In Defense of Belichek

• Posted by Adam Warner
• on November 16th, 2009

So if you fell asleep (or don’t watch football) you missed perhaps the most fascinating NFL game ever from a statistical perspective.

Here’s the story.

The Pats are beating the Colts 34-28, but face 4th and 2 on their own 28 with just over 2 minutes left. The Colts, having just rolled down the field in a blink to pull within 6 1 timeout remaining. Belichek decides to go for the 1st down, and they don’t make it. The Colts then proceed to drive easily for the go ahead TD, and kill the clock in the process. Good triumphs over Evil. Ewoks far away are dancing, as Yoda, Obi Wan and Anakin look on approvingly. Belichek of course gets pilloried for the decision. As Tony Dungy and Rodney Harrison point out that you have to punt the ball away and trust your defense, and that of course is the conventional wisdom in that spot.

Is it correct though? I would suggest no, or at least it’s not such a terrible call.

There’s no way to know the exact odds of everything here, but for argument’s and simplicity sake, let’s say the Pats have a 50% chance of getting the 1st. And if they don’t make it, the Colts have 100% chance of scoring. So in effect, the entire game rests on whether the Pats make it or not, and since it’s 50/50, it’s an expected gain of 0. But what if the Pats punt? The Colts get the ball on about their own 35 needing the same TD. They have 2 minutes, 1 timeout, and Peyton Manning. If the odds of scoring there are over 50%, then Belichek actually made the statistically correct decision, it they’re under 50%, then Belichek messed up. There’s no way to get an exact call on this , but I will say that in effect, Belichek only gambled 35-40 yards or so, the net gain from the punt. Do those 35-40 yards tilt the odds of the Colts getting a TD from 100% to under 50%, with the clock really not a factor? I would guess not. Belichek flied wildly against NFL conventional wisdom as there are few coaches who would even remotely consider trying for the 1st down. But without the benefit of hindsight, hard to say this was such a bad idea.

I didn’t totally agree with that Game Theory article in The Economist from a few week’s back, but Belichek’s decision was a living example of what football stats wonks say coach’s should do, but don’t because they put themselves on the line (if the Pat’s punted and the Colt’s scored, we wouldn’t hear a wisp of suggestion Belichek should have gone for it). Consider the Eagles-Packers playoff game from 2004. The Pack faced a similar decision late in the game, except in more favorable circumstances. They had a 4th and 1 on the Eagles 41, up by 3. Like the Pats yesterday, a 1st down all but ices the game. They punt it away, and netted all of 21 yards. The Eagles drove down for the tying FG, and then won in OT. The error of NOT going for the 4th was way worse than going for it yesterday imho. Belichek was accused of arrogance, but was it that or just weighing the odds a bit?

Just to balance the sort-of praise, we need to also criticize the decisions that led up to 4th down. The Pats blew their last 2 timeouts on that drive with the ball and the lead, ruining both their chance to review the spot on 4th down, and the ability to preserve some time on the clock if the Colts score. They also should have likely run on 3rd and 2 if they had a notion of going for it on 4th down.

As for Tom Brady, he’ll move on.

The information in this blog post represents my own opinions and does not contain a recommendation for any particular security or investment. I or my affiliates may hold positions or other interests in securities mentioned in the Blog, please see my Disclaimer page for my full disclaimer.

blog comments powered by Disqus