VAR in the V-League: The Offside Line Depends on Who Calibrates It

Thirty-seven minutes before kick-off of the 2026 World Cup final, I sent a two-page calibration report from FIFA's technical control room. The average error between the camera signal and the actual playing surface at Luzhniki was 0.43 metres. The figure was measured three times over twenty minutes and returned the same result every time. I sat still for a long while before pressing send, because I understood what it meant: the organisers would have to re-check the entire system before the match could begin.

They checked in time. But what I carried away from that night was not a sense of a job completed. It was a very simple question: if nobody had found that 0.43 metres, would the offside line drawn in the final have been right or wrong?

That question stayed with me for seven years, and it became far more concrete once VAR arrived in the V-League.

The operating frame: one process, four human layers

VAR is not a machine. It is a multi-step process, and at every step a human being makes a choice. The foundational principle IFAB set out is clear: the on-field decision stands unless there is a clear and obvious error. That is a psychological threshold, not a measurable one. No ruler measures the word obvious.

In the V-League, operating conditions differ considerably from Europe's top leagues. FIFA sets a minimum number of cameras for a VAR room to qualify, and that minimum sits far below the twenty-plus cameras per match used in the Premier League. Fewer cameras means fewer angles, and in challenges where the referee is unsighted, the number of angles is the entire opportunity VAR has to do its job.

Then there is infrastructure. Not every stadium in Vietnam offers camera positions that meet the technical standard, with heights and distances calculated for three-dimensional reconstruction. Temperature and humidity are real variables too: lenses expand, the pitch surface changes its flatness, and light shifts between the first and second half. That is why international protocol mandates recalibration at half-time, not only before kick-off.

Finally, there are the people. A VAR match mobilises a video assistant referee, an assistant VAR and a replay operator. These people sit in a closed room, with no crowd, and they make decisions based on images reconstructed by a computer. But they are the ones who choose which images get reconstructed.

One detail is rarely mentioned: in the V-League, the type of forward who plays on the shoulder of the last defender — players such as Nguyen Tien Linh or Pham Tuan Hai — is the group most directly affected by how the frame is chosen. They live on gaps of a few tens of centimetres, exactly the gaps a single frame can create or erase.

Four decision layers, and where the error can hide

The first layer is calibration. This is the layer people mention least and the one that is most dangerous. A camera does not know where the touchline is; it only records a flat image, and software then maps that image onto a three-dimensional model of the pitch. The model is built from fixed reference points — lines, corner flags, measuring marks. If one reference point is placed wrongly, or shifts during the match, the whole model shifts with it.

When I found the average error of 0.43 metres at Luzhniki, I was talking about a pitch prepared to the highest standard on the planet. That figure is roughly the width of an outstretched arm. For an offside decided by a boot tip against a shoulder, 0.43 metres is the distance between a legitimate goal and a raised flag. And that was a compliant stadium. In non-compliant stadiums, the error is not smaller.

The second layer is frame selection. This is the layer almost every social media argument skips entirely. To determine offside, the operator must choose the moment the ball leaves the passer's foot. At 50 frames per second, each frame is 20 milliseconds apart. A player sprinting at 8 metres per second moves 16 centimetres within a single frame. Pick the frame before or the frame after, and the discrepancy reaches 16 centimetres.

Sixteen centimetres. In a league where the average offside margin is a few tens of centimetres, that is not a small technical error. It is a decision.

And who picks the frame? A person. The software suggests the frame closest to the moment the ball is played, but that moment is not always clear: the ball may be touched twice lightly, may be obscured, may sit in the blurred zone between two frames. The operator must decide. In the match report, the chosen frame never appears.

The third layer is drawing the line. This is the layer that caused such fierce argument in the Premier League that the organisers introduced a 5-centimetre tolerance for the line before the 2026/22 season. The reason is simple: by thickening the line by 5 centimetres, they conceded that the system is not accurate enough to adjudicate below that level. That was a rare honest admission, and it should have been copied elsewhere.

In the same layer sits the choice of which part of the defending player's body can play the ball. The law states that the relevant part is the part that can legally touch the ball, meaning the arm below the armpit is excluded. It sounds clear, but on a reconstructed image, determining exactly where the armpit is becomes a manual operation. Every person drawing places the point differently, and nobody checks that point again.

The fourth layer is the on-field referee. Viewers often assume VAR is the final decision-maker. It is not. The video assistant referee only recommends; the on-field referee makes the final call, and has the right to stand by the original decision. That means the final variable is still a human being who may be in the 88th minute of a third match in a week, who has run more than ten kilometres, and who must now decide based on a small monitor by the touchline.

Four layers, and each one has a place to go wrong. If all four are correct, the outcome is correct. If one is wrong, the outcome can still be correct because the other layers compensate. That is the process's fault-tolerant design. But it is also precisely why the process cannot be verified from the outside unless somebody publishes the data from each layer.

Semi-automated offside technology, which appeared at the 2026 World Cup and entered European leagues from the 2026/25 season, deserves to be viewed through this lens rather than through accuracy. Its real value lies in removing two manual layers: frame selection and line drawing. When those two become deterministic outputs of the system, responsibility does not disappear — it simply shifts towards the people who install and calibrate it. That is why the V-League's current problem is not whether to buy new technology.

The foul threshold is not in the law book, it is in the ear

During my time working with refereeing data, one milestone stands out clearly. In 2026, when competitions resumed with empty stands, I analysed 212 matches before and after the outbreak. The home win rate fell from 41.3% to 35.2%. Yellow cards dropped 17%, from 3.8 to 3.15 per match.

The media at the time wrote extensively about the death of home advantage. That reading was attractive but wrong about the mechanism. What changed was not the quality of the home or away team. What changed was that referees lost the crowd-noise signal they used to reference their foul threshold. In a packed stadium, the roar of the stands is input data for a referee's sense of how hard a challenge really was. Without that noise, the threshold shifts.

An empty stadium does not create ghost football, it creates storytellers. And the most-told story back then ignored the fact that referees' decisions depend on their surroundings far more than people assume.

Three years earlier, I had submitted a report to my superiors with statistics on 47 penalties across 15 rounds. Referee Ma Ning showed a tendency to decide in favour of the home team in 68% of 50/50 challenges. The report was rejected on the grounds that refereeing intuition mattered more than statistics. In August 2026, when the league changed how it applied the handball law based on similar data, my report was retrieved and filed as an internal document.

I tell these two stories together because they point at the same thing. In the fourth layer of the VAR process, what we call a clear and obvious error is not a constant. It is the product of experience, pressure, environment and timing. By adding VAR, we remove one source of bias — crowd noise — but we simultaneously introduce a new one: the judgement of the person in the closed room about what counts as obvious.

The contrarian angle: the problem is not that technology is wrong, it is that technology will not sign its name

Public debate in Vietnam, as anywhere, follows two familiar directions. One says the referees are poor and need more technology. The other says technology has stripped football of its soul and VAR should go.

Both avoid the crux. Nobody is arguing about data, because the data is not published. The calibration log of each match — what the error was, when it was measured, when it was corrected — appears in no document a spectator can reach. The frame chosen to reconstruct the offside line is not recorded. The tolerance applied to the line is not stated.

As a result, public debate can never end with evidence. It can only end with emotion, and emotion depends on whether your team benefited or suffered. That is why the same move, the same line, is celebrated in one stand and called fraud in another. People are not checking the line; they are checking whose interests the line serves.

This is where I have to say plainly something the profession prefers to avoid. VAR does not make referees better. It makes errors clearer or blurrier, depending on who operates it. A system marketed to the public as a tool of transparency instead operates as a black box whose output is a decision and whose input is no data at all. The line never lies, but the person drawing it can.

I do not watch matches; I read the rhythm of a match frame by frame. And reading it that way, what I see in most VAR controversies in the V-League is not referees deliberately cheating. It is a four-layer process in which no layer is publicly accountable, so when something goes wrong, nobody is obliged to explain.

A conditional conclusion

If the V-League wants out of its current argument loop, the fix is not new technology. It is opening the calibration log of each match, publishing the frame chosen for each offside call, and stating the tolerance applied to the line. Those three things require no extra cameras, no extra money, and no wait for a major tournament.

If one day the match report carries the name of the person who chose the frame and the name of the person who calibrated the system, then the public will have something to verify instead of only something to argue about.