This is Throwbacks, a newsletter by me, Michael Weinreb, about sports, history, culture and politics, and everything in-between. Welcome to all new readers/subscribers, and if you like what you’re reading, please join the mailing list and share, on social media or through e-mail or however you feel comfortable sharing. (It’s still FREE to join the list: Just click “None” on the “subscribe now” page.) The best way you can help out is by spreading the word and sharing with others. I have set up payment tiers, if you wish to chip in and allow me the time to do a little more research on on these posts. You’ll get full access to the archive of more than 100 previous posts, as well.
I.
If you want to identify the precise moment when America’s most jockish sport began to wrestle with its inner nerd, you might as well start on an otherwise ordinary Sunday night in November of 2009. By then, we were six years removed from the publication of Michael Lewis’ Moneyball, and three years removed from the first MIT Sloan Sports and Analytics Conference, and all these wonkish and counterintuitive ideas had long trickled out of the coasts and into the real world.
Still, in 2009, it was hard to get our heads around the idea that professional football could have a working relationship to advanced math. Pro football was the sport that hoisted advanced math by its Fruit of the Looms and hung it, flailing like a Tri-Lambda, on a hook in the locker room. Football was pure id. Football was Dick Butkus and Marion Motley and Lester Hayes; football’s ethos was contained in the yawning gap between Jack Lambert’s front teeth. Football was John Facenda narrating as toothless young men braved frigid temperatures in order to render each other into limping old men.
And then, on that Sunday night in 2009, Bill Belichick—the coach of the moment, a cranky genius who was so utterly focused on the intricacies of the game that he could not suffer fools or sweatshirt sleeves—decided to go for it on 4th-and-2 in his his own territory against Peyton Manning and the Indianapolis Colts.
This was the first time that football analytics slapped us hard in the face. The Patriots were leading by six points; there were two minutes and three seconds left on the clock. The ball was at the Patriots’ own 28-yard-line. Every instinct in our reactionary souls screamed punt. But Bill Belichick, of all people, did not punt. Bill Belichick listened to the advanced math nerd in his soul. He chose to go for it, and Tom Brady threw a short pass to Kevin Faulk, and Faulk came up just short of the first down, and Peyton Manning led the Colts to the game-winning touchdown. The gambit had failed, and in response, Bill Simmons lost his mind.
You can feel Simmons, then the most popular writer at ESPN, wrestling with the future in that column—with the very notion that an entire realm of fresh thinking was about the eclipse the old paradigm. The greatest coach in the sport had assured us that football would never quite be the same; it belonged to a realm of revolutionary ideas now, some of them good, some of them bad, but all of them standing in opposition to what we believed were the hard truths about football itself. The paradigm had changed. And 15 years later, a lot of people are still struggling with it.
II.
In 2009, according to The Sporting News, NFL teams went for it on 4th down 557 times, which was a modern record. The number dipped slightly in future seasons, but then the number rose again and kept on rising, as high as 793 in 2021 and (by my rough calculations) nearly 800 this past season. At the top of that list were the Detroit Lions, who went for it on fourth down 34 percent of the time during the 2023 regular season, the highest rate of any team this century.
The Lions are coached by Dan Campbell, an overtly emotional and extremely likeable metalhead given to colorful speechifying. You want to see football personified? This is football personified:
But Campbell is also smart enough to figure out a way to marry old-school badassery with analytical theory. Balls to the wall and analytics, in this case, went hand in hand. Going for it on 4th down was married with a Detroit-ish ethos of coming back from adversity, again and again…
…and Campbell took it and ran with it, all the way to the NFC Championship, where it blew up in his face.
After Campbell chose to go for it twice and failed twice in a 34-31 loss to San Francisco 49ers, the backlash was inevitable. “Enter analytics, the effort by mathematicians to infiltrate NFL front offices (and to get NFL paychecks) without possessing traditional football skills,” wrote Pro Football Talk’s Mike Florio. “It worked in baseball, so they decided to move it to football. Even if it doesn’t work nearly as well in football as the mathematicians insist.”
This is the kind of worn-out argument that, as New York’s Will Leitch wrote, should have long ago been rendered ridiculous, the way it has in baseball. The difference, of course, is in the culture. Baseball has always appealed to numbers nerds. But there’s a faction of people who still expect football to wear its reactionary soul on its sleeve, which is why even something as harmless as a tight end dating a pop star has somehow become viewed as a catastrophic display of the sport somehow caving in to its soft side. And those same people insist that coaches like Campbell cannot simply bypass the sport’s conservative tendencies by buying into the math nerds’ thinking—because if it works, then the mathematicians own pro football, which these people view as the last bastion against America’s slow curdling into socialism.
III.
Fox’s Greg Olsen is the best color commentator in professional football. He also has a knack for uttering Zen-like statements that really only make sense if you meditate on them.1 After Campbell went for it on 4th down against the 49ers and failed a second time, Olsen said, “You can’t judge it because it doesn’t work,” and at first I thought to myself, Dude, what the hell does that even mean?
But after a while, I realized exactly what it meant. I started thinking about the inevitable pile-on of Campbell on social-media and in the sports-talk industrial complex; I started thinking about how, as The Atlantic’s Derek Thompson wrote, the fundamentally judgmental structure of our markets has essentially scared off so much original thought in the past decade or so. And I presume this is is what Greg Olsen meant: That sometimes, even superior ideas fail, but it doesn’t mean they aren’t still good ideas. It’s that our fear of failure has crowded out so much room for experimentation that we tend to dismiss them without a second thought.
Maybe the problem, ultimately, is not that Dan Campbell went for it. Maybe the problem is that we are so fearful of change that we don’t go for it often enough.
This newsletter is a perpetual work in progress. Thoughts? Ideas for future editions? Reply directly to this newsletter, contact me via twitter or at michaeliweinreb at gmail, or leave a comment below. If you enjoyed this newsletter, please join the list and/or share it with others or consider a paid subscription.
From the Archives
On the 1985 Super Bowl at Stanford University, those Apple-branded seat cushions, and the Super Bowl (potentially) becoming staid.
Think Different (January, 1985)
This is Throwbacks, a weekly-ish newsletter by Michael Weinreb about sports history, culture and politics. Welcome to all new readers/subscribers, and if you like what you’re reading, please subscribe and share. (It’s still free: Just click on the free option on the “subscribe now” page. If you do wish to contribute, see below.)
Earlier in the game, with the Lions in the midst of a slow collapse, Olsen uttered something about how in football, when things get bad, they tend to get worse. It was both cliched and prescient.
Another excellent column - thank you. In any sport (or business), whether to “go for it” is and has always been a risk assessment, right? But now there is more data to support the decision than there used to be. I like the aggressiveness of the Lions and it reminds me of when the Steelers started going for two with some regularity (back when they used to actually score touchdowns 😁). One significant aspect to this decision-making process should not only be what the percentages say about your likelihood of making it but also what the downside risk is if you don’t make it. The late game run by Detroit is a great example. I’m sure the data showed they had a great chance to make it. But did they give enough thought to the downside risk if they didn’t? We all know they had to burn a timeout, and the lack of that timeout later was a major factor in them losing the game. In the last couple of years I’ve seen teams in college and in the NFL go for two when they’re behind by two scores. An extra point would make it a one score game but if they miss the two point conversion, then it’s still a two score game. And teams have gone for two and that’s a data driven decision that I don’t understand. If you get the PAT, you can keep playing. But if you miss the two, the game’s over. One of life’s mysteries!