My work on football tends towards the analytic side, and that carries forward to other aspects of my life. I have had an interest in economics since taking Denslow and Rush‘s classes at the Warrington College of Business back in 2003-04. With little football news other than pro day, I’m going to tie in a concept from economics to football to try to help deepen your understanding of how analytics can affect the sport.
You probably don’t think about prices much except when you have to pay them. Gas has been getting more expensive lately, so I’m sure that’s been on your mind. Either you’re wincing a little bit more at the pump, or you’re getting a little more smug by the day about your Tesla purchase.
Economists think about prices all the time. They are one of the fundamental building blocks of the discipline, and there’s an enormous amount of study devoted to them.
If the phrase can be used as the title of an album by an easy-listening critical darling, then you certainly know about supply and demand. You no doubt know how those conspire to make prices move. If supply rises or demand falls, prices tend to go down. If supply falls or demand rises, prices tend to go up.
Beyond just that, though, prices contain information. You know this intuitively, and probably explicitly. Products of higher quality tend to cost more. Therefore when shopping for something you don’t commonly buy, you probably expect things that cost more to be the better ones and vice versa. There’s a reason why the word “cheap” can mean both low in price and low in quality.
Beyond that, prices can act as informational signals. If the price of something rises, it sends a signal to the marketplace. Consumers will adjust by consuming less of that thing if possible, and the prospects for charging higher prices will attract new people and firms to produce that thing.
Take gasoline again. A sustained price increase is a signal to car manufacturers to produce more fuel efficient models. When combined with the drastic drop in lithium ion battery prices, it signals that they should make more electric vehicles. A sustained drop, meanwhile, means it’s time to go all-in on crossovers, trucks, and SUVs. The actions of the car makers ends up coordinated based on price information alone without explicit collusion or direction from a central planner.
This might sound obvious, but if it is, it’s only because the insight is so spot-on that it has permeated the culture. Researching and fleshing out how price signals work in the 1930s and ’40s was largely why economist Friedrich Hayek won his Nobel Prize, and many have built on his work since then. The value of price signals for market efficiency is part of why most capitalist economists oppose government policies that dictate prices like rent controls or price ceilings and floors.
So how do we apply this to football?
The game produces a lot of numbers, and many of them have the ability to send signals in the same way that prices can. People’s ability to pick up on those signals varies over time, but they do slowly adjust.
For decades and decades, yards per pass attempt has been higher than yards per rush. That should have been a clear signal to teams that they should pass more and run less. Service academies have their reasons for running niche triple option attacks, but most everyone else should throw more.
The 1980s and ’90s saw an increasing prevalence of pass-oriented offenses like the West Coast, Run n’ Shoot, Fun n’ Gun, and Air Raid, but it’s only been in the last decade or so that almost all teams in the major college and pro ranks have decidedly swung towards the pass. In 2004, Dan Mullen said he liked to spread the field to run; in 2020, Dan Mullen had the quarterback with the most pass attempts in the country. The signal has gotten through to nearly everyone. This shift has happened almost everywhere, and no centralized authority like Roger Goodell or a conference commissioner commanded it thus.
What signals are still out there, waiting to be heard? Play action is the one talking the loudest right now.
The most cutting edge research is showing that teams don’t use play action enough. Play fakes make the pass game more effective. They do this whether the team is good at running or not, and whether it runs frequently or not. There is some evidence that, if anything, play action might help the team’s rushing attack a little.
You probably do need to employ it in an intelligent way, of course. If you almost always use it on 3rd & long and basically never run in those situations, it’ll probably stop working so well. That was a story with the 2017 Gators.
But in general, no team has yet to hit the threshold of using play action too often. Everyone should probably use it more until some actually finds where that ineffectiveness horizon lies.
But we knew this sort of thing already, right? The NFL is famous for being a “copycat league”. A Google search for that phrase in connection with “NFL” yields more than 27,000 results.
Perhaps, but it’s not too hard to still find teams that should throw more, or go for it on fourth down more. Probably no one uses play action enough, and everyone could probably profit by going for two more often.
Kevin Kelley of Pulaski Academy in Arkansas, the coach famous for basically never punting, always goes for two on his team’s first two TDs of the game. He figures he can get at least one of them, and if he does get just one, he’ll still have the same 14 points that kicking two extra points would’ve given him. However if his team converts them both, he could give up a touchdown and still jump out to a two-score lead at 16-7 and put extra pressure on the opponent early. Kelley has nine state titles to back up his methods.
The increasingly pass-oriented nature of football shows that listening to signals found in analytics is profitable. The teams that succeed in the years ahead will, in part, do so by listening to more such signals.