The Analysis With No Data: When Football Has to Relearn How to Say I Do Not Know

An A4 sheet sat on my desk in Marseille for three days. It was a 22-page analysis sent by a data firm: title page, table of contents, nine professional assessment sections, charts, and even a scale for source reliability. The opening page stated that the report processed the output of a first-stage data extraction. By page four I opened the raw data frame. The field for information points was empty. The field for entities involved contained a single instruction: identify them from the information points above. No club, no player, no match, no scoreline, no single fact to cross-check.

And yet all nine sections were complete. Conclusions included. Risk warnings included. A disclaimer at the foot of the final page.

The template was ready before the content arrived. For anyone who has sat in press rooms for twenty years, the scene is not unfamiliar. Only the speed has changed: where it once took an evening to assemble an analysis containing nothing, a system now does it in seconds.

An industry that manufactures conclusions faster than it manufactures evidence

Professional football entered this decade with more data than ever before. A single match in a major European league generates millions of positional data points; players wear chips; every passage of play is sliced into dozens of events. Running alongside that stream is another, far cheaper one: text. Transfer briefs, key-battle previews, pre-match verdicts, metric leaderboards. All of it can be finished before kick-off.

The temptation is not fabrication. The temptation is that the template is always available. An analysis needs six subheadings, three tables and one concluding sentence. If data arrives, fine; if it does not, the frame still stands, because the frame does not require truth to exist. Anyone who has worked a newsroom knows this: the deadline does not wait for the source.

The irony is that football has built a fairly strict regulatory system at the financial level. In November 2026 Everton were docked 10 points for breaching the Premier League's profit and sustainability rules; by February 2026 the sanction was reduced to 6 points on appeal. In March 2026 Nottingham Forest received a 4-point deduction. Since February 2026 Manchester City have faced 115 charges of breaching financial regulations. Juventus were docked points in a capital gains case and had part of the sanction returned through the appeals stages. FIFA banned third-party ownership in 2026, and Article 19 of the transfer regulations still limits the trading of minors.

At the money level, evidence is accepted as a precondition of judgment. At the analysis level, that precondition barely exists. That is the central paradox of the industry: one sport, two entirely different standards of verification.

Heat maps, pressing metrics and the new divination

A heat map shows that a central midfielder spent 62 per cent of his playing time in the left channel. It does not show whether he was there on instruction, because his left-sided team-mate kept abandoning his position, or because he simply wandered out of the middle. Position can be measured. Role cannot. A heat map records where a player was, not why he was there — and the gap between those two questions is exactly where conclusions get inserted on a whim.

The PPDA metric behaves the same way. It counts the passes an opponent is allowed before your team performs a defensive action; the lower the figure, the higher the pressure. Read alone, it is evidence of pressing. Read with context, it can be evidence of a team that is losing and forced to push up, or of a team facing opponents who pass so poorly that the ball comes back with little effort. The same fact, two opposing stories, and only footage can adjudicate.

Based on my experience watching Ligue 1 matches, I learned that the correct order is always this: observe first, quantify second, conclude last. When the press-room door closes, I begin to hear the match more clearly. In the autumn of 2026, in a corridor at the Vélodrome, I stood taking notes for four hours. Morgan Sanson left the pitch in the 62nd minute without saying a word to anyone, but the way he set his water bottle down on the bench told more than any stat sheet about his night. Details like that sit in no data frame, and for that very reason they are the hardest thing to invent.

In the transfer market the mechanism is more brazen still. Every contract begins with an evening on which somebody whispers into a telephone. The whisper may come from an agent pushing a price up, from a club applying pressure over a renewal, or simply from a reporter who needs a story. Within two hours it is a headline on forty different outlets, and none of them has an additional source. When the deal collapses, nobody publishes a correction, because corrections do not get read.

The fault is not in the machine

The laziest explanation is to blame artificial intelligence. This industry produced empty analyses long before language models existed. The reported-from-the-hotel genre — the author who was never at the ground but writes as if standing on the terrace — has been around for decades. So has the quote nobody said, attributed to a manager who never held a press conference. Machines merely inherited a habit that output quotas had already legitimised.

The deeper problem sits with the audience. An article that asserts confidently is always shared more widely than one saying there is not enough data to conclude. I do not know is the phrase the market punishes hardest, even though in very many cases it is the only accurate answer available. A null result is still a result: when the extraction stage returns zero, the correct finding is not that this team presses badly, but that there is nothing yet to analyse. Football analysis has no place for that kind of finding.

What to watch

This season I will be watching how clubs and league organisers publish raw data rather than only summary tables — when provenance is open, inventing numbers becomes far more expensive. I am also waiting to see whether any newsroom dares to install a hard gate: no primary source, no publication. And I am waiting to see how audiences respond to an admission of uncertainty.

The Moscow night taught me that the truest sources often carry no business cards. Each season is a heartbeat, and I am only trying to catch the right beat. Will an industry that lives on certainty agree to relearn how to listen for the rhythm before it writes the verdict?