The ending of last night's Super Bowl couldn't've been more perfect as a demonstration of the point I was making about scientific thinking in football (and, you know, in that book I keep flogging...). First, on the positive side, you have New England's Malcolm Butler making the key play:
"I knew what was going to happen," said Butler, an undrafted free agent out of Western Alabama who said New England was the only team that gave him a chance to reach the NFL. "I don't know how I knew. I just knew. I just beat him to the point and caught the ball."
Perhaps Butler knew because he had seen the play before. During a scout-team practice, backup quarterback Jimmy Garoppolo beat Butler on the same play for a touchdown to Josh Boyce.
"We did and Josh got a touchdown in practice on it," Garoppolo confirmed in the Pats locker room. "It got him ready and he knew what to expect. Jumping the route like that, that's very impressive. That's all instincts."
That's the successful sort of mental model-building and testing. They had practiced against this, Butler recognized the set, and guessed where the ball was going to go, and guessed correctly. It happened very fast, but the process underlying that is essentially scientific: recognizing a pattern, making a model based on that pattern of what will happen next, and using that model to predict the future.
And then, there's the other side, namely the decision to throw that pass in the first place:
[Seattle head coach Pete] Carroll said the call for a quick slant to Lockette was dictated by the Patriots deploying their run-stuffing package.
"We were going to run the ball to win the game — just not on that play," Carroll said. "They had sent in their goal-line people. They had guys on the line of scrimmage. So we thought we'd spread them out with three wides. ... We had three downs and we had a timeout.
"This one didn't work out for us. In retrospect, we could have run it."
That one happened a little more slowly, but again, the underlying process was scientific: They saw who the Patriots had on the field, made a model of what they expected New England to do, and called a play that they thought would work based on that expectation.
In retrospect, it looks like a bad decision, but then again, had Malcolm Butler not been thinking like a scientist, ESPN might be full of talking heads banging on about how the wily Pete Carroll wrong-footed the Patriots on the key play, completing a pass when everybody north of New Jersey was looking for a run. In science, as in football, you win some, and you lose some...
So, a bit of an I-told-you-so for a snowy morning after the big game. There are a bunch of other places where the scientific look-think-test nature of the game was on display-- ESPN had a great clip of an interception Tom Brady threw, showing a linebacker for Seattle making a last-second adjustment to get in front of a pass to Rob Gronkowski. But the end of the game on both sides was as nice a demonstration as I could possibly have asked for of how football players and coaches are using their inner scientists.
Just don't ask me to explain the halftime show. Some things are still beyond the reach of modern science...
Here's a nice .gif of the play showing a head-turn that might've tipped the play.
"n retrospect, it looks like a bad decision, but then again, had Malcolm Butler not been thinking like a scientist, ESPN might be full of talking heads banging on about how the wily Pete Carroll wrong-footed the Patriots on the key play"
That's what I keep thinking. It was only a bad call because it failed. If it had worked, it would have added to Carroll's coaching mystique.
I wonder if a fade pass to the corner of the endzone would have been more effective. But a slant is an *extremely* safe pass. Carroll is being abused when really people should be congratulating Butler on a fantastic defensive play.
And, of course, everyone forgets about the boneheaded interception that Brady threw in the first half in the shadow of the Seattle endzone, which was an even worse decision.