Valuable but immeasurable

People who matter aren’t always easy to quantify

Paolo Maldini was a nightmare for statisticians. Ask anyone who followed football, and they’d almost certainly agree he was a great defender. But he didn’t show it in ways that could be easily measured. When he played for Milan and Italy, he averaged one tackle every other game. He was so good at helping his team control the game, he didn’t need to make that many tackles.

People who are crucial to a team aren’t always the ones that might initially spring to mind. For example, who was the most important Tottenham Hotspur player in 2013? The obvious answer is Gareth Bale, but one analysis found that it was Mousa Dembélé whose absence had the biggest impact on results.

Beyond football

In the aftermath of the COVID pandemic, I’ve had several discussions with colleagues globally about what – and who – made things work well in the heat of the crisis. If you talk to people involved in the response, you’ll notice that many of the names who get regularly mentioned aren’t necessarily the ones highlighted in public coverage. Put simply, they don’t fit into the common metrics used to judge ‘importance’ in public consciousness.

Similarly, much of the work that was crucial to the response – from data curation to real-time analytics – didn’t fit neatly into academic reward structures like high profile journal papers on a CV, as we noted in this 2020 piece. Reliance on simple statistics might make evaluation faster, but overlooking key contributions can hinder future team performance.

On field and across fields

So how can we do better? One key component is a better shared understanding of what work is valuable and why. For example, in recent years, football analytics groups have developed much better tools for tracking player movement, and hence can move beyond simplistic metrics like tackles made to understand more deeply how teams control space on the pitch. Gradually, the measured statistics are starting to reflect the intuition about why players like Maldini were so good.

Knowing what matters can give teams a major advantage. When I was working on The Perfect Bet, I noticed that several groups who’d honed their skills on betting predictions were now moving into analytics to inform team management. In the years that followed, two of the teams that had the strongest links to such analytics – Brentford and Brighton – would rise to the Premier League.

Whether in sport or science, it’s worth the effort to move beyond simple metrics, and understand the value of contributions that might traditionally have been hard to quantify. It might be harder in the short term, but it will make things much easier in the long run.