Belechick Understands Probability Better than Media
Like him or loathe him, Bill Belechick flies his own path. His late fourth quarter decision to go for a first down rather than punt on 4th and 2 from his own 28 has sent media columnists into a Monday morning tizzy of epic proportions. CBS Sportsline’s Pete Prisco writes about the “unusually dumb decision.” SI.com’s Peter King seethed
All in all, I hated the call. It smacked of I’m-smarter-than-they-are hubris. Let Manning, with the weight of the world on his shoulders and no timeouts under his belt, drive 72 yards in two minutes, with his mistake-prone (on this night) young receivers and the clock working against him. Sure he could do it. But let him earn it. This felt too cheap. It was too cheap. Belichick’s too smart to have something so Grady-Littlish on his career resume, but there it is, and it can never be erased.
Of the major media outlets I looked over, only Yahoo’s Shutdown Corner link even flirted (and too quickly rejected) any real analysis that goes beyond the conventional — you don’t go for it from your own 28 — kind of “analysis.”
In contrast, Advanced NFL Stats breaks down the key elements, showing that the decision makes sense quantitatively:
With 2:00 left and the Colts with only one timeout, a successful conversion wins the game for all practical purposes. A 4th and 2 conversion would be successful 60% of the time. Historically, in a situation with 2:00 left and needing a TD to either win or tie, teams get the TD 53% of the time from that field position. The total WP [win probability] for the 4th down conversion attempt would therefore be:
(0.60 * 1) + (0.40 * (1-0.53)) = 0.79 WP
A punt from the 28 typically nets 38 yards, starting the Colts at their own 34. Teams historically get the TD 30% of the time in that situation. So the punt gives the Pats about a 0.70 WP.
One can quibble with the exact numbers, but the framework for the analysis (rather than the media’s reliance on mere convention) is correct. What the ANS framework highlights that the media misses is that their is a risk to punting just as there is a risk to going for it. Pushing a risk farther down the line doesn’t make it disappear or make it less. As ANS notes, the numbers probably work out more in favor of going for it when customized for this particular game.
I went back and looked at the 4th quater stats: the Colts had already gained 175 yards in the quarter with two drives longer than what would have likely been needed here in clock times of 2:04 and 1:49 (without employing any special clock-stopping strategies). Prisco and former Patriot Teddy Bruschi thought it “insulting” to the defense — well, exactly what about it’s 4th quater performance would have inspired confidence. Just as important, the Patriots had already gained about 470 yards of offense in the game, nearly 7 yards per play.
Rather than “stupid,” “dumb,” or “insulting,” this is is the kind of decision making that has made Belechick better than most NFL coaches. Risk aversion, media response (even if coaches deny it), or lack of analytical skill drives many coaches toward applications of “conventions” even when those conventions don’t make sense. Belechick is willing to go with the analytics and live with it. After all, it’s not how it turned out after the fact that makes it a good or bad decision, it’s the likelihood going in.