When a player suffers an injury, it not only damages their team’s chances of winning, it also affects the bottom line: a club still needs to pay an injured player while they are sitting on the bench.
Rugby Union team the Leicester Tigers is tackling the issue of player injury using predictive analytics. It does this by collecting as much data about each player’s health and performance as possible, and analysing it to spot the trends that usually precede injuries.
“The idea is that we’ll be able to see what leads to an injury in a particular individual, and when that starts to happen again, to avoid it,” says Andrew Shelton, the Tigers’ head of sports science.
The club is using a dedicated sport analytics package from specialist software provider Edge10, which is based on IBM’s SPSS statistical analysis program.
It analyses metrics about each player’s physical fitness, as well as data about their performance in training sessions.
“We bring together objective and subjective fatigue tests, how players slept, how their muscles are performing, how hard they trained, how much weight they lifted today,” Shelton explains.
“We also look at player’s physical movements during training sessions," he adds. "We get data from GPS tags on each player, collision data from accelerometers which the players wear on their backs, and a performance analyst who tags all the collisions in video.”
The system has only been in place for around six weeks so far, but Shelton hopes that by revealing the warning signs of possible injuries, it will bolster the club’s performance and cut the avoidable cost of paying injured players.
In future, he says, the club will incorporate more datasets into the statstical analysis. These could include psychological profiles of the players and even their genetic make-up. "We’re talking to various universities about looking at which genes seem to be important for which positions in rugby," Shelton says.
So far, the focus of the project has been to cut the risk of injury, but the same analyses could be used to improve the performance of the team in competitive matches.
Open source statistical programming language R is making in-roads to the enterprise
Leicester Tigers are participating in a study with Chester University that involves fitting players with GPS trackers during the game, and monitoring their moves on the pitch. At the moment, the club can only access that data if the other team is also participating in the study, but Shelton hopes to be be able to track player GPS data in all matches from next year onwards.
Of course, the ability to monitor the movements of the opposing team’s players would extremely valuable to the Tigers, but Shelton does not expect there to be much data sharing between teams. "We’d want to see their data, but we really wouldn’t want them to see ours," he says.