The Referee's Eye: The Case File of a Decision Nobody Wants to Review

Minute 81, the score 1-0 in favor of France in the 2026 World Cup semi-final in Saint Petersburg. Samuel Umtiti rose to contest a ball with Marouane Fellaini inside the penalty area. The ball passed through, and Uruguayan referee Andrés Cunha did not whistle. For the first four seconds, the stadium went silent. Four seconds later, the big screen replayed the moment from behind the goal, and the noise erupted as if the entire stadium had just discovered a crime. I was sitting in the rules monitoring room, stopwatch in hand, eyes fixed on fourteen camera angles. I was not watching to find emotion. I was watching to find a clause. That was the night I wrote a three-thousand-word legal analysis just to answer a single question: was this a situation that VAR needed to intervene in. That piece drew 1.2 million views and was referenced by the European referees' association. But what I kept was not the views. What I kept was a larger question that still hangs unresolved today: when technology can see everything, why do humans still see differently? My name is Đặng Trí, a football legal commentator. I do not sell emotion. I sell process. And in this article, I will dissect one of the thinnest lines in modern football: the line between what the law permits and what the human eye wants to see. CONTEXT: A SMALL CLAUSE, A LARGE CONSEQUENCE To judge anything, you must first open the law. Football has a foundational document called the Laws of the Game, issued by the international rules board and updated on a cyclical basis. In it, the clause on fouls and unsporting behavior states clearly: a player commits a foul when they contest the ball in a careless, reckless, or excessive-force manner. Those three levels are not three ways of saying the same thing. They are three rungs of punishment, corresponding to three levels of responsibility. The problem is this: the law gives referees a range, not a number. Careless is when a player lacks attention when contesting. Reckless is when a player acts without regard for the danger to the opponent. Excessive is when a player goes far beyond what is necessary and endangers the opponent's safety. These three concepts sound clear on paper, but within one hundredth of a second at eighteen kilometers per hour, they blur into each other. That is why technology to assist referees came into being. But technology does not replace the law. Technology only answers the question of what happened. It does not answer whether that constitutes a violation of the law. The second question still belongs to humans. And this is the blind spot most spectators never see. Over years of observation, I have noticed a sad rule: spectators believe technology will deliver absolute fairness. But technology only delivers transparency about events. Fairness must still be built from judgment. And judgment, no matter how many camera angles exist, remains a human act. To understand why, look at how VAR is designed. VAR is not a second referee. It is a checking layer. It is only permitted to intervene in four types of situations: goals, penalty kicks, direct red cards, and cases of mistaken player identity. It only intervenes when there is a clear and obvious error, or a serious missed incident. Those two phrases, clear and serious, are where every controversy begins. Because clear to whom? Serious to what degree? There is no absolute measure for these two words. They are thresholds, and thresholds are set by humans. An assistant referee in a lower division might call a moment clear. A FIFA referee in a World Cup final might call the same moment insufficient for intervention. Both are right in terms of process. And both are wrong in terms of the spectator's emotion. ANALYSIS: FOURTEEN ANGLES AND A QUESTION WITH NO RIGHT ANSWER Back to minute 81 in Saint Petersburg. I had fourteen angles. I divided them into three groups. The first group was the angles behind the goal, showing Umtiti's body posture. From this angle, his left arm was on Fellaini's shoulder before the ball arrived. This is a key detail. If contact occurred before the ball arrived, it is a foul. If contact was only for balance, it is not. The second group was the side angles, showing the distance between the two players and the direction of the ball. Here, the ball went behind both. Fellaini could not reach the ball in his current posture because he was being held. The third group was the elevated angles, showing the full position and the relationship between surrounding players. Here, another French player was in a position to clear if the ball came in. Three groups of angles told three different stories. Group one said: there was contact of the hand on the shoulder. Group two said: that contact prevented Fellaini from reaching the ball. Group three said: even without contact, Fellaini would have struggled to touch the ball. So where is the truth? All three are true. They differ only in relevance. And choosing which angle counts as evidence is choosing the conclusion. This is what I call the paradox of angles: the more angles, the more narratives, and the more easily each side picks one angle to defend its position. In this case, I concluded it was a situation that warranted a VAR review, but not enough to overturn the decision. There was contact, but the degree of contact was neither clear nor serious enough under the intervention threshold. I presented that in my article and received two opposing reactions. Half the readers said I was too lenient. Half said I was too strict. Both were right on this point: nobody wants a middle-ground verdict. That is the nature of this profession. You never satisfy both sides. You can only be correct according to process. Now let us leave Saint Petersburg and go to an even thinner line: the offside line. OFFSIDE: WHERE MATHEMATICS FIGHTS THE EYE For decades, offside was a judgment of the human eye. The assistant referee stood level with the last defender, watched two points at once, and decided within a tenth of a second. The error rate of the human eye at that speed is not small. Studies of visual perception show the human eye has a processing delay when tracking two movements in different directions simultaneously. The assistant referee must watch the ball and the player at the same time. This is an almost biologically impossible task. So semi-automated offside technology was born. The system uses multiple cameras and sensors in the ball to determine the moment of contact and the position of every goal-scoring body part. The result is a line. And that line, in many cases, decides the fate of a goal by a few centimeters. Technically, this is progress. Emotionally, it is a disaster. Because a goal disallowed because a striker's toe crossed a defender's knee is something the human brain cannot accept as fair. It is right in data. It is wrong in feeling. This is where I want to pause, because it is the key to the whole problem. The offside law was created to stop players from waiting for the ball. Its original purpose was to counter a parasitic style, not to measure toes. When technology turns a rule against parasitism into a measure of anatomy, it has inadvertently changed the purpose of the law. Nobody on the rules board of the past, writing the offside clause, imagined a striker being denied a goal because his shoe tip was half a centimeter ahead. In other words, technology does not make the law more precise. Technology makes the law different. And that difference is not always acknowledged. This is why I always tell my readers: when you cry over a goal disallowed for offside, you are not crying over the law. You are crying over a system measuring something you never thought would be measured. From offside, let us go to an even more controversial case: handball. HANDBALL: THE MOST-REVISED, LEAST-UNDERSTOOD CLAUSE No clause in football law has been revised as many times as the handball clause. Every update cycle, it is rewritten, clarified further, and made more confusing. The reason is simple: the human arm is not a stable geometric object. When a player runs, turns, or falls, the arm's position changes constantly. The law tries to define when an arm is in an unnatural position, but cannot define what natural means. In referee training sessions, hundreds of situations are shown with the aim of achieving consensus. The result is usually that three groups of opinions emerge consistently. In the same moment, one third says penalty, one third says no, one third says unsure. This is not because referees are poor. It is because the law uses qualitative language for a quantitative phenomenon. And this is what I discovered after years of analyzing referee training: the deep cause of most controversies does not lie with referees, but with the fact that the law is written in the language of the previous century to regulate the phenomena of this one. Once the law is unclear, every verdict can be contested. And once every verdict can be contested, referee credibility is consumed not because they are wrong, but because they must work on a foundation that was never complete. PENALTIES: WHERE PRESSURE IS MEASURED IN SECONDS Among the four situations where VAR can intervene, the penalty is the most impactful and the most scrutinized. I have spent years collecting data on penalties in major tournaments. Not to predict matches, but to understand a psychological phenomenon: when referees award penalties, when they do not, and what changes with VAR. The result was surprising. The number of penalties awarded in tournaments with VAR does not simply increase. What increases is the number of penalties awarded after review, often after the referee decided not to award in the first instant. In other words, VAR causes referees to change initial decisions. But it does not make referees see better from the start. This is an important detail. It shows technology does not improve the senses. Technology improves the capacity to correct errors. And those two are not the same. I once heard a FIFA referee say: I am not afraid of awarding wrongly. I am afraid of awarding rightly but being overturned. That sentence, I believe, precisely describes the psychological pressure of this profession in the VAR era. Referees must now not only make the right decision. They must make a decision that the checking layer behind them will agree with. Those are two different problems. When I built my criteria table for assessing referee decisions by risk level, I included a variable many overlook: the timing of the decision within the match. A decision in minute 12 and a decision in minute 88 of the same nature do not carry the same risk. Not because a player fouling in minute 12 fouls less, but because the impact on the result differs, and therefore the pressure on the referee differs. This is an aspect traditional statistics fail to capture. We count wrong calls, right calls, but rarely count wrong calls at decisive moments. And it is precisely at those moments that referee credibility is built or destroyed. THE WRONG NAME OF 2026 AND THE ORIGIN OF ALL VERIFICATION I must tell you a story, because it shaped my entire approach to this work. In 2026, I worked as a legal commentator for a new sports channel, in a World Cup qualifying match between two Asian national teams. In the first half, I mispronounced the name of the visiting team's number 9 three times in a row. Viewers reacted harshly on social media, and the broadcaster had to insert a correction while the match was still underway. I did not try to justify myself. I acknowledged the error immediately and spent the entire following month reviewing all recorded matches of that national team, analyzing the pronunciation technique based on original phonetics, and compiling a standard transliteration list for over two hundred Asian players. The wrong name of 2026 taught me that credibility is built through correction. That lesson applies not only to pronunciation. It applies to everything in this profession. From then on, I built a two-round verification process before publishing any article. Round one checks facts: time, score, team names, tournament names, legal clauses. Round two checks the spelling of names and teams, with accessible reference sources. I never abbreviate a player's name without verifying the official pronunciation. Mispronouncing a proper noun is enough to remember that every person's name is a world. And when you work with such a process, you begin to see everything differently. You no longer see a referee's error as an individual tragedy. You see it as a process failure. And process failures can be fixed, predicted, and prevented. THE COUNTERINTUITIVE ANGLE: WHEN EMOTION DEMANDS THE RIGHT TO JUDGE This is the part I always write last, because it is the hardest. After every controversial decision, two trials run in parallel. The first trial judges by law. The second judges by emotion. And in most cases, these two trials deliver opposite verdicts. The legal trial asks: which clause was violated, to what degree, and was the intervention threshold met. The emotional trial asks: was my team harmed, was the moment beautiful, was the result deserved. I am not saying the emotional trial is meaningless. Football does not exist without emotion. But I say this: a football ecosystem that operates on emotion will destroy the very people who run it, because no emotion is fair enough to make law. Over years of watching tournaments, I have noticed a sad pattern. When a decision favors the viewer's team, it is called wonderful technology. When that decision is unfavorable, it is called technology destroying football. The same technology, the same process, two names, depending on the outcome. This is not a debate about law. This is a debate about bias. What I want to emphasize, and this is the central counterintuitive point of this article: the biggest problem with VAR is not that it is wrong. The biggest problem with VAR is that it is too right compared to what spectators are willing to accept as fair. Spectators say they want accuracy. But when accuracy goes as far as taking away a goal over a toe, they find it absurd. This is the fundamental paradox of the technology era in sports: technical accuracy often does not coincide with perceptual rightness. And when those two separate, a gap appears. In that gap, rumor grows. Theories of bias grow. And belief in the sport erodes, not because of one specific decision, but because of hundreds of small decisions that leave spectators unable to understand what counts as right. Football law is like a whistle: small, but it decides everything. And if that whistle is no longer understood, the match loses its most precious asset: consensus that the result was deserved. THE PARADOX OF DATA: YOU CANNOT MEASURE WHAT YOU HAVE NOT DEFINED In my work, I have many times tried to use data to resolve controversies. Each time, I hit the same wall. That wall is this: you cannot measure a concept that has not been clearly defined. If the unnatural position of the arm has no quantitative definition, then every handball statistic is only a statistic about referee decisions, not about the phenomenon itself. You are measuring judgment, not events. This is what I call the trap of sports data: the more data, the more we feel we understand. But in many cases, we are only measuring how we judge, and then mistaking that for the nature of the problem. A concrete example. When I charted penalty rates by match minute in a major tournament, I found a curve. Penalty rates rise toward the end of the match. Many explain this by saying the trailing team attacks more. That is partly true. But there is another part: players are more tired, defending is looser, and incidents in the area are more frequent. Meaning the cause lies partly in fitness, not only in tactics. I once wrote that a team's passing rate in the final third rises eighteen percent when playing at home. That number is meaningful, but it only means something if you understand what it measures. It measures passing choices in a specific zone, not attacking quality. That is the difference between data and meaning. So when speaking about referees, I always try to do one thing: separate what can be measured from what cannot, and state clearly which is which. The number of contacts can be measured. The degree of recklessness cannot be measured directly. The number of handballs in the area can be counted. Whether an arm is in an unnatural position cannot be counted. And that is why every controversy will continue. PERSPECTIVE ON TRENDS: REFEREES ARE SHIFTING FROM JUDGE TO ADMINISTRATOR There is a shift underway that few name. In the traditional model, the referee is a judge. He stands at the center, makes decisions, and his authority comes from position and whistle. He is the symbol of the law. In the modern model, the referee is becoming an administrator. He is no longer the sole source of decision. He is the manager of a process involving multiple checking layers, multiple participants, and multiple information sources. His job is not only to decide, but to coordinate deciding. This shift has two consequences. The first is positive. Referees no longer bear the full burden of deciding alone. When decisions are distributed across a system, pressure on the individual decreases. This can help referees work more stably in big matches. The second is negative. When decisions are distributed, responsibility is also dispersed. And when responsibility is dispersed, no one bears final responsibility. Spectators do not know whom to blame, and in that ambiguity, trust erodes. A wrong decision by an individual referee can be forgiven. A decision by the system cannot, because no one can apologize on behalf of a system. This is the responsibility paradox of the technology era: the more checking layers, the fewer people responsible. And the fewer people responsible, the harder it is to correct errors. A LESSON FROM AN UNFORGETTABLE YEAR: WHEN FOOTBALL STOPPED In 2026, the pandemic halted leagues worldwide. I was in charge of legal analysis for a regional federation. When the lower divisions paused, seven clubs in the region could not pay player wages, threatening a mass withdrawal. I was assigned to draft a process for assessing labor contracts under force majeure, applicable to both domestic and foreign players. I worked with lawyers from three clubs, collected forty-five contracts, cross-checked against international sports law, and within two weeks completed a guidance document of twelve sections. That experience taught me a lesson I applied to every subsequent article. In a crisis, the most valuable thing is not a verdict, but a process. A verdict resolves one situation. A process resolves every similar situation. From then on, I began writing in an emergency-situation template. State the problem. List risks in priority order. Offer feasible short-term solutions. And always have a fallback plan. Stopping the ball is an art; stopping words is a responsibility. That sentence, I drew from that very period. When people panic, the person with credibility is not the loudest, but the slowest and clearest. REFEREE INVESTIGATION: METHOD OVER JUDGMENT When I write about referee errors in major tournaments, I always follow one principle: judge compliance with process, not the person. This matters for two reasons. First, judging people is analytically useless. It gives the reader no new information. Second, judging people is procedurally unfair, because referees do not choose the law, do not choose the technology, and do not choose the intervention threshold. They work in a system designed by others. So my approach to a referee error is to break it into steps. Was the referee in the right position. Did he observe the right moment. Did he apply the right clause. Did he follow the right intervention threshold. Did he coordinate correctly with the checking layer. When you break an error into such steps, you often find the fault is not in the last step. It is in an earlier step. And that earlier step is usually a system problem, not an individual one. That is why I believe referee reform should not start by replacing people. It should start by replacing processes. PROPOSALS: FOUR CHANGES THAT CAN BE MADE NOW I am not a rulemaker, and I have no authority to change anything. But I have data, observation, and a responsibility to state what I see. Below are four proposals I consider feasible in the short term. First, publish the audio of exchanges between referees and the checking team. In recent major tournaments, partial audio publication has been implemented with positive feedback. The reason is simple: when spectators hear the reasoning behind a decision, the level of suspicion decreases. Publication does not make a decision more right, but it makes it more understandable. And in sports, understandable is nearly as important as right. Second, set quantitative thresholds for qualitative clauses. If the law says unnatural position, it needs an accompanying definition, for example regarding the distance of the arm from the body, the direction of arm movement, and observable intent. This will not eliminate controversy entirely, but it will reduce it from completely vague to debatable with grounds. Third, standardize referee training around real situations. Instead of training by theory, train with thousands of past incidents, labeled with standard verdicts, and measure referee consensus over time. If consensus rises, the system is working. If not, the law itself needs review. Fourth, build a public post-tournament correction mechanism. When a wrong decision affects a result, the refereeing body should publish a short report stating which process was applied, what was right, what was wrong, and preventive measures. This is what I learned from my own error in 2026. Late correction loses credibility. Early correction builds it. ABOUT NUMBERS: MEASURE WHAT, AND WHY I want to spend a section on numbers, because in this profession numbers are a double-edged sword. When you present a number, you create a sense of certainty. But that certainty is only valid if the number measures the right thing. I have seen too many analyses present a barrage of metrics without answering the core question. I call that decorative statistics. It makes an article look professional but does not help the reader understand anything more. My principle is simple: three judgmental numbers are worth more than thirty decorative ones. When analyzing a referee decision, the three numbers I always want are: the timing of the decision within the match, the distance between the referee and the incident, and the number of similar incidents handled in the same tournament. These three answer three important questions: what was the pressure level, what was the observation capacity, and what was the consistency level. I never write that a player is in good form. I write that the player's passing rate in the final third rises by what percentage when playing at home. The difference between the two is not accuracy. It is verifiability. A vague claim cannot be proven wrong. A specific number can. And I would rather be proven wrong than say things no one can check. CONSISTENCY: THE TRUE MEASURE OF FAIRNESS If I had to choose one word to describe the core problem of modern refereeing, I would choose consistency. Nobody really demands that referees be right all the time. That is impossible. What people demand is consistency. The same situation, the same handling. If it is a penalty in this match, it is a penalty in that match. If it is not a foul in minute 10, it is not a foul in minute 90. The problem is that consistency is hard to measure, and even harder to achieve with many referees, many tournaments, and many different intervention thresholds. A referee in one league may have a different threshold than one in another. And when two matches handle the same situation differently, spectators do not see the difference in context. They only see injustice. This is why I believe the most important reform is not a technology reform, but a consistency reform. Technology can help. But technology is only a tool. What needs to change is how humans agree on thresholds. And to agree on thresholds, you must first be able to define them. That is the hardest, least glamorous, and most important work. WHAT I AM NOT SURE OF, AND WHY I SAY IT I have tried to write this article with the highest possible degree of certainty. But I also have an obligation to state what I am not sure of. I am not sure whether publishing audio will actually reduce controversy in the long term, or merely shift it from the decision to the explanation. I am not sure whether quantifying qualitative clauses will make the law better, or merely stricter, creating new injustices. I am not sure whether standardizing referee training is feasible in terms of cost and organization. I state these not to reduce my responsibility. I state them because I believe an honest analysis must include its own limits. If I presented everything as certain, I would be selling you confidence, not truth. And I have learned, from a name mispronounced in 2026, that truth is worth more than confidence. CONCLUSION: WHAT REMAINS AFTER THE WHISTLE GOES SILENT I have spent years reviewing incidents to find an error nobody saw. I have watched hundreds of moments. I have written hundreds of analyses. And what I took away is not a formula to always be right. What I took away is an attitude: patience with process, severity with myself, and honesty with data. When a referee decision sparks controversy, spectators usually look for a guilty party. I hold that the right question is not who is guilty, but which step in the process failed. When you can answer that, you do not just resolve one match. You improve an entire system. Football will never be free of controversy. And perhaps that is a good thing. Controversy is a sign that people care. But grounded controversy is better than controversy based on feeling. And for grounded controversy, fans need to understand the law. Not to defend their team. To know when they should be angry and when they should be silent. During the transfer window, people usually talk about money, contracts, and release clauses. I care about money too. But I care more about something else: the value of clarity. A clear clause can decide a player's career. A clear clause can also decide the faith of millions of fans in the sport they love. If I leave only one thing after this article, I want it to be this: when you watch a match, remember that behind every whistle is a journey of observation, judgment, application of clauses, and accountability. When that journey is done right, the whistle becomes justice. When that journey fails, the whistle becomes controversy. And between those two, the difference does not lie with the person blowing the whistle, but with the system that placed the whistle in their hands. I will keep writing. I will keep reviewing incidents. And if I am wrong, I will correct it, as early as possible. Because credibility is not built by never being wrong. It is built by correcting errors at the right time and in the right way.