Kazan 2026, Luis Diaz's disallowed goal, and three kinds of empty data in the VAR room

In the 90+3rd minute at Kazan, on 27 June 2026, the assistant referee raised his flag. Kim Young-gwon had just put the ball past Manuel Neuer, and the flag drew a familiar line: offside. Nobody on the pitch reacted immediately. Three minutes later the referee confirmed the goal, because the ball had touched Toni Kroos before reaching Kim Young-gwon, and under IFAB Law 11 a player who deliberately plays the ball restarts play for the opponent. South Korea led 1-0. On 90+6, Son Heung-min rolled the ball into an empty net. Germany left the 2026 World Cup bottom of their group, eliminated in the first round for the first time since 2026. I was in the technical room at the Kazan stadium that night. What I recorded was not the shot, but the distance between what the assistant referee could see and what the law required him to see. In 2026, the whole football world turned upside down, and I learned to stand still and observe inside the storm. VAR appeared at a World Cup for the first time, and it forced the entire industry to redefine something that had seemed obvious: how much data is enough to make a decision. Russia 2026 was the first World Cup to use VAR in all 64 matches. FIFA published figures showing accuracy on match-changing decisions rising from 95.2 percent to 99.3 percent. At the same time, the tournament produced 29 penalties, the highest tally in World Cup history up to that point. Those two facts travel together, and they say something few people want to admit: when cameras see more than the human eye, the number of decisions that must be made goes up, not down. The VAR protocol issued by IFAB allows intervention in only four categories: goal or no goal, penalty or no penalty, direct red card, and mistaken identity. The threshold is a "clear and obvious error" or a "serious missed incident". Most spectators never read that paragraph closely. VAR does not go looking for mistakes. VAR only corrects a mistake when it is large enough to be seen from several camera angles. Between those two things lies a grey zone, and almost every argument in the stands sits inside it. People watch a match with their hearts. I watch it with lines that have already been drawn. From Kazan to now, I classify refereeing errors into three types, and all three are data problems. The first type is data that does not exist. The assistant at Kazan stood at the corner of the pitch, his sightline blocked by a wall of defenders, and he simply could not see the ball touch Kroos. No technology can repair what the eye never saw. Only another angle can be added. That is why FIFA brought semi-automated offside technology to the 2026 World Cup: 12 cameras tracking 29 data points on each player's body, calculating positions within about half a second, and sending a signal to the VAR team. According to FIFA's published figures, the average handling time for a complex offside situation fell from roughly 70 seconds to about 25 seconds. Saving nearly a minute is not about speed. It is about a decision that is less distorted by emotion. The second type is data that exists but is not processed. On 30 September 2026, at Tottenham Hotspur Stadium, Luis Diaz scored for Liverpool. The assistant raised his flag for offside, the VAR team checked and reached the wrong conclusion. The audio of the exchange between officials, later released by PGMOL, shows the phrase "check complete" being spoken when the check should have continued. The camera was not wrong. The camera was right. The failure was in the protocol: one signal misread between two people, and nobody in that chain had enough time to ask again. The third type is data that is entirely empty. This is the hardest kind, and the one the refereeing profession is not allowed to dodge. Some matches are played on pitches without enough cameras. Some incidents are completely obscured. Some youth competitions have no VAR at all. The referee still has to blow the whistle. Walking out alone with a whistle, sitting in front of a monitor with a straight-on angle, in the end I am still the man in the middle of the storm. In 2026, aged 37, I commentated on Arsenal against Tottenham and mispronounced Harry Kane's name three times in the first half. Listeners complained loudly. That night I watched the whole recording back and realised the error was not in my pronunciation. It was that I had gone on air with an empty information sheet. I had taken the commentary seat without preparing enough data on FIFA's new offside law. From then on I built a 27-point technical checklist before every broadcast, and the first rule on it is this: never begin from a blank cell. Based on my experience following matches in the Premier League and eight World Cups, the problem in modern football is no longer a shortage of data. The problem is that empty data is still routinely treated as full data. A VAR team with three camera angles but missing the decisive one can still conclude with confidence. A recruitment department with ten recorded matches of a 19-year-old can still assemble a complete profile. A long-term representation contract can turn a player into a media product, and then his words are written by somebody else. All of these are large decisions taken on a thin base of evidence. There is a line I still repeat to referee trainees in London: a decision made on empty data is still a decision. It does not become meaningless just because you lacked information. It simply becomes a decision where you must disclose your level of certainty. A head referee is not someone who never errs; what matters is what you do after the mistake. This is where I want to push back against the popular belief that technology will remove controversy from football. Seven years of evidence since Kazan says otherwise. VAR does not erase argument, VAR relocates it. People used to argue over whether the referee had seen the incident. Now they argue over where the threshold of "clear and obvious" is drawn, and who draws it. The number of contentious decisions has not fallen. Their address has changed, from the pitch to the technical room. There is another blind spot that is rarely discussed. When technology sees more, teams start playing in ways that avoid technology. Defenders hold the offside line half a step deeper. Strikers time their runs to the count. At the 2026 World Cup I counted a large number of goals ruled out for gaps of a few centimetres, distances the naked eye cannot resolve in real time. Football is teaching players to play to the machine's line. That is the tactical consequence of an administrative decision, and it has not been analysed enough. The club side is no different. Load management is presented as a science, with GPS metrics, running volume and heart-rate data. But the commercial friendly calendar is still what ultimately decides who rests. No amount of data can change a contract that has already been signed. A penalty can change the fate of a match; a contract can change the fate of a club. So I propose one small change that could be made immediately. Every time a VAR team intervenes, publish the confidence level of the decision alongside the conclusion. Three camera angles, one blocked, medium confidence. If a referee could say that sentence on camera, the argument would shift from "you were wrong" to "did you have enough data". The Premier League took the first step when it began releasing VAR audio monthly from 2026, under Howard Webb's Match Officials Mic'd Up. The next step is to publish the missing data too, not only the data that was available. Every argument in the stands, every argument outside the pitch, has an answer sitting in some camera angle. And when there is no camera angle at all, the most honest answer is to say that there is no answer yet. An empty data sheet is not a conclusion. It is a state that needs to be declared. Football only improves when the person holding the whistle dares to say: this is my decision, made with this much information.