Can data help us achieve more Olympic gold?

We caught up with Nick Brown, Deputy Director for Performance, Science and Innovation at the Australian Institute of Sport (AIS), who has embraced a data culture to help the AIS’s elite athletes in their journey to bring home gold from Rio.

Q: Tell us about the mission of the AIS?
A: Our goal at the AIS is simple: to get more out of our elite Australian athletes – to get them on more podiums, more often.

Q: Can you tell me what innovation looks like at the AIS?
A: Innovation comes in many forms at the AIS – from the way we deliver programs for sport and work with colleagues in universities, to the way we develop and deliver technology for athletes and coaches.

Q: How does technology help you today?
A: We capture a whole lot of information from 2,000 athletes each week – including about 300 data points per athlete, or if you like, 600,000 data points a week – who are preparing for the Olympics and Paralympics in Rio. We collect data on how much they trained, how they felt, how well they slept, their physiological data for the day and physiotherapy information. And all that data is stored in one place – our Athlete Management System.

Q: How are you using data and technology to help predict and hopefully prevent athlete injury?
A: We can lose up to 20 per cent of an athlete’s training time due to injury and illness. And when they don’t train, we know they’re often unable to meet performance targets. We partnered with Microsoft and BizData to help answer a question – how could we know if an athlete was likely to be injured in the next three days?

We used predictive analytics and machine learning to better understand the relationship between training loads and injury and illness. We now know that an athlete needs to maintain a high to moderate chronic training load. Basically, they need to stay fit long-term and not have any sudden peaks or troughs in their training.

Using the data and analytical approaches we’ve developed, we can have athletes training and competing more consistently, and losing fewer days to injury and illness.

Q: How does it work?
A: The power of predictive analytics allows us to estimate when an athlete might perform well, or become injured or ill.

The Microsoft and BizData project is still in the pilot testing phase, but it is showing promising results. For the pilot test, the data gets uploaded each night through an Azure SQL Database connected to Azure Machine Learning. It is run through our custom algorithm, with the Power BI Dashboard updated at six o’clock in the morning. Azure ensures the data is secure and used solely for the pilot test.

We’re still learning, but we are now seeing true positive signs on athletes who may be injured in the next three days. This will allow us to get up-to-date information to athletes’ coaches so they are able to make training decisions that morning. A rollout to coaches and physiotherapists appears to be a reality in the near future.

Q: Can you imagine the future role of data for elite sports?
A: Data will help inform decision-making. We have a lot of really experienced individuals at the AIS, so data will help our coaches and athletes make better, informed decisions and perhaps shift some paradigms. Sport is steeped in tradition, which is important, but we can always improve. We’re helping inform what training practices will look like in the future and learning how we can help reduce injury and illness in the future.

Tags: , , , , , , , , ,

Related Posts