The dangers of consequentialism

Everyone is talking about Bill Belichek's decision to go for it on 4th-and-2 from the Pats' own 28 with around two minutes left (sorry Mike). And most people (including the NBC commentators speaking three-and-a-half minutes after the game) have concluded it was a bad decision. But the only reason they offered as to why the call was bad is that it did not work. Had it worked, it would have been called gutsy and a brilliant decision.

This is the problem--in both sports and law--of pure consequentialism, in which the a priori wisdom of a decision is evaluated solely by the outcome. But the decision to go for it cannot be right or wrong based solely on the result. The result is good or bad; but the decision must be evaluated independent of the outcome. Evaluating a decision as right or wrong must be based on the quality of the reasoning that went into it. Since 2001, the Pats have converted 63.5 % of attempts on 4th-and-2-or-less, a higher percentage when Brady is the quarterback. And on a day in which the offense racked up more than 400 yards and generally had moved at will, those sound like pretty good odds. Plus, in the situation, the Colds defense would be particularly worried about jumping offside, so their aggression may be ever-so-slightly-restrained. All-in-all, it strikes me as a highly unconventional, but hardly unreasonable or reckless decision. And, in fact, the play worked, except Fault did not catch the ball cleanly, thus losing forward progress as to the spot.

Gregg Easterbrook writes the Tuesday Morning Quarterback feature for espn.com and he is constantly arguing that coaches should go for it on 4th-and-short, particularly around midfield and deep in opposing territory. I am looking forward to what he has to say about this one.

Update:

The folks at Advanced NFL Stats, who know stuff about mathematical analysis that I don't, say that Belichek made the right decision. (H/T: Deadspin) The success rate on 4th-and-2 is 60% and teams score a touchdown from the opponent's 28 with 2:00 remaining approximately 53% of the time. This puts the Pats in a statistically better position than punting would have.

Further Update:

Easterbrook weighs in: Belichek was absolutely right to go for it (although he questions some of the other calls and moves, particularly the call on 3d-and-2). Easterbrook also takes on one of the sillier memes about this--Tedy Bruschi saying Belichek showed a lack of faith in his defense--by pointing out that what Belichek did was to show faith in his offense to get two yards on a day in which it averaged more than 6 yards a play.

The "lack of faith" meme rests on the assumption (which Easterbrook has been fighting) that going for it on 4th down is so far out of the norm that it is justified only in special circumstances. The assumption is that an offense really only has three downs to get a first down ordinarily and to use an additional down shows desperation of some sort--here, lack of faith in the defense. But if the mindset is that four downs means four downs and the percentages favored New England, there was nothing insulting to the defense here, just a faith in the offense within the normal rules of the game (four downs to try to get a first).

Posted in: