VAR's Blind Spot: When the Footage Is Sharp but the Reader Is Not

VAR's Blind Spot: When the Footage Is Sharp but the Reader Is Not

Anfield, December 2026. The trial VAR room sat behind the west stand, barely fifteen square metres, with three screens arranged in an arc and a stopwatch I still keep today. On 73 minutes of the Merseyside derby, inside the penalty area, a player in blue went down after a challenge with a centre-back in red. The assistant referee stood to the right, roughly eleven metres from the point of contact. The referee was about seventeen metres away, his line of sight blocked by a shirt. I pressed the stopwatch, rebuilt the phase at 0.25 frames per second, measured each man's eye-scan angle, and calculated the figure that later became my signature: the man with the whistle had 0.4 seconds of clean sight.

The footage was frighteningly clear. No blur, no grain, no skew. And yet for two weeks the city argued. Nobody argued about what the footage showed. They argued about what the footage meant.

The first lesson, and the biggest: technology does not erase controversy, it simply moves it from the question "who is right" to the question "who gets to define". Every eye has a blind spot; the only difference is whether we dare look into it.

From IFAB 2026 to Moscow 2026: a revolution nobody prepared for

In March 2026, the International Football Association Board (IFAB) approved trials of a video assistant referee system. Two years later, at the 2026 World Cup in Russia, VAR appeared at a major senior international tournament for the first time. In 2026/20, the Premier League adopted it at scale. Three years between those milestones — too short to write a training manual, let alone train an audience.

I have covered eight Olympic Games and eight World Cups. Across that span I have seen the laws of football change at systemic level twice. The first was when the offside law abandoned the 1990s concept of "level is onside", turning every phase into a binary question. The second was when VAR made every phase a binary question you could rewind. Both times, what changed was not football. What changed was how much error people were willing to accept.

In 2026, at the World Cup in Qatar, FIFA deployed semi-automated offside technology: twelve cameras mounted under the stadium roof, recording at 500 frames per second, tracking 29 data points on each player's body and sampling 50 times per second. The ball carried a motion sensor. That is a citable fact — the source is FIFA's technical documentation published ahead of the 2026 World Cup. The number is not there to show off technology. It is there to answer one question: if the machine measures to the millimetre, why are people still shouting at each other?

From 2026, the Professional Game Match Officials Limited (PGMOL) shifted to a strategy of public disclosure. Audio of conversations between referees and the VAR room is released periodically, with explanations from the head of the organisation. From 2026, IFAB approved a rule restricting in-match access to referees to team captains only, trialled at the Euro 2026 finals and then rolled out across several domestic leagues.

Those three layers of reform — technology, transparency, conduct control — are all technically correct. None of them has reached the root of the problem. I will show that root through four cuts.

The "clear and obvious" threshold and the microscope paradox

The founding principle of VAR is the "clear and obvious error". That principle was written for the human eye. It assumes a buffer zone between right and wrong, wide enough that we do not have to adjudicate millimetres.

The problem is this: when you hand a person the ability to rewind footage at 0.25 frames per second, you have not handed him a tool. You have handed him a truth. And once a man holds a truth, he will not accept being told to keep quiet about it.

The principle says: do not re-referee the match. The tool says: you can see the truth. The tool always wins.

The result is a paradox I have tracked across many seasons. The intervention threshold in the written text has not changed. The intervention threshold in practice keeps falling. Offside lines are drawn to the millimetre. Contacts on the edge of the box, once classified as "the referee's call", get dissected frame by frame.

On 17 April 2026, at the Etihad, Manchester City hosted Tottenham in the Champions League quarter-final second leg. Deep into stoppage time, Raheem Sterling scored and the stadium erupted. The goal was disallowed by VAR because Sergio Agüero was offside in the build-up. Tottenham advanced on the away goals rule. It is the perfect illustration: the machine produced an answer that cannot be argued with on the data, and that very unarguability is what hurt most.

When a referee errs and you lose, you lose to misfortune. When a machine is right and you lose, you lose to a law of nature. Misfortune is forgiven. Natural law is resented.

Law 12 and the greyest zone in modern football

Of football's seventeen laws, Law 12 is the only one that cannot be written in numbers. Offside is geometry. Ball out of play is geometry. Handball is psychology.

Before 2026, Law 12 revolved around the word "deliberate". A referee had to read intent in a human being running at thirty kilometres an hour within a fraction of a second. In 2026, IFAB added two more objective criteria: an arm in an unnatural position, and an arm beyond the shoulder line. In 2026/22, IFAB revised again: accidental handball by a teammate in the build-up is no longer an offence, leaving only the case of the scorer himself handling immediately before scoring.

Three amendments in four years. That is the signature of an unstable area of law, and referees are the ones standing on shifting ground.

I remember the pre-match meeting ahead of the 2026 World Cup semi-final between France and Belgium in Moscow. FIFA had invited me onto its media advisory panel. In that room, Pierluigi Collina — a man the entire refereeing profession looks up to as a monument — argued that Samuel Umtiti's handball on 51 minutes was not deliberate. I disagreed. My grounds sat in the text of Law 12 itself: the arm was raised beyond the shoulder line, and that arm blocked the path of the ball. Two objective criteria. No need to read minds.

I cited data from fourteen qualifying matches to show that his reading was inconsistent with how the same law had been applied in the same tournament. The argument ended in favour of my position. L'Équipe reported it.

The detail worth noting is not that I won an argument with a legend. It is that even the most authoritative man in the profession read the same law differently from me, at the same moment, watching the same clip. Legends tell stories with reputation; I tell stories with the match record.

When the written law still permits two reasonable readings, every claim of "clear and obvious" becomes a claim about power rather than truth.

The geometry of a decision

Based on my experience watching matches across many seasons, I keep my own tracking table of reaction times and referee positioning. The method is crude: log the position, measure the sightline, time the interval from the ball leaving the foot to the end of the phase.

| Tracked indicator | Ideal threshold | Common reality | Gap | |---|---|---|---| | Distance to point of contact | 10–15 metres | 15–22 metres | 5–7 metres over | | Clean observation time | Above 0.8 seconds | 0.4–0.6 seconds | 0.3–0.4 seconds short | | Sightline blocked by players | Under 20% of phases | 35–45% of phases | Double | | Deviation from standard diagonal | Under 3 metres | 5–8 metres | Double |

The table says something few want to hear: most refereeing errors are not errors of perception. They are errors of geometry. The referee did not see wrongly. He stood in the wrong place.

And this is what VAR cannot fix. VAR adds a pair of eyes in a closed room, but it cannot move the legs of the man with the whistle. The on-field referee still has to make the first decision — the decision that unlocks the entire process. If that first decision falls inside the "not clearly wrong" zone, VAR is not permitted to touch it. An entire three-tier error-prevention system depends on one second of judgement from the person closest to the ball.

In football, there are runs only the referee sees; everyone else sees only the outcome.

Empty stadiums, noise and 27 per cent

In May 2026, when football restarted after lockdown, I proposed a data project to the Merseyside referees' association: collect cards, stoppage time and referee-player exchanges from matches played without crowds. Colleagues called me outdated. Three months later I had results from thirty-two matches.

The yellow-card rate rose 27 per cent against the same period the previous season. Average stoppage time per contested phase fell. The number of players surrounding referees dropped markedly.

The easy reading: crowds make players hot-headed, so without crowds the game is cleaner. The second reading, and the one I choose: without noise, referees decide by law rather than under pressure. More cards does not mean a more violent game. It means that previously, some fouls went unpunished because blowing for them would have provoked seventy thousand people into a roar.

The report was published in full in a specialist refereeing journal. What I took from it was not the 27 per cent. It was this: for decades people assumed crowd noise was part of football. It turns out it was a variable never entered into the model.

Empty stadiums still produce data; it is the noise that distorts the verdict.

The counter-intuitive view: we are wrong about the question, not the answer

This is what I believe, and I know it will irritate people.

Refereeing accuracy has improved. Fans' sense of fairness has worsened. Those two curves run in opposite directions, and we have spent nearly a decade arguing about the first one.

The conventional explanation is that fans are biased, fans do not know the laws, fans are led by social media. That explanation is both true and useless, because it reduces everything to the audience's fault.

My explanation differs. Error in football before VAR was random error. It was distributed evenly, aimed at nobody, and therefore forgiven. Error after VAR has an author. When a decision has been reviewed by three people on four screens, it is no longer an accident. It is an act. And acts have accountable agents.

Trading random error for authored error is a bad deal emotionally, even when it is a good deal technically. Fans do not buy accuracy. Fans buy the feeling of being treated justly. Those two things are close but not identical.

At a higher level, clubs understand this mechanism perfectly. A team beaten three-nil can still spend a full media week talking about a VAR call on 68 minutes. That noise is not aimed at fixing the system. It is aimed at diverting attention from a failing midfield, a failed winter signing, wage-bill pressure. The referee becomes a free shield, and that shield is always available.

I have written repeatedly about risk-aversion in modern football. It does not exist only on the coaching bench. It exists in the VAR room too: a referee knows that if he intervenes and is wrong, his name makes the front page; if he does not intervene, his name never appears. That incentive structure pushes human beings toward silence.

The match does not end when the whistle goes; it ends when we face up to our own error.

What should happen next

Three things, in order of priority.

First, invert the intervention threshold. "Clear and obvious" was designed for the human eye and is obsolete now that the tool has become a microscope. Instead, adopt a challenge system: each team gets two review requests per match. A team that believes it was wronged pays with its own request. The burden of proof moves from the VAR room to the club. This frees referees from the pressure to intervene constantly and kills the culture of endless complaint.

Second, publish referee positioning data in post-match reports. Not to shame individuals, but to move error from the emotional zone into the measurable zone. The same referee standing twenty metres from the point of contact three matches running is a coaching problem, not a moral one.

Third, open referee microphones live during matches, not only in retrospect. The captains-only rule points the right way, but it still leaves the crowd guessing. Half-transparency is still a form of darkness.

A whistle can change a club's fate, but it must never change the conscience of the man blowing it.

And the question I leave for the people running football: if we have invested enough money to measure a player's movement to the millimetre, why have we not invested enough thinking to explain to the audience that a millimetre is not justice?