Numbers never lie, only sometimes

The web has been abuzz with intelligent discussion about Bill Belichick's decision to go for it on fourth-and-2 with 2:06 left. The Pats missed it and lost to the Colts, 35-34. In a word, they were daggered.

Many people far better with numbers than us have concluded Belichick gave himself a better chance to win going for it than he would have punting. Here's one of the best arguments.

It has definitely forced Mr. Buckner and I to reconsider. However, I tend to side with Buckner for the following reasons, some of which he has already stated:

The numbers factored in consider these were both typical offensive series (Pats and Colts). These WOULD NOT have been typical series. The Pats have a great offense. Also, with it being their last chance, the Colts would have four downs to gain 10 yards, not three, thus increasing their probability to advance the ball downfield. Typical probabilities of them gaining 70 yards vs. them gaining 29 must be adjusted accordingly.
Momentum is not a word statisticians love. It's similar to the idea of being clutch, which many of us agree is a fool's crutch. However, in this case momentum cannot be denied. The Colts had it by tearing down the field on the previous possession. It only increases once they receive the boost gained by working with a short field.
For an adequate answer to this questions, probabilities must somehow factor in several intricacies. Did the Pats have the option to run on fourth down? (Not really.) Why? Well, were they missing their top short-yardage back (Sammy Morris)? Yes. Was their defense gassed? Probably. Etc., etc.

So, yes, while many smart people have continued the debate today, I'm not sure simple numbers answer whether Belichick made the correct decision or not. It's much like baseball. Numbers can tell us the correct decision to make over the course of a long season or a long series. In baseball, with so many games, so many at-bats, so many pitches, those decisions should be uniform. But in football, with so few games, so few plays and so few decisions like these, more must be taken into account.

The situation has a much greater impact in football than it does in baseball. Is Brett Favre better than he was 10 years ago? No. But in this situation, is he having a better season? Absolutely. But don't Adrian Peterson, an enormous line and a stout defense help swing the numbers his way?


  1. It's tough to argue when the argument presented is entirely in the realm of the intangible. As for me, I'll always side with the tangible. Individual statistics in football are obviously extremely context dependent. I still believe that given 3rd and 2 or 4th and 2, the Pats convert 60% of the time easy. We're talking about a team that had put up 34 points on the night, and as long as you concede that the number is somewhere around 50%, Belichik made the right decision.

  2. It's funny how a mere 24 hours after I was ready to destroy Belichick, I now find myself thinking that people doing that are a bit shortsighted.

    I still think it was the wrong decision, but I think it's more wrong to say it was ridiculous than it is to defend it.

    That said, I want to draw attention to three reason why I think the 40 yards in field position (the expected gain from the punt) was more critical in this particular scenario than just about any other scenario where you gain 40 yards of field position:

    1) The clock. If the Colts had the ball on their own 30 at the two minute warning with one timeout, they would have been slightly pressed for time. Running the ball would not have been an option, and if the drive didn't get off to a fast start, the clock would have become an even more important factor. Instead, having the ball at the Pats' 28 at the two minute warning allowed the Colts to bleed the clock down at their choosing.

    2) The use of four downs. The shorter the distance you need to go, the bigger the factor of the four down availability. In all liklihood, needing only 28 yards is only going to mean two first downs. Having four downs at your disposal to get two first downs is a very favorable situation.

    3) The shock factor. We can't measure it. We can just observe, and here I think we can use a little bit of Monday Morning quarterback analysis. The Patriots defense looked totally helpless on that drive, and I think part of it was because they were truly unprepared for that situation. Would they have put up more of a fight (regardless of the 40 extra yards of cushion) had they had the entire field to defend? I think so.

    All that said, I really wish Belichick would expand a little bit more on his thinking. So often, you can't even blame coaches and players for offering cliches.

    When media members like Suzyn Waldman ask Joe Girardi questions like "Was Hideki Matsui locked in against Pedro Martinez?" you can't even expect anything interesting.

    Here there really was a chance for Belichick to possibly offer some specific insight that could have been fascinating. Not surprisingly, he disappointed once again.