The Data Gap in V-League VAR: When Referees Are Judged by Noise

Part 1 — One Moment, and One Void

On the stand, a man shouts something the rain swallows. On the pitch, the referee puts his hands on his hips and waits for a signal in his earpiece. The VAR team, in a control room hundreds of kilometres away, is rewinding a four-second clip at three different speeds. Fourteen cameras around the ground, but only two of them have an angle through the outstretched leg.

After the match, a supporter opens his phone and looks for the incident. He finds seventeen videos on social media, all shot vertically from the stand, all blurry, all with someone swearing in the background. He does not find an official record. No decision log. No VAR report. No name of the video referee. No start and end time of the review. No reason.

That is where every argument in the V-League begins, and where every argument ends. We argue about something for which we hold no data.

The moment the naked eye misses, the data never forgets. But here, the data was forgotten before the moment even happened.

I am not writing this to say Vietnamese referees are wrong. I am writing because I have no way to prove they are right or wrong — and that is a larger problem than any single decision on the pitch.

Part 2 — Context: What V-League VAR Has, and What It Doesn't

VAR first appeared in the V-League in the closing phase of the 2026 season, after years of preparation with technical support from the Asian Football Confederation. Running a single VAR match in Vietnam is estimated in the tens of millions of dong per game, covering equipment rental, transmission lines, operating staff and video referee training. That is not a small figure for a league where many clubs still operate on annual budgets of a few tens of billions of dong.

So VAR in the V-League is not deployed uniformly. It is deployed selectively, round by round, on fixtures deemed high-risk. There is a practical reason: not enough equipment, not enough trained people, not enough money. But it produces a technical consequence few discuss: one league, one rulebook, but two coexisting standards of adjudication. A match with VAR and a match without VAR do not differ merely in whether someone reviews the incident. They differ in whether the referee knows he will be supported, and that awareness changes how he referees.

Here I need to state a professional principle: I do not believe in luck. I believe in the number that repeats a hundred times. One missed penalty in one VAR match proves nothing. A hundred VAR matches, fully recorded, prove something. The problem in the V-League is the second half of that sentence: we have the first match, but we do not have the hundredth.

Imagine a league body trying to answer the simplest question available: how many VAR interventions occur per match on average, and what proportion overturns the on-field decision? In Europe's top leagues, that question has a public answer, updated every round. In the V-League, it has no public answer. Not because anyone is hiding it, but because nobody has built the table to fill in.

In nine years of watching Vietnamese football, I have learned something about missing data: it is rarely a conspiracy. It is usually a process that was never written down. But the consequences are identical — when there is no record, people fill the void with emotion.

And emotion, in a closed stadium, always wins.

Part 3 — Core: Cameras Do Not Lie, But Cameras Are Not Enough

3.1. Fourteen cameras and the physical limits of truth

A VAR-equipped stadium typically has twelve to sixteen cameras: some fixed around the pitch, some mobile along the touchline, and at least one high-speed camera for offside. That sounds like a lot. Set it against another reality: a challenge inside the box can involve six human bodies obscuring the sightline within a one-metre radius.

Every slow-motion replay has its own truth. My job is to find the one that cannot be argued with. And in most incidents, that truth does not exist.

This is what viewers are rarely told: VAR is not an all-seeing camera. It is a finite set of viewpoints, recorded at a finite frame rate, interpreted by a finite human being. When an incident happens at the edge of the box, eighteen metres from the nearest camera, with a player standing between the lens and the point of contact, the question of whether a foul occurred stops being technical. It becomes a question about how much certainty the system permits.

The IFAB Laws are explicit: VAR may intervene only for a "clear and obvious error" or a "serious missed incident". The phrase "clear and obvious" is a threshold, not a description. It says: if you have to watch four times and still hesitate, the on-field decision stands. That is a wise piece of design. But it is only wise if someone records that you watched four times.

In the V-League, we have the rule. We do not have a record of how often it has been applied.

3.2. Offside: where technique meets physics

Offside is the category where VAR is most technically effective, because it is a relative measurement of two points on a plane at one instant. Yet even here there are three sources of error the viewer never sees.

The first is frame rate. At 50 frames per second, consecutive frames are 20 milliseconds apart. A player sprinting at 8 metres per second travels 16 centimetres in that interval. If the offside line is drawn on the nearest frame rather than the exact frame of the pass, the error can exceed the entire width of a foot.

The second is camera reference. Cameras are never perfectly perpendicular to the pitch, and every projection from image to plane requires calibration. Where semi-automated systems exist, calibration is done with synchronised cameras and a geometric model of the pitch. Where they do not, drawing the line remains a manual operation by the video referee.

The third, and most important, is the definition of the moment of the pass. In an incident where the passer makes contact twice in one motion, or the ball is deflected by an opponent, identifying the reference frame becomes a judgement. And a judgement, however well made, is still a judgement.

I want to tell a personal story here. In 2026, aged sixteen, I was a volunteer statistician at a national U19 match. I was assigned to log 47 fouls and 12 offside calls. I found that the referee had missed a foul inside the box in the 78th minute, leading to a contentious goal. I built a comparison table against the IFAB Laws and sent it to the organisers. There was no reply.

I learned from the U19 league that a mistake is less frightening than nobody measuring it. One referee missing a penalty is the story of one referee. A system in which nobody records that miss is the story of an entire football culture.

3.3. Time: the most neglected metric

There is one VAR metric almost nobody in Vietnam tracks, despite being the easiest to measure: time.

A single VAR review costs between one and three minutes on average, including the referee walking to the monitor, reviewing, deciding and explaining. In complex incidents it can exceed four minutes. Those minutes do not disappear. They are added to stoppage time, and they change the nature of the match: a team that has just scored is forced to wait in limbo; a team that has just conceded gains four extra minutes to reorganise.

In the 2026 study I joined on the effect of empty stadiums on match outcomes, I collected data from 18 rounds and found home win rates fell from 42% to 31%. My supervisor advised me to compare against five previous seasons. That showed me the change was not large enough to support a conclusion: the sample was too small and squad effects too dominant.

The empty stadium taught me that noise never scores a goal. But it taught me the reverse as well: silence does not automatically produce truth. If you only have one season, you do not have a trend. You have one season.

I wrote myself a rule from that: never use one season to assert a trend. In today's V-League, we do not even have enough data to talk about one season.

3.4. The table nobody has built

Let us list the minimum a professional league must publish so that stakeholders can argue on the same data footing. The list is not long.

One: number of VAR matches per season, by round and by fixture.

Two: number of VAR reviews, broken down by incident type — goal, penalty, red card, mistaken identity.

Three: number of interventions leading to an overturned decision, and number of reviews that upheld the decision.

Four: average duration of a review.

Five: stoppage minutes added because of VAR.

Six: list of appointed video referees, with years of experience and number of VAR matches worked.

Seven: audio recordings of the exchange between referee and VAR team, published after the match or at season's end.

Seven items. None requires new technology. None demands meaningful budget. All of it is information the organisers already hold in the course of operating — it simply has not left the control room.

And here is the point I want to press: the referee's decision is only the endpoint. The real journey lies in every camera angle. If you publish only the endpoint, you have published nothing. You have issued a conclusion without evidence and asked others to believe it.

In every other field, we do not accept that. Nobody accepts an audit report that says "checked, correct". Nobody accepts a laboratory publishing results without publishing methods. Football has no reason to be exempt.

Part 4 — Core (continued): What the Data Has Said, Elsewhere

4.1. The 2026 World Cup and the principle of context

In 2026, aged seventeen, I built a spreadsheet tracking all 64 World Cup matches in Russia. I logged successful passes, possession share and sprint counts for every team. The result surprised me: the champions ranked only seventh in possession.

When I published the analysis on my blog, the first reply was: "He doesn't understand football." I spent several nights rewatching all ten of that team's matches, checking every number against every passage of play. What I found was not a defensive team. I found a team willing to concede the ball in harmless areas and win it back in areas where goals are scored.

Football does not change because you look at it more closely. Football changes because you look at it correctly.

That lesson applies directly to how we judge referees. A referee with a high VAR intervention count is not a bad referee. A referee with a low count is not a good one. The raw number means nothing. The contextualised number means everything: match type, tension level, fixture, whether the teams are collision-heavy.

This is why I strongly oppose judging referees on a single moment. A slow-motion clip circulating online is not data. It is a sample of size one. And a sample of size one, in any discipline, permits no conclusion at all.

4.2. Why inverted wingers make refereeing harder

There is a tactical trend in Vietnamese football over recent seasons that I believe connects directly to the volume of contentious decisions in the box: the prevalence of the inverted winger.

When a winger starts wide and drives diagonally inside, the opposing full-back must choose between following him into a crowded area or handing him off. Either way, the probability of a physical contact at the edge of the box rises substantially. Those contacts occur precisely where cameras see worst and referees see worst: the boundary of the penalty area, with bodies blocking and movement running diagonally.

I do not have enough V-League data to prove this link. I have only repeated observation: teams using many diagonal inside runs generate more incidents requiring referee judgement in the 18-yard box. And when the number of judgement calls rises, the number of contentious decisions rises — not because referees got worse, but because the judgement workload grew.

This is the kind of argument I want to make publicly, with a warning attached: it is a hypothesis, not a conclusion. Turning it into a conclusion requires positional incident data across at least three seasons. That data does not exist publicly. Which is the second problem.

4.3. The non-numerical grey zone

A referee does not only process laws. He processes human beings at their emotional peak.

Some things cannot be measured by any index, and I must discuss them, because otherwise data analysis becomes a sophisticated fraud. The sense of injustice a player carries after three uncalled fouls. The temperature change in a derby after the opening goal. A referee's ability to step between two arguing groups and say exactly one sentence that silences both.

No metric captures those. But they exist, and they decide matches more often than an offside line drawn two centimetres off.

What I want to say here is not "trust emotion". It is this: when data is missing, that grey zone gets filled by the loudest voice, not the most informed one. In today's V-League, the grey zone is being filled by noise, because we have nothing else to fill it with.

Part 5 — Contrarian: Emotion Is Not the Enemy. Silence Is

Here I have to argue against the image people usually attach to me.

People assume a VAR analyst, a rationalist, will say emotion is football's enemy. That shouting fans are the problem. That player protests are the problem. That if everyone stayed quiet and accepted decisions, football would be better.

I do not believe that. I think that belief is dangerous.

Fan emotion is a signal. It is not evidence, but it is a signal. When forty thousand people react identically at the same instant, that is raw data worth recording — not to conclude the referee was wrong, but to establish that an event requires explanation.

The V-League's problem is not that fans react too much. It is that after they react, nothing happens. No explanation. No record. Nobody says: we reviewed it, this is what we saw, this is why we upheld the decision. The silence after the roar is what destroys trust, not the roar.

I once thought the opposite. At nineteen, researching empty stadiums, I believed football would be fairer without crowds. The data showed home win rates falling, and I nearly concluded that crowds were a confounding variable to eliminate. My supervisor pointed out I was confusing eliminating a variable with understanding one.

Football does not exist to be measured. Football exists to be played, and to be watched. Fans are not noise. They are part of the system. And a system that cannot explain itself to its participants is a system with a design fault.

There is a line I keep in my head in front of the monitor: a referee who does not whistle is useless. Likewise, a VAR system that does not explain is useless. Making the decision is only half the job. The other half is making it verifiable by someone who was not in the control room.

And here is the final paradox I want to place on the table: in leagues that publish VAR audio, the number of controversies does not fall. In the short term it rises, because people gain material to argue about. But the quality of the argument changes. People argue about how the referee interpreted the law, rather than whether he was bribed. The first argument can lead to reform. The second only leads to bitterness.

Part 6 — Takeaway: Three Things to Publish, and One to Change

I have no authority to propose policy. I have a spreadsheet, nine years of notes, and a belief that what can be measured can be improved.

First, publish a VAR statistics table every round. The seven items above need no new technology and no large budget. They need one person to sit down after each round and fill in a form. After three seasons, we would have what we currently lack: a trend line.

Second, publish the audio of the exchange between referee and VAR team, with a reasonable delay. I understand the concern about pressure on referees. But the current pressure is far greater, because it has no shape and no limit. A published recording is a pressure that can be measured, answered and improved. A wave of outrage on social media is a pressure that cannot be answered.

Third, publish the list of video referees with matches worked. This is the most basic element of accountability in any profession. Nobody demands a video referee be perfect. But the public has a right to know who is making the decision, and how long they have been doing it.

And the thing that must change, the only one that does not fit on a form: how we talk about error.

Today, a refereeing error in the V-League is handled as a scandal. It requires a person responsible, a punishment, an apology. That handling sounds strict, but it actively prevents learning. Because if every error triggers a sanction, nobody wants to record errors. And if nobody records them, nobody learns from them.

At the U19 tournament in 2026, I sent my comparison table to the organisers and received nothing back. I was furious. Years later, I understood that the organisers had no process for receiving a data table from a sixteen-year-old volunteer. They did not reject me. They simply had nowhere to put my document.

That is what Vietnamese football is missing, and it is not VAR. VAR is already here. It is a place to put the data.

Next season, there will be another incident in the 78th minute. There will be another referee standing fourteen metres from the point of contact. Forty thousand people will stand up again. And the question I want to ask is not whether the decision was right.

The question is whether, after the final whistle, someone will open a file and record that the moment happened.

If the answer is no, we will argue in the dark again, with noise as our only evidence.

And noise, as I learned from an empty stadium, has never scored a goal.