ColumnTipping Point

Sport has moved on from emotive, skin deep, judgments of every kind

In team sports, there is more structure than ever before, and more patterns, and more data-led decisions

In sport, ignorance used to exist on a shorter spectrum. Whatever sport you followed it seemed like there were fewer things to be ignorant about. Nobody knew everything, of course, but there were always experts who had answers and other people who had insight and in that pyramid of knowledge the rest of us made up the broad base. Shouting, usually.

In sport, insight was as much a gift as an intellectual property. It could be shared, but not learned, or acquired. People were blessed with it. They could see things in players, or in play, before they were obvious. How could they tell? We didn’t know that either.

It’s different now. The mystery in sport, and the part played by chance, is under siege. The top coaches and the top teams have no desire to be undermined by capricious elements outside of their control, or to trust wholly in the naked eye. So, in team sports, there is more structure than ever before, and more patterns, and more data-led decisions, and a persistent crackdown on chance, and there are more things to know.

All of which means, for you and I, the mystery is as much to do with ignorance as wonder. What just happened there? Why? Really????

READ MORE

So in hurling and football now players will turn up in places that you would least expect, and your first thought is that they’re lost or they’ve gone rogue, and your second thought is that somebody who knows more about this thinks it’s a good idea. It used to be the case in the GAA that if a player was loose, then another player with whom he was paired was responsible for this dereliction of duty, and you could direct your outrage accordingly.

Now? The chances are you will be convicting an innocent party. In the modern era, not all players are paired. Some players are meant to be alone, honouring a greater plan. You don’t know the plan? The plan is none of your business.

You wonder what they’re thinking. You wonder what really matters. You wonder how little you know and, by deduction, the sheer scale of your ignorance. Before the Limerick hurlers played Galway in The League of Lies back in February the Limerick analysis team set up their station next to the press box, at the back of the stand in Pearse Stadium.

It was a team of four, led by Sean O’Donnell, one of the sharpest minds in hurling. One of the computer screens was the size of a desktop and there must have been three laptops too. We couldn’t see what in-game data was being logged, but if Limerick were devoting this kind of man power to harvesting information from an inconsequential League game in February they must believe it’s important; it must shape their thinking.

Whatever we saw that afternoon, or whatever we thought we saw, or whatever we believed was important, we were looking at the game through a different lens. They don’t think like the rest of us. That would be foolish.

Many years ago Donncha O’Callaghan showed me an internal performance report from a Test match he had just played for Ireland. Mervyn Murphy was the IRFU’s lead analyst at the time and the detail in the report was staggering. O’Callaghan knew exactly what his performance targets were and whatever was said in a TV studio, or whatever marks he was given out of 10 by a newspaper, or whatever anybody said in the stands, or in the pub, the ultimate arbiter of whether he had played well or not were the numbers contained in this report.

No other opinion mattered. Every other view would have been based on impressions, or things that caught your eye, or something that stuck in your memory, or assumptions about what O’Callaghan was supposed to achieve in the game. Some of that would have been valid and accurate, but it would also have been incomplete. The game had moved on from emotive, skin deep, judgments of every kind.

Since the turn of the century, professional soccer has entered into an ever-expanding relationship with data analytics. In his book Expected Goals, Rory Smith of the New York Times paints a fascinating picture of the pain-staking labour in collecting the data, and how it is made fit for purpose.

Recruitment, for example, is no longer just the domain of chief scouts and their vast network of boots on the ground, standing at schoolboy matches, trusting their jeweller’s eye for a diamond. It is a numbers game. Smith profiles a German company called Impect who, among other services, “furnishes its customers with more than 1,200 ways to assess a player,” he writes. Bayern Munich and Paris Saint German are among their clients.

When Liverpool completely overhauled their recruitment processes under Fenway’s ownership, their research department included an astrophysicist, a chess champion, a former researcher on the Higgs boson and a polymer physicist from Cambridge. These were the boffins who convinced Jurgen Klopp to sign Mo Salah from Roma, after the Liverpool manager had initially rejected the striker.

They based their recommendation on covariance, a statistical term that measures the relationship between different elements. In this case, the data convinced them that Salah and Roberto Firmino would complement each other. Firmino had already been signed on the basis of his numbers.

“Data has suffused almost every aspect of how football is played, how it is coached, how it is scouted, how it is consumed,” writes Smith.

So, a concept like xG (expected goals) has entered the idiom of football analysis, and will pop up in the TV coverage of games, and if you asked the casual fan what exactly it means and how those numbers are calculated, you would get some fuzzy answers. On the Opta Analyst website you will find a really interesting explanation that runs to about 1,600 words, accompanied by funky graphics. But how much of that do you really need to know? And anyway, goals should always be a surprise.

If you watch top level sport it was never easier to know next to nothing. But you’re still allowed to wonder why the hell nobody is marking up.