Heat Maps and the Trap of Surface Data

The night before Vietnam – Iraq in the 2026 Asian Cup qualifiers, I sat in front of my screen with a tidy belief: Iraq's diamond midfield would be broken by a high press. The match ended 1-1, but Iraq produced 23 shots, three times what I had imagined. I was not wrong about the scoreline; I was wrong to trust a diagram drawn on paper without ever checking it against a single concrete moment on the pitch. All that night I re-watched Iraq's last fifteen matches, going over each transition again and again, noting where each player received the ball. The 2026 mistake did not disappear; it became the yardstick for every prediction I make.

Years later, as heat maps and touch metrics poured into every analysis, I realised I was looking at a more sophisticated version of that same old error. A handsome heat map, glowing red down the right flank, looks a great deal like evidence. It gives us the feeling of having verified something. But it only answers where a player stood; it does not answer what he stood there to do, for whom, and at which moment of the move. I call this surface data: a label that sounds highly technical while containing no tactical content at all. A match can be tagged "dominant possession" while that team only passes sideways in its own half — the label is literally true and football-wise meaningless.

This is where I part ways with most of the content drifting around online. Data is not a conclusion; data is raw material that still has to be cooked. A heat map does not tell me why the right-back pushed high, which player covered the space behind him, and how the opponent reacted in the next two seconds. To know that, I have to watch the move at least twice: the first time as a fan, the second time tracking only one player or one zone.

Take the example I keep returning to: France in the 2026 World Cup final. While the stands turned toward Mbappé, what caught my eye was Deschamps dropping Griezmann deep to form a five-man plane with the midfield. Croatia wanted to press but could find no anchor to hold on to, because France kept the distance between their lines at a level that made any isolated press meaningless. I see a team as a drawing, and the biggest surprise comes from the attacking plane. On the heat map, Griezmann is just a patch of colour in midfield. In the geometric drawing, he is the hinge that jams Croatia's whole machine.

Then came the 2026 World Cup and Morocco. The media revelled in the fighting spirit of the first African team to reach a semi-final. I was drawn instead to a drier detail: how Walid Regragui shifted from a 4-3-3 in possession to a 5-4-1 out of it, with Hakimi and Mazraoui working as twin drills. When they locked down Portugal's Cancelo and Walker, the heat map would record both Morocco full-backs dropping very deep. But reading only that map would make us miss the entire intent: they dropped deep to pull the opponent out of position, then burst out on the right counter-attacking beat. I no longer name the best player; I name the most effective space.

The summer of 2026 gave me a rare laboratory. When the Bundesliga played in empty stadiums, home advantage all but evaporated and the away win rate rose by about 12%. That made me believe many of the metrics we habitually read as tactical traits were really just noise from the stands. The summer of 2026 gave me the answer: football without crowds is left with nothing but technique. When the outside pressure is removed, what remains on the pitch is what is real — and most of the heat maps we had been quoting so eagerly were not part of that "real".

In the same spirit, at Euro 2026 I refused an assignment asking me to write that Germany would crush Hungary. I argued that Joachim Löw's Germany left a defence too open to counter-attacks, while Hungary were the best massed-defence side in the tournament. The match ended 2-2 and Germany nearly went out. The lesson was not that I got it right; it was that I forced myself to state the conditions under which my prediction would be refuted. A judgement with no refutation condition is just a belief dressed up in jargon.

And here is the counter-intuitive point. People tend to think the danger of modern analysis is too much data. I think the danger lies elsewhere: enough data to create the feeling of understanding, but not enough to force us to understand. A handsome heat map lets us end an argument without rewatching the move. It rewards laziness with the appearance of rigour. The most insidious trap is a framework that is complete in form yet empty in content: room for every field, and not one field with real evidence. I produced exactly that in 2026 — only back then it was written as a diagram rather than a table.

What keeps me most on guard is the satisfaction of an analysis that looks complete. When every box is filled, we easily forget that some boxes never had data to fill them. Every match is a miniature model; I only point out where it heats up if you are willing to look calmly. And to look calmly, we have to accept saying "not enough information" instead of forcing a pretty conclusion.

So in the next match, I will test this with a very concrete question: when the team I am watching loses the ball on the right flank, which player is the first to drop back and cover the space behind the full-back, and does he do it before or after the opponent receives the ball? If the heat map cannot answer that question, it is still just a pretty patch of colour. And if it can, I will be the first to rewrite how I read data. Defeats in a match usually happen when we start praying instead of adjusting — and praying, in analysis, often wears the disguise of a heat map.