Related Articles
A meaningful data revolution in football probably wasn’t possible in 2002 as Billy Beane’s methodology later labeled Moneyball changed baseball’s talent evaluation, player recruitment and valuation efficiencies forever by going against conventional wisdom. The complexity and fluidity of football held the sport back in that regard while other segmented sports adapted the model. But 15 years on, technology has drastically advanced the possibilities, and data science is tackling the world’s game like never before.
The Numbers Game: How Data is Changing Football, a documentary released Friday by FourFourTwo Films, features the Oakland Athletics executive vice president of baseball operations and subject of author Michael Lewis’ bestselling 2003 book. It addresses the growing influence of data-driven analytics in football – a sport that’s not exactly foreign to Beane.
“It’s a very dynamic sport,” said Beane, part of the conglomerate that purchased Championship side Barnsley at the end of 2017. “Baseball is very stop-start and it lends itself to measurement, but on the flip side there are a lot more events going on during a football match than there are in a baseball game – and anybody who is well versed in modelling, whether from a computer science background or a mathematics background, will tell you the more data you have, the better you are able to put the models together.”
That’s something another of the film’s subjects knows something about. FourFourTwo turns to STATS director of data science Patrick Lucey to weigh in on the future of big data and deep learning in football.
“Football has actually been collecting the most data for the longest time,” Lucey said. “But football is the most complex sport. It’s low-scoring, it’s continuous, it’s time-varying. It’s very strategic. It’s very subjective.
“So, just say you and I were analysing the game. We could come up with different opinions. When you compare it to other sports like basketball – high-scoring. Tennis and American football, they’re segmented. Baseball, it’s segmented. It’s very easy to do the analysis. You have a lot of data points.
“So the key for football is actually to come up with the right language and ask the right question for specific things. How was our formation? How did we press? How were we on set pieces? Did we attack via the counter attack? All these different things, we have to learn directly from data.”
That’s particularly true for clubs without the luxury of bulging pocketbooks. The film explores the use of data by clubs such as Forest Green Rovers, a STATS partner playing in English Football League Two, and acknowledges that the high-stakes nature of promotion and relegation in football makes things that much more consequential.
“The recruitment side for a small club is really, really key,” Forest Green manager Mark Cooper said. “It’s important that we’re different. In January, every club will be after the same players, and probably we can’t compete for those players that everyone’s after, so we have to find other types of players. We have a different way of playing, and we have to find players that can fit into that. And we have to use the data for that.”
Looking forward, what’s already been achieved is just the beginning with ever-advancing methods of predictive analytics and player assessment in the works.
“There’s lots of cool stuff that people haven’t thought about,” Lucey said. “The idea of ghosting – being able to simulate plays that you haven’t seen before.
“You could have an example of a play and you can say, ‘Well, how does this team defend in that situation? What happens if I switch that player with another player? How does the outcome change? In terms of just body shape – where’s the player facing? Are they making the right decisions? In terms of injury analytics, player load, fatigue, how’s their technique changing over time.’ Now, using deep neural networks, we can actually simulate these things.”
It amounts to the world’s most popular sport finally being able to fine-tune the problem solving Beane took on nearly two decades ago.
“The goal gets paid for in today’s world. Which, shouldn’t the guy who created all those things?” Beane said. “Measuring those things is really the challenge. Giving proper credit to player performance is what we’re all trying to achieve, not just in baseball but in every sport.”