The Data Gap: Where the Transfer Market Is Easiest to Fool

In my office in Liverpool, there is a spreadsheet I open every morning like a ritual. It holds no transfer fees, no expected-goals figures. It holds empty cells — and beside each empty cell is a short note explaining why I have not yet allowed myself to enter a number there.

By mid-August this year, that spreadsheet still had more than forty empty cells. I have looked at it long enough to realise something I was not conscious of fifteen years ago, when I first entered the profession. My job is not to fill in numbers. My job is to determine which cells are not permitted to be filled, because the data behind them is not yet ripe enough to become a conclusion.

Every transfer window is the same. The greatest pressure does not come from deals already completed, but from deals still sitting in the grey zone — where there is no contract, no confirmation, nothing but a rumour attractive enough that people want to believe it. And it is precisely there that an empty spreadsheet becomes the most dangerous thing.

A data gap is not data of zero. It is a dangerous invitation for people to fill it with their own imagination.

I learned this lesson later than I would like to admit. In 2026, writing a World Cup special from Russia for a sports site, I built a homemade xG model and analysed all sixty-four matches. The model gave me a tidy conclusion: France would win, averaging 2.4 xG per match — the highest in the tournament. I published that prediction from the group stage, confident to the point of arrogance.

Then Croatia reached the final. My model called it luck, and the internet called me a machine. I was exhausted, and hid in a library for two weeks to review all the data. My error was not in the number. My error was that the model ignored corner kicks — a data gap I had filled with the assumption that it did not matter. I did not enter a wrong number. I entered a number where I should have left the cell empty.

From then on, I added a section to the end of every analysis called “Limitations of the Analysis”. It makes my writing longer, makes me look less decisive, and irritated a few editors. But it is the only way I know to keep my spreadsheet honest.

xG is a revolution, but every revolution needs time before people accept it.

The transfer market today lives inside a paradox I observe more clearly than ever. There has never been more data — every player tracked across hundreds of metrics, every pass recorded to the centimetre — but precisely for that reason, data gaps have become harder to notice. People assume that with enough figures, we no longer have to guess. In reality, with enough figures, we easily forget where we are guessing.

Take a concrete example. In 2026, João Félix moved from Benfica to Atlético Madrid for a fee of 126 million euros, at nineteen years old, with only one full season in the Portuguese top flight. That figure made him one of the most expensive players in history at his age. My point is not whether that deal was right or wrong. My point is that to value a nineteen-year-old off a single season of peak data, you are forced to fill in a great many empty cells — and every cell you fill is an assumption about the future, not an event that has happened.

In my trade there is a term few outsiders know: the “panic premium”. It is the surcharge a club is willing to pay when the clock counts down to the final day of the window. In data terms, the panic premium is nearly impossible to predict, because it depends not on the player's quality but on the buyer's fear. A defender valued at twenty million euros in June can become thirty-five million euros on the thirtieth of August — the same person, the same set of metrics, differing in one variable alone: time.

And this is what troubles me. When such a deal closes, people immediately fill the data gap with a story. They say the player is the “perfect piece”, the “deal of destiny”. But behind those stories, no data proves he will succeed in the new league. Only belief, carefully packaged to look like a conclusion.

That is why I refuse absolute predictions. I give only probabilities, and I state clearly which assumptions they rest on. When a club asks me whether it should pay a hundred million euros for a player who has not played fifty top-flight matches, my most honest answer is: I do not know, and no one knows. What I can do is point out that the price is reflecting an expectation, and that expectation has never been tested against data — because that data does not yet exist.

Every number in the transfer table is a fate waiting to be written.

In 2026, at forty-two, I was working as a transfer-market administrator in Liverpool, watching Juergen Klopp take the club into the Premier League Top 4 with seventy-eight points through a frenzied pressing game. I calculated Liverpool's average PPDA at 8.2 — the lowest in the league — while a conservative side like Manchester United stood at 15.7. I wrote a long piece on the concept of gegenpressing and posted it on my personal blog. The reaction was not what I expected: people called me a machine, someone turning football into addition.

I remember the 4-3 win over Manchester City on the nineteenth of January, 2026. Sitting in the stands, I saw the numbers I had calculated take physical form: every movement, every pressing rhythm dipping then surging, every time an opposing player touched the ball in panic. Data never lies, but I learned that it only tells the truth when we place it in the right spot.

Data whispers; those who listen hear miracles.

In March 2026, when the pandemic halted world football, I went through a crisis of faith I have never described publicly. Liverpool were twenty-five points clear of Manchester City, all but certain champions, and the season was suspended. I asked myself: if data cannot predict a pandemic, what does data mean? I wrote three drafts and deleted all three.

When football returned in June with empty stands, I realised something I had never accounted for. The home-win rate fell from 46% to 39%. Not because players ran slower, but because what was called “home advantage” came largely from the crowd — from the roar, from the invisible pressure on referees, from the sense that eleven men on the pitch were not alone.

An empty stadium does not distort the data, but it makes the truth feel hollow.

I had to spend weeks alone redefining my model. From then on, I developed a concept I call “data context” — meaning every metric must be read alongside the conditions that produced it. A PPDA figure recorded in a full stadium cannot be compared directly with the same metric in an empty one.

Euro 2026 taught me another facet of the same problem. Through a chance connection, I gained access to the internal training data of the Italy national team. They ran an average of 112 km per match — not the highest. But another metric was outstanding: ball-circulation speed. I wrote a piece arguing that Italy were not a defensive team but a motion machine. It was shared more than ten thousand times.

What I took from it was not that I had been right. What I took from it was that for the first time, I had enough data to fill cells I had previously been forced to leave empty. Not because I was smarter, but because I had access to something others did not. It was a reminder of humility: most data gaps do not exist because no one can measure them, but because we are not permitted to see them.

Now I want to address another data gap, one I consider larger than any empty cell in a transfer spreadsheet: the space of subjective judgement inside VAR.

People believe technology will erase controversy. But the way the law operates tells a different story. VAR intervenes only for a “clear and obvious error”. That very phrase is a legalised data gap. Clear to whom? Obvious to what degree? From one camera angle, a challenge may be a foul; from another, it is a fair contest. No algorithm can answer that question, because the answer depends on a judgement no one can digitise.

I once spoke with a referee about this. He told me the greatest difficulty is not seeing the foul, but deciding whether the degree of error crosses the “obvious” threshold. That threshold exists in no document. It lives in each person's head, and it differs from person to person.

This is the blind spot I want to stress. We build data systems to reduce subjective judgement, yet at the very moments that matter most, subjective judgement returns — only dressed in the clothing of technology. An empty cell in a spreadsheet and an “obvious” threshold in the VAR law are the same problem: we must fill a value that data cannot supply, and we call it fairness.

Those who are right before their time always pay the price of solitude.

So what is the signal for the next round? One concrete thing comes to mind. When the transfer window closes, look at the clubs that chose to stand still. Not because they lack money, but because they are patient enough to keep the empty cells in their spreadsheet. In a market where everyone wants to fill in numbers, those who know how to leave blanks may understand the value of each number best.

As for me, this morning I opened that spreadsheet again. The forty empty cells are still there, and I have not rushed to fill any of them. I learned at Anfield that belief is also a variable — and sometimes the most honest variable is the one we choose not to fill in.