By Matt Barbini//National Team High Performance Consultant
Data is a challenge. It’s our job to provide our athletes and coaches with every available resource to be successful, but information alone is a marginal resource and practical applicability is essential to its utility.
Earlier this year I had the opportunity to attend the MIT Sloan Sports Analytics Conference, which was an incredibly valuable experience for many reasons, but particularly with regard to how and why we present data to our coaches and athletes. An important principle that was reinforced for me throughout the conference was that in order for data to have value, I need to present the information in such a way that it means something to the recipient. Otherwise, it’s just numbers on a page and conversion to performance gains is next to impossible.
In all cases, context is absolutely critical. Check out an example of one of our race stats report below:
As you can see, this report includes plenty of information. But in a vacuum, without relevant context, it doesn’t have a tremendous amount of value. Furthermore, to make decisions or race execution changes based on such limited information might be detrimental. However, if viewed in direct comparison to another performance, their best three times in that event for example, the data has much more practical application.
For us, a common practical use for this type of data is to examine what the best in the world are doing in a specific event, or component of an event. Recently, we’ve written articles about the women’s 200 breaststroke and men’s 100 breaststroke utilizing situational data to draw relevant conclusions. In the case of the women’s 200 breast stroke rate, data allowed us to confirm that the best in the world are dramatically accelerating their tempo on the last 50. In the men’s 100 breast, incremental splits from the world’s top performers indicated that American athletes are losing ground in the middle 30 meters of the race.
In short, data can be compelling and suggestive. But proper utilization requires a clear understanding of the context, a comparative standard, and a realization that numbers on a page are only part of the equation. Data, no matter how in depth, is not an end-all-be-all indicator of performance quality. The value comes from being able to answer, “So what?”