V-League 2026-2027 On-Pitch Violence: The Card Table and the Gap That Makes Players Believe They Will Get Away

At minute 63 on the Đồng Nai pitch, defender Lê Văn Sơn drove an elbow into striker Nguyễn Đình Bắc. Not long afterwards, centre-back Huỳnh Tấn Sinh — also of Đồng Nai — kicked Nguyễn Đình Bắc from behind twice. Two actions, two moments, one target. Referee Lê Vũ Linh stood within a radius wide enough to make a ruling. The match data recorded the final outcome: one yellow card for Tấn Sinh, and no red card shown.

Numbers carry no bias. Bias lives in those who lack numbers. I reopened the data from the Đồng Nai versus Công an Hà Nội (CAHN) match of 20 September not to find someone to blame, but to answer a narrower and harder question: if on-pitch violence is a variable, what is feeding that variable?

The answer does not lie in any individual's temperament. It lies in the gap between the level of risk the law threatens and the probability a player actually pays a price.

Context: two rounds, four incidents, one meeting

V-League 2026-2027 has only passed two rounds but has already recorded four serious on-pitch violence incidents. In round 2, centre-back Đoàn Văn Hậu (CAHN) broke the leg of midfielder Doãn Ngọc Tân (Thể Công Viettel). Also in round 2, foreign player Ezequiel Santos Da Silva (Nam Định) stamped on the abdomen of Võ Hữu Việt Hoàng (SHB Đà Nẵng). In the following round, it was the turn of Lê Văn Sơn and Huỳnh Tấn Sinh to target Nguyễn Đình Bắc.

Four incidents in two rounds. Expressed as a rate, that is a far more alarming figure than the early-season averages I have recorded in previous campaigns. I have spent nearly three decades watching football, including almost nine years hosting and producing nightly programmes, and I have learned one thing: what repeats within the first two rounds is usually not random. It is the signal of a system that permits it.

The organisers' response was swift. Representatives of the V-League Organising Committee, the Vietnam Professional Football Joint Stock Company (VPF) and the Vietnam Football Federation (VFF) met directly with the VFF Referees Board. At that meeting, the parties demanded that referees apply the law correctly, maintain the competition's discipline, and prevent violent acts on the pitch.

The phrase "apply the law correctly" sounds solid. But in the language of a data analyst, a demand only has value when it comes with a measurement mechanism and an enforcement mechanism. "Apply the law correctly" here can be understood as not allowing missed fouls or missed cards. The problem is this: missed cards already happened, and they happened in the very match that opened the sequence.

The Tấn Sinh incident: a double error, not a single one

In the incident where Huỳnh Tấn Sinh ignored the ball and went for the man, targeting Đình Bắc, referee Lê Vũ Linh used only a yellow card. He did not use a red card, even though Tấn Sinh kicked Đình Bắc off the ball twice.

This is where I want to pause, because the data here describes not one action but two. Two kicks at the same player, in the same off-the-ball situation, are two separate actions. In the assessment framework I use for transfer reports and player risk reports, this is a "double error" structure: the first action reveals intent, the second reveals the active decision to sustain that intent when the ball is gone.

My risk scorecard divides unsporting behaviour into four tiers. Tier one is an over-committed challenge on a live ball — high injury risk but explainable by timing. Tier two is a late challenge. Tier three is an action aimed at an opponent after the ball has moved on. Tier four is a repeated action aimed at the same target within a short window.

Tấn Sinh's incident falls into tier four, not tier three, because of repetition. And here is the legal crux: a tier-four action in an off-the-ball situation is no longer "a foul in playing the ball"; it is an action aimed at a person. A red card is the corresponding handling threshold for that tier, not an optional level of punishment.

Showing only a yellow card generated two kinds of data. The first is visible data, what spectators see in the disciplinary record. The second is hidden data, what every player in the league reads: a repeated, clearly targeted off-the-ball kick still sits within the yellow-card threshold. This signal spreads more widely than any press release.

The physical line and an excuse that no longer holds

In my tracking files I always place two columns side by side: the column of unsporting actions and the column of the physical state of the player who committed them. The reason is simple. Many violent challenges late in the second half are not pure malice but the consequence of a player crossing his own biological physical line: movement control degrades, reaction time lengthens, and the body chooses the shortest option instead of the correct one.

Croatia 2026 taught me that heroes also have biological limits. But the 2026 story also taught me the opposite: biological limits can explain a failed collision, they cannot explain an action aimed at a specific player twice in succession. Fatigue blurs the boundary; choice draws it.

That is why I challenge every emotional explanation. No index in my dataset supports the hypothesis that two kicks at the same player within minutes are the product of tiredness. Fatigue does not repeat twice at the same target. Fatigue has no address.

VAR: the question is not "whether" but "who carries responsibility"

Speaking to reporters, former VFF Vice President for professional affairs and former head of the VFF Referees Board Dương Vũ Lâm offered a very clear analytical frame: "Referee Lê Vũ Linh missed a card in that incident. Huỳnh Tấn Sinh ignored the ball and went for the man too blatantly, yet received no red card."

More importantly, Mr Dương Vũ Lâm set out a comparison principle: "If Tấn Sinh is to be retrospectively punished with a suspension for this error, then referee Lê Vũ Linh should be suspended in the same way. In addition, the responsibility of the VAR team must be clarified."

This is the structure I call, in process analysis, the "four-link chain of responsibility": player — main referee — VAR team — organisers. A system only works when all four links each carry their own probability of sanction, not when all risk is dumped onto the final link.

Under Mr Lâm's framing, if the VAR team (comprising Võ Hoài Thương and Nguyễn Thành Sơn) advised the main referee but was overruled, then the VAR team fulfilled its duty. Conversely, if the VAR team did not advise a red card, the VAR referees must also be sanctioned.

From a data standpoint, this is a textbook verification structure. It produces four mutually exclusive scenarios, each leading to a different disciplinary conclusion. A system that cannot draw these four scenarios is a system that cannot handle the truth, because the truth in VAR situations always lies in the question "who said what, and when".

A missed card is more dangerous than a wrong card, and here is why

There is an objection I have heard many times: that showing more cards only fragments the game, that referees should keep play flowing. I once spent three months testing that hypothesis on a Ligue 1 sample and reached the opposite conclusion.

When I analysed Marseille versus PSG in October 2026, PSG won 3-0 but the expected-goals data showed Marseille created the more dangerous chances (1.94 versus 1.21). I received hundreds of critical comments, most aimed at my gender rather than my method. I rebuilt a framework of 23 Ligue 1 matches and showed that PSG were winning heavily on abnormal conversion efficiency. Three months later PSG's indices fell and they lost 1-2 to Lyon.

The lesson I drew was not "I was right". The lesson was: a small sample can always be misread, and the only way not to be misread is to widen the sample and state the method clearly. I apply that principle to the card story.

In football there are two types of refereeing error: the wrong card (showing a card when undeserved) and the missing card (not showing a card when deserved). The two are not symmetrical in consequence. A wrong card disrupts one match. A missing card changes the behaviour of an entire league, because it changes the probability a player calculates before committing an action.

A risk model saves no one, but it gives them a chance. If a player believes a tier-four action carries an 80 percent probability of being sent off, he will not repeat it. If that probability drops to 30 percent, the calculation changes instantly: a yellow card in exchange for removing the opponent's key striker is a profitable trade.

That is why a missing card is more serious than a wrong card.

The enforcement gap: what the table does not show

There is one detail in Mr Dương Vũ Lâm's remarks that I consider the most important and the least noticed: "Normally, specific disciplinary rulings for referees are not widely published, but just watch the referee list in the following rounds and you will know whether the referees involved have been suspended. If they are not appointed in the rounds immediately after, the likelihood is high that they are being disciplined."

This is an indirect enforcement mechanism. It works to a certain extent, but it creates two data problems.

The first is measurability. If disciplinary rulings are not published, there is no public data to assess the consistency of the system. A referee not appointed in the next round may be suspended, may be rotating by schedule, or may be resting for personal reasons. When these three possibilities cannot be distinguished from outside, every analysis of the system's severity becomes speculation.

The second is deterrence. A disciplinary ruling that is not published still works on the person punished, but has almost no effect on those who are watching. In every behavioural model, deterrence operates through observation, not through direct experience. The number of players who have been suspended for on-pitch violence is far smaller than the number of players currently weighing whether to commit such an action. The second group is the decisive one.

I understand the reason for discretion. Publishing referee disciplinary rulings places personal pressure on people working under enormous stress. But between protecting an individual and letting a system lose its deterrent capacity, I choose clarity: a minimum disclosure mechanism, in aggregate form, without naming individuals, but publishing the number and type of sanctioned errors each round.

The counter-intuitive angle: harsher crackdowns do not make violence disappear

Here I must argue against myself. The natural reaction after four incidents in two rounds is to escalate: demand referees show more cards, punish harder, suspend longer. The data do not support that reaction in any simple way.

If I pursue the hypothesis that "showing more cards reduces violence", I must accept an unwanted consequence: showing more cards also increases the number of clashes caused by frustration. A player wrongly booked will raise his intensity in the next challenge. The game is fragmented further, ball-in-play minutes fall, players lose rhythm and are more prone to late challenges. This is a feedback loop I have recorded in several leagues.

The real variable is not the number of cards but the consistency of carding. A league that shows few cards but does so consistently is safer than a league that shows many cards erratically, because players can predict the handling threshold. On-pitch violence is largely born of unpredictability, not of leniency.

This leads me to re-read the whole story. Four incidents in two rounds do not necessarily reflect a rise in violence. They may reflect a rise in a category of behaviour the system is handling inconsistently. In other words: the problem may not be that players are more aggressive, but that the handling threshold has become blurrier.

And here I must acknowledge the limits of the model. I can measure cards, collisions, ball-in-play minutes, distance covered, pressing intensity. I cannot measure what is happening in a referee's mind at minute 63 when he must choose between a red card that changes the match and a yellow card that keeps it under control. Every model has a blind spot, and the blind spot of this model is decision psychology under pressure.

Risk as probability: a scorecard for the next round

People see a comeback; I see a chart breaking. In this story, the breaking chart sits at the intersection of fixture density and the severity of handling.

If I had to build a risk scorecard for the next round, I would put three variables on the table.

The first is the fixture calendar. Match density in the early phase of a season is high, recovery windows are short, and as noted, physical state directly affects the quality of a challenge. The highest-risk group is not the dirtiest players but the ones logging the most minutes on insufficient recovery.

The second is the outcome of the meeting between VFF, VPF, the organising committee and the Referees Board. If the meeting produces a clear directive that all tier-three and tier-four actions must be handled at the red-card threshold, the probability of violence falls in the next two rounds — but the probability of card controversy rises sharply. If the meeting ends only with a reminder, the probability of violence stays the same and the probability of a more serious incident rises.

The third is the referee appointment list. This is the variable the public can observe and, in Mr Dương Vũ Lâm's phrasing, an indirect indicator of disciplinary rulings. If referee Lê Vũ Linh and the VAR team from the Đồng Nai versus CAHN match do not appear in appointments for the adjacent rounds, that is data showing the system is self-correcting. If they continue to be appointed normally, that too is data — pointing the other way.

I draw no conclusion here, because there is not enough data. But I set a test condition: if after two more rounds the number of tier-three and tier-four incidents has not fallen, the hypothesis "players are more aggressive" must be rejected, and the hypothesis "the handling system is inconsistent" becomes the primary one.

What I actually believe

Data is the only thing I trust after witnessing too many broken promises. But data protects no player. It only shows where the leak is.

In this story of four incidents and two rounds, the leak is not in a referee, a VAR team or a player. It is in the gap between the severity the law prescribes and the actual probability that an action is sanctioned. The wider that gap, the more reason players have to believe they will get away.

What worries me is not the elbow at minute 63. What worries me is the meeting three days later, where the demand to "apply the law correctly" was issued without a measurement mechanism and without publication of results.

A demand that cannot be measured cannot be sustained. And a system that does not publish its own enforcement data is stripping itself of its deterrent capacity in front of observers.

Readers may see a forgotten match at Đồng Nai. I see a changing variable: the probability a player calculates before kicking an opponent off the ball for the second time in the same incident.

If that probability is not pushed clearly higher in the next two rounds, we will have more data. Data no one wants to read.