This week’s Track Radio sport and tech discussion

Thursday 30th July 2026. Each week, I talk on Track Radio about technology and sport on Sonja McLaughlan‘s afternoon show (between 2.30 and 3pm). This week’s topic is about analytics, AI and scouting in football.

Question: What’s the link between Florence Nightingale and the Premier League?

Answer: analytics

Back in 2008, my research team was appointed as an innovation partner by UK Sport, tasked with helping our Olympic teams get new medals. We’d mostly been involved in structural aerodynamics, design and testing but what we were about to find out was that our Olympic sports didn’t want that from us any more. That was for the big teams like of BAE Systems, McLaren and Frazer-Nash. It seemed that most of the coaches had read Moneyball, the best-selling book by Michael Lewis and what they now wanted was analytics.

If you don’t know the Moneyball story, it’s about how the cash-strapped Oakland Athletics baseball team from California matched the spending power of the bigger and richer baseball teams. The hero of the story was Billy Beane, the general manager of the Oakland A’s. Baseball was run by those who made decisions through intuition, gut instinct and experience, those with the arrogance to stand in a room of tobacco-hawking men and stare them down.

Beane, however, had been introduced to the research of a baseball fanatic called Bill James who had manually collected baseball data and analysed it. He had one question: What does it take to win? He created an equation that predicted how many runs a team would get using just two pieces of information for each player. It was remarkably accurate. He quickly showed that the coaching dogma spouted by the baseball fraternity was all wrong.

Of course, he was dismissed by the power brokers because he threatened their control over the game. How could a stats nerd like James, who’d never played baseball in his life, know more than the seasoned pros who’d been in the game since forever? Despite this, Beane employed a Harvard graduate to expand on James’ research and help him assemble a new team based on these radical ideas. The A’s improved dramatically and went on to get the longest continuous series of wins of all time, 20 wins in a row. Beane rejected a $12.5 million job offer from the Boston Red Sox and his fame was secured when Brad Pitt played him in the 2011 movie of the book.

The Moneyball effect was swift and the rest of the sporting world woke up to the new world of analytics. My dictionary tells me that analytics is ‘the systematic computational analysis of data or statistics’; it might be new to sport, but people have been doing it for centuries.

19th century analytics: Florence Nightingale

You may have heard of Florence Nightingale – she was the nurse who looked after soldiers in the Crimean War in the 1850s. The classic image of her is as ‘the Lady with the Lamp’, wandering around darkened wards in the middle of the night, caring for frightened soldiers near to death. What you might not know is that she was also brilliant at analytics.

She collected data on the causes of death of her soldiers and showed the government that very few of them actually died of their wounds: most of them died because of the unsanitary conditions of the hospital itself. She presented the data as a beautiful pie chart so that it was easily digestible by the politicians and suggested they put in better sanitation. It worked. She then used the same approach over the next couple of decades to lobby for better sanitation and conditions back in Britain. The Public Health laws she promoted increased the lifespan of the average Briton by around 20 years.

Analysis is something scientists do every day of their lives, but what Florence Nightingale realised was that not everyone likes numbers. If you want to convince someone of a course of action, then you need to make the numbers palatable. Her pie charts were some of the earliest examples of what we now call infographics, one of the key components of modern analytics.

“The ability to learn faster than your competitors may be your only sustainable competitive advantage.”
Arie de Geus.

Coming back to the 21st century, Scott Drawer, head of Innovation for UK Sport back in the day (and now head of performance analysis at Ineos Grenadiers) was a key advocate of data: he introduced me to this quote by Dutch Business theorist Arie de Geus. It seems perfectly apt for the world of professional sport where winning means trophies and money. And when we talk about sport and money, Football is at the top of that particular leader board in the UK. Football has embraced analytics for a couple of decades now and is now starting to tinker with AI to make things work faster and get that quick competitive advantage de Greus talked about. Most Premier League teams have an analytics team, with many building their own proprietory IT systems. Arsenal’s relationship with their data system provider StatDNA was so good they bought the company.

Some data can be recorded using wearables either in training or in a match to give speed and position data. A lot of data is manually coded from video: passes, tackles, interceptions and so on. What is certain is that there is a lot of data and many variables to choose from. Undoubtedly, AI will help with the processing of some of this data but there is still a lot of manual input.

Train an AI system with enough real-world data and it can make predictions, but only within the bounds of the original dataset. Trying to predict the performance of professional women footballers is unlikely to work if the AI was trained on men’s data. The signing of Mo Salah and Andrew Robertson by Liverpool is often cited as an analytics-aided decision, although whether it was using an AI-driven system is unclear.

What is clear, is that a lot of data is being collected on players across the world. Teams like Brentford and Brighton appear to be able to punch above their weight (to mix in a boxing phrase) while Teams like Liverpool, Arsenal and Chelsea use it to predict how players might fit into their team’s systems.

I suspect that for the big decisions, a coach is unlikely to rely solely on data prediction, not when their neck is on the line. A scout’s task is not just about finding the right players, it’s about reducing regret.

As Florence Nightingale showed us, using data can help us choose wisely but it still needs a real person to make the decision.