VAR in the V.League: When a Verdict Is Built on an Empty Dossier

On the monitor inside the operations room, the frame freezes on the last defender's foot. The VAR operator rewinds four times, but the only usable angle is blocked by another player's shoulder. The referee stands in the middle of the pitch, hand on his earpiece, waiting for an answer nobody can give with certainty. The final whistle goes, the score stands, and in the stands thousands of people believe they have just been robbed of a goal.

In November 2026, at San Siro, I sat in a similar position. In the 56th minute, Gonzalo Higuaín scored to put Juventus 2-0 up against Milan. The feed in the VAR room showed the Argentine forward had been 0.2 metres offside. I saw it. I hesitated. And I did not recommend a review. Milan lost 0-2. After the match, the referee supervisor criticised me in front of the whole team in a tone so calm it was uncomfortable: a VAR operator is not permitted to stay silent merely because he fears being wrong.

What I learned that night was not a lesson in courage. It was a lesson about evidence. A decision can never be better than the information used to produce it. And when the input is empty, the whistle still has to sound — but it sounds in the dark.

Context: the chain of evidence and the break nobody sees

VAR operates like a chain, not like a switch. An incident happens on the pitch, a camera records it, the frame is calibrated, the offside line is drawn, the law is interpreted, the decision is made, and finally it is communicated to the crowd. Seven links. At every link, some information is lost: resolution, frame rate, camera placement, the distance between lens and the plane being measured, even the runner's speed at the instant the ball leaves the foot.

Viewers only see the final link. They see the referee make the signal, they see the score change or not change, and they draw conclusions about the entire chain behind it. That is why arguments about VAR rarely end: people argue about the conclusion, while the problem usually sits in the input data.

In analytical work, I encounter another version of the same disease. Dossiers that look complete: a title, headings, tables, a clear nine-part structure. But read cell by cell and every one of them says there is not enough information. No team name, no player name, no source, no timestamp, not a single verifiable fact. The form is intact, the content empty.

A conclusion can never be more accurate than its input data. This is not an aphorism; it is an operating rule. If the first link returns zero, the chain does not fail at the end — it fails at the beginning, and everything after that is theatre.

What is worrying is that Vietnamese football is producing such dossiers every matchday. It is just that they are not written on paper.

Core: the 37-point checklist and the price of silence

After the San Siro night, I locked myself away for a month to review 47 similar incidents. Not to find reasons to justify myself, but to answer one question: what made me hesitate? The answer turned out to be very concrete. I hesitated because I had no pre-written standard to check against. I had only a feeling, and a feeling cannot be defended before a panel.

From that I built a 37-point checklist. It starts with things that seem trivial: camera identity, frame rate, the moment the ball leaves the foot, the moment the foot touches the ground, the defender's body orientation relative to the direction of the pass. Then it climbs: whether the attacker is active or passive, whether the ball came from a deliberate touch or a deflection, the distance between two feet in the decisive frame, the system's permitted margin of error at that distance.

Every criterion carries a weight. When the aggregate score crosses the threshold, I recommend a review. When it does not, I stay silent — but this time it is an informed silence, not a fearful one.

This mechanism matters more than it appears. It turns a gut decision into a decision that can be re-examined. And it lets me say to myself a line I still use as a working principle: Before I blow the whistle, I review myself.

But a checklist only solves half the problem. The other half lies on the data side: whether there is anything to feed into the checklist at all.

When the data is complete: the Mbappé lesson and the limits of intuition

In June 2026, before France met Argentina in the World Cup round of 16 in Russia, Sky Sport Italia invited me to work as a VAR commentator. I spent three days before the match doing something colleagues considered pointless: reconstructing Kylian Mbappé's speed profile from his last 14 matches in Ligue 1 and the Champions League.

The result: his peak sprint speed reached 36.5 km/h, roughly 2.8 km/h above the average speed of Argentina's defenders over the same period. But that number, standing alone, means nothing. What interested me was structure: Argentina's back line held a high, narrow defensive line, with two older centre-backs who tended to drop with the ball rather than with the man. When an opponent plays the ball into the space behind, the gap between them and Mbappé widens over the course of the match, not within a single incident.

I wrote an analysis piece of about 1,200 words setting out a specific scenario: Argentina's defensive structure would break between the 60th and 70th minutes, when stamina drops and the distances between lines grow. The match unfolded exactly as the model suggested: Mbappé won a penalty and scored twice, France won 4-3. The article was later shared more than 5,000 times.

The point I want to make is not that I predicted correctly. The point is the structure of an evidence-based prediction: define the variable, establish the baseline, set the time window, then let the match test it. That dossier had complete input data. It was not a prophecy; it was a model capable of being disproved.

A model that can be disproved is an honest model. A conclusion that cannot be disproved, because there is nothing to disprove, is merely a statement decorated with a confident tone.

Systemic diagnosis: crowd noise is a variable

In October 2026, when Serie A returned in empty stadiums because of COVID-19, Milan went on a run of seven matches without a win. Public opinion blamed the sale of André Silva to Monaco for 35 million euros. I did not believe that was the cause. One departing striker cannot explain why an entire system lost its counter-attacking capacity.

I analysed transition data from the previous 14 matches and found an anomaly: Milan's back line lost roughly 42 percent of its counter-pressing capacity in conditions without crowd noise. It sounds absurd, but the mechanism is clear. Crowd noise is a signal. It tells players when to push up, when to drop, who has been left unmarked. When that signal disappears, the pressing system loses an input layer it had always implicitly relied on.

I wrote a 30-page report for the editors of La Gazzetta dello Sport proposing a three-phase recovery plan. It was published in full.

From then on I moved firmly toward systemic diagnosis. Before criticising a player, I have to answer one question: what mechanism produced this error, and is that mechanism repeatable? One failed action is an accident. A failed action repeated ten times is a design.

Applying it to the V.League: technology cannot fill a data gap

From the 2026-2026 season, VAR began to be introduced into V.League 1 in stages, with training support from international experts and the Vietnam Football Federation coordinating. This is real progress, not a slogan. But I want to talk about the hardest part of VAR, the part no trophy recognises: the dossier.

VAR demands three things that are not always prepared together. The first is equipment: number of camera angles, camera placement, calibration capability. The second is people: certified operators, trained in image calibration, drilled on the threshold for clear and obvious error. The third — and this is the most neglected — is the log: a record of every disputed incident, which angle was used, what the margin of error was at that distance, and the reasoning for recommending or not recommending a review.

Equipment can be bought. People can be trained. But a log only forms when a culture accepts that recording an important decision matters more than defending it.

When a league has no log, every dispute becomes an endless dispute. Nobody can say whether an error is isolated or systemic, because there is no data series to compare. The cameras may have recorded it, but if nobody stored it, the cameras never existed.

And this is what worries me most: in leagues like that, both the defenders of referees and their attackers are talking about an empty dossier. One side says the decision was correct because the referee is a professional. The other says the decision was wrong because their club was harmed. Neither has a single fact to verify, and therefore neither can be disproved.

I do not trust my eyes, I trust the slow-motion replay. But if the slow-motion replay is not stored and not published to a consistent standard, then I am left with my eyes alone — and the eyes, as the San Siro night taught me, are an instrument that can be deceived by the user's own fear.

The five layers of checking a football dossier needs

In daily work I apply five layers of verification before allowing myself to write a single assertive sentence. It does not require expensive software. It requires discipline.

The first layer is identity. Every dossier must name its entities in full: clubs, players, referees, competitions, governing bodies. A dossier about an unnamed club is a dossier that cannot be verified, no matter how many pages it runs.

The second layer is source quality. Not all information carries the same weight. A competition organiser's statement, a referee supervisor's report, a post-match interview and a social media post cannot be placed side by side as equals. When the source tier cannot be established, the conclusion must be downgraded one level of confidence.

The third layer is timing. Information without a timestamp cannot be judged as still valid or already stale. In football, a claim that is accurate in September can be entirely wrong in February, simply because the fixture list and fitness have changed.

The fourth layer is sample size. A player scoring three goals in two matches is not yet a trend. A team winning four in a row is not yet a system. I always ask myself: if this situation were repeated ten times, how often would the outcome differ?

The fifth layer is cross-checking. This is the most expensive layer and the most frequently skipped. Every conclusion must be compared against at least one independent source: event data, footage from another angle, or a qualified person with no shared interest with the original source.

A dossier that passes all five layers can still be wrong. But it is wrong in a way that can be corrected. And in an industry where every conclusion is re-examined seven days later, the capacity to correct matters more than the capacity to be right the first time.

Contrarian angle: the real enemy is not error, it is the phantom dossier

There is a widespread belief that VAR's biggest risk is wrong decisions. I think the bigger risk lies elsewhere: decisions that are made, defended, published and written into history — with no input data behind them.

I call this the phantom dossier. It has the shape of a conclusion: structure, headings, a confident tone, statistics cited without a clear source. It lacks exactly one thing — a verifiable fact. And because it lacks exactly that, it cannot be disproved, cannot be corrected, and cannot be removed.

A phantom dossier is more dangerous than an error, because an error leaves traces. A wrong decision can be reviewed, and the review itself generates data for next time. A phantom dossier leaves nothing but a conclusion. It contaminates the whole analytical chain behind it, like an empty link in a chain: everything after it still moves, but no force is transmitted.

Here I have to check myself. I am a person who believes in data, but that belief has one compulsory limit: Every verdict needs one review, including the verdict of the data. A model can be wrong because the input data is noisy, because a variable is mis-set, because the sample is too small, or because the modeller unconsciously selected data to confirm what he wanted to believe.

For Vietnamese football, this means: do not chase Europe's technology tier. Build the lowest tier first — a standardised incident-logging system, applied uniformly to every match, every round, regardless of whether the club is big or small. A league with a good log and three cameras is more trustworthy than a league with twelve cameras and nobody writing anything down.

Technology does not kill football; it kills blind faith. But technology does not create truth either. It only makes truth checkable — provided there is a dossier to check.

A closing thought: an empty dossier is not an ending, it is a signal

An empty dossier does not say the story does not exist. It only says the story has not yet been collected. Between those two things lies the entire difference between an honest conclusion and a decorated one.

After many years in this trade, I have realised that an analyst's greatest value is not how many judgements he issues, but whether he knows what he is missing and says so plainly. An analyst who says this dossier is not yet sufficient to conclude is more useful than one who produces a conclusion at any cost. The first leaves a gap that can be filled. The second leaves a conclusion that has to be dismantled.

On the V.League, I do not have enough data to say whether VAR here is working well or not. And I will not say it while the dossier is still empty. That is not evasion. That is professional ritual: before I blow the whistle, I review myself.

The open question is not for referees but for those running the league: if a dispute arises tomorrow, will there be a dossier to open and read — or will we start from zero, once again, in front of the same silent stand, waiting?