The old NFL overtime rule was not unfair

On Tuesday the owners of the NFL franchises agreed to change to the rules for how to deal with playoff games that end in a tie after “regulation” time. (On average, about one of the 11 games each post-season is tied after 60 minutes.)

The old rule was simple: a coin toss gave one team a right to elect to receive the kick-off in overtime; they would then play according to the usual rules until one team scored. It was what North American sports refer to as the “sudden death” rule. This rule has been used in NFL playoff games for decades, and in regular season games since 1974 (with the modification that in the regular season a game still tied after 15 minutes is recorded as a tie).

So what’s wrong with this rule? There is one perennial complaint: it’s unfair. And there’s a strong belief that fairness is a core value in sports, and that ensure fairness in the competition is the main rationale for many of the rules. Fair enough. But why was the old OT rule unfair, according to the critics?

Because too much was supposedly being decided by the coin toss. This was felt strongly by fans (of the losing team) and pundits (speaking for nobody in particular) whenever a team would win the toss, receive the kickoff, march 30 yards or so downfield, and kick a field goal. This looked unfair because the other team’s offense never even got a chance to touch the ball.

I’m sorry, but at the risk of sounding like an academic — or worse, a philosopher — the old system was not unfair. Who was it unfair to? From the point where the fourth quarter ends in a tie, the rule treated both teams in exactly the same way. It is not biased against either team. They have battled to a draw after 60 minutes (3+ hours in real-time). At that point they can both expect to have about 50-50 chance of winning. The use of a coin toss and sudden death is not biased against either team.

In fact, even a rule that simply decided the game with a coin toss would not be unfair. Nor would a rule (common in many sports, and used in the NFL in the regular season if the game is tied after four five quarters) that gave each team one point instead of awarding the winner two points and the loser none (or half-a-win instead of either a win or loss).

Now there’s no denying that something feels very wrong about deciding a sporting match with a coin flip. But that wrongness is not “unfairness.” And thinking more carefully about why it is wrong will help us (at least those of us interested in the mission of This Sporting Life) think more clearly about what is valuable or meaningful about spectator sports.

But let’s return to what, at the very least, was a perception of unfairness with the previous rule. Is there any data that would support the worry? It clearly matters in some sense for sports fan whether a coin flip, rather than the play of the game, decides the contest. So has the coin flip been the deciding factor? I haven’t been able to amass all the relevant data. It is complicated by the fact that our data on overtimes comes since 1974; but in 1994 the kickoff was moved back 5 yards. It appears that until 1994 the team winning the coin toss scored first 52% of the time, and since then about 60% of the time. This suggests that the simplest way to minimize the impact of the coin toss would be to move the kickoff back to the 35 yard line; or perhaps to give the coin-toss winners the option of starting their first possession on their own 15 yard line or to making the other team do so.

In the next post I will talk about the implications and justification of the new rule, but for the time being it is worth noting that the new rule will not allow the team winning the coin toss (Team X) to win on the first possession with a field goal. If X scores a field goal, the other team, Y, gets at least one possession to try to score; and if Y score a field goal also, the game reverts back to sudden death.

If what we are concerned about is minimizing the advantage for Team X, it is not clear that the new rule will make that much difference. If it doesn’t score a touchdown on its first possession (which would end the game under the new rules), it still has an excellent chance of having one more possession than Team Y at the point when the game enters sudden death. But I will discuss some of these probabilities in the next post.

We should also note that strategizing about the probability of winning in OT begins not after the coin toss, but at some point early in the 4th quarter of a tight game. Unless one team tied the game with a touchdown and a two-point conversion on the last play of the 4th quarter, at least one team, and probably both were making strategic decisions in the 4th quarter based on an expected probability of 50% of winning in OT. For example, a team that ties with a TD and extra point on the last play was thinking that its chances of success with a 2-point conversion were a little less than 50-50. (In fact, several teams have gone for two in this situation: some winning and some losing, because it too is about a 50-50 proposition.) The fact that both teams think that OT is a 50-50 proposition shows that they think it is fair, ex-ante. (Imagine how different the 4th quarter would be in a tight game if the coin toss for overtime were conducted before the start of the game…)

In the next post I will argue that the new rule is better. It’s not any fairer than the current rule. Both are equally fair. (By contrast, it would not be fair if the home team automatically got to receive the kickoff in OT; or if the teams owners got to bid for the right to receive the kick in OT in an instant electronic auction, with the advantage going to the richest or craziest owner’s team.)

So what is it people are reacting to about the old rule (which will continue to be the rule in the regular season)? It is not unfairness, but randomness. It is the idea that something that has nothing to do with football could decide the game. In fact, it is unsatisfying (and feels unfair) whenever something besides the effort, talent, strategy, preparation, and execution of the players and coaches decides the contest. Some of the “extraneous” factors that can decide a game are in fact unfair. If there are blown calls by referees, that adds an element of randomness that we don’t want to affect the outcome; but if the officials appear to be actually biased, or worse, if they are shown to be taking bribes or betting on the games they are judging, then that really is unfair.

Our distaste for random events and supposedly extraneous factors deciding games tells us more than simply how we should or shouldn’t reform the rules of a sport. It also no doubt plays a role in why some people are more or less attracted to different sports. Because, as a matter of fact, some sports are much “chancier” than others. We see this clearly with non-sports games: bingo and lotteries (including the ones where you get to pick your own numbers) are totally random; chess, rather less so. We might also say that a single game of baseball or soccer is less indicative of which team is truly better than a single game of basketball or American football. This is in part because of the varying roles of random events and probabilities in these different sports.

This is part of the reason some people may prefer one to the other; and also why we might want to argue that some sports are actually superior to others. But I emphasize at this point (since is an on-going theme in this blog) that the “taming of chance” in a sport is only one of the reasons it might be considered more worthy as a sport (in the same way that chess is more worthy as a game than bingo).

The distaste of randomness is also one of the reasons we tend strongly to minimize the role of chance and luck in our post-facto narratives of great sporting triumphs or defeats. The day after a big game we will speculate about how the outcome would have been different but for this or that chance occurrence (a dropped pass, a blown call by the referee); but within a week the dominant narrative will lean heavily on the supposed physical, mental, and moral qualities of the leaders on the winning and losing teams. We do the same with our narratives about many important events in our lives and societies.

In the next post I will argue that the new OT rule will not be any fairer than the old rule, and will not even necessarily reduce the randomizing effects of the coin flip. But it will have other consequences that any football fan can applaud.