Full Cover, Empty Core: When a Referee Report Has No Data to Verify

On 12 March 2026, as European football froze because of the pandemic, a young colleague sent me an analysis file. The cover was complete: "VAR Decision Report — La Liga, Matchday 27", with charts, the fixture name, and the line "source: personal database". I opened page three and sat still. Not a single number. Not a single error code. Not a single timestamp. The tactical table was empty, the financial table was empty, the risk table carried only headings waiting for content like milestones on a road no one had walked.

My first reaction was to call and demand an explanation. My second reaction — honestly — was to fill in the blanks myself. I had watched that match. I remembered a few incidents. And I knew exactly how dangerous that temptation was, because I had once fallen to it, not in a data file, but in front of more than four million listeners.

For anyone working as a referee, a document like that is more common than people think. Modern analysis runs through several layers: first the source article is broken into information points, and only then is the analytical framework applied to those points. When the first layer fails — truncated text, an extraction error, a source that will not load — the result is not a blank document. The result is a document carrying every label: correct field, correct format, correct section headings, but hollow inside. That is the worst kind of error, because it does not shout. It quietly wears the clothes of success.

I learned this from my own trade. Football law changes year by year, and each change leaves an old layer of memory behind in the mind of the practitioner. In 2026 I learned that a ball touching the armpit area was not handball. In 2026 the law changed to include the armpit. I did not check. In June 2026, in the France versus Australia match, I stated on radio that Josh Risdon had not committed a handball because the ball struck his armpit. I was wrong. The editorial desk had to issue a correction. For the first time in thirty years, I was contradicted in public.

The law does not live in memory; it lives in data. I got one sentence wrong and lost an entire reputation. If only I had known this back then. Since then I have kept a table of law updates year by year, and I write every analysis with a note on the clause and the date of its amendment. In August 2026 I began a personal project: recording every VAR decision in La Liga and the Champions League with error code, timestamp, distance and ball speed. By the time the pandemic stopped football in March 2026, I had 523 matches in hand. A single match is only a story. Five hundred matches are the law.

That hollow file taught a lesson I had to write three times before it sank in. When data does not arrive, the only correct choice is to leave it empty — empty in the sense of stating clearly "insufficient information", not empty in the sense of skipping over it. That is a professional statement, not an admission of defeat.

I sort every piece of information into three colours. Green is where the law exists, the minutes exist, the clause can be checked. Yellow is pending verification. Red is what I only half remember. A modern referee report needs seventeen categories, from tactics to club finance, from form to league standing, from rules compliance to dressing-room conditions, risk, media narrative and the transmission chain of the whole industry. Each category needs a minimum input to come alive. The tactical category needs at least a formation diagram or a playing-style label. The financial category needs one figure: transfer fee, wage, contract length, revenue, or a signal of owner funding. The form category needs a league position or a sequence of results. Without those kernels, every cell must sit at "undetermined".

The clearest example is the compliance category. A financial fair play checklist cannot be marked "safe" merely because no charge has appeared. The absence of evidence of a breach is not evidence of compliance. The correct status for such a cell is "undetermined" — not "low", not "none", but "undetermined". In the legal profession this is a survival principle; in football commentary it has been almost forgotten.

Based on my experience following matches, I once counted every phase of play to test the gap between process and results. A team can win through goals that the chance-quality data does not support, and lose matches in which they played better. To know that, you need both sides: actual points and process data. If you have only one side, every comparison is a fabrication of both. Verification is not the final step of analysis; it is the condition for analysis to exist.

Across my 523 matches, 74% of disputed offside error decisions took an average of 47 seconds to review. I know that number because I recorded each one, each error code, each distance, each ball speed. When I counted every phase of play, I understood that the law judges no one. It only waits to be applied correctly. A hollow report is a phase of play that has never been judged — and the only way to handle it is to state plainly that it has never been judged.

There is a paradox here that no one wants to look at directly. Readers do not want to receive an empty answer. Television viewers do not want to hear an expert say "I have not checked". The pressure to produce a conclusion within three seconds is real, and it comes from the stands, from the newsroom, from the very trade I have lived on for fifty-one years.

But that same pressure breeds an error worse than any refereeing error. When a hollow report is read as a full one, people begin filling the blanks with things that sound plausible: a mid-sized club, a fee that sounds about right, a familiar tactic, a well-known name. Each piece is correct as a template, and the whole building is wrong as a fact. Worse still, an empty risk cell is easily read as "no risk", when it should properly be read as "not assessed". Not assessable and no risk are two entirely different things. Referees do not need protecting. They need to be understood through correct data.

For those who work as referees, I propose one hard rule: if the input data is empty while the labels are full, throw an error instead of publishing the document as a success. A timestamp must be attached before any judgement, because when information exists matters as much as the information itself. At sixty-seven, I do not need to remember everything. I need to know how to find what is right — and sometimes, what is right is to stay silent until I find it.