The Empty Frame: A Lesson in Honesty from the VAR Room

Hai Phong, 21:47. The wall clock in the observation room runs about four seconds behind the clock on the pitch, and I have grown used to that gap. On the monitor, Ceres-Negros striker Fidelis Ikiri had just put the ball over the home goal line. The Lach Tray stands rose. At real speed the move looked legal: the away striker broke through, a Hai Phong defender raised his arm, the assistant referee kept his flag down.

I rewound it. Frame 25, then frame 30. The offside line was drawn from two points: the last defender, and the playable part of the attacker's body. The gap appeared in the measurement box: 0.3 metres. Thirty centimetres. Enough for the naked eye to miss inside one fifth of a second. I pressed the signal. Three seconds later the referee raised his arm and the goal was disallowed. The match finished 2-1 to Hai Phong.

The home coaching staff left the pitch to a wall of noise. No one knew what I had been doing in that room. No one at the press conference mentioned those thirty centimetres. There are offsides nobody sees, but the camera never blinks. That was the night I understood that my job is not to deliver a verdict, but to make sure the verdict has a foundation. Those are two very different things, and most of the arguing in the stands exists because nobody distinguishes between them.

From whistle to algorithm

In 2026 FIFA brought the video assistant referee into a World Cup for the first time. IFAB, the body that governs the laws of football, set a threshold called "clear and obvious error" — meaning VAR may only intervene when the mistake is large enough for everyone to see. Four years later, in Qatar, semi-automated offside arrived: twelve cameras under the roof tracking 29 points on each player's body, with alerts reaching the operations room within seconds. In the 2026-25 season the Premier League adopted the same technology. At the other end of the same trend, the ATP rolled out Electronic Line Calling across its entire tour from January 2026, and Wimbledon 2026 became the first edition in 147 years with no line judges on any court.

What these changes share is a belief that is very easy to accept: machines do not blink, machines do not tire, machines do not favour anyone. That belief is partly right. But it rests on a premise that is rarely spoken aloud: the machine must be running, and the data must have arrived. After years in this trade, I have come to see that almost every public argument about sports technology concerns the final step — the moment the referee raises his arm — while the decisive step sits upstream, where somebody is checking whether the signal has actually come through.

Three weeks and sixty-four matches

In June 2026, at the World Cup round of 16 in Russia, I was one of three assistants supporting the referee in the match between Spain and Russia. In the 42nd minute, Gerard Pique handled the ball inside the penalty area. I did not spot it. The referee was called to the touchline monitor, and Artem Dzyuba scored the penalty. The match ended 1-1 after 120 minutes, Igor Akinfeev saved two in the shootout, and Russia went through.

I blamed myself for three weeks. During those three weeks I rewatched all 64 matches of that World Cup, logged every VAR incident, and rebuilt the timing of each frame. I did not share how I felt with colleagues. Not out of fear, but because I wanted to know where I had gone wrong before telling anyone that I had gone wrong.

The result of those three weeks was not a list of errors. It was an observation about process: we were deciding faster than the data reaching us could support. A frame pushed onto the screen half a second earlier than the correct frame can place an arm where it never was. I wrote a series on my personal blog admitting my own mistake and proposing that the check be extended by two seconds before locking in a call. Two seconds sounds small. In a football match, two seconds is an entire counter-attack.

The biggest mistake is not blowing the whistle, but refusing to own your whistle. I learned that sentence by paying for it.

When the frame is empty

There is a harder situation than being wrong. It is having nothing to look at.

The feed from the stadium to the analysis room is not always stable. There were evenings when the screen simply went black, or worse, stayed lit but frozen on the third second while the match had already moved ten seconds on. The referee is waiting for your signal. The stands are howling. The commentators are guessing. And you are the only person on the planet who can say a sentence that carries weight: "I have it."

The natural human reflex in front of a gap is to fill it. The brain cannot tolerate silence, so it takes experience, memory, probability, and assembles a story that sounds entirely reasonable. In the VAR room that reflex has another name. It is called guessing. And guessing is the one thing that is forbidden.

Sometimes the correct conclusion is a diagnosis of the system's own limits, not a verdict on the incident.

I know that sounds like an evasion. But count the cost. If I fill the gap with a plausible conclusion, that conclusion goes into the match report, into the next morning's coverage, into the memory of the fans, and into the career record of a twenty-year-old player. It will never be corrected, because no one will remember it was born from an empty frame. An invented conclusion is more dangerous than an empty one, because an empty one can still be filled with real data, while an invented one has already become history.

The data analyst walks into the dressing room

Over the past decade a new class has appeared in professional football: the data analyst. They arrive with spreadsheets, models and expected-value metrics. In some ways they do what I do — read what the eye misses. But there is a fundamental difference that keeps me at a certain distance from that trade.

I once saw an internal report concluding that a team was playing well because their expected-goals figure was higher than their opponents' across four rounds. Three weeks later that team lost four in a row. The problem was not the number — the number was correct. The problem is that a number cannot measure rhythm. It does not know the holding midfielder has played four matches in ten days, that the right back has a sore ankle, that the coach has just changed shape and the whole squad needs two weeks to adapt.

A model built on past data will always answer the question you ask it. What it cannot do is tell you your question is wrong. The rhythm of a match sits outside the spreadsheet, and in the VAR room that rhythm decides how much time you have left to look.

Tennis, electronic line calling and a 3.6-millimetre tolerance

Tennis moved ahead of football in handing judgement to machines. When the ATP rolled out Electronic Line Calling across the whole tour from the start of 2026, the arguments largely vanished from close calls. Spectators no longer see a line judge crouch. Nobody argues any more. That is a genuine achievement.

But that technology also carries a tolerance, and that tolerance is usually not projected onto the big screen at centre court. Tracking systems work by reconstructing trajectories, and a tennis ball deforms significantly when it hits the surface. The published margin of error sits in the region of a few millimetres. Those millimetres are seen by no one, shouted about by no one, written about by no one. They sit quietly inside every close call in every major match.

A millimetre changes a team's fate; I have learned to live with that.

What I am saying is not that technology is bad. Technology beats humans at the business of repeating without fatigue. What I am saying is that every system has an edge, and an honest sport has to tell its audience where that edge lies. No system measures to infinity. There are only systems that admit their limits and systems that hide them.

A transfer contract is like an offside call

The transfer market is where I see the same mistake in a different shape. Big clubs buy data packages, hire analysts, build performance centres — an arms race of branding. They need to look modern more than they need to be right. Meanwhile the small clubs, whose budget covers one signing per window, have to be right. They have no room for error.

A contract is like an offside call: misjudge the timing by a beat and everything falls apart.

At Hai Phong I once tracked a seventeen-year-old during a period of financial crisis at the club. He had the technique, but his nerve collapsed whenever he was thrown on while the team was behind. The coaching staff wanted him off the list. I sat down, reviewed every touch he had taken in training, and sent a short report to the technical director. I did not judge whether he was good enough. I said only that he needed six months of separate work with the youth side.

Six months later he made his debut and scored a goal that kept the club up. I do not tell that story to claim credit. I tell it to point out that a correct decision in sport usually lies not in declaring who is good and who is bad, but in identifying the right moment and the right conditions. That is also the job of the person sitting in the VAR room.

The argument does not disappear, it moves

The popular belief is that more technology means more fairness. I think that is half true. Technology does not remove argument. It moves argument upstream.

Fans used to argue about whether a striker was offside. Now they argue about whether the chosen frame was at the right moment, whether the cameras were calibrated in sync, whether the offside algorithm handles a player's skeletal joints correctly. These new questions are far harder, and more importantly, the audience has no access to the answers. They are pushed further from the process while their sense of certainty increases.

At the deepest level there is a paradox I have never seen resolved. A referee is never praised for the decision he did not give. No stand has ever applauded a whistle that stayed silent. By the same logic, nobody praises a report for daring to say "insufficient data". The rewards in this trade always go to the person willing to assert, even when the assertion is built on an empty frame.

That is why I write. The referee is the only person on the pitch not allowed to pick a side — and I stand behind him. Standing behind him means being accountable for what nobody sees, even when nobody acknowledges it.

There is a less comfortable version of this story. When a wrong conclusion is published, the public goes looking for someone to blame. They look at the player, the coach, the referee. They almost never look at the person who supplied the input data. When everyone blames the nineteen-year-old who missed his penalty, the person in the VAR room has to stand up — not to defend the boy, but to state clearly which part of the failure belonged to the process and not to his feet.

One concrete proposal

I once proposed extending the check by two seconds. After several years I realised two seconds is not enough when the input data was thin to begin with. So I propose something else: a data integrity declaration.

Before every match that uses assistive technology, the organisers publish four figures: the number of operational cameras, the actual frame rate, the calibration status of the system, and the transmission latency to the operations room. Those four figures take about thirty seconds to read out. In return, spectators know exactly how much certainty underlies the thing they are about to argue about. If a frame is missing, they have the right to know before they start shouting.

It sounds technical and dry. But ultimately the entire ethical foundation of sport rests on spectators believing the result is real. That belief is built from the smallest details, at the hours when nobody is watching, in the rooms where nobody is filming.

I found that offside at two in the morning, after everyone had gone home. Nobody asked me to do it. But if a sport is only honest while the crowd is watching, then it was never honest at all.

Tomorrow, at some stadium somewhere, there will be another empty frame. The question is not who will see it. The question is who will have the nerve to say that it is empty.