When Data Is Insufficient: Esports Referees and the Principle of Silence Before Evidence

At the 2026 World Cup group stage, four of twenty-five semi-automated offside decisions took more than eighty seconds to reach a final result. I recorded that figure in the notebook I kept throughout the tournament, next to a line marked in red: "SAOT is fast, but not as fast as people think." The semi-automated offside technology was celebrated by Qatari media as a revolution — twelve cameras, a sensor inside the ball, a three-dimensional model rebuilt within seconds. But when I rewatched the footage frame by frame, what struck me was the silence. The silence between the moment the system returns a number and the moment the referee blows the whistle. In that silence, the stands had already passed judgment. And in that silence, I realized that the problem facing esports referees today does not lie in the machines, but in the human hands that operate them.

Six years after VAR debuted at the 2026 World Cup, the story of officiating in sport has shifted on its axis. In football, the referee is no longer the only person who sees the play; they share their field of vision with a closed room full of screens. In esports, that boundary is even blurrier. A League of Legends, Dota 2 or Valorant match unfolds in a digital environment where every click leaves a trace, but also where every trace can be misread without context. Servers record coordinates, timestamps, damage; servers do not record intent. Most esports disputes, from a contested objective to a technical pause, sit precisely in the gap between data and intent.

Based on my experience watching matches, most controversial decisions do not come from referees lacking data, but from them concluding too quickly while the data is still incomplete. Fans remember player names; I remember where the assistant referee was standing. That is why I always begin an analysis with the driest possible question: how much of the data do we actually have before we conclude?

I came into this profession through a notebook. In 2026, while the whole world talked only about goals at the World Cup in Russia, I sat recording every refereeing decision across sixty-four matches: two hundred and eighty-six yellow cards, four red cards, twenty-two penalties. In the France – Croatia final, referee Nestor Pitana whistled eleven fouls in the first half alone, and I wrote in the margin: "One day I will have to do this job." By August, the notebook ran to forty-seven pages, classifying one thousand two hundred and eight decisions according to a form I designed myself. Those forty-seven pages taught me one thing only: stay silent when you have not seen the evidence. Every passage of play is a line in a report, and I write none of it down incompletely.

In 2026, when the pandemic left stadiums empty, I rewatched forty-three matches of the Malaysia Super League played without spectators and found a figure that made me sit still for a long time: referees' bias in favour of home teams was eighteen point two percent lower than in the 2026 season. My first article on that finding went live in June 2026, right during Euro 2026 and the Tokyo Olympics. My blog traffic jumped from seventy to two thousand one hundred visits per week. What I remember most is not the readership, but a single comment: "So referees are swayed by the crowd after all." Correct. And precisely because of that, data must matter more than feeling.

That same summer, the Euro 2026 semi-final between England and Denmark produced a controversial penalty, with referee Danny Makkelie making the call. I spent three days reconstructing the incident from seven different camera angles, measuring contact distance, and cross-checking it against the "sufficient contact to fall" standard UEFA had published. The analysis drew three thousand two hundred reads overnight. But my conclusion was neither "right" nor "wrong". Referee data is not for convicting, it is for exonerating. I showed that from the main broadcast angle the contact looked heavier than it was; from the angle behind the goal, it was lighter than the penalty threshold. Both were visually true, and both were insufficient to conclude.

At the 2026 World Cup in Qatar, I followed referee Szymon Marciniak through the Argentina – France final. The score stood three-all after extra time, four-two on penalties. I tallied twenty-eight fouls, six yellow cards and two penalties across one hundred and twenty minutes. A final forgives no carelessness, not even a referee's. But what drew my attention most was how Marciniak managed the emotions of twenty-two players in a match where every minute could rewrite history. He did not brandish cards to display authority; he showed them to cool things down. That is a skill that appears on no statistical table.

Then came SAOT. The semi-automated offside system promised to eliminate human error by sending signals directly from the sensor inside the ball and motion-tracking cameras. But when I cross-checked twenty-five offside decisions at the group stage, four of them took more than eighty seconds. SAOT is a steel eye, but the operator is still a human hand. The system only returns positional data; it does not decide. Someone still has to choose the moment the ball leaves the foot, still has to define which part of a player counts as offside, still has to decide whether the passage of play is worth stopping at all. Technology does not remove judgment; it merely relocates it into a closed room where no one sees who is making the call.

This is where esports must learn from football, but in its own way. In a League of Legends match, when a team requests a technical pause over a suspected connection fault, the referee has no footage to review the way VAR would. They have server logs, ping, data packets. But server logs do not tell the story of a player who lost a crucial teamfight to latency rather than to a bad decision. That is an evidence gap, and the only way to close it is process: record everything, publish the criteria, and accept that some passages of play will never yield a conclusion that satisfies everyone.

If football took more than a century to standardise the offside law, esports has covered only a tiny fraction of that road. The rules of League of Legends or Dota 2 change by season, sometimes every few months. Each time, officiating standards must be rewritten from scratch. A play deemed legal this season can become a foul the next. That is why I always attach the patch number and the date to every note: a conclusion without a timestamp cannot be reused.

I live and work in Penang, reporting on esports for the Malaysian market. What I see here is a familiar paradox: domestic tournaments grow more professional by the year, yet the culture of judgment remains emotional. A young player who misplays a teamfight can be convicted by social media within hours, while the footage shows his team's connection had been flickering for ten minutes beforehand. Stories like that convince me that exonerating people with data is not merely a writing technique, but a moral duty.

Across seven years of watching this industry, I have noticed a recurring pattern. When a team loses, the community hunts for an individual to blame. When a player is criticised, few bother to reopen the footage to check whether the accusation holds. Emotion can tilt, but footage cannot. I once spent two thousand three hundred words analysing Lamine Yamal after his goal against France in the Euro 2026 semi-final, when he was barely sixteen. Colleagues called him a generational prodigy. I gathered data from fifty of his matches at Barcelona in the 2026-2026 season, cross-checking it against Lionel Messi in 2026, Kylian Mbappé in 2026 and Pedri in 2026. My conclusion: at least fifty more high-density matches are needed to establish "generational class". The article was cited by four newspapers, but it appeared exactly three days after my colleagues'.

Those three days are the whole story. In those three days, the community had finished writing the legend, finished building the statue, finished selling the belief. And three days later, with a little more data, the picture emerged a little more complex. I accept being slow. But I do not accept conclusions without evidence.

The irony is that technology itself, the very thing expected to soothe disputes, keeps generating new ones. VAR and SAOT lead spectators to believe every decision can be "proven" objectively. That belief overlooks one simple fact: data does not speak on its own. An offside line drawn on screen looks absolutely precise, yet it rests on a specific frame, an interpolation algorithm, a choice of moment that a human set beforehand. In esports, a damage figure on a scoreboard looks dry and neutral, but it can overlook that the player spent twenty minutes under gank pressure, sacrificing resources to open a path for teammates.

Spectators' emotions are valid data, but they are not evidence. A defeat makes people want to believe in a conspiracy, a script, an individual betraying the collective. My job is to replay the entire footage to cool down accusations of "match-fixing", "deliberate losing" or "carrying the team" that have no basis. Not because I believe any given player is clean by default, but because the burden of conviction must rest on the accuser, not the accused. Footage is one of many sources of evidence, not absolute truth; but while no better evidence exists, it is the fairest anchor we have.

The biggest blind spot in esports analysis is not a shortage of tools, but a surplus of conclusions. People conclude first, then go looking for data to justify it. I do the opposite, and that is why I am often seen as slow. But a referee does not hold the whistle to beat the clock; they hold it to blow at the right moment.

Esports is growing faster than any governance system can keep pace with. Every new tournament brings a new kind of dispute, and every kind of dispute demands a new standard. I believe the industry's future lies not in adding ever more technology, but in building a generation of referees willing to say the hardest sentence of all: there is not enough data to conclude. What I want the community to answer for itself is not whether machines can replace referees, but whether we have the patience to wait for a correct conclusion instead of a fast one.