Wrong Label, Right Whistle: When a VAR Data Packet Doesn't Match the Pitch

Two in the morning in Hai Phong. I pulled up frame 41 of a passage of play, and on screen the semi-automated offside line appeared neat and tidy: the visiting striker half a metre beyond the last defender. A crisp line. Round numbers. But at the very header of the data packet, the match ID did not match the match unfolding in front of me. The shirt numbers in the skeletal tracking layer did not match either. Nothing on the pitch in the 71st minute resembled what the system was drawing for me.

I found that offside error at two in the morning, after everyone had gone home. This time the mistake was not in the player's foot. It was in the label on top of the data packet.

There are offside errors nobody sees, but the camera never blinks. There are also errors the camera sees very clearly — except it is looking at the wrong match.

The work that happens before the line is drawn

After years in the VAR room, I have learned that the stands only ever see the final moment: the line appears, the referee gestures, the crowd reacts. The real work happens earlier, in silence. A semi-automated offside decision does not begin with a whistle. It begins with a data packet: the skeletal tracking stream of 22 players, the match clock, the calibration parameters of each camera, and a label identifying which match, which half, and which pitch that packet belongs to.

A label sounds like dull administrative detail. It is the single most important element in the entire process. The label is the root; the line is only the branch. When the root is wrong, the more beautiful the branch, the more dangerous it becomes.

What is frightening is that a failure at the label layer does not produce a broken image. It produces a perfect one. The system raises no alarm, shows no exclamation mark, changes no colour. It draws the offside line correctly, to the centimetre — for a different match.

Based on my experience covering matches in the V.League and the AFC Cup, I categorise three mandatory checks before a data packet is allowed into a decision. I call them the three shields: provenance, timeline, and physical plausibility.

The three shields

The first shield is provenance. Who supplied this packet, from which system, which software version, and is there an independent record confirming it? A data packet with no clear provenance is a packet that has no right to decide anything, no matter how persuasive the values inside it look. In 2026, in the AFC Cup group stage, Hai Phong FC faced Ceres-Negros and I was the VAR assistant. In the 78th minute, the visiting striker Fidelis Ikiri was 0.3 metres offside before scoring an equaliser to make it 2-2. I sent the signal up to the refereeing team, the goal was disallowed, and the match finished 2-1 to Hai Phong. The coaching staff never knew I had intervened, and I did not need them to. But had the label been wrong that day, my signal would have been correct in data terms and wrong in match terms. Nobody in the stands can tell the difference.

The second shield is the timeline. The packet's clock must match the match clock, match the fourth official's log, match the broadcast recording. When those three time sources diverge, that is the first warning sign. In one file I examined, the document header carried one timestamp, the body cited another, and both contradicted the data inside. Three fragments that cannot coexist in a single reality. In football, that pattern usually appears when a packet from match A is assigned to match B, or when an old record is blended into a live feed. To the viewer, the match flows on normally. To the person in the VAR room, everything has been off for a long time.

The third shield is physical plausibility. A player sprinting at 41.7 km/h for two continuous seconds does not exist. A player covering 14 metres in 0.8 seconds does not exist. A shot speed beyond the biological ceiling of the human body is a sign of calibration error, not of a wonder strike. When a value breaks the limits of physics, the right response is not admiration but a pause and a question: where did this data come from, and who is it talking about?

The lesson from the 42nd minute

In 2026, at the World Cup round of 16 in Russia, in the match between Spain and Russia, I was one of three VAR analysts supporting the referee. In the 42nd minute I failed to spot Gerard Pique's handball in the penalty area. After review, the referee awarded Russia a penalty, the score finished 1-1, and Russia won the shootout. I blamed myself for three weeks, quietly re-watching all 64 matches of the tournament and taking notes on every VAR incident. I did not tell my colleagues.

That mistake taught me something unrelated to eyesight. It was about tempo. I had watched fast enough to see, but not slowly enough to be certain. From then on I proposed a rule: before any decision, spend two extra seconds verifying the provenance of the packet in front of you. Those two seconds will not save a goal. They will save the fairness of an entire match.

What I learned at a deeper level: errors in the VAR room are rarely errors of the eye; they are usually errors of the source. A referee misreading a passage of play is a moment's problem. A mislabelled data packet is a system's problem, and it repeats itself until somebody notices.

Who is responsible?

The instinct of the crowd is to blame the referee, or the technology. Both are ways of blaming the eye. But when a line is drawn from a faulty data packet, the referee is not wrong — the referee is deceived by a system that appears objective. The technology is not wrong either — it simply draws exactly what it was given.

The blind spot sits upstream: data governance. Who applies the label, under what procedure, who re-checks it, and who is told when a mismatch is found? In most of the controversies I have followed, that question is never asked. People argue over centimetres while the real problem is the name of the match.

In 2026, when the pandemic halted football, I was doing VAR analysis for a club in financial crisis. A 17-year-old talent, Nguyen Van Truong, was written off as mentally fragile after a few poor matches. The public reflex then was identical to the public reflex before a controversial VAR call: find the youngest, weakest person and put everything on them. I chose differently — I sent a report on his strengths to the technical director and asked for him to train separately with the U19s. Six months later Truong made his debut and scored the goal that kept the club up. Patience, sometimes, is a form of evidence.

There is a paradox I believe to be true: more cameras do not make us more correct; they only make us more confident. A system with 12 cameras and a wrong label will deliver a wrong conclusion with enormous persuasiveness. A system with six cameras and a correct label, with someone cross-checking, will deliver a humbler but more trustworthy verdict. In those same years, external data analysts began drawing conclusions about players from spreadsheets detached from the actual rhythm of matches. They too are applying labels — they simply do not call it that.

What I propose

A two-second procedure. Before any semi-automated offside decision, the operator reads out three items: match ID, match clock, camera calibration code. Those three must match the fourth official's log. If they do not, stop. No exceptions for packed stadiums, no exceptions for the final minutes.

The referee is the only person on the pitch not allowed to pick a side — and I stand behind them. My job is not to make the line look prettier. My job is to make sure that line is being drawn on the right pitch, for the right match, at the right moment.

A millimetre changes a club's fate; I have learned to live with that. But a wrong label can change the fate of an entire season, and nobody will remember the name of the person who applied it. Perhaps it is time we started recording those names — not to point fingers, but so that fairness has something to stand on.