Nine Sections Complete, Zero Data: The Hole in Football Analysis

In August 2026, in Sochi, I said into a microphone that Pepe's 88th-minute handball was a deliberate offence. That was a wrong statement of the law. That night I downloaded the VAR data from the first twelve World Cup matches, logged every decision into a two-thousand-row spreadsheet, and spent a full month checking it against FIFA's original Laws of the Game.

It took me seven more years to see a different kind of error, and a worse one: being wrong when there is nothing to check against. Not misjudging a real incident, but constructing an incident that never happened and describing it in the tone of someone holding evidence.

This week, a nine-dimension analysis of transfer-market football passed through my system. Title: absent. Source: absent. Publication date: absent. Information points: empty, not a single entry. Every field carried the same sentence — insufficient information to assess.

That report still completed all nine sections. It still had tables, arrows, risk checkboxes, and a glossary of technical terms at the end.

The system is built in two layers. Layer one decomposes the source article into discrete event units: title, source, club name, player name, timestamp, source reliability. Layer two takes exactly those units — and only those units — as its permitted foundation, then applies nine analytical dimensions: tactics, finance, results, league landscape, rules, dressing room, risk, media narrative, industry transmission.

The anti-fabrication rule sits here: when a field is empty, the answer must be insufficient information, never a plausible-sounding guess.

This time layer one returned an empty array. No competition. No club. No player. Exactly one populated field remained: the domain is football.

And layer two still ran to completion. It still produced nine sections, each with headings and tables, each cell reading insufficient information to assess. Technically correct. Substantively meaningless.

This is the transfer window. This is when newsrooms run at full capacity, hundreds of lines a day, every line needing an analysis. It is also when an empty array is easiest to fill.

The problem in football analysis is not fake news. It is empty content presented in correct format.

A wrong article can be caught. People cross-check, argue, point out the absurdity. But a report with complete structure, complete sections, complete terminology, and no factual content inside is far harder to catch. It states nothing false, so there is nothing to refute. It simply occupies space.

In football we built an entire apparatus against the first kind of error. VAR arrived at the 2026 World Cup and spread into domestic leagues. The original aim was narrow: correct clear mistakes visible on replay. But alongside VAR, another apparatus grew — the content machine around every decision. Every incident reviewed ten times, each time with an opinion attached.

The volume of opinions has grown faster than the volume of evidence.

Law 11 of the Laws of the Game covers offside. In 2026, a colleague asked me to analyse a disallowed goal by Alan in the AFC Champions League semi-final between Guangzhou Evergrande and Urawa Red Diamonds. The referee flagged offside. I watched the footage forty-seven times, measured the defender's running angle, and concluded Law 11 had been applied incorrectly. The six-thousand-word piece with fourteen frame captures reached more than two hundred thousand shares.

What I learned was not that I was better than the referee. It was this: a conclusion stands only when every link in it can point to a specific frame. Watching forty-seven times was not perfectionism. It was the minimum threshold for a judgement to be allowed to exist.

The paradox: football has internalised this evidentiary standard for referees, but has never applied it to itself. Referees are assessed by the share of correct decisions. Commentators are assessed by nothing at all.

I tested that by counting. Over three weeks tracking transfer news in a major European league, I logged the provenance of every item: club announcement, agent confirmation, journalist with direct access, or merely a source close to the situation. The last category dominated. And when a deal collapsed, nobody went back to check where the original claim came from.

Here I want to describe a more specific hole, inside the financial rulebook itself.

UEFA's financial fair play rules cap the losses a club may record, and they bite on transfer fees and wage bills. A transfer fee is a published number with a contract, a date, and third-party confirmation. It sits inside the monitoring perimeter.

Signing fees for free agents do not. When a player's contract expires and he joins on a free transfer, the money that would have gone to his former club is split into a signing-on fee, agent commission, and loyalty bonuses. The nominal transfer fee is zero. On paper, the deal is cheap.

In reality it is not cheap at all.

This is exactly the same error as the empty report, only at a different scale. A data field is blank, and the system defaults to treating its value as zero. Nobody assigns a number to the gap. Nobody asks how much the gap conceals.

Data analysts are flooding into dressing rooms with spreadsheets and models. Most of the conclusions they bring are mathematically sound and football-wise skewed. Because football runs on rhythm, and rhythm lives in no column of a spreadsheet. A midfielder who misplaces four passes in the first fifteen minutes may be the best player on the pitch, if those four passes were four attempts to break the defensive line that failed. The model records four failures. The person in the stand sees an intention.

But when the data is entirely empty, there is not even rhythm left to read.

The familiar reaction to this problem is to blame social media, speed culture, algorithms. To blame the audience for clicking.

I think the cause lies elsewhere, and it is far less comfortable.

Our systems are designed to always produce output, regardless of input. A process permits only two states: complete, or error. The third state — complete but empty — is undefined. And what is undefined is not blocked.

So an empty array travels the whole pipeline, triggers no alert, and reaches the reader as a polished document. It is counted toward articles analysed. It lifts the performance metric. It exists.

My mistake on live television became the foundation for a new system. I abandoned confident assertions and moved to a multi-branch structure: if clause X applies, the conclusion is A; if clause Y applies, the conclusion is B. But a multi-branch structure has value only when every branch has data. Without data, multiple branches become multiple ways of saying the same thing: I do not know.

And I do not know is the correct answer. It is simply not the rewarded one.

An empty stadium is a referee's finest laboratory. When the noise from the stands is gone, the match exposes its pure structure. The same holds for data: when every field is empty, you see clearly which systems genuinely need data, and which need only the appearance of data.

What I want to take from this story is not a call to verify information. That has been said many times and has never changed anything.

What I want is a gate at the input: if the count of information points is zero, the process stops. No output. No metric credit. No publication.

As someone who once stated the law incorrectly in front of millions of viewers, I know the price of a conclusion without evidence. That price is not paid once. It is paid gradually, every time a reader trusts a report with nothing inside it.

In this transfer window, with hundreds of deals described daily, readers should ask one question of every analysis: which specific fact in it can be checked again in December?

Any piece that cannot answer that question never existed.