The Evidence Standard Inside the VAR Room: Why Referees Cannot Rule on Guesswork

Minute 68 at Lach Tray, and the whistle cut across the crowd. Referee Nguyen Van Kien went straight to red for centre-back Nguyen Huu Phuc. I was on the stand, about forty metres from the incident, and within three seconds I heard the verdict from the rows behind me: match-fixing. That night I rewound the footage seventeen times, checked it against IFAB Law 12, and wrote an analysis before morning. The piece drew 45,000 reads, nine times my average at the time. But what I carried out of that match was not the figure. It was the realisation that a football culture can convict a man in three seconds, on a standard of proof lower than any courtroom would accept.

Five years later, in December 2026, I sat in the VAR operations room in Lusail for the World Cup semi-final between Argentina and Croatia. In front of me was the skeletal-tracking screen, semi-automated offside running in real time, the offside error margin printed at 0.1 metres, down from 0.4 metres on the previous generation of technology. Nobody in that room had to guess. The distance between those two worlds is the distance between a football culture governed by process and one governed by emotion.

Context: where the standard of proof sits inside the Laws

The VAR protocol was approved by IFAB in 2026 and first appeared at the 2026 World Cup. The protocol itself is not about right and wrong. It is about the intervention threshold. Two lines are set in bold in the original text: minimum interference, maximum benefit; and clear and obvious error. At the 2026 World Cup I collected data across 64 matches and recorded an average of 3.2 minutes of dead ball per match arising from referee interventions. The final between France and Croatia, refereed by Nestor Pitana: 31 fouls awarded, only four yellow cards, roughly eight fouls per card. That was a match managed by a tolerance threshold, not by a reflex to book.

Then Euro 2026 arrived and cracked that confidence. In the quarter-final between Spain and Switzerland, Spain's opening goal stood although Ferran Torres was 0.3 metres offside — an error by the VAR technician, who failed to draw the offside line on the frame. When I compared data from the 48 Euro 2026 matches against the 2026 World Cup, the VAR error rate was 1.8 times higher. My writing at the time forced UEFA to issue an email explaining its new procedure to media outlets. I mention this not to boast. I mention it to show that technology does not generate a standard on its own. The standard comes from the people who write the procedure, and in Vietnam, those people have never published their document.

Core: the burden of proof sits with the reviewer, not the referee

This is the point almost every V.League argument skips. The VAR protocol places the burden of proof on the VAR team, not on the on-field referee. The system's default is that the on-field decision stands. To overturn it, the VAR room must demonstrate a clear and obvious error — an evidentiary threshold, not a probability threshold. In other words, the Law does not ask how likely it is that the decision was correct. It asks whether there is a frame that proves it wrong beyond argument.

The asymmetry here is enormous and rarely discussed. The on-field referee sees the incident once, at live speed, from one angle, sometimes twenty metres from the point of contact, while twenty-two players are shouting at him. The VAR team has twelve camera angles and the right to slow playback to 50%. On the data, the person in the VAR room knows more. Yet the protocol still hands the final decision to the on-field referee, because there is a kind of data no camera records: the rhythm of the match. A collision in the 12th minute, after three consecutive fouls, is a different story from the same collision in the 88th minute of a decided game.

A referee is not convicted by noise from the stands, but by a frame that can be rewound — and when no such frame exists, the only procedurally valid ruling is to let the on-field decision stand.

That leads to a deeper problem in Vietnam. VAR has been trialled in V.League since 2026 at selected matches, but it arrived alongside unsynchronised infrastructure: limited camera counts, non-standardised camera positions, frame rates not perfectly aligned between machines and, most importantly, no published audio of the exchange between the on-field referee and the VAR team. When the entire decision process is sealed, spectators are left with one source of evidence: the broadcast replay, usually the angle chosen by the production team. From there, every argument stops being an argument about Law. It becomes an argument about trust.

I have seen this football culture do the opposite. In 2026, when V.League was suspended indefinitely after round five, the contract dispute between Song Lam Nghe An and striker Michael Olaha over termination on grounds of force majeure created a problem no existing clause answered. I drafted a 15-page handbook summarising the contract provisions affected by distancing orders, and the Vietnam Football Federation sent it to 28 member clubs as an official reference document. A federation can write a standard when it is forced to. The problem with refereeing sits somewhere else: nobody is forced to draft.

A pandemic does not create gaps in the Law — it only knocks on the gaps that were already there. The same holds for VAR. Technology did not create the errors of Vietnamese referees. It simply shines a light into gaps that have sat inside our training, appointment and assessment systems for years.

The contrarian angle: more intervention does not make football fairer

The default reaction of spectators to a controversial decision is to demand that VAR intervene more. That demand is emotionally reasonable and systemically wrong. If a VAR team accepts overturning decisions that do not meet the clear and obvious threshold, the first consequence is not more fairness — it is that the on-field referee loses control of the match. Players work that out within ten minutes. After that, every contest becomes a lobbying exercise, and every match runs fifteen extra minutes of dead ball, not because of the Law, but because pressing the referee has become a tactic.

There is a second risk few people name. When we judge referees only on outcomes, a referee who gets a decision right through a lucky guess is praised exactly as much as one who gets it right through process. Nobody audits the process. At the 2026 World Cup I noted that referees were assessed on video after every match, incident by incident, position by position. In V.League that assessment exists, but the results are never published. A closed assessment system can only produce two kinds of referee: the protected and the abandoned. Neither gets better.

And here is the final contrarian point. We tend to assume Vietnam's VAR problem is technology. It sits elsewhere. We imported IFAB's clear and obvious threshold wholesale without writing a Vietnamese interpretation of it. A V.League referee is applying a standard he cannot articulate, in a league whose speed, contact levels and fixture density differ sharply from Europe. People see the red card; I see a clause written in too much of a hurry. A standard that is never written down cannot be taught, cannot be assessed and cannot be improved.

A refereeing error is never a single event — it is the scorecard of the entire Law book.

Takeaway: three documents that must be written

If the VFF refereeing board wants to move the argument off social media and into a meeting room, three documents need to be drafted and published in the coming season. First, a Vietnamese edition of the VAR intervention threshold, defining precisely what counts as a clear and obvious error under current infrastructure conditions, with examples drawn from matches already played. Second, publication of the audio between on-field referee and VAR team after every match that uses the technology. Third, a per-match referee assessment report in condensed form, stating how many incidents were reviewed, how many stood, how many were overturned, with causes classified.

A good referee is not one who never errs — but one who forces the Law to question itself. A football culture matures when it stops convicting in three seconds and starts demanding evidence across thirty pages. Amending a Law takes ten minutes; admitting a Law is wrong takes ten years. In V.League, we have had those ten years. It is time to start.