Expert comment: What really went wrong with VAR in the Manchester derby?

The controversy over Erling Haaland's winning goal in the Manchester derby is being described as another VAR failure. But there is a more interesting story here: the technology appears to have worked, while the human decision-making around it failed.

Haaland was ultimately shown to be onside, with Manchester United's Patrick Dorgu playing him onside. The more difficult issue concerned Manchester City's Enzo Fernández, who was in an offside position and attempted to play the ball but did not touch it.

Professional Game Referees (Pro Ref) subsequently acknowledged an “error of judgment,” saying VAR failed to recognise the likely impact of Fernández's position and should have recommended an on-field review.

This makes the incident a useful case study in human decision-making, the limits of VAR technology and how officials interpret complex visual information.

The technology answered the easy question; but not the difficult one

There were several different questions within the incident, requiring different types of decision:

• Was Haaland offside? (Objective/technology assisted decision)
• Was Fernández in an offside position? (Objective decision)
• Did Fernández touch the ball? (Objective decision)
• Did Fernández's action affect an opponent? (Subjective decision)
• Should the referee have reviewed that question? (VAR process)

This distinction is important.

The technology can establish where players are positioned, but it cannot completely automate the interpretation of whether an offside player's movement has interfered with an opponent.

That is where human judgment remains central.

VAR decisions depend not only on the technology showing an event, but also on how humans perceive and interpret the visual information presented to them.

That distinction becomes particularly important when a decision moves from an objective question (where was the player?), to a subjective one (what did the player's action do to the opponent?).

Did VAR become too focused on whether Fernández touched the ball?

This is perhaps the most interesting explanation for what happened.

Fernández was clearly in an offside position and attempted to play the cross, but he did not make contact.

If the review becomes:

“Did Fernández touch the ball?”

the answer is straightforward: no.

But that is not necessarily the whole offside question.

The relevant issue is whether his action affected an opponent. Fernández was attempting to reach the cross in an area where United defender Lisandro Martínez was also trying to defend.

The key question therefore becomes:

Did Fernández's movement influence Martínez's ability to play the ball?

That is much more subjective.

This may explain why Pro Ref's admission is significant. VAR did not necessarily need to make the final subjective call itself. It should have recognised that there was a subjective question and sent referee Michael Oliver to the monitor.

So, the potential error was not simply “VAR missed an offside.”

It was potentially:

VAR identified the offside player but failed to correctly interpret his involvement and failed to escalate the subjective question to the referee.

The bigger issue: human factors and VAR complexity

The incident raises a broader question about what happens when humans are given increasingly sophisticated technological information.

VAR was introduced to reduce human error. But once a decision goes to VAR, officials are dealing with multiple camera angles, slow motion, tracking information, communication between officials and an intervention threshold.

More information does not automatically produce better decisions.

There is a risk of paralysis by analysis: officials become focused on a technically precise question and lose sight of the wider footballing context.

There could also be anchoring. The assistant referee initially flagged Haaland for offside. Once technology established that Haaland was actually onside, officials had to reset and consider whether another player was responsible for an offside offence. Whether the original decision actually influenced the review is unknown, so this should be treated as a research question rather than an established cause.

A structured review process, like the below, could potentially help.

A simple checklist for a complex offside review

1. Who is in an offside position?
2. Who eventually plays the ball?
3. Did another offside player attempt to play it?
4. Did that action affect an opponent?
5. Is that assessment objective or subjective?
6. Does the referee need to make the final judgment?

The bigger VAR lesson

The Haaland incident shows a fundamental limitation of technology-assisted refereeing:

The technology can tell us where Fernández was.

The difficult question is what his movement meant for Martínez.

That is why the Manchester derby incident is potentially more interesting than another debate about whether VAR should exist.

It raises a broader research question:

When technology gives officials more precise information, do we also need better processes for helping humans interpret that information?

For me, the key lesson is that improving VAR cannot only mean improving cameras, tracking systems or offside lines. We also need to understand human perception, cognitive bias, information overload and decision-making under review conditions.

The technology may reduce one type of error while creating new opportunities for another.

In this case, the most important failure may not have been that VAR couldn't see what happened.

It may have been that the officials didn't correctly interpret what they were seeing.