One sentence from a recent conversation with Professor Franco Impellizzeri really stayed with me:

“The data tell you that something happened — not exactly what happened and why.”

And perhaps that is one of the most important lessons for everyone working with data in sport.

We now have GPS, heart rate, power, accelerometry, AI and machine learning. We can measure almost everything.

But data do not automatically tell us:
→ what actually happened
→ why it happened
→ whether it matters
→ and certainly not what we should do next.

Impellizzeri gives a great example from football. Based on the GPS data, a training drill appeared not to produce the desired amount of high-intensity running. The initial conclusion: the drill doesn’t work.

Until they looked at the video.

During the drill, the coach had moved towards the middle of the pitch. As a result, the players gradually built up their speed instead of sprinting maximally from the start. The data were not “wrong”. But without context, the interpretation was.

For me, this highlights an important distinction:

Collecting data is not the same as understanding data.

And perhaps that is where the future value of the sports scientist, coach and performance professional lies.

Not in collecting even more data.

But in knowing which data matter, how reliable they are, what they actually mean, and how they relate to the context of the athlete.

That also requires something that is not always popular in sport: accepting uncertainty.

No metric is perfect. No model tells the whole story. And even an individualised approach is still based on generalisations and averages.

So critical thinking does not mean rejecting everything.

It means continuously asking yourself:

“What do I actually know — and what do I merely think I know?”

A fascinating conversation for anyone working in load management, sports science, data and performance.

And perhaps even more relevant for anyone who believes that collecting more data will automatically lead to better decisions.

This post was generated with the assistance of ChatGPT based on the original article, with the aim of translating the research into insights for coaches and support staff.