Empty Data: When Football Builds Legends on Sourceless Numbers

For three weeks after the 2026 World Cup quarterfinal between France and Uruguay, I sat rewatching footage of fourteen matches, logging every shot into a squared notebook. What kept me awake was not the 2-0 scoreline, but a line on the television stat board: France, 39% possession. The number was boxed in red with a downward arrow, as if Didier Deschamps's team had just committed a crime. But when I counted it myself, France generated 2.1 xG against Uruguay's 0.4. The attack led by Kylian Mbappé and Antoine Griezmann was not weak. It simply played differently.

That was the first time I understood something that became a professional principle: a number with no traceable source is more dangerous than a wrong number. A wrong number can be caught. A sourceless number simply drifts, gets shared, gets cited, and finally becomes the default truth of the crowd. Data does not make a revolution. It only strips the paint off the legend.

Today, sitting in Guangzhou and writing for a Chinese audience about European football, I still see that scene every week: a table of numbers floating across social media, clean, compact, persuasive, and without a single line of source.

Context: the era where everyone cites numbers

Modern football no longer lets anyone argue without data. A striker with two goals in five games gets called a “breakout star”. A team that concedes three times gets dissected with xG, PPDA, and key passes. Terms once reserved for analytics departments now appear in headlines, in social captions, in commentary by people who have never opened a raw data table.

On the surface, that is progress. The problem is that numbers spread faster than the ability to verify them. A figure is born somewhere — perhaps inside a club's internal model, a scraper site, an anonymous account — then copied through five layers of intermediaries. By the time it reaches the reader, the origin has evaporated. The reader receives the conclusion, not the method.

I graduated from a journalism academy and began my career at a sports newsroom, where I learned a simple discipline: no source, no story. When I moved into data analysis, I realised that discipline has almost vanished in the world of numbers. A goal is verified by footage. An xG figure is often trusted absolutely, even though it depends on the model, on the definition of a “clear chance”, on the person labelling it.

I work as an analyst for the Chinese market — a vast market, hungry for data, prone to sanctifying every new metric. The pressure of speed there is enormous. You must publish within an hour of full time or lose your readership. In that race, verification is always the first step cut.

That is why I adopted a rule of my own: every number I publish must have at least two independent sources, and I must state which model stands behind it. It sounds extreme. But the times I broke that rule were the times I was wrong.

Core: three stories, one lesson

The first story comes from Euro 2026. Federico Chiesa scored two goals and added one assist in five matches. The media called him the tournament's “breakout star”. I too was swept up while watching live — Chiesa's pace, his driving runs into the defence, the feeling he could produce something at any moment. But when I sat down with the data, the picture changed colour.

Chiesa's xG across the tournament was just 1.8. He scored twice, meaning he outscored the quality of chances he created. His shot-on-target rate was 41%, below the average of top European wingers. In other words, Chiesa played well, but those two goals inflated how sustainable the performance was.

I wrote a 2,000-word analysis for my personal blog, arguing Chiesa's form would be hard to sustain. I cross-checked the numbers against FBref, Understat and StatsBomb before publishing. Three sources, three models, one conclusion. The following season, Chiesa tore his cruciate ligament and declined. I was not glad about it. But it confirmed something more important: caution is not pessimism; it is a way of respecting the truth. Chiesa did not break the data. He broke the way we read the data.

The second story comes from the 2026-2026 season, when the pandemic turned stadiums into empty stands. Liverpool lost five consecutive home games at Anfield — something that had never happened under Jurgen Klopp. The media called it a “collapse”. I do not use that word. Collapse is an emotional label; analysis needs a mechanism.

So I separated the data. Liverpool's PPDA — the passes they allowed opponents before pressing — rose from 8.2 the previous season to 12.5 during the no-crowd period. That figure says something very concrete: Liverpool pressed less, slower, less fiercely. And the cause was not fitness. It was psychology. Without the roar from the stands, a high defensive line lost the drive to push up. When 53,000 fans fall silent, the numbers start to speak.

That is the lesson of separating randomness from structure. Five straight defeats sounds like a disaster. But when you isolate each variable — home, away, rest between matches, the presence of a crowd — you find a mechanism, not a curse. The empty stadium taught me that noise is data. Even Mohamed Salah and Virgil van Dijk, two pillars of that squad, could not compensate for the void the stands left behind.

The third story returns to France-Uruguay. It is the example I use most when explaining to colleagues the trap of a single metric. 39% possession is a correct number. But it means nothing on its own. Placed beside 2.1 xG against 0.4, you understand at once: France were not overwhelmed, they deliberately surrendered the ball and counter-attacked. The same fact, two readings, two opposite conclusions. Every number tells a story. The story is not inside the number.

What do the three stories share? Not “bad data”. The data was right. The error lies in interpretation and, worse, in transmission — when a metric is pulled out of the context of its model and turned into universal truth.

The same logic applies to the transfer market. Signing fees for free agents are often praised as “smart business”, but in essence they are costs pushed off the main books, sidestepping financial fair play oversight. No source line states that fee clearly, so nobody can verify it. The transfer market is where impatience gets priced.

And this is where I want to be blunt about what I call “empty data”. In my trade, a data-extraction table sometimes returns blank fields — no source, no date, no unit of measurement. Newcomers treat that as a technical glitch and move on. But a blank source field is no small detail. It is a signal. When a number dares not declare its origin, that silence is itself a fact.

I have seen stat tables shared tens of thousands of times with a completely empty source column, and nobody — not even accounts with hundreds of thousands of followers — questioned it. The problem lies not with the data, but with reading culture.

In sociology, we call this “legitimation through numbers”. A claim accompanied by a figure is automatically treated as objective, regardless of where the figure came from. The number becomes a stamp, no longer evidence. And once a number becomes a stamp, it no longer serves the truth — it serves whoever holds the stamp.

Contrarian angle: when too much data is also a trap

Here I must argue against myself, because that is what I always demand of colleagues.

If I spend this whole piece saying “verify every number”, then I fall into another trap: paralysis through verification. Some things in football cannot be measured by xG. A player's fear after an ACL injury does not appear in any metric table. A midfielder who hesitates half a second before a tackle — because the subconscious still remembers the pain — will be recorded as “out of form”, not as “not yet healed”.

And this is what I believe but rarely write: rushing back from an ACL injury is destroying the second phase of many players' careers. The body can recover on a medical timetable. Fear has no timetable. The problem is that nobody measures fear, so nobody reports it, so it does not exist on paper. The absence of data does not mean the absence of a problem.

Data does not erase emotion. It explains why emotion exists. But when we trust only what can be measured, we inadvertently declare that what cannot be measured does not matter. That is the greatest blind spot of an entire generation of analysis.

What to watch in the next round

In this annual-season cycle, the signal I watch most is not the league table, but the gaps in public data. When a team suddenly stops publishing its pressing metrics, when a club hides a key player's recovery timeline, read that silence before you read the match. The empty stadium once taught me that noise is data. Now I have learned one more thing: so is silence.