This is a review of the Injury Risk Mitigation System (IRMS) research conducted by Calham Dower, Abdul Rafehi, Jason Weber, and Razali Mohamad.
Player injuries are a key concern for all teams across all sports. Injuries can cost teams thousands of dollars every day and negatively affect the teams’ competitiveness. Over the years several programs have been developed in an attempt to estimate injury risk, but to date none have been able to present precise and accurate data. The Injury Risk Mitigation System (IRMS) was designed to solve this problem. The IMRS is able to take advantage of the fact that teams are collecting more data than ever before. It helps analyze this information in order to produce estimations of an individual’s injury risk on any given day by looking for patterns in the historical data.
The human body is an extremely complex system and therefore it requires a complex program that is able to analyze data for patterns leading up to any variety of injuries. An Artificial Neural Network is capable of modelling very complex systems and determining how different variables relate to each other and whether those variables are independent or dependent upon each other. Testing the program using data from two current Australian Rules Football teams from 2012 to 2017, illustrated that teams can now more accurately forecast the risk of injury for the upcoming two-week period than any other approaches currently being used.
The more data is entered into the Injury Risk Mitigation System, the more accurate its predictions. Using only one team’s data means it could take years for the system to make accurate predictions. IMRS is able to minimize this issue by using data from different teams. It is able to do so in such a way that the data remains anonymous as to where it came from. This allows the system to begin making accurate predictions much sooner and help teams minimize injury risk among their players. As training patterns change from year to year, the model must be updated on a regular basis. As long as a team consistently uploads new data IMRS will be able to provide accurate predictions for the next two-week period.
IMRS does not predict injuries, but instead predicts the risk of injuries. Predicting the risk of an injury, rather than injuries themselves, is very valuable in formulating training plans for players, as these plans can be adjusted in order to minimize the chance of those injuries actually occurring. One caution to remember is that IMRS’s ability to predict injury risk is more accurate during the season than in the preseason.
In summary, a player’s risk of injury ebbs and flows over the season. Teams can recognize the periods when a player has an increased risk of injury. This will aid the team in managing their players in order to maximize their performance and minimize their risk of injury. This model is most useful towards the end of a season, helping teams balance performance versus risk of injury and allowing them to maintain healthy athletes as they approach the end of the season and championships.
How mature is your team’s analytics program? Take the Sports Analytics Maturity Assessment.
Learn about the Groundbreaking Sports Analytics Model coaches and sports analysts are talking about!
Learn all about Sports Analytics here.