise, which includes handling reviews of a wider variety of plays.

"I think that doing it this way probably improves their ability to make the right call," he said. "So I don't agree this is a significant area of concern for us."


The technology gap was a major issue when replay started, and players, managers and coaches still wonder about it. Does a Wednesday afternoon game in Milwaukee utilize the same cameras, angles and technology as a Sunday night game in Boston?

The answer is not quite. Because every park is laid out differently, Major League Baseball feels it can't mandate where cameras should be located. But every team receives recommendations on camera locations and necessary angles. The differences, from game to game, are now described as "minimal" by one source familiar with the process.

Nearly all games now have two "super slo-mo" cameras on hand, which can cover plays at every base in the digital detail needed for optimal replay reviews. Early on, some parks didn't have even one of those on site.

"That has always been the goal: that every ballpark would have the same level of cameras, the same amount of cameras and the same quality of cameras," Schuerholz said.


It wouldn't be accurate to say the new slide rules are totally responsible for the 35 percent increase in reviews this season. But baserunning plays, mostly at second base, appear to account for the biggest chunk of that increase. It also appears that unsuccessful challenges of those plays are responsible for another fascinating replay trend: a sharp decline in the percentage of calls that are overturned.

In 2014 and 2015, nearly half of all reviewed calls were reversed (47.3 percent in 2014 and 48.9 percent in 2015). This season, that percentage has dropped to 42.5 percent. Meanwhile, confirmed calls are way up, from 24.3 percent in 2014 and 23.2 percent last year to 31.9 percent this year. (A call stands if there is not enough clear and convincing evidence to reverse or confirm the call on the field.)

Why? One executive said you can connect almost all those dots to the slide-rule reviews.

"I thought we were close [to a perfect replay operation] at the end of last year, but we've added too much," the executive said. "I'd say take away those reviews at second base -- the slide rule and the neighborhood-play rule -- and make them not reviewable. Let the umps call it at game speed. It's too fast a play."

Concerns about these rules go beyond replay, but let's stick to the topic. Would dialing back the "Utley Rule" reviews make for fewer replays? No doubt. But Schuerholz has a difficult time accepting the logic that these plays shouldn't be reviewable simply because they might look misleading in slow motion.

"The objective is to get plays called correctly and have no games lost because of an inaccurate call," he said. "So it seems to me it would be even worse to have the technology to see a call that was missed and not correct it. ... We have to make a choice. Do we want the game called correctly or not?"

We know there are people out there screaming: "No! We miss the human element." We hear from those people. We believe they're sincere, and they really do hate those 1 minute, 55 second reviews so much that they'd rather get calls wrong than watch umpires stand around with their headphones.

As always, we respect your opinion. But please, take a step back and think about this. Replay is not ruining this sport -- far from it. This isn't the NFL. The average team challenges two calls per week. An average game has only a 66 percent chance of producing one replay, and that one replay lengthens the average game by only two and a half minutes, including the time it takes the manager to decide to challenge.

In the meantime, very, very, very few calls that get reviewed are resulting in blatantly wrong decisions. MLB actually reviews the reviewers. Of the 1,338 plays reviewed last season, Schuerholz said, only 10 were determined to have been decided incorrectly.

Yes, baseball needs to keep searching for ways for this system to function more quickly, because as Schuerholz said two years ago, it's important to preserve "the rhythm and flow of our great game." But how much more quickly? That debate never ends.

"We really ought to figure out what's an appropriate amount of time [for a play to be reviewed]," Schuerholz said. "And we're probably not there yet. I don't know if 1:50 is the right answer or 1:35 or 1:22. If we're at 1:50, everyone is going to say, 'That's too long because it's a pause in the action.' But I would maintain it's a pause for a good reason.

"We can't please everyone, and we know that. But I also know this: We're way better off than we were two years ago."