|Notes and Observations, Week 2: Defense Shines as Patriots Defeat Vikings 30-7||Connelly Top Ten: Patriots Average Team – 10-6||Analyzing the Patriots Strong Safety Rotation||College Football Week 3 Roundup: BC Upsets USC, UMass Loses Heartbreaker|
This is in response to John, who thinks that going for it on 4th and 2 is too risky, and in accordance with George, who praised Bill Belichick for the move.
The math has been done on Belichick’s already-infamous decision to go for it on 4th down on Sunday night. But, there are still plenty of doubters out there, clinging to conventional wisdom and factors they claim are un- or under-accounted for. As a stats-oriented fan, my initial inclination is always to demand proof – to call out methodology and make a stand for logical consistency.
Unfortunately, that’s not a very convincing or compelling case, and it’s a bigger argument than we need to be having here. What I’m really calling out is a sports world-view. I think that, in all walks of life, but especially in sports, discussion needs bedrock. It needs a common set of agreed-upon truths. Those are/should be the truths of mathematical and statistical analysis. Most people would rather analyze with their gut, or with their experience, or with their long history of watching sports (note that those are all the same thing), because it’s easier and more accessible.
The truth is that if you’re trying to sway a conventional sports fan on the topic of Belichick’s 4th down decision, you’re probably trying to convert them to a stats-oriented sports world-view. Even if some of what you say takes, you’re not going to win that argument. Nobody accepts change that quickly.
If you want to convince a conventional sports fan that conventional wisdom is wrong, then you need to argue conventionally.
In that vein, I’m not going to cite numbers, or fiddle with some model, I’m going to provide a conventional response to conventional wisdom.
Belichick’s 4th down decision is this simple: Did the Patriots have a better chance of gaining 2 yards on 4th down and keeping the Colts from gaining ~30 yards in 2 minutes if they didn’t make it, or attempting to punt and keep the Colts from gaining ~70 yards in 2 minutes. Obviously you can play with the numbers based on how good you think the punting situation would be, but this is the proper construct for the debate.
Well, the Pats make two yards at a pretty good rate. Even with the line stacked on a 4th down, this is thought to be the best offense in the league by some. They’d already scored 34 points and were able to run and pass the ball all night.
On top of that, even if they don’t make it, we’re talking about 30 yards and a touchdown in 2 minutes. That’s not a gimme for any team. Momentum might swing a bit if the ball is turned over, but does that actually affect the play on the field? To a measurable degree?
If you punt, you leave them with 70 yards and a time-out. The thing to remember is that the last 30 yards of that 70 are likely going to be the toughest. The defense loses the prevent and uses the back of the end zone as a 12th defender. They’ve got to gain that 30 either way, although there’s less time constraint and that morale boost when they do it via a turn-over.
With that conventional analysis, let’s draw a conventional conclusion: The Patriots offense is really good and was playing well all night. With the difference between the two outcomes being 40 yards of prevent defense, which had already been shred in the two previous drives, it’s tough to imagine any scenario where the attempt to go for it was less likely to produce a win than punting.
I’d also like to make a point about “risk.” “Risky” is probably the adjective most commonly (mis?)used to describe this decision by Belichick. How is it risky? I suppose it’s risky in the sense that Belichick was going against the norm by going for it on 4th down, and so he sort of put his neck on the line (as if he would ever be in danger). I also suppose that the decision produced a single higher-leverage play than would otherwise have occurred in the game (although I think a few of the downs on the subsequent drive by Indy might have actually been even bigger in terms of win probability). But, the context it seems to be regularly used in is that the move itself was risky. It was risky to go for it. That this decision introduced more risk for the Patriots than there otherwise would have been. Conveniently we’re not really going to identify what that risk would be, but the move was “risky”.
I don’t understand that. Why isn’t anyone saying it would be risky to punt the ball away to a super-hot offense without at least trying to let your own super-hot offense win the game for you?
By definition, the decision that introduces more risk of losing the game is the inferior decision, so by calling the move “risky,” what we’re really seeing analysts do is make an end-run around actually having to make an argument about why the decision was wrong. If you can call it “risky” by virtue of the fact that it’s not done very often, and then use that charged adjective to imply a lower chance of success (or more accurately, a higher probability of failure), then you’ve already slanted the argument in favor of the conventional wisdom. This behavior is how things get perpetuated.
In the end, the biggest problem is unoriginal thinking. Conventional wisdom exists because people don’t want to continuously re-travel the same paths that lead to conventional wisdom’s teachings. As a result, it becomes short-hand for the lazy that want to use it as a postulate. Sometimes that’s okay, and sometimes it’s even necessary if we want to have basic discussion.
But it doesn’t relieve us of the need to, every now and then, eliminate our assumptions and biases and actually step through the decision making process free of short-cuts.