How to use data to answer key coaching questions

Many who have a deeply geeky interest in the dynamics and numbers behind sport would have watched or read ‘Moneyball’ and fist-pumped the air in delightful exasperation.

I was not one of them. Back in 2010 I was just over a decade into my career in rugby union and my role was morphing slowly from being a sets-and-reps strength and conditioning coach to being a more performance coach with a wider lens on everything from development to optimising performance. I had been at the sharp end of sport long enough that I had become institutionalised by the landscape. I had found myself nodding in agreement with traditional viewpoints regarding the bubble or island in which sport exists and how subjective opinions based on lived experience ruled decision making.

This was also distorted, and reinforced hugely, by success. The team I was working with at the time had grown from having its annual home at the middle-to-bottom of the league to regularly reaching the quarters and semis of the rugby equivalent of the Champions League, winning Europe once and having a win ratio in the upper 70%s. I considered myself “in the know” enough having “spent enough time in the sport”.

But if you don’t know what caused your success you have no blueprint on which to continue, or replicate it - which is where a wonderful coincidence threw itself forward. Firstly, GPS had become a part of the language of sport, and it had mapped through the journey of the team from being bottom feeders to being the big fish. Also, everything from force platform testing in the gym to early performance analysis systems had bridged the gap between training metrics and performance metrics, and I started taking an interest in the deep dive of stats and numbers I had collected since 1999. 

The second part of the coincidence was I started sharing coffee conversations with performance role staff in other sports – football, athletics, cycling - which is when the penny dropped. The trends not only in the creation of talent but in the maximisation of optimal performance were ubiquitous through all sports. In other words, if used as an ally – and not a boss - data was literally giving us the answers to the exam questions on a cheat sheet. 

What happened next in sport was predictable. A thousand Excel files on a thousand laptops formed an arms race to let data play a part in analysis. Data correlations were being thrown forward as reasons to change training direction, which inadvertently led to a lot of sample and research bias used as a means of reinforcing predetermined preferences. It even reverted down to certain gym exercises being cornerstones of performance programmes.