Skip to Content

Revolutionizing Football

New computer modeling software could make gridiron coaches rethink their decisions and look to science for guidance.
August 31, 2006

A startup venture, EndGame Technologies, has designed novel computer modeling software to assist National Football League coaches with critical play-calling decisions–the kind that often determine the outcome of the game. Should a team punt on fourth down–or go for it? Or attempt a two-point conversion after a touchdown? The software, called ZEUS, is designed to answer such questions by calculating the consequences of each decision in a matter of seconds.

Football could be transformed by new software programs designed to determine the best play-call in tight situations. (Credit:

Head ZEUS researchers Chuck Bower and Frank Frigo want to revolutionize football by changing the way plays are called. Already, their statistics on past NFL games have revealed that teams consistently lose approximately one game per season by making the wrong play calls in critical situations. And with only 16 games in an NFL season, each win or loss is of paramount importance.

Football is a game of strategy and risk management. Each coach goes into a game with a plan, explains Dwight Smith, head football coach at MIT. Although every variable is considered beforehand, adjustments have to be made as the game progresses. ZEUS is meant to help coaches make those on-the-fly decisions. The software was created by mining historical NFL data and developing distribution curves of success rates for individual actions, such as how far a running back carries the ball.

Coaches could use ZEUS at any point in the game, by entering a set of variables, such as the score, field position, possession, down, and time remaining on the clock. Then the user enters two play-call options and the software analyzes each one separately, playing the game to its conclusion 100,000 times. During each iteration, ZEUS considers a different scenario.

All the decisions about play calls that occur after the initial play are based on historical NFL data. This data isn’t specific to a particular coach, though; it’s based on records from a wide variety of coaches. (The researchers call this algorithm the “generic” NFL coach.) In a matter of seconds, ZEUS displays the Game Winning Chance–the play that gives the team the highest probability of winning.

Unlike interactive football simulations, which require about the same amount of time to play out as a real game, ZEUS performs its calculations and recommendations in real time, displaying the computer’s results before the next snap.

According to Bower, the challenge for them now lies not in the technology, but in getting NFL teams to adopt it. “We have shown it to 10 NFL teams, [everyone] from managers to head coaches,” he says. “But at this point no one has stepped forward and said we are ready to pay for the product.”

The problem: ZEUS is still illegal under NFL guidelines. The league doesn’t permit computers on the sidelines or in the coach’s booth on game day. Although there are no restrictions on using high-tech analytical tools off the field, the traditional elements of strategy continue to be preserved on game day.

“We are very resistant to changing for the sake of changing,” says Brain McCarthy, NFL spokesperson. “Part of the appeal of the NFL is that it is man against man against elements and unscripted drama. When you add technology that could directly influence play on the field it has the potential of detracting from the overall product and enjoyability of the game.”

Bower and Frigo aren’t the only ones pushing for a technological revolution in the NFL. KC Joyner, called “The Football Scientist,” and a regular contributor to ESPN Insider, uses game film to track, tabulate, and analyze nearly every measurable statistic in an NFL game. These statistics, complied and explained in Scientific Football 2006 (pdf), use a performance-based metric system with the goal of “quantifying everything and putting it into perspective.”

Joyner believes that if an NFL team doesn’t take advantage of the latest technology, it will be hurt in the long term. “Nontechnological teams can still get good players, and some things will work, but as they get further behind the curve, it is going to catch up to them at some point,” he says.

Whether or not the league accepts ZEUS and teams decide to use it remains to be seen. For now, it can be implemented only in practice situations.

Bower and Frigo recently added customization–the ability to enter the characteristics of specific teams–and also developed a second application, ZEUS PPV (“Player Position Value”), which determines the value in incremental wins/losses per season of individual position players.

“Once [NFL] teams begin to embrace technology, the entire league will advance and be more successful,” says Joyner, “It is going to take one successful coach looking for an edge, willing to take a chance.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.