VAR in the V.League: The Technology Arrived, the Measurement System Did Not

VAR in the V.League: The Technology Arrived, the Measurement System Did Not

Autumn 2026, Stand B at Thong Nhat, and a sheet of paper that never got a reply

I was sitting in the twelfth row of Stand B, holding a printed sheet with two tables on it. The match was U19 Hanoi against U19 SHB Da Nang in the final round of the national U19 championship, played in Ho Chi Minh City. I was sixteen, working as a volunteer statistician, and my job came down to two columns: count fouls, count offsides. After ninety minutes I had 47 fouls and 12 offsides.

What I still remember today was not in either column.

In the 78th minute, a U19 Hanoi player came in from behind inside the penalty area. I marked the foul column, circled it, and wrote "inside the box" next to it. The referee waved play on. Thirty seconds later the ball was in the net. The goal stood, and the match turned in a direction my sheet had not predicted.

That night I opened the IFAB Laws of the Game, worked line by line, built a comparison table between what had happened and what the law said, and sent it to the organisers. No reply came. I do not think the organisers were acting in bad faith. It is simply that in 2026, at a youth tournament, nobody had been assigned to read that table. What I sent had no place in the process.

What I learned from that U19 tournament is this: an error is less frightening than an error nobody measures.

Nine years later I still do the same job: re-read the incident, check it against the law, and look for where the error is. Vietnam has VAR now. The technology arrived. What did not arrive is the measurement system around it.

Four reviewable categories, one interference threshold, and a boundary everyone misreads

The IFAB VAR protocol permits intervention in four categories only: goal or no goal, penalty or no penalty, direct red cards, and mistaken identity. Four. No more. Every argument outside those four is outside VAR's scope, even when the stadium is screaming that the referee got it wrong.

The interference threshold is narrower still. VAR may recommend an on-field review only for a "clear and obvious error" or a "serious missed incident". That threshold is deliberately high. The whole system is built on a minimum-interference philosophy. VAR exists to correct obvious mistakes, not to replay the match from a control room.

There are two different procedures, and the crowd almost always merges them. The first is an on-field review, where the referee must reach the final decision after watching the footage. The second is a VAR-only review, where the referee simply receives a signal through the earpiece. Offside usually falls into the second category, because offside is a factual question: a player is either in that position or he is not.

But "factual" needs a technical definition. Where is the measurement point? Knee, shoulder, elbow, toe? IFAB had to add guidance that the hand and arm do not count towards an offside position, and that the shoulder boundary runs down to the bottom of the armpit. Every time semi-automated offside technology is deployed at a major tournament, the argument shifts from "the referee was wrong" to "how is the system measuring this?". The argument does not disappear. It relocates.

Every slow-motion replay has its own truth. My job is to find the truth that cannot be argued with.

For Vietnamese viewers, the hardest part to accept is the delay. An incident that looks obvious from the stand still takes tens of seconds to process. That delay is not laziness. It is a sign that the system is trying to be more precise than it can afford to be, because behind it there is no database thick enough to shorten the verification step.

How VAR reached Vietnam, and the cost of a match with VAR

VAR entered V.League 1 from the 2026-2026 season, under coordination between the Vietnam Football Federation and the Vietnam Professional Football Joint Stock Company, with technical support and training from FIFA. The rollout was designed in stages: some matches at some stadiums, expanding round by round, season by season.

That approach makes technical sense. It also creates a problem few people discuss: when a league runs VAR unevenly across matches, the data collected cannot be compared directly. A team punished in a VAR match and favoured in a non-VAR match experiences two different standards of fairness inside one season. That is a consistency problem in application, and it is worse than having no VAR at all.

Behind every VAR match sits an operational chain. You need a full VAR team: the video referee, two assistant video referees, a replay technician, a transmission operator, a communications operator. You need enough camera angles and enough frames per second to catch the instant the ball leaves the foot. You need stable transmission between the camera positions and the control room. You need a room dark enough, with enough monitors and enough distance that the person inside is not swayed by the noise of the crowd.

The list of VAR-licensed referees is a short list. A referee only sits in the VAR chair after completing training and being assessed as competent. That resource cannot be expanded by administrative decision. It expands with time, and time is the one thing a running season does not have.

Cost is another variable rarely mentioned. A VAR match is more expensive than a non-VAR match across several line items: equipment, personnel, transport, and the opportunity cost of concentrating the best referees on a handful of showcase fixtures. In a league where broadcast rights income remains thin and clubs lean heavily on sponsorship and owner funding, every decision to expand VAR is a budget allocation decision, not a purely technical one.

From my own tracking of matches, the gap is very specific: leagues publish results, but they do not publish the log of interventions. An analyst can look up who scored, who was booked, who is suspended. There is nowhere to look up how many times VAR was called last season, in which category, whether the decision was overturned, and how many seconds each review took.

The data I counted myself, and a lesson about sample size

In 2026, as a nineteen-year-old second-year student, I joined a research project on the effect of playing without crowds on results in the V.League. I collected data from 18 rounds and produced a number that excited me: the home win rate fell from 42% to 31%. Eleven percentage points. A clean, tidy figure, very easy to turn into a headline.

I took the report in. My supervisor read it and asked one question: had I compared it against the previous five seasons?

I had not. When I did, the picture changed completely. Seasonal variation within a single league can be large enough that an eleven-point gap sits comfortably inside the normal range. On top of that, several clubs changed their squads mid-season, and squad factors explained more than crowd factors did. The scientifically honest conclusion had to be: not enough evidence to claim anything.

I do not believe in luck. I believe in repeating something often enough that a trend separates itself from the noise.

Since then I have kept one rule: never use a single season to assert a trend. That rule applies intact to the VAR story in the V.League. Three seasons with VAR is a small sample. One VAR match that malfunctioned is an observation, not a model. And one slow-motion clip going viral on social media is a media event, not statistical evidence.

But there is an important distinction. A small sample blocks conclusions about results trends. It does not block conclusions about process. How often VAR is called in a season, its distribution by category, the overturn rate, average handling time, the interval between errors by the same referee: those are process indicators, and process can be observed from the very first match. The problem is that they are not being recorded systematically.

Three indicators worth measuring more than accuracy rate

The first is the distribution of VAR triggers. If most interventions come from a single category, that tells us about the quality of training in that category, not about the technology. If most interventions are offside, for example, the problem lies in how assistant referees and referees coordinate their positions, not in their eyesight.

The second is the overturn rate. Too high, and the system is intervening too often. Too low, close to zero, and there are two possibilities: either the on-field referees are performing very well, or the interference threshold is being set higher than the protocol allows. Telling those two apart requires a secondary indicator: how often VAR reviewed an incident and decided not to intervene.

The third is latency. The average time from incident to final decision is the most direct measure of how prepared the infrastructure is. Short latency is not only better for the viewing experience. It is better for accuracy, because the longer a decision drags, the more the noise of the crowd seeps into the control room.

These three indicators share one property: they can only be measured if somebody is responsible for recording them. And where there is no record, argument always wins, because argument does not need data to survive.

Tactics changed before the laws did

The most interesting thing about VAR in the V.League is not in the control room. It is on the pitch, in the adjustments coaches made before any law was amended.

When VAR arrived, the cost of an offside trap went up. Previously, a high defensive line gambled that the assistant referee might miss a legitimate goal. Now that chance has fallen sharply, so the reward for stepping up remains but the risk has grown. The natural response from coaches is to shift the timing of the step: one beat later, waiting for the pass rather than for the receiver.

That shift sounds small, but it changes the whole spacing between the lines. When the defence waits an extra beat, the midfield must drop to avoid leaving space behind it. When the midfield drops, the forwards drop with it, and the block becomes deeper. A change in the VAR interference threshold can push an entire league a few metres back.

Set pieces changed too. Once every goal from a corner can be reviewed for blocking, coaches start teaching players to block with the body rather than the hands, and to stand where the camera does not create an unhelpful angle. These are adjustments that appear in no tactics textbook, because they only exist when a slow-motion system sits behind the game.

On the other side, VAR created a new incentive to go down in the box. A player knows that if the incident is reviewed, the chance of being caught deceiving the referee is real. The law is explicit that using deception to mislead the referee is punishable. But there is always a gap between the rule and its enforcement, and that gap is measured by how many cards are actually issued for simulation in a season.

I have never seen a league publish that indicator properly. Not even where VAR infrastructure is at its best.

A counter-intuitive angle: the inverted winger and the abandoned ground

There is a trend larger than VAR and larger than the V.League flattening football, and it belongs here because it shapes how decisions get made.

The left-footed winger playing on the right and cutting inside has become the default template in almost every academy. The reasoning is sound on paper: cutting inside creates a shot with the stronger foot from a dangerous zone and opens the lane for an overlapping full-back. It is an efficient model, and it has spread to the point of becoming a selection standard.

The consequence is that the traditional winger, the one who hugs the touchline, beats his man and crosses, is being written off wrongly. Written off because he is being judged by the criteria of a different role. A touchline winger does not generate many shots, so he is labelled inefficient, when what he actually generates is space for others and pressure on the opposing full-back. Neither shows up in a basic stat sheet.

At youth level the problem is sharper. Youth coaches under pressure to deliver results favour physicality and running. An U18 who is fast and strong in duels gets picked ahead of one who handles the ball well but is half a step slower. That choice is rational at U18 level and harmful at first-team level, because by the first team speed is no longer an absolute advantage, while close-control technique still is.

I have seen this repeat often enough to call it a pattern: every generation of Vietnamese players leaves behind a technical group, not because they were not good enough, but because the evaluation system had no column for what they did well.

Money, contracts, and what VAR cannot see

In a transfer window, noise drowns out signal. Transfer rumours have far shorter life cycles than contracts, and the player becomes the centre of a story he does not control.

The contract structure is the real story. When a Vietnamese player moves abroad, the true value of the deal sits in the clauses behind it: the release clause, the sell-on percentage, and the training mechanism. FIFA operates a system that pays clubs which trained and developed a player, based on the years he was registered between the ages of 12 and 23. For an exporting football nation, that is real money flowing quietly for years.

Doan Van Hau's 2026 loan to SC Heerenveen, Nguyen Cong Phuong's move to Incheon United the same year, and Nguyen Quang Hai's June 2026 contract with Pau FC were three important markers, not only for the deals themselves. They set a benchmark for how a Vietnamese player is valued abroad: on development potential, not on domestic achievement. The gap between those two valuation methods is the problem every V.League club faces at the negotiating table.

In the other direction, money flowing into the league still comes mainly from owners and sponsors. That creates a structure in which a club's stability depends on the decisions of one person or a small group, rather than on operating revenue. In that structure, spending on refereeing and technology infrastructure is the easiest line to cut, because it does not score goals and does not sell tickets.

Noise and measurement

The most common argument against VAR is that it kills the emotion of football. I understand that argument. I have stood in a stand and felt the explosion when the ball hits the net.

But I should be explicit about what I believe, because I do not want to be read as someone defending technology at any cost.

Empty stadiums taught me this: noise has never scored a goal.

When the V.League played without crowds during that special period, I watched a lot of matches with one screen and one spreadsheet. No cheering, no drums, no crowd pressure. And football was still exactly what it always is: a sequence of human decisions within limited space and time. What disappeared when the stands emptied was not football. It was the emotional overlay we use to interpret football.

In the other direction, I do not believe adding cameras automatically makes football fairer. Fairness in football does not come from equipment. It comes from applying one set of standards consistently, standards published in advance. VAR is only a tool for checking that consistency. If the standard is not published, the tool has nothing to check against.

That is why I call the VAR problem in the V.League a measurement problem, not a technology problem. The technology is here. The cameras are here. The monitors are here. What is not here is a public record that lets anyone, including the most sceptical fan, verify independently whether the system is operating consistently.

There is another risk worth stating plainly. When a football culture has no habit of publishing data, public pressure fills the vacuum. A handful of fast-spreading slow-motion clips shape the perception of an entire season. Referees are judged by how often they appear in clips, not by how often they got the decision right. At that point we are building a reward system for avoiding controversy rather than for making correct calls.

I have watched that mechanism operate in many parts of the world. It is not unique to Vietnamese football. But where data is public, the mechanism is restrained. Where it is not, the mechanism grows freely.

For supporters, this has a very concrete consequence. You cannot evaluate a referee, because you have nothing to evaluate beyond what you remember. Memory is selective. It remembers errors and forgets correct calls. That is not the supporter's fault. It is the fault of a system that provides no substitute memory.

What to publish, what to measure, what to be patient with

I have no authority to propose anything to anyone. I have my spreadsheet and a habit of cross-checking.

But if I were handed a short list, I would start with the cheapest item with the largest effect: publish the VAR intervention log after every round. A simple table with four columns: category, initial decision, outcome after review, and handling time. That table requires no new infrastructure. It requires one person responsible for writing it down.

I would also be more patient about sample size. Three seasons is far too short to talk about trends in refereeing quality. But three seasons is more than enough to talk about process, if the process is recorded. Distinguishing those two things is a basic skill for anyone working with data, and it should be a basic skill for anyone commenting on football.

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

I still keep the comparison table I wrote that night in Ho Chi Minh City. It sits in a stack of paper, handwritten, never answered. It reminds me that football's problem has never been a shortage of people spotting errors. The problem is a shortage of places to file them.

VAR has given Vietnam slightly more than it had before: it created a process, and a process can be measured. What remains does not sit with the cameras. It sits with the decision of whether to record what the cameras see.