Big Data: The Winning Formula In Sports

The 2011 film Moneyball (and the earlier Michael Lewis book it was based on) stoked public interest in the way data are used in sports, but since we learned about the way  Oakland Athletics general manager Billy Beane used analytics to reverse the fortunes of his ailing MLB team, development of data approaches to sports has continued to grow and evolve.

Arsenal’s English defender Calum Chambers (Photo credit Oli Scarff for AFP/Getty Images)

Teams and the analytics providers have have come up with increasingly sophisticated ways of monitoring and capturing ever-growing volumes of data. Cameras, sensors and wearables record every aspect of player performance. Managers, coaches and athletes are using data to dictate calorie intake, training levels and even fan interaction in the chase for better performance on the field.
Dallas Mavericks coach Rick Carlisle, announcing that its analytics department had been expanded with new recruits, recently said: “We’re going to be a better team this year – we know that by the analytics.”


In the UK, Premier League soccer team Arsenal has recently invested millions in developing its own analytics team to make better use of the data it is now collecting. One important data stream comes from 8 cameras installed around its stadium to track every player and their interactions. The system by sports analytics provider Prozone tracks 10 data points per second for every player, or 1.4 million data points per game. The system is also used to monitor 12,000 soccer matches around the world, which are all analyzed using automated algorithms as well as manual coding of every interaction with the ball to increase the accuracy and value of the analysis.

Coaches and players can now have access to information such as ‘all passes by Lionel Messi that were unsuccessful’ or ‘all successful tackles by Christiano Ronaldo’. Of particular interest is the analysis of ‘off ball’ events. Most players only spend a fraction of the game directly interacting with the ball and the vast majority making runs, getting in dangerous positions or disrupting the flow of the opposing team. A lot of untapped potential can be found in that data.

In addition to video analytics, wearable devices are increasingly used to track performance even more closely. In American football or rugby for example, injury levels have been reduced in the professional game due to wearable sensors that monitor the intensity of activity and impact of collisions, and compare this to historical data to determine when a player might be in danger of overexerting or injuring themselves. Teams in many sports have found that the cost of implementing analytics programs can quickly be recovered if they save the team having to pay for expensive players to sit out a season with injuries.

At present, the NFL does not allow GPS trackers to be worn during games, but in training sessions they are a standard piece of the kit for every player. The same is true for soccer, where the international governing body FIFA had long opposed the use of wearables during games. However, FIFA has recently announced changes including allowing players to wear monitoring equipment during matches for the first time. The International Tennis Federation also recently relaxed rules about the use of new sensor-equipped tennis racquets such as the Babolat Play during tournament.

These racquets record the power, degree of spin and point of impact involved every time they come into contact with the ball. All of the data can then be downloaded to a computer or even a smartphone and compared with other users’ data to help a player get insight into how they can improve their game. The Swingsmart works on similar principles but is designed to be attached to golf clubs.

Data collected during professional games is also immensely valuable to the armies of sports scientists, nutritionists and medical personnel involved in the industry. Having detailed access to records of player performance and activity helps to assess the impact of training schemes and diets, and predict recovery times after injuries.

I have recently worked with an Olympic sports team that also analyzes how well their athletes sleep. Data is collected from wearable devices athletes wear at night and is then correlated with track performance. At important competitions, the coaches can now assemble their team not only on past performance but also on the level of sleep team members had the nights before the event.

It isn’t just elite athletes who are being monitored – its fans, too. We’ve got use to venting our frustrations on Twitter, and following and interacting with our favorite stars on Facebook. A partnership between HP and NASCAR involves monitoring social media activity of fans during races. NASCAR says that one race generates an average of a million social media posts, and that these can be mined for insights which will help the sport as a whole as well as individual drivers or teams.

Some critics of the union between Big Data analysis and sports have said that it takes away from the fundamental principle that sport is about humans competing on the track or field. Isn’t there a danger that it will become more about techies competing in the analytics lab?

But the truth is the days when any major international sport was purely played on the field are long gone. Due to the money up for grabs by the winners, teams and athletes have long been supported by an entourage of coaches, physiotherapists, psychologists and experts of any flavor to give them a competitive edge.

Why should they not have statisticians and analysts at their disposal too? Particularly when they have the ability to unlock insights which could improve the work of all the other experts involved.

The fact is, the real action will always take place on the pitch. Statistics might tell us which player is most likely to correctly make the split-second decisions which are necessary to win in top flight elite sports. But it is never going to take the swing at the ball for them. And for the most part, those who head the teams and make the big decisions are never going to entirely give up on their gut instincts – so Big Data is likely to inform the sport industry, rather than dictate to it. Which should lead to more skilled play on the field, higher standards of sportsmanship and a more dynamic experience for the fans.