Analytics in Sport: Making the Invisible, Visible

This post is inspired by my beloved Essendon Bombers.  Who, after showing significant promise at the start of the season beating the top contenders for the championship, have since suffered injury after injury after injury, with nothing short of humiliating defeats in the lead up to the finals.

Analytics in the world of sport is not a new concept, and yet many teams are still to realise its full potential.  Much like the world of business, sporting teams around the world are looking to analytics to make the patterns and trends in human performance and game strategy that may be invisible to the human eye, visible.

One area in which analytics is making a big difference is in the field of injury prevention.  An injured player isn’t the only one who feels the pain – it can also negatively impact team morale, game results, and match attendance.

For this very reason, Leicester Tigers rugby team turned to IBM predictive analytics to identify the key risks and avoid injury in their players.  Rugby players are fitted with sensors to measure and monitor physical performance, collisions and fatigue.  In combination with psychological data, the Tigers were able to rapidly analyse and predict each individual player’s risk of injury, taking action to modify training programs for those players at risk.

Of course not all data is easily captured and measured in manageable data stores.

In the world of basketball, box scores were the key statistics used for analysis.  Described as a “dance”, there is so much more to basketball that lends itself to be analysed, assuming it can be captured to start with.  Take for example the way players interact on the court, their moves, counter-moves, the location of where shots are taken etc.  All of this information is key to understanding the best strategy to win the game.

IBM, in conjunction with the Computer Science Department at the University of Southern California, took on that very challenge – innovating with optical tracking technology to capture information about moves on the basketball court and using analytics to derive new insight.  For example, the fact that taking a long 2-point shot not only has a low probabilty of success, but is also very difficult to retain possession on the rebound.

In both sport and business, the fundamental question remains the same: How do you translate data into something that can help you win?  Using analytics, we can derive never before seen insight into hidden trends and patterns, making slight shifts in strategy and the moves we make to achieve success.

For more information on analytics in sport visit ibm.com/sports