VAR Diary: A 14-Second Crack in the Mirror of the Rules

Minute 67, matchday 29 of the 2026 K League Classic. Seoul World Cup Stadium. Lee Dong-gook put the ball in the net and the stands erupted. In the VAR room a few dozen kilometres away, I saw what forty thousand people did not: his right foot had crossed the last defender's shoulder by exactly 0.3 metres.

I started rewinding the rear-angle feed. Three times. Then four. Then five. By the time my signal left the room, the clock had run to 14 seconds — double the 7-second standard FIFA writes into its operating guidelines. The referee on the pitch had long since awarded the goal. There was nothing left to fix. I was 23 years old, and that was the first VAR error of my life.

Three nights later, I was still rewinding that footage. Not to find the referee's mistake. To find the mistake of the observation machine itself.

That is why I started keeping a log.

Context: when VAR became an institution

The K League Classic began operating VAR in the 2026 season, earlier than most Asian leagues. In that period, every VAR room in South Korea ran on a shared protocol: an alert signal had to leave the room within 7 seconds of a suspect incident being confirmed; the on-field referee could only review when a signal arrived; and the final decision belonged to the whistle-holder on the pitch.

It sounds clear. In practice, it was not.

The problem was this: the 7-second protocol assumed the person in the VAR room needed only one camera angle to decide. In reality, every offside situation requires at least three angles — from behind, from the side, and from above — to reconstruct the relative positions of foot, shoulder and ball at the exact moment of contact. Three camera angles multiplied by loading time, rewinding time and cross-checking time is enough to push a person from 7 seconds to 14 without any laziness at all.

The error in Seoul that year belonged to an assumption, not to a person. The assumption that the observation tool was fast enough to keep pace with the match.

And this is what I learned over the following nine years: every VAR error is a crack in the mirror that reflects the laws. The crack is not in the person holding the mirror. It is in the fact that the mirror was polished with a kind of light that does not exist on the pitch.

The core: data from three studies

In 2026, I was sent to Russia as a VAR analysis assistant for a South Korean broadcaster. My main job was to record every handball incident so it could be checked against IFAB's new law, rather than to commentate on matches.

I collected 27 handball situations across the tournament. I then classified them on three criteria: the position of the hand relative to the shoulder, the direction of the ball's travel, and the player's reaction distance. The result cost me sleep.

Only 31% of those 27 situations were handled consistently under the same reading of the law. The other 69% were handled in three different directions, depending on where the referee stood, how many seconds they reviewed, and — most importantly — which team was benefiting from the state of the game.

I wrote a 40-page report and sent it to my editors. They ran exactly one small chart, with no caption. I set up a personal blog and published the entire dataset without asking anyone's permission. The piece drew 50,000 reads from referees, sports lawyers and plenty of fanatical supporters.

The trap of 2026 was not in the hand; it was in the belief in a definition that does not exist.

IFAB writes the handball law in language that tries to cover every case: "the hand or arm has made the body unnaturally bigger". But "unnaturally" is a comparative word, and nobody defines what it is compared against. Every referee compares it against a template in their own head — a template built from the match they watched last week, from the referee they admire, from the pressure in the stands.

Two years later, when the 2026 pandemic pushed global football into a state without spectators, I had the chance to test this at a much larger scale. The broadcaster cut my contract for budget reasons. Instead of worrying, I locked myself in a room for six months and analysed 1,247 VAR decisions from five European leagues — the Bundesliga, the Premier League, La Liga, Serie A and Ligue 1.

Two findings emerged.

First, with no spectators, referees' VAR review time fell by 22%. That sounds logical: no shouting, no pressure, faster decisions.

Second, and this is the notable part: the rate at which the original decision was upheld rose by 15%. In other words, referees did not change their decisions more often — they defended their decisions more often, even with less pressure on them.

This breaks a popular assumption. People often say VAR fails because the stands apply pressure. My data says otherwise. The stands do not make referees wrong. The stands make referees hesitate. Remove the stands and they hesitate less — but they do not become less stubborn.

The noise of the stadium is not written into the laws, yet it carries legal weight.

Then came 2026, when, as a mid-level analyst at a consultancy, I built a player-evaluation model based on VAR data. The model showed that defender Kim Min-jae committed 0.73 fouls per match in Serie A — a "high card-risk" level. I advised the company not to recommend signing him.

Napoli signed him anyway. Kim Min-jae became a pillar of the side that won Serie A in 2026. At the end of that year, I wrote a ten-page self-criticism and removed the model from the system.

Where did I go wrong? I ignored two variables. First, the covering ability of teammates — a defender commits many fouls in a loose defensive system, but far fewer in a system where someone shields him. Second, Italian referees read the law on contact very differently from South Korean referees. The same challenge, in Serie A, is a fair contest; in the K League, it can be a foul. My model measured the player's behaviour but could not measure the refereeing environment in which that behaviour took place.

I call that lesson the natural position.

The counter-intuitive angle

After nine years, I believe most VAR controversies do not originate in referee error. They originate in a false expectation placed on the tool.

Supporters imagine VAR as a judge holding the whole truth in their hands. In reality, VAR is a set of cameras placed at positions chosen by human beings, operated by people with limits on their time and perception. When one camera angle cannot see a hand, it does not follow that the hand did not touch the ball. When an offside line is drawn one frame out, the margin of error can reach several dozen centimetres — larger than the distance that needed to be determined.

This is the biggest blind spot in the public debate: we argue about conclusions, but not about the resolution of the data that produced those conclusions.

A wrong decision does not ruin a match; the silence after it is what ruins trust.

In the K League, the organisers make a habit of publishing the VAR report after each match. In Vietnam, domestic competitions usually publish only when a formal complaint is filed. In Europe, UEFA publishes selectively, usually for incidents that have already caused enough controversy that staying silent becomes more expensive than speaking up. Three ways of handling the same question: when does the public have the right to know the reasoning.

None of those three answers is "natural" in an objective sense. Each is an imposition by a system of power on an identical law. The only thing I can compare fairly is the degree of consistency: a federation that publishes regularly will be scrutinised more, but it also learns faster.

What we seek on the pitch is not justice, but an excuse to stop arguing.

A forward-looking thought

I no longer believe VAR can be fixed by training better referees. I believe it must be fixed by publishing the raw data — response time, number of camera angles used, rate at which decisions are upheld — for every incident, immediately after the match.

Supporters do not need to know whether the referee was right or wrong. They need to know what was seen, and what was obscured. When the distance between those two things is laid bare, the argument changes direction: from "he is blind" to "the system lacks a camera at that angle".

VAR was born from the fear of error, but it nurtures the fear of a late truth.