The NFL's Groupthink Problem Is on Full Display at the Draft
In a spacious auditorium on Georgia Tech’s campus back in February of 2019, Alex Wainger, a data engineer from Facebook, pitched a panel of NFL experts on a good idea: Eliminate the rule on punts where the kicking team has to remain on the line of scrimmage until after the kick.
He looked through the previous three years of GPS data on every punt in the NFL and correlated that information with injury data over a similar period of time. The elementary version of his conclusion is that this rule was added at a time when the league was somewhat desperate for attention and wanted to microwave scoring on punt returns. It doesn’t need that anymore. And by getting rid of the rule, you’re going to have more fair catches. By having more fair catches, you get fewer injuries. Fewer injuries are better than more injuries (in practice, it looks a little like what the now-former XFL did on kickoffs).
He won what was called the NFL’s “first and future” competition for this. The league stages these competitions as a way to open source good ideas and innovative products (and, as one contestant pointed out, probably save a lot of money since the cash prizes are less than what it would cost for someone with data experience or industrial engineering experience to actually spec it out). It’s the less-famous version of the NBA’s data hackathon, which Wainger has also won before for a data-driven exposé on the way players actually tend to make out-of-the-norm choices during the playoffs as opposed to the regular season (or “hero ball” as they call it).
After the presentation, at an open bar reception, Wainger told me he was approached by some engineers from the league who were curious about his findings and excited to take the information back to 345 Park to run the numbers. While it’s not going to become a rule tomorrow, or next week, or even in the next five years, it may someday become one. Even if the NFL wasn’t the first league planting its flag on data-driven decisions, it seemed, at least to him, genuinely interested in doing it now.
“You normally think of them as a little further behind baseball or basketball in terms of using analytics for stuff,” he said, when asked if he was surprised the NFL was holding a data analytics competition in the first place.
This is a good thing. As leagues like the NBA and MLB go through their analytical renaissances (or second renaissances), the NFL could use a boost from an optics perspective there.
We’ll get to more of that in a minute.
But first, let’s talk about how owners are wrecking football.
***
Back in November I was lucky enough to write an essay for SI’s Innovations issue on the state of the NFL as an innovator, and I made calls across the spectrum: coaches, players, data wonks and others. The end product centered on the very pressing issue of technology and reviewable pass interference, which left some of the best conversations I had about the actual future of the NFL on the cutting room floor. Instead of condensing them for the space of a magazine article, I decided to expand on them here.
University of Chicago professor Richard Thaler, a Nobel Prize-winning behavioral economist, and Cade Massey, a professor at Penn’s Wharton School of Business whose expertise lies in why we make certain decisions, wrote an influential research paper in 2005 called “The Loser's Curse: Decision Making & Market Efficiency in the National Football League Draft.” It’s the reason a lot of good teams value mid-round picks today. They famously pitched their idea to Washington owner Dan Snyder and then watched him torch the team’s draft capital to trade up for Robert Griffin III in 2012.
But Thaler and Massey have individually challenged a lot of the previously held notions that we’ve had about the NFL. For example, there really is no general manager who is better than another at picking players. Tropes like these—the idea that there are personnel wizards and that higher picks are better—have rusted to the point of corrosion and are bringing down teams that could be facing a sort of breaking point. As Massey points out, it’s so much easier for an organization to get criticized when it operates in a way that is different from the norm, so many teams’ behaviors tend to cluster together. And this thought process affects everything.
Thaler cautions that there is stupidity everywhere, across major entities, both sports and non-sports related—some baseball teams still bunt, basketball teams have just recently realized that a three-pointer is worth more than a two-point shot, and Apple continues to roll out disastrous iOS updates—there is simply more of it in the NFL.
His process researching and presenting “The Loser’s Curse” gave him some first-hand insight into the broken loop that many NFL teams are operating in, which almost always starts at the top with a new class of owner and thinker. Thaler says the typical owner today is someone who purchased the club with a “spare billion dollars he made doing something else” and often vacillates between being not involved at all and way, way too involved. Think, then, about the atmosphere that creates.
You have two organizational pillars, the general manager and the head coach, who are left to maintain vast swaths of real estate by themselves. The likelihood that either of them try something different and risk losing that high ground of authority in the eyes of the owner is small. For example: The general manager who bundles and trades his high draft picks for more chances at finding better, cheaper talent in the middle rounds runs the risk of the head coach telling the owner that he’d be able to win if that crazy person in the front office didn’t trade away all of his chances to select “premium” talent at the top of the draft. And the coach, who constantly goes for it on fourth down and rarely punts, runs the risk of the general manager telling ownership that he’d be able to win with the players he drafted if the crazy person down on the sidelines knew how to manage a football game.
“I don’t know any CEO who decides what the IT system ought to be,” Thaler said, reasoning that coaches and general managers should focus more narrowly on their best skill sets and delegate the rest of their responsibilities. “They’ve got a guy for that. When they can’t turn a computer on, they call the IT guy to come in.”
Both trading back to accumulate more picks and going for it more often on fourth down in the right situations (broad examples) are analytically sound and have proven successful in the rare occurrences where ownership, coaching and personnel are tightly aligned. The Ravens, for example, have not only invested in an offense that, by NFL standards, may not be totally conventional, but have talked openly about their desire to learn more about fourth down strategies and have won on some critical middle-round draft picks. And while Massey and Thaler’s work on the draft has become more accepted over time, there are still countless teams interested in making a splash on draft day, willing to sacrifice valuable future capital for a momentary swoon. The Chiefs, Bears, Eagles and Rams have all traded up to draft first-round quarterbacks of varying abilities in recent years. And watch an NFL game each Sunday where a team is trailing by 30 points and still, inexplicably, punts on fourth down.
“I’ve been around a couple of teams closely enough to watch what is going on and the coach and the GM both make end runs to the owner and both of them know they’re going to get fired at some point,” Thaler said. “What is it that John Maynard Keynes said? It’s better to be wrong conventionally than to be right unconventionally.”
Massey, whose expertise lies in the way people make decisions, brings up the example of the weatherman. Every day, the weatherman knows whether he or she has won or lost based on the accuracy of their prediction. A huge loss—a prediction of a sunny day that turns out to be a torrential downpour—forces them to reconsider the process. They’re calibrated for day in and day out success.
Football teams win and lose every week, but it is far more difficult to compute the granular, day-to-day data on how an individual in a position of power is performing. Plus, a general manager can blame a bad draft pick on poor coaching. Or, a coach can blame a bad season on poor talent acquired by the general manager.
And at the end of the day, how many owners are actually committed to putting their hands in the muck and sifting through all of the rubbish to find the truth? Thus, how many owners are committed to building the type of stable environment where innovation can blossom and flourish, and how many just opt for the quick fix by firing one of the two people in power?
“It’s keeping score,” Massey said. “You can’t learn without keeping score.”
***
Considering what we know, is it scary at all to think that the same people who struggle to internally police their organizations (or lack the desire to) are the same ones who simply hire new head coaches and general managers every few years hoping they’ll fix the problem?
Here is a list of the number of head coaches remaining from each coaching hiring cycle between 2013 and 2016:
2013: 1
2014: 2
2015: 0 (!)
2016: 1
The point of this is to show that, both from a macro level and a micro level, the NFL is in danger of some serious stagnation.
Which is why we come back to Wainger and the NFL’s first and future competition. Over the years, the competition has played host to some seriously bright people, people who have taken a data-driven approach to solving big problems staring the NFL down.
While it’s unclear the ultimate influence they’ll have, the continued process of creating a forum for these ideas and, at least internalizing them, could go a long way toward combatting the current state of play: A cadre of owners sit across the ideological spectrum, both in terms of their vision for the league and level of involvement in their respective teams, aggressively pushing each other to the middle.
More often than not, the result of decisions they make on a grand scale are blasé compromises that don’t get to the heart of a problem (see: pass interference reviews). Those processes trickle down into their respective teams and, eventually, trickle down into your life, as you consume a mediocre on-field product that could be so much better.
Like the rule that Wainger proposed on punts, things may not change next week, next month or next year. There really isn’t much of an incentive to change at the moment because, despite all of its failings, the league is still an unstoppable behemoth. But when that creeping stagnation eventually strangles football, at least they’ll have a good idea of where to turn to save it.
• Question or comment? Email us.
