“Let ’em Play”: Concluding Thoughts

I started on Monday with a puzzle – what might be said in favor of enforcing at least some rules of sports less strictly at crunch time? – and tried to develop a solution. That solution turned out to be two solutions, or two variants of a single solution.

All competitive sports, I have claimed, share a core interest that the outcomes of contests reward competitors’ relative excellence in the performance of the sport’s fundamental athletic tests. To further this interest, each sport has reasons – weighty but not decisive – (1) not to enforce penalties on infractions when, for contextual reasons, the penalty would be unusually over-compensatory, and (2) to sometimes disregard the rule-like form or surface of some norms in favor of the standard that underlies it.

These arguments are tentative and partial, only first steps toward a solution to the puzzle. But whether they ultimately justify the temporally variant enforcement of particular rules of particular sports, all things considered, is not greatly important to me. Think of this study as a search for what Robert Nozick called a philosophical explanation: not a defense of the thesis that temporal variance in sports is optimal, but an account of how that could be.

Philosophical explanations are not always the right goal. Often we want to know what some agent should do. In this case, however, I’m satisfied to identify factors and analytical devices that might prove useful for theoretical projects across reaches of law and sports.

For example, the analyses here might helpfully illuminate the lost chance doctrine in torts; the granting of equitable relief, near contest’s end, from rules governing municipal and corporate elections, or appellate litigation; the difference between genuine “jurisdictional rules” and mere claim-processing rules; and possibly much else.

Those are just promissory notes at this point. So I’ll conclude by offering one final non-obvious lesson – albeit one for gamewrights, not for legislators or judges. It concerns soccer.

Here are two much-noted problems with the beautiful game: there is too much diving, and refs make too many errors. The latter is partly a consequence of the former, but it’s also a consequence of there being only a single referee and FIFA’s refusal to introduce any form of instant replay review. (Plug: my thoughts on instant replay are here.)

While these are familiar criticisms, I maintain that soccer harbors a third defect, one that works as a multiplier, exacerbating the first two problems and exacerbated by the fact (not itself a problem) of low scoring. That problem concerns the red card – in particular that it results in ejection of a player for the remainder of the match without allowance given for substitution.

This is an unusual complaint. But if it’s a surprising charge, its connection to the issue of temporal variance might seem obscure.

Here’s the connection. A central assumption undergirding the argument that basketball referees should “let ‘em play” is that, presumptively, the competitive impact of a penalty should bear a stable relationship, over the course of a contest, to the competitive impact of the infraction that the penalty penalizes. We saw, however, that (holding closeness of contest constant) a contest event has a greater impact on outcome the closer it occurs toward contest’s end. Non-enforcement of the penalty at crunch time aims to rectify this imbalance.

I’m not going to suggest that soccer’s red card should be brandished more reluctantly at crunch time. Unfortunately, that’s not because soccer ensures that the red card exerts a constant competitive effect regardless of when issued. It’s because red cards exert a greater competitive effect the earlier they are awarded. Because a red card results in ejection of the offending player and a ban on his being replaced, it entails that the offender’s team play short for the remainder of the match (or until the opposition is red-carded too).

So the more time remaining at point of infraction, the greater the penalty. In effect, a red card awarded at minute 15 reads “play shorthanded for 75 minutes” whereas one awarded for the very same infraction at minute 85 reads “play shorthanded for 5 minutes.” The red card thus violates the sensible principle of game design that, presumptively, the same infraction should call forth the same penalty regardless of the time of occurrence.

This disparity in the effective magnitude of the red card sanction should occasion little concern if the optimal penalty for committing a red-card offense (serious fouls, spitting, handling the ball to deny an obvious goal-scoring opportunity, etc.) were to be shorthanded for 90 minutes. In that event, the sanction would never be too high, and the fact that it would generally be too low would be unavoidable. But that’s not plausible.

To be sure, what would be an optimal period of shorthandedness is extraordinarily difficult to determine. But the basic parameters are plain: Because a red card is awarded for a serious offense, the offending team should incur a significant penalty, one that meaningfully affects its prospects for victory. Yet we don’t want the penalty to be virtually outcome-determinative – all the more so given the prospect (exacerbated by the prevalence of diving, by the presence of a lone referee, and by the absence of replay) that some red cards will be issued in error.

Nobody would seriously entertain a proposal to replace the penalty of ejection with the award of two goals to the opposing team. Given soccer’s very low average scores and margins of victory, a sanction of such magnitude would threaten to convert the sport into an extended exercise in penalty avoidance. Similarly, we might expect that sending off a player in, say, the 10th minute is apt to have such a significant impact on game outcome as to contravene the competitive desideratum.

The obvious solution is for soccer to unlink the penalty of ejection from the penalty of shorthandedness. Soccer already decouples the consequences of a red card for the player involved from the consequences for his team: The player is sent off for the remainder of the match and is disqualified for the next game too, but the team plays shorthanded only for the remainder of that game, not for the next.

Soccer’s governing bodies should consider taking this decoupling further. That the offending player may not return does not entail that his team should play shorthanded for the rest of the contest regardless of when the foul occurred. Many sports, not only hockey, allow a team to substitute for an ejected player after some period of penalty time. Perhaps soccer should follow their lead.

To require a team to play shorthanded for nearly a full game is draconian even when the offense really warranted dismissal. But it’s heartbreaking when—as happens disappointingly often in this otherwise beautiful game—the red card should never have been issued.

Figuring out what would be an appropriate period of shorthandedness would prove challenging. I’ll leave that to the econometricians. I claim only that the current system that makes the competitive impact of a red card so radically dependent on its time of issuance is unlikely to dominate the alternatives, and therefore that further investigation is warranted. More to the point: that we should think harder about soccer’s red-card system is only one among the many and diverse lessons to be learned by reflecting on the puzzle of temporal variance in sport.