VAR and Handball: The Shoulder-Arm Boundary Is Still the Most Costly Blind Spot

On 16 June 2026, in the 55th minute of France against Australia in Kazan, referee Andrés Cunha stopped play, walked to the monitor, and awarded France a penalty after the ball struck the arm of Josh Risdon. I was sitting in a Valencia radio studio with my microphone already open, and I answered as if there were nothing to look up: the ball hit his armpit, there is no offence to call. The colleague beside me corrected me on air: under the 2026 amendment, the armpit region sits inside the definition of the arm. More than four million listeners heard that wrong answer, and the editorial team had to broadcast a correction. In thirty years behind a microphone I had been contradicted live exactly once, and I lost that day because I remembered the law instead of looking it up.

The law does not live in memory. It lives in data.

The law text and the times Article 12 corrected itself

Article 12 of the IFAB Laws of the Game does not speak about the hand the way spectators do. The text speaks about the arm, and the arm is counted from the point where it joins the torso, with the armpit as the upper boundary. The 2026 amendment put that boundary into the wording. The 2026/20 edition added the criterion of the arm making the body unnaturally bigger, and settled that a goal scored directly with the arm, even accidentally, is not awarded. The 2026/22 edition adjusted the description of the body silhouette when the arm is in a running position. Four revisions in six years, each one tighter, and each one leaving behind a group of viewers who learned the wrong version.

Across more than half a century of watching matches from the stands and from the studio, one unhappy pattern holds: handball arguments rarely live in the incident. They live in the version of the law each side is carrying. Spectators apply the edition they saw on television a decade ago, referees apply the current edition, and commentators apply memory.

Technology made the picture harder. VAR arrived at the 2026 World Cup. Semi-automated offside came in from the 2026 World Cup, together with a ball fitted with a sensor that fixes the exact moment of contact. The incident lasts two hundredths of a second. The argument lasts days.

Three variables decide one whistle

The conclusion that can be verified from hundreds of hours of footage is this: a handball decision is governed by three variables, and only one of them is geometry.

The first variable is anatomical. Where the arm begins relative to the shoulder. That is a geometric question with a correct answer.

The second variable is posture. Whether the arm's position is justified by the movement of the body. A player sprinting, turning, falling, jumping to contest a ball. The same arm angle is forgiven in a running stride and punished in a standing block. The criterion of the body being made unnaturally bigger sounds objective, yet it depends entirely on what the referee imagines the underlying movement to be.

The third variable is the origin of the ball. Whether it came from an opponent's shot, from the player's own miscontrol, or from a header half a metre away. The law does not list distance and ball speed as formal criteria, but match management does. In my own records of handballs penalised inside the penalty area, the rate of penalisation rises clearly as the ball travels further, meaning when the player has time to react.

Three variables, one of them measurable. The other two are judgement. That is why the same incident, two experts, two conclusions, and both of them honest.

A camera angle does not show what viewers believe they see

Broadcast cameras sit level with the halfway line, eighteen to twenty metres above the turf. In the far goal area, the parallax between the real shoulder line and the shoulder line on screen can reach several centimetres, occasionally nearly a hand's width when the incident happens at the edge of frame. That is why VAR needs several angles, and why a conclusion built on a single angle does not deserve to be called wrong.

In my own files, the most bitter handball disputes inside the penalty area come from incidents with only two usable angles, both behind the player. Nobody, not even the most experienced referee, can reconstruct an arm position from behind with certainty.

523 matches and 47 seconds

In August 2026, after the shock of 2026, I started a personal project: logging every VAR decision in La Liga and the Champions League with an error code, a timestamp, ball distance and ball speed. By the time the pandemic stopped football in March 2026, I had 523 matches. One finding forced me to rewrite my own outlook: 74 per cent of heavily disputed offside decisions took an average of 47 seconds from freeze-frame to announcement, and most of that time was not spent measuring. It was spent persuading each other.

I published a 48-page report proposing a 30-second cap on each review. The Valencia football federation invited me to advise on process reform. For two years I treated it as the proudest achievement of my life.

One match is only a story. Five hundred matches are the law.

My 30-second proposal has a blind spot

Now the hardest part: a time limit is the right tool for the wrong rule. Offside is a geometric problem with a binary answer. A 30-second ceiling forces the crew to conclude, and the fastest conclusion is always to uphold the on-field decision. For offside that is acceptable, because the line can be published with the frame.

Handball is a different animal. There is no line to publish. Forcing a judgement call into a time ceiling only pushes the referee toward the safest option, and the safest option in football favours the team the stadium protects. When I went back through my own data, handball incidents in the penalty area concluded in under 35 seconds were unevenly distributed between clubs: the group with the largest crowds and heaviest media pressure enjoyed a clearly higher share of favourable outcomes. I do not have a sample large enough to call that systemic. I have a sample large enough to call it a trend, and a trend has to be counted.

Referees do not need protecting. They need to be understood through correct numbers.

The emotional layer the law cannot measure

Spectators judge by the body silhouette in a tenth of a second. Referees judge by the wording of Article 12 after four replays. The two sides use different standards of evidence, and neither is methodologically wrong. What worries me is that when the two standards drift far enough apart, trust in referees collapses before the professional argument is finished.

I once wrote a piece defending referees using a flawed data set. That was the day I betrayed my own principle, and I apologised publicly on my blog after re-checking the original footage. My mistake in 2026 was not forgetting the law. It was being certain.

What should happen next

The game already publishes offside lines to spectators. For handball, the equivalent is entirely feasible: publish the freeze-frame, draw the torso axis, measure the arm angle against that axis, and state which edition of the law applies. Do that, and the argument shifts from whether we trust the referee to whether we understand the criterion.

At 67, I do not need to remember everything. I need to know how to find what is right. And if I had to pick one task for the coming season, I would measure the arm with a ruler rather than measure trust with shouting.