Because of the money involved in modern professional sport, it’s an area in which people are willing to embrace new science and technologies in the pursuit of that all-important competitive edge. Increasingly, this includes data. It wasn’t always this way.
When the Oakland Athletics baseball team started using rudimentary data analysis to evaluate potential recruits in the early 2000s, general manager Billy Beane had a hard time convincing people to move away from existing methods, which relied heavily on received wisdom and subjective opinion. The on-field success that followed inspired first a book then an Oscar-nominated film and the world of baseball rushed to follow their example. Pretty soon, every team in the NBL was incorporating the same kind of data into their scouting and recruitment.
In the years since, advances in technology and the constant need to stay ahead means that sport is one area where the use of big data has exploded. The technology and our understanding of its potential have come a long way since The A’s tentative first steps into evidence-based recruitment. Social media, biometrics and GPS are three areas that have enabled huge leaps forward. Below is a look at how big data is used in sport today, as well as where it might go next.
Scouting and Recruitment
Recruitment is still a major application for data analysis in professional sport. The sums at stake mean there is no place for guess-work, and the need to minimise uncertainty puts data right at the centre of the process. Analysing performance data is now standard practice, but the level of detail and sophistication has increased exponentially since the early days. With the sheer quantity and breadth of metrics available, it’s now possible to objectively analyse a team’s style of play, the standard of teammates and the quality of opposition, and use them to predict a potential recruit’s future performance within a team.
It’s not just performance metrics either. High-profile transfers can be undermined by off-the-field matters or issues in a player’s personal life. Teams now profile players in great detail away from their sporting endeavours. The rise of social media has meant there is now an abundance of data that can give an accurate picture of character, motivation and background, and help determine how a potential target will respond to adversity and how they might integrate into an existing group.
Wearable tech has hugely improved injury prevention. It’s now possible to measure hundreds of biometric signs per second in real-time to monitor things like effort, fatigue and physical impact. This data lets coaches and players analyse body mechanics, so they know when they are more likely to sustain an injury.
Wearable tech also provides a huge array of information about an individual’s technique, which can enable coaches and analysts to make tiny adjustments and minimise the risk of injury that way.
Professional sports as an industry is constantly seeking ways to attract fans and capitalise on the fierce loyalty that comes with team affiliation. Using large-scale sentiment analysis of social media feeds, professional teams can get huge amounts of data about how people react emotionally to events during games, and then factor this into the match-day experience.
Using data in novel ways to engage fans is also gaining traction. Giving spectators access to information such as heart-rates and distance covered in-game can give them new insights into player performance and enhance the viewing experience that way.
The Future – Tactics?
The tactical domain is an area where big data analysis is notable by its absence. Because of the lack of precedent and the range of hard-to-measure aspects like technical skill, individual physiological performance and team formations, tactics are still seen as more art than science. This could change in the near future, as advances in machine-learning, AI and the ever-increasing datasets available mean the scope for tactical analysis is already much greater than it was a couple of years ago.
Professional sport looks likely to continue pushing innovation in big data through increasingly inventive applications, to meet the many and various demands it throws up. Further proof, if it were needed, that big data can help solve problems, model outcomes and improve performance, irrespective of the industry or setting.