When Football's Analysis Machine Returns a Blank Page

On the night of 30 June 2026, in a small pub in Liverpool, the whole room screamed one name: Kylian Mbappé. I sat in the corner, my eyes fixed on the shortest man on the pitch — N'Golo Kanté. He touched the ball 58 times in the match France won 4-3 against Argentina, and by my notes that night, he never once lost possession under pressure. Nobody in the pub mentioned him. The spotlight always lands on the scorer, while the man who keeps the whole match from collapsing stands in the dark.

Years later, I did something strange. I fed a football analysis engine a blank file — no title, no data, no players, no scoreline, no date. I waited for it to throw an error. It did not. It returned a nine-dimension report, the full skeleton: tactics, finance, dressing room, rules, risk, media, the transfer chain of an entire football ecosystem. Every box marked "N/A". Three thousand words. About a match that never existed.

That was the moment I understood what the football analysis industry has been hiding for fifteen years.

From around 2026, football entered its digital era. Opta, StatsBomb, and later dozens of data companies turned every pass into a pixel and every movement into a number. Clubs built their own analytics departments. Broadcasters started showing heat maps after every match. We were taught that to understand football, you must read the data.

But there is a paradox few are willing to name. The more data there is, the more identical the analysis becomes. Open ten articles about the same match and you will read the same frame ten times: possession, shot count, heat map, then a conclusion so safe it means nothing. That frame is a cookie cutter. Put anything in and the same shape comes out. Even when you put in... nothing at all.

The industry is a whole ecosystem: data companies sell numbers to broadcasters, broadcasters resell them to viewers, viewers take them online to argue. With every loop, a number is repeated once more, and with every repetition it loses a little context. By the end of the chain, people argue with numbers nobody remembers the conditions of.

I have followed football across eight World Cups and eight Olympic Games, plus many major cycling tours. In every sport I see the same disease: people trust the template more than their own eyes. And football, the most measured sport on the planet, has it worst.

Let me tell you about the three false gods this industry worships.

The first god: the heat map.

The heat map is sold to viewers as an X-ray. You look at it and think you see a player. You see nothing. You see where he stood, not why he stood there.

Take a full-back. His heat map glows all over the opponent's half. The quick verdict: "He attacks well." But wait. Is he there because the system pushes him up, or because he abandons his position? A heat map cannot tell those apart. It gives you a smear of colour and lets you invent the story yourself.

This is what I call the new divination: the heat map does not explain a player's real role in the system; it only redraws where he once passed through and lets the viewer's imagination finish the job. A fortune-teller does exactly the same — offers a blurry image and lets you fill it with your own fear.

In 2026, at seventeen, I wrote a two-thousand-word piece arguing that Trent Alexander-Arnold was not a right-back but a disguised midfielder. At the time he had played three games for Liverpool and was being torn apart for his defending. I used data from five matches: he created twelve chances, the most of any defender in the squad. The piece was mocked, then shared by a big tactics account and reached fifty thousand views. The lesson was not that "data wins". The lesson was: data only has value when it comes with a reading the crowd does not yet have. The number on its own is mute.

Another example, this time about goalkeepers. A goalkeeper's heat map is almost empty — he stands in one spot all game. By the logic of positional data, he barely exists. But one save in the 88th minute, with the score at 1-0 and the whole stadium holding its breath, is worth more than a hundred sideways passes. Positional data cannot measure the moment. It measures where he stood, not what he dared to do.

A striker's heat map deceives in a different way. A centre-forward who touches the ball twenty times can look invisible. But if eighteen of those twenty touches happen inside the box, he is a lethal threat. A centre-forward who touches the ball sixty times mostly in midfield is... helping his team pass. Same position on paper, two completely different players. The heat map will not tell you who is more dangerous. It only tells you who touched the ball more.

The second god: possession percentage.

If I had to pick the most deceptive metric in football, I would pick possession. It is the easiest to measure and the least informative. A team with 62% possession may be controlling the match, or may be lulling itself to sleep with meaningless sideways passes between two centre-backs.

Look at Spain. In the round of 16 at the 2026 World Cup, on 1 July, they held around 75% of the ball against Russia, took more than twenty shots, and still went home after a penalty shootout. Their possession looked like an advertisement. The result did not.

And here is where I want you to stop. High possession is usually the sign of a team that does not know what to do with the ball once it has it. Sideways passes do not open space; they only push responsibility onto the next man. A team that truly wants to score accepts risk: forward passes, passes into the feet of a marked man, losing the ball. That 62% does not say which team is better; it says which team is more afraid of losing the ball. The ball does not roll by calculation. It rolls by the fear of being left behind.

The third god: xG.

xG — expected goals — is the tool I use most and the tool that is most abused. It answers the question: this shot, under average conditions, has what percentage chance of scoring? It does not answer: this player, tonight, with tired legs and a head full of fear, will he score?

We turn xG into a court of law. A team that wins on xG but loses the match is called "unlucky". A team that loses on xG but wins is called "lucky". But football is not played on paper. A shot from a position with xG 0.08 can be the finest moment of the match, while a penalty with xG 0.78 can be missed because a hand shakes. The average calculation knows nothing about a shaking hand.

What is frightening is that these three gods do not stand alone. They form one machine. Feed it data and it produces a report. Feed it a blank page and it still produces a report — only every box reads "N/A". The template does not need the truth. It only needs to be filled.

Try a small experiment. Give the machine two matches: a dull 0-0 and a breathless 4-3. It will return two reports with the same structure, the same length, the same tone. The 0-0 is no less "analysis" than the 4-3. That is the clearest proof that the template cannot tell emotion apart. It can only tell whether there is data to fill in.

That machine also swallows the transfer market. In August 2026, Chelsea paid Brighton around 115 million pounds for Moisés Caicedo, a British football record at the time. A defensive midfielder. You can read that number two ways. The first: evidence of inflation and madness. The second, mine: it is a story being priced. Transfers are not for buying players. They are for buying a story nobody has written yet.

And when that story fails on the pitch, the machine goes back to the old template: heat map, possession, xG. It never asks the real question: why does a player who shone at his old club struggle at the new one? The answer lies in fear, in expectation, in waking up each morning with a price tag hanging over your head. No metric measures that.

Then there is the dressing room, where the template also rules. A manager who wins three games is a genius. Three defeats, and he is finished. Nobody reads the space between those two lines: a squad losing belief, a dressing room splitting into factions, a star who no longer wants to run. The analysis machine can count the kilometres a player runs, but it cannot count the reason they stopped running.

And this is what annoys me most. The data machine is good at detecting what happened, but poor at detecting what almost happened. A midfielder runs into space to open a gap for a teammate — the ball never reaches his feet, so no metric records it. A defender drops at the right moment to block a passing lane — no duel, so nothing to count. Those quiet heroes live in the part the data does not look at. That is why I always go looking for them: not because they are undervalued, but because the very way we measure has made them invisible.

I followed esports long enough to see the same thing there, only faster. An esports player's career is far shorter than a footballer's, and both the youth pipeline and post-retirement support are close to zero. There, the number is worshipped even more, because the number is the only thing organisations will look at. And there, a twenty-four-year-old player pushed out is still recorded in a beautiful stat sheet — then disappears.

Now the part where I might be wrong.

I use data every day. Without xG, without touch counts, without passing maps, I am blind. The heat map is not evil. The evil is letting it replace the eye. A doctor does not refuse an X-ray; he just does not let the film practise medicine for him.

And I admit: sometimes that nine-dimension frame is useful. It forces questions about finance, about rules, about the dressing room — things the pure eye overlooks. The problem lies elsewhere: that frame can run when there is nothing to analyse, and that is when analysis becomes theatre. When a machine writes three thousand words about a blank file, it is not lying about football. It is telling the truth about us — people who built an industry that can say a great deal without needing to know anything.

The most hated man is simply the one willing to stand before the mirror everyone else avoids. Perhaps I am that man. Perhaps I am equating the emptiness of a machine with the emptiness of an entire industry, when that industry still has people who read matches with their eyes and their hearts, and they are still right. If I am wrong, I am wrong for letting the noise of templates drown them out. But I keep one rule: every contrarian argument must survive one question — does it help you see the match more clearly? If not, it is just expensive noise.

I am not against data. I am against laziness dressed up as science. There is a gap between "having numbers" and "understanding a match", and most of this industry lives inside that gap, believing it has already reached the far shore.

There is one small detail in that empty report I keep thinking about. In the "risk" section, the machine noted that the greatest risk was "process risk". It recognised it was analysing nothing. That was the only moment in three thousand words when the machine told the truth. And the only moment it made me stop.

Based on my experience watching matches over many years, I have drawn one rule: whenever someone opens an argument with possession percentage, that argument will almost certainly end with nobody learning anything. But when someone opens with a specific passage of play — which minute, who was where, and why — the conversation will go somewhere. The difference is this: data summarises, while the moment explains.

An empty stadium, the noise dead, and something presumed dead breeding again.

The next revolution in football will not come from more data. It will come from knowing when to throw the template away. People call that a shock. I call it the first time football spoke straight into my face: do not ask what the machine did. Ask what it has to say when there is nothing at all. And if the answer is three thousand words of pure "N/A", then the problem was never the machine. It was in the people who believed in it.