Instant Replay Is an All-or-Nothing Affair
In Super Bowl XL, Ben Roethlisberger scored a dramatic one-yard touchdown just before halftime. On that play, it was initially ruled that Roethlisberger carried the ball into the end zone, and that call was upheld upon review. Many people, myself included, thought it was very clear upon review that the ball did not break the plane of the goal line, and therefore the run should not have been ruled a touchdown. But it was, and the Steelers went on to win the Super Bowl.
Scenarios like this illustrate how instant replay plays an important role in professional football in the United States. Oregon's controversial onside kick recovery against Oklahoma in 2006 in which review ultimately gave Oregon the win shows that replay's importance extends to the college ranks, as well. Cases like these abound and stir controversy about instant replay's role in football.
Some people suggest that instant replay belongs and allows us to ensure that the game is played according to the rules. Others suggest that instant replay ruins the flow of the game. Another reason replay is controversial is that some plays are reviewable, but others are not. The distinction between reviewable and nonreviewable plays is not as clear as one might hope, and the aim of this chapter is to suggest that the distinction is seriously problematic. Once we see that this distinction cannot be made in any interesting way, we shall see that either every play must be reviewable, or no play is reviewable.
This discussion will center around the NFL's use of instant replay. But what is said here about instant replay can also be said regarding the NCAA's use of replay, and of any other football organization that employs instant replay.
Consistency
Were a member of the officiating crew to call holding on one player over and over again throughout the course of a game but not call it on other players who are blocking in exactly the same way, we would think that there was something very wrong with the way that official was calling the game. The problem would be that the official was applying a rule inconsistently. The official would, in effect, be maintaining that some instances of blocking in a particular way are holding, while other instances of blocking in that way are not. That would be like a police officer claiming that sometimes driving seventy miles per hour in a school zone is speeding, but other times it is not. We think that rules should be applied consistently.
Not only should the rules themselves be applied consistently, but the motivations for the rules should also be applied consistently. For instance, we would find it odd if the NFL were to make it illegal to tackle the quarterback out of concern for player safety but still allow wide receivers going across the middle to be tackled by an opponent they do not see coming. To make rules like this would be to apply the motivation for a rule (in this case, player safety) inconsistently. But when the NFL makes so-called horse-collar tackles illegal because there are serious safety concerns about such tackles (rule 12, section 2, article 1 [12.2.1])1 and also makes hitting a passer below the knees illegal (12.2.12), we can appreciate that in both cases the concern is for catastrophic lower-leg injuries to players. Terrell Owens was the victim of a horse-collar tackle late in 2004 that resulted in his missing most of the playoffs (though, famously, not the Super Bowl). Carson Palmer suffered a serious knee injury in the playoffs against the Steelers (as they were on their way to Super Bowl XL) when Kimo Von Oelhoffen fell into his lower leg. Both of the rules mentioned were new for 2006, due in part to these injuries’ making salient ways in which players can unnecessarily be in harm's way during games. The point is that these rules are motivated by the same consideration, and we expect that motivation to be applied consistently.
The Rules
The rules governing instant replay in the NFL are succinct. They appear in section 9 of rule 15. An important aspect in the statement of the rules is the distinction between reviewable and nonreviewable plays. The plays that are reviewable are carefully spelled out, so that plays not covered by the description of reviewable plays are nonreviewable. Reviewable plays are broken down into three categories: (1) plays governed by sideline, goal line, end zone, and end line, (2) passing plays, and (3) other plays. The first category concerns the usual issues, like whether the ball broke the plane of the end zone on an apparent touchdown, whether a ball carrier stepped out of bounds, and so on. The second category concerns things like whether a pass was completed, whether a quarterback fumbled or threw an incomplete pass (because his arm was moving forward when the ball left it), and the like. The third category is new for 2006 and governs things like whether a “runner was ruled not down by defensive contact.”
Again, anything that is not a reviewable play as articulated by the rule counts as a nonreviewable play. The rules explicitly mention nine nonreviewable plays, including the status of the clock, the proper down, penalty administration, force-outs (whether a receiver would have come down with the ball in bounds had he not been pushed out by a defensive player), and field goals.
Important to keep in mind here is that this distinction concerns whether plays are reviewable. We're not here concerned to specify exactly which plays are to be reviewed but only which plays it is permissible to review. Many plays that could be reviewed are not. These rules concern which plays could be reviewed.
“Get the Calls Right!”
In a 2003 Sporting News article on replay, Paul Attner provides reasons for thinking instant replay is a good idea. He says, “Because football once existed before replay—and before television, for that matter—we still must deal with the grumpy-old-men argument that we shouldn't remove the human element from the game. All that means is we know the officials will make mistakes, but let's not do anything to correct the errors. Nonsense.”2 The idea seems to be that instant replay makes it more likely that the right call will be made; at least, it allows for the correction of officiating mistakes. Since we have definitive rules according to which the game is to be played, why not do everything possible to ensure that the rules are being followed? If we needed a slogan, it could be this: Get the calls right.
But notice that if this is the motivation for including instant replay at all, it does not motivate the NFL's treating such things as field goals as nonreviewable. After all, it would be very easy to tell in a replay whether the ball went over the crossbar and between the uprights. And since it is not at all uncommon for a game to be decided by a field goal, it is certainly important to get those calls right.3
Put more generally, if getting calls right is the motivation for instant replay, then we should allow replay on every play. Not to make every play reviewable is to apply the motivation inconsistently. And as we saw above, inconsistent application of the motivations for rules is unacceptable.
One might object at this point that getting the calls right is not important all the time, just on certain calls. But to say this is to allow that inconsistency in the application of the rules is acceptable, which, as we saw above, it is not. So to be consistent, given the motivation for using instant replay, all plays should be reviewable.
Reasons against Instant Replay
In opposition to Attner's position, Steve Greenberg argues that the NFL should give up on instant replay. He says, “There's a blissfully simple explanation why replay should be beanbagged: It's not fun. You know what's fun? Hut-hut-hike, violent tackle. Replay keeps us from that.”4 Greenberg's complaint is a common one among those with whom I watch football: instant replay ruins the flow of the game. One's favorite team can be in the midst of a stimulating drive, which can be disrupted with the throwing of the red flag. Replay, on this view, adds an unnecessary stoppage to the game. Keeping the game exciting is more important than making sure every call is made correctly.
Notice that this line of thought applies to all plays: allowing that any play be reviewed would be to allow the flow of the game to be ruined. So no play should be reviewable. If part of the NFL's motivation for maintaining that certain types of plays are not reviewable because reviewing them would be too time consuming (thereby ruining the flow of the game), their application of the motivation is inconsistent; any replay ruins the flow of the game and so should not be allowed.
Is There a Middle Ground?
So far, we've seen that the reasons for having instant replay are reasons for maintaining that every play is reviewable. We've also seen that a common reason for thinking that some plays should not be reviewable holds for all plays as well. But perhaps there is a middle ground here. Perhaps there is a distinction to be made between those types of plays that are reviewable and those that are not that would allow us to say that the reviewable plays are such that it is more important to get them right than it is to ruin the flow of the game, while the nonreviewable plays are such that it is more important not to ruin the flow of the game than it is to get them right. This would make sense of the way the NFL's rules on instant replay are articulated and would allow the NFL to maintain that the motivations for the rules are indeed consistent. Let us consider, then, what the differences might be.
Judgment Calls
The most commonly held way to distinguish those plays that are reviewable from those that are not is by whether the call is a “judgment call” or not. Consider pass interference penalties (which, being penalties, are not reviewable). Pass interference is defined as follows: “It is pass interference by either team when any player's movement beyond the line of scrimmage significantly hinders the progress of an eligible player or such player's opportunity to catch the ball” (8.2.5). So when an official is considering whether to call pass interference, the official must determine whether a player's movement “significantly hindered” another player's chance at making a catch. But what constitutes significant hindering is up to the official, so this type of call crucially depends on the official's judgment.
Reviewable calls, on this view, do not crucially depend on an official's judging things in a certain way. Only calls that are independent of an official's judgment are reviewable. So consider things like whether the ball has broken the goal line on a short run. This involves no judgment on the official's part, since either the ball has or has not crossed the goal line. Crossing the goal line is not like “significantly hindering” one's chance to make a catch.
There are two problems with this way of distinguishing reviewable from nonreviewable plays, however. The first is that this does not capture the distinction as it is practiced. Surely, for instance, there is no judgment call involved in a field goal; either the ball goes above the crossbar between the uprights or it does not. Field goals are nonreviewable plays that do not depend on there being a judgment call involved. Thus, field goals show that the distinction cannot be made entirely on the basis of whether there is a judgment call involved.
The second and more serious problem with this way of making the distinction is that it is a superficial distinction: on close inspection it becomes difficult to make sense of it. Recall Ben Roethlisberger's one-yard touchdown run in Super Bowl XL. On that play, it was initially ruled that Roethlisberger carried the ball into the end zone, and that call was upheld upon review. Its being upheld means at least that there was insufficient video evidence to overturn the call (as per the rule).5 Many people, myself included, thought it was very clear upon review that the ball did not break the plane of the goal line, and therefore the run should not have been ruled a touchdown. Somebody is mistaken. It cannot be the case that it was clear that Roethlisberger did not score, and there was insufficient video evidence for overturning the touchdown that was called on the field during the game. How are we to explain this? The best explanation for the controversy surrounding this call seems to be that whether the ball breaks the plane of the goal line is a judgment call.
This might strike some people as shocking. Either the ball broke the plane or it did not, end of story, and the story does not involve anyone's judgment. I am happy to grant that either the ball broke the plane or not, but that isn't really what is at issue. At issue is whether one can determine whether the ball broke the plane. That certainly looks like a judgment call.
One might still be puzzling over the claim that plays like touchdowns involve an official's judgment in the same way pass interference penalties do. To help motivate this idea, let us consider other types of reviewable plays. What becomes apparent is that there are many types of plays that involve a judgment call.
So consider first fumbles. A fumble is defined as “any act, other than a pass or legal kick, which results in loss of player possession” (3.2.4). The crucial notion in the definition of a fumble is that of possession, which is defined as follows. “A player is in possession when he is in firm grip and control of the ball inbounds” (3.2.7). So when, exactly, does a player who fumbles cease having a “firm grip and control” of the ball? This question is especially pressing for situations in which a replay is initiated because there is a question about whether a player was down before losing possession of the ball. If a player is down (7.4.1.e) before losing possession, the loss of possession does not constitute a fumble, and the ball therefore cannot change possession. It is easy enough to imagine situations in which there will be widespread disagreement about whether a player lost control of the ball before going down. This makes it look like whether a player fumbled involves a judgment call.
Or consider whether a player makes a catch, which is defined as occurring when “a player inbounds secures possession of a pass, kick, or fumble in flight” (3.2.7). But on nearly every weekend during the NFL season there are a number of plays about which there is disagreement concerning whether a player had possession of a pass. This suggests that whether a player makes a catch involves a judgment call.
Less frequently, but no less tellingly, there are issues concerning whether a player is down. A player is down when he “is contacted by a defensive player and he touches the ground with any part of his body except his hands or feet” (7.4.1.e). This can be very difficult to determine. For instance, in the second quarter of the 2006 Rose Bowl, Vince Young lateraled the ball to his running back, who took the ball in for the score. Many people thought that Young's knee was down before he pitched the ball. The play was not reviewed, but it is unclear whether the play would have been overturned. Plays like this suggest that determining whether a player is down involves a judgment call.
The last example I offer concerns when a player is out of bounds. Players often step very close to the sidelines, and any contact with those lines constitutes the player's being out of bounds. But it is often very difficult to determine whether a player did or did not make contact with the sideline. Such cases involve a judgment call.
The point of these examples is, again, to show that the notion of a judgment call is not able to make the distinction the NFL would like with replay. Given the nature of an official's imperfect connection to the truth of exactly what happens on the football field, all calls are essentially judgment calls. Some types of calls (like pass interference) are clear less often than other types of calls (like when a player is out of bounds). But that doesn't mean that the two types of plays are essentially different. So to maintain consistency, the NFL cannot motivate the distinction between reviewable and nonreviewable plays in terms of whether there is a judgment call involved.
Correctable Calls
Another way in which one might try to motivate the distinction is in terms of whether a call is correctable; that is, whether a play's outcome could be corrected. Imagine a situation in which an official blows the play dead because the official rules that the player stepped out of bounds, when in fact the player did not.6 Suppose further that the runner had plenty of room to run and was likely to have gained at least twenty yards on the play.
There is no way to tell exactly what would have happened had the official not blown his whistle. Therefore, there really is no way to correct the play, and so there is no point in reviewing it. Plays like whether the ball crossed the goal line are correctable, and so are reviewable.
This distinction will not work, either. It suffers from two defects. The first is familiar: this distinction does not capture what the NFL actually does. Field goals are surely correctable, even though they are not reviewable.
There is yet a more serious problem: this distinction also fails. There is no such thing as an uncorrectable call. Some calls are easier to correct than others, but every call is correctable. One might wonder how the imagined scenario above, in which the player is ruled out of bounds when he is not, might be corrected. After all, the ball could not be moved to where the runner might have wound up, because we simply cannot determine where that is (the runner might have broken all the tackles and scored, or the runner might have tripped and fallen out of bounds). So such a call is not correctable. The NFL's rules allow for such plays to be corrected, though not in the same way other plays are corrected. Suppose that the running play imagined was a reverse and the ruling was that the runner stepped out of bounds behind the line of scrimmage, resulting in a loss of yards. A mistaken call like this could well prove costly. And while the team on offense cannot regain whatever result would have happened on the play, it surely need not be made to suffer the losses it incurs because of the errant call. Rule 7.4.3.a says that if an official inadvertently blows his whistle (thereby ending the play), “the team in possession may elect to put the ball in play where it has been declared dead or to replay the down.” When rulings are mistaken and seem uncorrectable, the play could be treated in exactly the same way an inadvertent whistle is treated. This would allow the team to minimize the effects of a bad call. They need not lose a down, as it could be replayed. And they need not lose yardage, as the down could be replayed. So such plays are indeed correctable, even though the correction is not as obvious as it might be in other types of plays.
Well worth noticing is that every play is so correctable. No play is such that it couldn't be corrected in at least this manner. But if this is so, then whether a play is correctable would not allow us to distinguish reviewable from nonreviewable plays.
Whether a play involves a judgment call or is correctable will not allow one to draw the distinction between reviewable and nonreviewable calls. There seems not to be any other way to make the distinction, so to maintain the distinction on either the motivation of getting calls right or to keep the flow of the game going is to be inconsistent with respect to the motivation for the rule. As we saw above, this is unacceptable. Instant replay, therefore, is an all-or-nothing affair: either every play is reviewable, or no play is.
A Glance at a Different World
One last stumbling block for this position, in some people's eyes, might be the logistics of exactly how it would work for every play to be reviewable, or for no play to be reviewable. In actuality, neither situation is too difficult to imagine.
Consider first it being the case that no play is reviewable. Since football was played for a long time before the use of replay, we are quite familiar with football without replay. There are issues about getting calls right, but for many, consistency and maintaining the flow of the game are well worth the cost of a few bad calls. Football and football fans survived for years with bad calls, so there is little reason to think that a world without replay would be all that chaotic.
A world in which every play is reviewable is not that difficult to imagine either. In the NFL, during the last two minutes of each half and overtime periods, replays are initiated by a replay assistant who is in a booth watching video of the game, rather than being on the field. And as rule 15.9 indicates, “there is no limit to the number of Referee Reviews that may be initiated by the Replay Assistant.” If all plays are reviewable, the replay assistant's job becomes much more involved. The replay assistant could initiate reviews from the booth throughout the game, which would help to ensure that calls are made correctly.
Conclusion: All or Nothing
There is no middle ground to take regarding instant replay. Either all plays are reviewable or none are, and in either case, the change to the game need not be that dramatic. But considerations of consistency should motivate the NFL and other organizations that use or are considering using instant replay to adopt an all-or-nothing approach.
Notes
I am indebted to George Schumm, David Merli, Nicholaos Jones, Ryan Jordan, and Mike Austin for helpful discussions about instant replay and comments on earlier drafts of this work.
1. All reference to rules follow the order rule, section, article with only the numbers mentioned. All such references are to the Official Rules of the NFL for 2006. See Larry Upson, ed., Official Rules of the NFL (Chicago: Triumph Books, 2006).
2. Paul Attner and Steve Greenberg, “Should NFL Instant Replay Stay or Go?” Sporting News, 17 November 2003; available at http://findarticles.com/p/articles/mi_m1208/is_46_227/ai_110314522.
3. Recall Phil Dawson's game-tying field goal against the Ravens on November 18, 2007. The initial call was that the kick was no good, and the Ravens thought the game was theirs. After a lengthy conference, the officials changed their decision, and the game went to overtime. The Browns wound up winning.
4. Attner and Greenberg, “NFL Instant Replay.”
5. The referee might well have thought it was clear, however, from the replay that Roethlisberger did indeed carry the ball into the end zone. But for present purposes, that is neither here nor there.
6. Plays in which the runner is not ruled out are reviewable, but the type of play in question is not.