Debut Post: Fixing Instant Replay

Hi! Welcome to the inaugural post of The Spiel, my new sports blog. Auspiciously enough, I’m starting this blog right as people are proclaiming the end of blogging, but hopefully that only applies to politics blogs.

I want this to be a blog about the kinds of sports discussions that tend to get bogged down in fruitless, repetitive arguments between, well, teams: purists versus moderns, traditionalists versus statheads. In these arguments, the advocates for each side tend to share the same points among themselves and then repeat those the same points back to their opponents; each side gets so used to responding to the other side’s tired arguments that the discussion calcifies. My point in this blog will not be to say that “the truth is somewhere in between”—personally, I lean towards the moderns nearly every time—but to point out the kinds of more nuanced points that get missed by this kind of polarization, in the hopes of making discussions fresher, more interesting, and more complex than they so often tend to be.

One good example of the kind of debate I’m talking about is the debate over instant replay review. Granted, this may not seem like the most timely time to enter this discussion, which—at least in football and baseball, the two biggest American sports—would appear to be resolved: both sports have replay review and they’re not going back. But in fact I think the discussion is as timely as ever, since the implementation of replay in both sports—in football since 1999, in baseball since just last year—has shown replay’s potential to both enhance and harm the entertainment on the field, which is after all the purpose of all rules and officiating: to keep the game as entertaining as possible. If this sounds like a radical idea, it’s only because we wrongly treat “entertainment” and “spectacle” as synonyms; conventional sports entertain by giving us a fair competition governed by clear and enforced rules. Replay aids in this enforcement, but whenever its benefit is outweighed by the harm it does to other aspects of the game’s entertainment, it should be considered a failure, and reworked.

It seems to me that the most common arguments for and against replay review have lost sight of the goal of entertainment: the anti side ignores the ways in which officiating mistakes hurt the product on the field, while the pro side sees correctness, not entertainment, as the ultimate goal. After a fairly lengthy consideration of the issues involved here, I’d like to suggest an idea that I haven’t seen anyone suggest before: a time limit on replay reviews, to distinguish between calls which are merely incorrect (and should be reluctantly tolerated) and calls which are bad (and should be eliminated at all costs). Almost everything wrong with replay, it seems to me, could be fixed by making this distinction.

2014 was baseball’s first season of allowing replay review of most plays on the field, and while there are quibbles about how it’s been executed, it’s clear that baseball can’t go back: once you have the power to correct mistaken calls on the field, it’s hard to imagine ever being without it. I admit that, having been replay-skeptical before last year, I spent a few weeks grasping at any evidence that replay was a disaster—more from the take-sides instinct that I mentioned above than from anything substantive. But now I’m resigned to the necessity of replay review. In Game 1 of the 2014 National League Division Series, which I listened to on the radio, there was a crucial call early in the game that went against the Giants (I’m a Giants fan). When the radio announcers began saying that it looked like the call was wrong, I felt relief: it was going to be overturned. I remember the feeling of impotent frustration and anger that used to accompany a blown call going against my team, and the idea of going back to the days when we had no recourse for that feeling is a non-starter.

However, replay comes with a cost, one that was widely predicted (and that was already on view in football): the game grinds to a halt while the replay officials study the video to make a determination. I only attended two baseball games in 2014, which is what having a toddler will do for you, but both of them featured lengthy replays, and both are worth considering, if you can stand to spend that much time watching nothing but umpires wearing headphones. The first, White Sox at Giants on August 13, turned almost entirely on a crucial 7th inning at-bat, in which the Giants batter, Joe Panik, trying to get the Giants’ first run home from third with one out, hit a weak grounder that led to a play at the plate. The Giants runner, Gregor Blanco, was pretty clearly out, with the ball beating him by about a second…but the Giants challenged, and the call was overturned based on the controversial new Rule 7.13, establishing that a catcher can’t block the plate before the ball arrives. How long did it take to determine that the catcher had blocked the plate? Oh, about ten minutes. Click this, I dare you:

Not only was the elapsed time unconscionable, but overturning such a crucial call in the late innings (a tie game with one out, as opposed to a one-run deficit with two out), on such a ticky-tack rule, is not how a game is meant to be decided. Even Giants fan blogger Grant Brisbee, in his recap of the game, declared, “I’m not okay with that”:

… I feel unclean, completely gross. The Giants only scored runs because of a technicality, a loophole, a deft legal maneuver. They were going to get shut out, yet they scored seven runs because Joe Panik hit a baseball just horribly enough. I’m all for the new rule … but I’m also in favor of the umpire getting the opportunity to use his discretion. This opens up an Angel Hernandez Wormhole of Stupid, but I’m okay with an umpire looking at the replay to see if the runner really couldn’t score because of the catcher’s positioning.

Granting that the delay here is due at least as much to the rule, not replay itself, replay is still a necessary cause: with an unlimited amount of time to review the call and (as Brisbee points out) no room for discretion, the rule’s confusing implementation is no longer just a problem for the umpires who have to enforce it, but is inflicted directly on the fans, in the form of replays like this (and like those in the Fangraphs link above).

The second game I attended, Nationals at Dodgers on Labor Day, featured four replays; three of them were nice and short—and, not coincidentally, confirmed that the umpires had been correct in the first place—but the most significant one took three minutes (which is not ten minutes but is still a long time to watch nothing happen), overturned a call at a significant moment in the game (it turned a leadoff single by the fastest player on the team into an out, in a game ultimately decided by one run), and did so on marginal, even dubious evidence:

If Dee Gordon was out on that play, it’s because the tip of Gio Gonzalez’s glove brushed the tip of Gordon’s heel just before Gordon reached the base, and I don’t think the TV broadcast shows an angle that proves that that even happened. As with the Giants replay above, no one on earth would have objected if the original call stood. So all the replay gave us was a result that is just as uncertain as the original call, at the cost of three minutes of dead air.

The thing that is wrong with replays like this is not just that they are a boring waste of time, but that they completely shift the focus of the action from the players to a bunch of people nobody cares about, and in particular a group of replay officials in New York who are making these decisions sight unseen (hence the umpires’ headphones while they wait for the decision). Watching this in person, I knew it was boring, but I couldn’t know how absurd it was for the TV viewer. Seriously, I know none of you watched those videos—not with me droning on about how boring they are—so at least go back and watch the second one, the Dodgers one. The announcers start calling the play by sharing what we’re interested in, relevant descriptions of the action (the pitcher, Gonzalez, is late covering first base, which allows a fast runner like Gordon to apparently beat the tag)—but once the play is being reviewed, there is nothing to narrate except the replay…which they dutifully do, for two minutes, and by the time the call is made they have understandably made themselves believe that the replay is the action:

And the runner is…OUT! What a challenge by the Nats’ video coordinator, Eric Dalton, and then out to Matt Williams via Randy Knorr!

It sounds like a celebration of a breathtaking triple play, rather than of one guy in the clubhouse watching video and then telling a coach to tell the manager to tell the umpires to tell the guys in New York to review the call.

In cases like these, replay review, which is intended to reduce our frustrations with poor officiating, actually compounds one of the major ones: that bad calls make the officials, not the players, the center of attention; that officials should “let ‘em play,” by making only the strictly necessary calls and otherwise staying out of view.  Keith Law (I believe) coined the hashtag #umpshow to describe umpires who, through a combination of bad and unnecessary calls and theatrics, make themselves the center of the action instead of the players; well, what is this kind of review, if not an #umpshow dragged out for three or five or ten minutes? Is it less egregious because the #umpshow was commissioned by the coaching staff of one of the teams?

Since it may seem that I’m putting an enormous amount of weight on just two replay reviews over the course of an entire season, I should acknowledge: of course these are just anecdotes, and it’s true that the average time for baseball replays has been under two minutes (though this doesn’t factor in the time it takes for the manager to decide whether to challenge the call in the first place). But it’s the extreme cases that we need to talk about, because those are the ones that are doing harm, and, as I’ll argue below, the ones that are most easily avoidable.

So if replay reviews are not entertaining, but the alternative—just accepting bad calls with no recourse—hurts the entertainment value of the sport too, how do we incorporate replay in the least disruptive way possible? We have to reframe our goal for replay, not as improving the accuracy of calls per se—a task both endless and boring—but as improving the sport’s entertainment value. This means that the goal of replay should not be to get calls right, but to avoid bad calls (or blown calls, if you like). Correct calls are good in themselves, like all correct things, but they are not so great a good that they are worth sacrificing the entertainment value of the game. By limiting our use of replay to bad calls, we can keep replay’s most important gains while minimizing the harm it does to the sport.

So what is a bad call? When we say that an official made a bad call, we don’t, I think, simply mean that the call is wrong. We mean that the call is clearly wrong upon watching the replay, or even in real time (as with this infamous call, which I think almost singlehandedly led baseball to adopt replay a couple seasons later). When the result on the field clearly doesn’t conform to the rules, the results are arbitrary instead of governed by accomplishment, and arbitrariness is not interesting. So the more glaring the officiating error, the more the sport’s entertainment value suffers.

Okay, that’s obvious. But the corollary is that the less obvious the wrongness of the call is, the less sports fans care, all else being equal. Of course I know that all sports hinge at times on seemingly tiny margins and arbitrary rules, and sports fans are dutifully obsessed with those margins and rules. But I don’t think it can be denied that when replay reveals that an official missed an extremely close call, all but the most unhinged fans react with resignation, not fury.

So the ambiguous calls are the ones with the least upside, the ones that fans least care about getting right (or, shall we say, they are the least outraged if the call is wrong). Not only that, there are multiple disadvantages to subjecting such calls to replay review. The main one is that the ambiguous calls take the longest time to review, as officials have to zoom in to the millimeter level from several camera angles, and then deliberate a very difficult judgment call. Overturning an ambiguous call also is frequently more intrusive, often hinging on obscure or complicated rules (as with the Giants clip above). And finally, the ambiguous calls are the ones that are most likely to produce inconclusive replays, meaning that the benefits of replay (the assurance that the call is right, or the power to correct a clearly incorrect call) don’t really accrue to them anyway; in fact, ambiguous calls produce a risk that the replay official will get the call wrong, which has twice the frustration of an ordinary blown call but takes ten times as long. (Dez Bryant’s overturned catch against Green Bay in the NFL playoffs last month is an example of how frustrating this kind of replay review can be, turning a thrilling athletic accomplishment into minutes of waiting followed by days of legalistic arguing.)

This is what comes from the premise that the purpose of replay is to get the call 100% correct. Once you make that concession, there is no way to avoid wading into the swamp of a long, inconclusive review. In fact, the more inconclusive the replays are, the more time the officials spend on them, in the hopes of seeing something that they can call conclusive. This is perverse.

On the other hand, if we redefine replay’s job as to avoid bad calls, and to otherwise stay out of the game’s way, the solution is simple: replay review should take only as long as is needed to determine whether the call was blown or not. In other words, a time limit: if two minutes, say, have passed and the replay officials still can’t make up their minds, the review is ruled inconclusive and the original call stands. Not because a two-minute delay is barely noticeable while a two-and-a-half-minute delay is an eternity (though obviously less time is better than more time), but because if it takes a long time to tell if the call is wrong, then it wasn’t a bad call.

“Don’t make the perfect the enemy of the good” is a truism in politics, but not in sports, where the relentless pursuit of perfection is considered a basic virtue. When it comes to designing the rules and their enforcement, though, we don’t have the luxury of striving for perfection. We should not tolerate umpire error—that is, an umpire botching the call, missing something obvious—but we should tolerate umpire fallibility, not as a virtue in itself, but as an alternative to indefinite delays in pursuit of a perfection we’ll never achieve.

I know I’ve spent a lot more time on the problem than on the solution here, but I would love to discuss the solution in the comments. I think it’s a simple and effective fix. Unfortunately, I don’t think a lot of baseball cognoscenti would go for it—but that’s a subject for another post.

Image via


9 thoughts on “Debut Post: Fixing Instant Replay

  1. Great first post but I have to say I disagree. I think your goal is worthy but I just want to get the play right and if it takes a a few minutes, I’m fine with that. I think your example of the replay of the play at the plate is an outlier as Rule 7.13 is a problem and MLB should do something about it. But I don’t think taking some time to try and correct a call is a problem. The Dez Bryant play was a tough one but as far as I can tell the refs made the right call, according to the rules. Isn’t that what we want? If the rule is wrong or dumb that’s a different conversation but I have no problem with them analyzing it from 50 different angles if that’s what they need to do to get it right.

    I think a time limit, and the challenge system, is just another “gimmick” that blocks us from getting what we want, which is the correct calls. I think college football has the best system where all plays are eligible to be reviewed and are often done so fairly quickly. There’s nothing more frustrating than watching a baseball game and having the manager walk out to talk to the umpire and then standing there while he waits for a signal from the dugout as to whether he should challenge or not. Just have someone watching in NYC and if they see a close play they buzz the umpire and tell him to stop the game for a minute or two while they check it out. I think most reviews would be over in less than a minute.

    Definitely enjoy the discussion on this. Looking forward to more posts.


    • Cory, thanks so much for the comment! First off, certainly different fans are going to prioritize these things differently. What seems like an interminable delay to me might be no big deal to you; what seems like an acceptable level of official error to me might be frustrating for you. One of the reasons I find this topic interesting is that it brings to the surface differences in priorities between fans, differences that aren’t always obvious.

      That said, let me defend the merits of what I’m saying a little bit. I definitely agree that Rule 7.13 is a problem in itself. For that matter, it seems, so is the rule on completions/incompletions that governed the Dez Bryant play: the call may well have been right, though a lot of people said it was wrong (I don’t know NFL rules well enough to judge, and I got lost in the discussion of whether “goes to ground” is the same thing as “going to ground”), but one thing a lot of people seemed to agree on was that it’s a badly-written rule. But that’s part of my point: there are always going to be unclear or ambiguous rules or rule situations, and those are bad enough without spending several minutes appreciating just how unclear they are. The problem with replay as it’s currently constituted is that once it starts, it has to , even in situations (like the two we’re discussing) where With rules like these, no result can be truly satisfying, so there’s no hope that the time the replay takes will be productive…but once the machine starts you can’t stop it until you’ve reached your unsatisfying conclusion. Of course, one might fairly prioritize the importance of the rules above these concerns, and you could even argue that exposing the silliness of badly-written rules is a benefit of replay in disguise. It just seems to me that a playoff game—or even a regular-season game—is not the time to stop everything to legislate these kinds of rules.

      I completely agree about the silliness of the manager challenge—as I plan to discuss in a future post, we get the worst of both worlds, since under this system replay doesn’t even eliminate the manager’s trip out of the dugout. And I’ll have to think more about the example of college football. I do wonder, though, about the workability of stopping to review every close play: think of how close so many pickoff throws are, for instance, or think about the complexity of what you do when (say) the lead runner could have taken an extra base had the call been correct. For all its complexity, football is nice and linear in a way that baseball isn’t. But maybe I’m not using my imagination enough here.

      Anyway, thanks again for commenting! I hope you’ll stick around for future posts.


  2. Hey Tom, as an avid Giants fan, I am all too aware of how instant replay played a role in the teams success this year. I am lucky enough to live in the Bay Area and get to watch or listen to almost every game as well as pre/post game and all of the other conversations that occur the next day on KNBR. The discussion at the beginning of spring training was very suspicious of instant replay, whereas discussion at the end of the postseason was almost entirely supportive. Marty Lurie, in particular is a great listen, if you get the chance. As the sportscasters opinions go, so do the listeners.
    I think that since you are discussing the entertainment value, it is important to take into account the entertainment value of the replay itself. For myself, as well as many others, the discussion about the replay, during the replay, is highly entertaining. The strategy of the use of replay also gives an added element of entertainment, and has truly shown which managers are masters of strategy and which are still learning. Granted this comes from a baseball LOVER, not just a passing fan. Rob Manfred is on the hunt for new ways to get new fans, and pace of play is a big deal right now. I wholly support, shortening the process, but I really do believe that will come with time.
    Related note: would love to know your thoughts on the home plate collision rule…


    • Katy, thanks for the comment! You make a good point about the entertainment value of replay; I should have mentioned that. (It is a pleasure that is pretty much only open to those watching on TV; live or on the radio it really does feel like dead time.) That said, it seems to me that in most cases there are diminishing returns: the longer the replay, the less fresh it is. In the days before instant replay review, broadcasters would still show close calls on replay; the difference was that the game had moved on, so the announcers could drop the topic once it was no longer interesting. If we capped the length of replay reviews, the broadcast would have the discretion to do the same thing, rather than being chained to the replay no matter how long it takes.

      As far as Rule 7.13, I’m no kind of rule expert, but some of the problems it’s caused seem to be the simple function of a lack of nuance: the rule says “the catcher cannot block the pathway of the runner as he is attempting to score,” but doesn’t clarify what “block” means—is the plate “blocked” to the runner if the runner is still several feet away?—or what the catcher is allowed to do in the process of receiving the ball. Other rules that hinge on distinction like this have been clarified on these kinds of points; think of the obstruction call in the 2013 World Series,, which outraged many people until they realized that the rulebook had specified exactly how the rule should apply in that situation. In other words, the plate-blocking rule needs to replace its difficult-to-define absolutes with practical, game-tested standards. As with obstruction and interference, we ultimately need to give the umpires more discretion here—something that a lot of fans are (with some justification) pretty reluctant to do. This might be a good topic for a future post!


  3. Hey, Tom!

    Nice read. I have a counter-offer. Rather than a time limit, how about a view # limit? Admittedly, I care very little about baseball games being longer or shorter. I absolutely agree about the loss in entertainment value. From the football perspective, calls can only be overturned if the video evidence is “indisputable.” That, for me, provides a foundation to impose a limit to the number of times that a play can be replayed. If the judge can’t make the call after watching the thing 5 times, or seeing 3 angles, then whatever evidence the video might provide is not indisputable. After X replays, the replay hood shuts down, and you live with the call on the field. I am willing to argue for more or fewer viewings.

    Liked by 1 person

    • Hi Jonathan! That makes a a lot of sense, and solves the same problem I’m aiming at from a different angle. I agree that the point is not the time itself, but the diminishing returns (in terms of accuracy and entertainment) of more and more replay. What’s needed is the ability to say, “Enough; we’re not getting any closer than this, and it’s not worth trying.”


  4. Pingback: Pete Rose and the Fallacy of Forgiveness | The Spiel

  5. Pingback: Embrace the Wrongness: What Replay Is and Isn’t For | The Spiel

  6. Pingback: The Sport of Politics | The Spiel

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s