Manchester Derby VAR Stood Down: One Error, Two Layers of Failure, and the Price of Trust

On the night of the Manchester derby, when the ball left Erling Haaland's boot and crossed the line, I was sitting in a broadcast booth in Manchester and heard a strange silence. Not the silence of a goal. The silence of a stadium waiting. Waiting for the line. Waiting for the editor. Waiting for someone in a closed room to confirm whether this goal was permitted to exist.

I have worked in English football journalism long enough to know that, in the Premier League this season, a goal is no longer an event. It is a proposal awaiting approval.

The next morning, the football press reported that the VAR team from the Manchester derby had been withdrawn from the following round of fixtures. VAR Matt Donohue and assistant VAR Blake Antrobus were not assigned. The refereeing body admitted they had been wrong to allow the goal.

An error admitted before anyone could interrogate it. That is what made me sit back down.

The Manchester derby ended 1-0 to Manchester City. Haaland scored the decisive goal. On the field, the assistant referee initially raised the flag for offside. The VAR team then intervened and overturned the decision to onside. The goal stood. City took three points from United.

Then came the turn I seldom see at this level: the refereeing governing body — named Pro Ref in the source — issued a statement admitting the VAR team had erred. Specifically, they said VAR had failed to take into account the impact of a player in an offside position, affecting the opposing goalkeeper or defender's ability to reach the ball before Haaland struck.

Matt Donohue was the VAR. Blake Antrobus was the assistant VAR. Both were removed from the assignment list. The decision was made by Howard Webb — chief refereeing officer — and the management team. Their expected return is no sooner than after the international break.

There is one detail here that demands the reader stay alert. The statement mentioned a player with a very familiar name: Enzo Fernandez. The problem is that Enzo Fernández — the Argentine midfielder — plays for Chelsea. He has no connection to Manchester City or Manchester United. This detail is almost certainly a source or extraction error, and it lowers the reliability of the document. Someone who reads statements for a living, as I do, cannot overlook that. I flag it here, and I will return to it later.

What actually happened in that moment?

The IFAB offside law — Law 11 — contains a clause known as "interfering with an opponent". A player in an offside position may be penalised if they affect an opponent's ability to play the ball. This is the most subjective clause in the entire rulebook. It cannot be drawn with a line.

The trap Donohue and Antrobus fell into has a very particular structure. Initially, the assistant referee raised the flag — meaning he saw offside. The VAR team overturned it — meaning they saw onside. Two parties looked at the same passage of play and read two different conclusions. That is not rare. What is notable is that, in overturning it, VAR appeared to check only the position of the scorer, while ignoring the role of a second player in an offside position.

This is the crux. The error at the Manchester derby was not about the line, but about the role. The VAR team verified the question "who is where" correctly, but answered the question "who is doing what" incorrectly.

In other words: this is a two-layer error. The first layer is the on-field decision. The second layer is the reversal in the booth. And the second layer is harder to correct than the first, because it was made by technology. When a machine tells you that you are right, you carry that feeling off the pitch.

Since VAR arrived, English football has been waging a quiet war between two things: instinct and evidence. The assistant referee raises the flag by instinct. VAR overturns it by evidence. But which evidence? One frame? Two frames? Thirty frames per second, and how many of those frames contain a player's "intent"?

I have hosted major sports events, and I learned something about audiences: they forgive a referee who makes an error far faster than they forgive a referee who explains. When a referee points to the spot, the stands react and then move on. When VAR takes three minutes to draw a line, the stands do not react — the stands freeze. And that freeze accumulates round after round.

Technically, this error belongs to the hardest category to automate. Semi-automated offside technology can solve the positional question very well: the line, the toe, the shoulder, the knee. But it cannot solve the role question: did that player block the view, did he distract, did he force the goalkeeper to choose wrongly. That is why every time football applies more technology, it exposes more of the human part — and that human part is not in the VAR booth, it is in the millisecond a player has to decide.

On a broader view, the refereeing body's response has a notable structure. They did not wait for external pressure to escalate. They admitted the error first. They withdrew two officials first. Howard Webb and the management team put their names to that decision. Administratively, it is an assignment decision, not yet a formal disciplinary penalty. But in football's world, being removed from a round of fixtures is a form of discipline that needs no name.

I have enough experience to tell self-defence from responsibility. Here, I see both. The refereeing body protected institutional credibility by acting before public opinion acted. That is a smart move. But a smart move is not the same as a solution.

Now, back to the Enzo Fernández detail. I do not want to use it as a joke. I want to use it as a test. An official statement named the wrong player in one of the most-watched matches on the planet, and what does that say about the precision of the whole system? There are two possibilities. Either this is a simple clerical error, and the fault lies in the communication chain. Or there is another player sharing the name, and the story is more complex. Whichever it is, it forces me to downgrade the reliability of the entire account and verify each detail myself.

That is the discipline I learned from years of writing about crisis clubs. Sunderland were relegated, and I still hold my view: going down is a way of going up. But going up from where? From admitting precisely what you do not know.

In terms of the result, the allowed goal reshaped the table. Manchester United lost at least one point, possibly three. In a season where the gap between a European place and seventh is only a few points, a wrong decision can be the boundary between singing in a European away ground and watching on television at home. I use no specific table figures here, because the report does not provide them. But the principle is clear: a wrong goal does not merely steal three points from one match, it rewrites a chapter of the season.

The contrarian view I want to offer sits here: the suspension of two referees does not prove the system is working well. It proves the system is working fast enough to protect itself.

Picture a stage. The audience boos the actor. The director instantly replaces the actor for tomorrow's show. The audience applauds. But the script is unchanged. The fault at the Manchester derby lies in a clause written with vague words like "interfering", and in a VAR process designed to answer a binary right-or-wrong question while the actual passage of play is a continuous grey band.

When I hosted a major event recently, I mispronounced a player's name three times. The audience called in to complain. I did not turn it into a tragedy. I did not turn it into a joke either. I simply noted: my error was an error of detail, not an error of thinking. At the Manchester derby, the VAR team's error may likewise have been one of detail — a clause applied without sufficient consideration. But unlike me, they cannot say "sorry, I will be more careful" and carry on. Because they operate something that is supposed to guarantee fairness.

I once stood in an empty Etihad during the football shutdown, and I heard my own applause most clearly. The silence at the Manchester derby that night echoed in a similar way. When there is no audience left to react, people hear only the sound of the system auditing itself.

What is worrying is the risk of recurrence. This type of multi-player offside error cannot be erased by more training. It can only be erased by changing how the question is framed. And changing the question is the work of lawmakers, not of whistle-holders.

Football was always a play of mistakes — I merely helped make it more watchable. But when mistakes become a systemic process, we are no longer watching a play. We are conducting an audit.

The question I leave is not whether Donohue and Antrobus deserved to be removed. The question is: how many more times will we replace the actors before we dare to rewrite the script?