No Data, No Verdict: 523 La Liga Matches and the Lesson of a 2026 World Cup Mistake

On 16 June 2026, at the Kazan Arena, Uruguayan referee Andrés Cunha stood about fifteen metres from the Australian penalty area and did not blow his whistle. It was the 55th minute of France v Australia in World Cup Group C, and Josh Risdon had gone into Antoine Griezmann inside the box. The ball stayed live. Play continued. Then Cunha stopped, put his hand to his earpiece, exchanged a few seconds with the VAR team in Moscow, and pointed to the spot. In the 58th minute Griezmann converted — the first penalty awarded through VAR in World Cup history.

At that same moment, four thousand kilometres from Kazan, I was sitting in a studio in Valencia with an open microphone and a tangled headset cable. The station had brought me in as its rules expert. In the commentary around that incident I talked about the boundary between shoulder and arm, the armpit zone, and I stated with certainty that the law placed that zone outside the handball definition. I was relying on the version of the law I had learned in 2026. The colleague next to me corrected it live on air: since 2026, the armpit zone has been inside the definition. More than four million listeners heard me get it wrong. The newsroom had to broadcast a correction at the end of the programme.

Thirty years in the trade, and it was the first time I had been contradicted directly, publicly, and correctly. I once got one sentence wrong and lost an entire reputation. If only I had known this back then.

Last month a young editor sent me a long analysis of a La Liga match. The document had all nine required sections: tactics and technique, club finance and the transfer market, form and the public-opinion cycle, league landscape and team positioning, rules and governance, management and the dressing room, risk profile, media narrative and expectations, and the industry transmission chain. Beautiful structure. Correct headings. But under every heading, an empty field. No team name. No player name. No figure. Not one quote. No date. No source.

He messaged me: I couldn't retrieve the source data, so I was going to leave it blank and note that the signal was weak. I replied within two minutes, and I remember writing almost exactly this: a blank document is not a weak signal. A blank document is a failed result, and the only way to handle it is to run it again from the start, not to fill it in with guesswork.

In football we meet a version of this situation every week. A referee has three seconds to decide. A VAR team has roughly forty seconds, twelve camera angles and an offside line. A commentator has ninety minutes of airtime and no right to stay silent. Those three levels do not share the same standard, but they share the same pressure: produce a conclusion.

The laws of football are built to withstand that pressure, and they succeed at exactly one point: the law always has an effective date. Law 12 of the IFAB Laws of the Game, the section on fouls and misconduct, has had its handball definition rewritten season after season. The amendment effective from 2026 moved the shoulder-arm boundary down below the armpit. The 2026 update clarified the concept of an unnatural arm position. An old version of the law still reads fluently, still sounds reasonable, still persuades. It is wrong at the only point that matters: timing.

That is why I keep a colour-coded law reference. Green means the clause is in force and I have verified the original text. Amber means the clause was recently amended, is in a transitional period, and must be read alongside the federation's guidance circulars. Red means the clause is obsolete and must never be cited. Every incident analysis I have written since 2026 carries at least one footnote stating the law version and its publication date. The law does not live in memory. It lives in data.

In August 2026 I began a project I then regarded as a personal pastime. I logged every VAR-involved decision in La Liga and the Champions League: error code, minute, the gap between the moment of the incident and the moment the referee finalised the call, the number of camera angles used, and the added dead-ball time. I logged it by hand, in a spreadsheet, twelve to eighteen rows per match. By March 2026, when the pandemic halted European football, my file held 523 matches.

The first indicator that stopped me was not the number of wrong decisions. It was time. Within the group of disputed offside decisions, seventy-four per cent of reviews ran too long, with an average delay of forty-seven seconds measured from the moment the VAR team began drawing the line to the moment the referee announced the final outcome. Forty-seven seconds of dead ball is not a problem of law. It is a problem of process. One match is only a story. Five hundred matches are the law.

I wrote a forty-eight-page report, attached the original spreadsheet so that anyone could open it and check every row, and proposed a threshold: thirty seconds per offside review. Beyond that threshold, the referee reverts to the on-field decision. The reason is not that the on-field decision is correct. The reason is that a process which cannot be completed in thirty seconds is itself generating a new kind of error: the error of delay.

The Valencia football federation invited me to advise on reforming the process. I accepted, on two conditions. I take part in no meeting concerning referee appointments. And every figure I produce must be published unedited, including the parts that work against my own proposal.

Let me be clear about the 523-match file: it answers one narrow question, and only one narrow question. It measures time, error codes, camera angles. It does not measure bias. I receive a great many letters asking whether I have evidence that big clubs benefit. I answer by stating my limits plainly: across those 523 matches I hold no variable recording crowd pressure, no indicator of media pressure, and no way to separate the pressure of a seventy-thousand-seat stadium from the actual quality of a defence.

That answer is honest, not evasive. It leads to the central principle of my trade: a null result is a valid result.

That principle has a name in technical documentation, three words the football analysis industry almost never uses: not assessable. Using it is treated as an admission of weakness. Yet in any respectable measurement system — medicine, aviation, refereeing — it is a formal output, ranking equal with yes and no.

I realised something more important after rereading that blank report: it was a correct report placed inside a template designed to always contain content. The nine-section template was born from an assumption that data always exists. In modern football that assumption is almost always true, and precisely because it is almost always true, it has become dangerous. When a template always has a field to fill, the person filling it will always find something to fill it with, even when that something does not exist.

Every chapter of my two books must be written three times. The first pass to understand what I want to say. The second to check whether I have data for it. The third to look for whether the data is arguing against me. Most of the cut material sits in the third pass. I call it the data-refuted section, and it is always longer than the printed one.

Across those 523 matches I found a notable pattern of behaviour in myself: my most certain moments were my least checked ones. It sounds like a psychological observation, but it takes a numerical form. I recorded the time I spent checking before forming a judgement, then tracked the share of judgements I later had to amend. In the group that checked for under a minute, the amendment rate was far higher than in the group that checked for over ten minutes. I have not published the specific figure from this analysis, because the sample is still small and collection conditions were not consistent month to month. When the sample is sufficient, I will publish it with the original file, exactly as I always do. At sixty-seven, I do not need to remember everything. I need to know how to find what is right.

There is a simple economic reason why those three words, not assessable, have almost vanished from television. Viewers switch on to hear a conclusion. An expert saying I do not yet have enough data creates three seconds of silence, and three seconds of silence on live television is more expensive than any analytical error. Newsrooms rarely fire people who say the wrong thing. They fire people who say nothing.

I understand that pressure, because I lived inside it, and I lost to it exactly once. My failure in 2026 did not come from not knowing the law. It came from daring to be certain. The distance between I am not sure and I state that is one verb, but it is the entire difference between an expert and someone reading out a bulletin.

Compare that with refereeing, and the structure is identical. The VAR protocol states that an on-field decision is overturned only for a clear and obvious error. The VAR team does not re-referee the match. The system is designed to accept a defined error margin rather than pursue perfection at any cost. A system that demands absolute correctness will always be late, always interrupted, always distrusted. A system that knows its acceptable error margin survives.

That is why I object to framing the debate as VAR right or VAR wrong. The more useful question is: which process costs VAR the least time while holding an adequate level of accuracy. Dry, I know. But the answer to it is what changes the viewer's experience, not arguments about a single decision.

The same logic applies to another subject I have tracked for years: injury and return. When a player comes back from a long injury, public opinion demands almost immediately that he prove himself. Framing it that way creates pressure to go full throttle in the very first match, and that pressure is a medical variable: it raises the probability of re-injury. Minutes played across the first three matches after a return is something we can count, publish, and use to protect the player. Demanding that a human being prove his worth by risking his body is cruel, and cruelty is not a strategy.

I also dislike the phrase protecting referees. Protection is a political act. What referees need is to be described accurately. Referees do not need protecting. They need to be understood through correct data. When a decision is criticised, the job is to open the tape, count the angles, note the minute, and check the law version in force at that moment. The job is not to write a collective apology.

In 2026 I wrote a piece defending a La Liga referee using a metric on how often he had decisions overturned by VAR. Three days later, going back through the raw file, I discovered I had sampled wrongly: I had included cases the VAR team reviewed without intervening. The metric in the article was wrong. I did not correct it quietly. I wrote a correction published in the same slot as the original piece, stating exactly what was wrong, what the right figure was, and how I had miscalculated. I wrote a defence of a referee using bad statistics. That was the time I betrayed my own principle.

After that I set a hard rule for every piece: no figure goes to print without three things — the competition, the collection period, and the collection conditions. Without those three, the number stays in the file. Readers have the right to open the file and recompute it themselves. If I do not give them that right, I am not analysing. I am persuading.

For Vietnamese football and Vietnamese audiences, this has a specific consequence. You consume a large volume of analysis about competitions you do not follow directly, mostly through translations and short extracts on social media. In that transmission chain, the things most easily lost are dates and sources. A line quoted from 2026 can circulate through forums for seven years while keeping its persuasive sheen, even though the version of the law it refers to was replaced long ago. The only defence is the habit of asking in reverse: where did this come from, which date, which competition.

I do not think the problem lies with the fans. Fans have no obligation to verify. That obligation belongs to the people paid to speak, myself included. When someone like me gets it wrong and does not correct it, the error does not stop at one sentence. It enters the shared memory of a generation of viewers, and there it outlives every legal text.

What I propose is not large. In the official match report, add a field called not assessable, and let referees, VAR teams and technical committees fill it in when an incident lacks data. A blank field that is recorded is worth more than a field filled with guesswork. For media analysis, add one publication rule: every figure must carry its competition, its period and its collection conditions. And add one process: when an error is found, correct it in the original slot, at the same font size, with the same airtime. A correction buried in a footnote is not a correction. It is a ritual for saving face.

The big tournament cycle is approaching, and I know I will again sit before a microphone with millions listening. I will again have to answer within three seconds. What I carry into the studio is not a better memory. It is a spreadsheet already open, a colour-coded law reference, and one sentence I have learned to say without fear: I have not yet been able to verify that.

More than five hundred matches have taught me that a correct verdict is not the fastest verdict, nor the one most people agree with. It is the verdict that can be opened, reread and checked line by line. When I counted every passage of play, I understood that the law judges no one. It only waits to be applied correctly. If one day audiences demand that from the people sitting in the studio, football will not need more heroes. It will only need more people who read the law correctly. A shocking decision is not recklessness, if it is built on five hundred foundations.