V.League and the Signals That Never Reach the Data Export

On my screen in Singapore, the second leg of the 2026 ASEAN Championship final kicked off at 20:00 on 5 January 2026. I had four windows open at once: a live stream, a match-data feed updating in real time, a spreadsheet I had built from the group stage onward, and an empty notes file. In the first half, Nguyen Xuan Son scored. The data feed added a row. Around the middle of the match he went down after a challenge and did not get straight back up. The data feed added nothing after that. In the export file I received after the match, there was no column called "injury". There was no column called "what a team has left when it loses one man". I sat looking at that file for a long time, and remembered a line I wrote in 2026: there are numbers that never make it onto the stats sheet; they live between two touches of the ball.

Vietnamese football is in a period when data appears more densely than ever, but most of it is event data — the kind that records what has already happened, not what is happening. V.League 1 publishes a basic set of metrics per player per round: minutes, goals, assists, cards, shots, pass accuracy. That is enough to write a news item. It is not enough to understand a match. There is no xG model calibrated for the league, no frame-by-frame positional tracking, no post-shot pressure metric, no distance between the lines at the moment possession is lost.

VAR arrived in V.League 1 from the 2026/24 season and is now a familiar part of the round. It solves one specific class of error — clear offsides, handballs in the box, missed red cards — and it creates a new class of dispute. Not because the technology is poor, but because the technology forces people to state standards that used to live only inside a referee's head.

"Clear and obvious error" is a phrase. It is not a numeric threshold. Everything behind that phrase is judgement.

I work as a data consultant for football clubs and I report on football for the Singapore market. My readers do not watch V.League every week. They see results, see goals, see a few slow-motion replays. My job is to tell them the rest. And the rest, in Vietnam, is emptier than in any league I have worked in.

Based on my own experience of watching these matches, one thing repeats in almost every round: people argue about what was recorded and ignore what decided the result.

Let us start with the most misjudged position on the pitch.

Goalkeeper is where Vietnamese data is poorest, and where transfer decisions go wrong most often.

In Europe a goalkeeper is judged on two separate axes: stopping goals and participating in build-up. The second axis is measured by long-pass accuracy under pressure, touches outside the box, decision speed when pressed. The first is measured by post-shot xG — the probability of a goal given the shot's direction and destination, compared against what actually happened.

In V.League neither axis exists on paper. What exists is goals conceded and clean sheets. Both depend more on the back four than on the keeper, and neither says anything about the quality of the shots he faced.

Nguyen Filip took the Vietnam number-one shirt in 2026. Dang Van Lam held it for years before that, with a spell in Japan. The transfer conversation around both men is almost always told through feeling: who commands his box better, who organises the defence better, who carries "presence". Nobody tells it through save rate adjusted for shot quality.

The problem is not that the metric is hard to compute. The problem is that nobody records shots in a way that would allow it.

A V.League club may have three cameras covering a match, in three fixed positions. To compute post-shot xG you need the shooter's position in pitch coordinates, ball speed, the angle of the goal the ball is heading towards, and the keeper's position at the moment of contact. With three fixed cameras and no tracking data you can estimate two of those four. The other two you cannot.

I once heard a goalkeeper describe how she reads a shooter's belly, something that never appears in a data export. She said that before a penalty, what decides the outcome is not reaction speed but seeing the shooter's hip rotate before his foot meets the ball. A signal that appears roughly two-tenths of a second before the strike. No data file records it. It decides matches.

In V.League, dozens of goalkeepers do exactly that every week, and nothing on the stats sheet acknowledges them. People call them lucky when they save and call them old when they concede. Same action, same quality of reading the situation, different outcome by a tenth of a second.

And here is the market consequence: a goalkeeper with mid-level distribution skills who has been built up as "good with his feet" will typically carry a higher transfer value than a goalkeeper who has simply stopped shots well for several seasons in a row. That sounds counter-intuitive in a league where most teams do not build from the back. But it holds, because what is sold to the public is image, and image is easier to construct than data.

If a V.League club wanted to filter goalkeepers, it would not need expensive technology. It would need one person to rewatch every shot a given keeper faced across a season and record two things: shot location and ball destination. After roughly three hundred shots, that club would have a better profile than any scouting report currently on the market.

Nobody does this work. It costs time rather than money, and nobody pays for a correct decision.

A season is not the sum of 38 matches; it is the repetition of 17 forgotten passes.

That brings me to the biggest story in Vietnamese football over the past eighteen months.

Nguyen Xuan Son is the case where Vietnamese data both proved its power and exposed all of its holes.

Rafaelson arrived in V.League, naturalised, and became Nguyen Xuan Son. In the 2026/24 season he scored 31 goals for Thep Xanh Nam Dinh, a figure never previously reached in V.League 1, and helped the club to the first title in its history. At the 2026 ASEAN Championship he scored 7 goals, finishing as the tournament's top scorer, including a goal in each leg of the final.

Everyone saw the 7. Fewer saw the structure behind it.

I took the event data for the whole tournament and separated Vietnam's attacking sequences into three groups: attacks in which Nguyen Xuan Son was directly involved, attacks without him that still targeted the box, and attacks without him anywhere near the ball. The conversion rate of the first group was several times that of the other two. That is unsurprising. What is surprising is the difference in where players received the ball.

With him on the pitch, Vietnam's midfielders received possession higher up the pitch on average, and the number of passes into the box rose markedly. With him off it, the midfield dropped, and that number fell. In other words, he did not only score. He changed where his team-mates were willing to stand.

That is a type of influence for which no stats sheet has a column.

Then came 5 January 2026. He suffered a serious injury, required surgery, and was lost for a long period. In that match's export file, he left the pitch and his minute count stopped. No column recorded that the national team had lost an attacking structure, not merely a player.

I tried to do the work somebody should have done long ago: rebuild Vietnam's recent matches, split by time with and without Nguyen Xuan Son on the pitch, and count the team's touches in the box. The gap was very large. But when I tried to check it against a second dataset, I had nothing to check it against. No positional dataset has been published for those matches. I had only my own notes, and one person's notes are not evidence.

That is the humility I am forced to keep. I believe my conclusion. I cannot prove it with public data, because public data does not exist.

A football nation that does not record position cannot answer the question of dependency. It can only return to that question every time a player goes down.

Now I want to talk about the hardest part, the blind spot I consider the biggest in Vietnamese football today.

In a corridor, if you only look towards the light, you will miss what is standing in the dark.

VAR is that light. And for four years people have stared at it so hard that they no longer see the rest of the corridor.

I am not saying VAR is unnecessary. I am saying it is being judged by a yardstick that cannot measure it.

When a decision is reviewed and found to be a "clear and obvious error", what is being examined is not the incident itself. What is being examined is the gap between what the on-field referee saw and what the VAR room saw. That is an important distinction, and it is misread almost every time.

An offside can be determined to within a few centimetres. To do that, a line is drawn, and every line has thickness. A player moving at speed can travel several centimetres between two frames. If the system captures at 50 frames per second, the interval between frames is two hundredths of a second. For a sprinting player, that can be fifteen centimetres.

So when the offside line is a few centimetres wide and the running stride introduces fifteen centimetres of error within a single frame, what exactly are we measuring?

We are measuring the accuracy of the system. And that accuracy is not in the hands of the person making the decision. The VAR official receives a still image with a line drawn on it. He does not see the frame before and the frame after. He does not know that between those two frames the player moved fifteen centimetres.

This means that in certain incidents, what goes on trial is not the player but the physical limits of the equipment. And the party that pays is the club.

In V.League, where the number of cameras per match still varies from stadium to stadium, the gap between "truth" and "data" is wider still. A match supported by VAR at a stadium with enough angles is a different match from one in the same round at a stadium with too few. Same league, same laws, two different levels of accuracy.

I have spoken with people who work in refereeing, and what they told me was not about technology. They told me about having to make a decision within about sixty seconds, on incomplete imagery, and then being judged all week by people who watched the incident ten times at home.

That is another injustice, and it has nothing to do with the clubs.

My view is that "clear and obvious error" is a regulation so vague that it allows two referees to reach two opposite conclusions from the same frame, with both able to justify themselves from the wording of the law itself. There is no wrongdoing here. There is only a space for judgement wider than people imagine.

And that space is not measured, not published, not debated with numbers. It is debated only with emotion.

There is one more spot in the dark corridor, and I think it matters most in the long run.

The satellite-club system is turning young talent in smaller leagues into invisible assets, and data is the perfect instrument for it.

The Hoang Anh Gia Lai academy is the clearest example of a golden generation — one cohort growing up together, promoted to the first team together, producing a group of national-team players. That is the old story, and it is a good one.

The model replacing it is less good.

A big club does not need to develop its own youth if it can sign an agreement with a small club, send young players there, and retain a buy-back option. The small club receives the player cheaply, plays him, develops him. If he turns out good, he returns. If he does not, the loss sits with the small club.

In that model the big club breaks no domestic-training rule. It simply optimises around it.

And the instrument of that optimisation is data. Whoever has better data on an eighteen-year-old in a lower division buys him more cheaply. In Vietnam almost nobody has that data. There is no tracking in the lower divisions. No scouting reports are published. There are eighteen-year-olds playing twenty matches in a lower division who leave not a single line of data on the internet.

That is where talent is most forgotten, and also where it is cheapest to find.

I have long given a third of my writing space to veteran players and to competitions few people watch. I do it for professional reasons, not moral ones. Those players are where the gaps in data are widest, and the widest gaps are always where the most unmined information sits.

Clubs dissolve, football stops. But the data never stops telling the story.

Now comes the part where I have to be most careful.

Everything I have written above shares one weakness. I am telling a story about overlooked signals, and that story slides very easily into a claim that everything hidden matters.

That is not true.

Plenty of things go unrecorded and matter not at all. A defender who runs many metres is not automatically a good defender. A midfielder who completes many passes does not automatically set the tempo. A player arriving in the right position may be reading the game, or may simply be standing still while the ball happens to roll towards him.

When I say positional data is missing from V.League, I am not saying every transfer decision in V.League is wrong. I am saying those decisions are being made with less information than they need, and that as a result their outcomes carry more variance. High variance does not mean wrong outcomes. It means outcomes that cannot be repeated.

This is the point I want to underline, because it is the boundary between analysis and belief. When I write that Nguyen Xuan Son changed where his team-mates were willing to stand, I am describing a correlation. I have no evidence that he himself is the cause, in the sense an experiment could demonstrate. Another variable may have moved alongside it — how opponents defended, how the coach chose his personnel, the physical condition of the whole squad in each phase.

Correlation is not causation. And a football nation short of data is one in which correlation is almost always read as causation, because there are not enough samples to separate the two.

I stand by my conclusions on Nguyen Xuan Son, on goalkeepers and on VAR. But I hold them with a larger band of uncertainty than a confident piece of analysis usually permits. That is the price of analysing a league whose data is not yet thick enough to contradict you.

If someone handed me a complete positional dataset for V.League and it showed I was wrong, I would rewrite this article. I mean that seriously.

So where should we look in the next round?

I will track a single metric, and I think it is worth more attention than any league table: the number of passes into the box by each team in the first fifteen minutes of the second half. This is the phase in which fitness starts to separate from tactics, and the phase in which Vietnamese teams change structure most. If a team's figure falls steadily across rounds, that is a fitness signal. If it rises, that is an adjustment signal.

And if it does not change across a whole season, whatever the results, that is a signal that the team is playing one way and one way only — and waiting to be figured out.

A small metric. Nobody publishes it. Nobody sells it.

I will count it myself.