The Handball Law and the Armpit Boundary: What 523 Matches Say About Contentious Decisions

On June 16, 2026, at Kazan Arena, referee Andrés Cunha walked back from the monitor area and pointed to the penalty spot in the 58th minute of France against Australia. The ball had struck the arm of Josh Risdon inside the box. Antoine Griezmann converted, opening the scoring, and it became the first penalty in World Cup history awarded through VAR. Seven minutes later, at the other end, Samuel Umtiti handled the ball and Mile Jedinak equalised from the spot. In the 80th minute Paul Pogba settled it at 2-1. Three goals, two penalties, one match torn into countless arguments that ran for months.

As for me, in Valencia, nearly two thousand kilometres from Kazan, I sat in a radio studio with an open microphone and told listeners that the decision for France was wrong, because the ball had struck Risdon's armpit. I spoke in the confident tone of a man who had followed football law for more than forty years. Forty seconds later, my editor leaned in and whispered: the law revised for the 2026/19 season had placed the armpit inside the arm. I sat silent for the rest of the match. The station had to read a correction after the final whistle.

What I lost that night was the right to be believed.

From that night I built a law reference updated year by year, each clause tied to its publication date and its date of effect. No analysis of mine is allowed to cite a law without a version note. The law does not live in memory; it lives in data. I wrote that on the cover of my first notebook, and it is still there.

To understand how a contact at the armpit can bring down an entire broadcast, you have to go back to where the notion of deliberate began. For decades, IFAB's Law 12 carried one short sentence: a player commits an offence when they handle the ball deliberately. The key word was deliberate. A referee must read intent inside a moment lasting less than two hundredths of a second, while the stands roar, both sets of players crowd in, and sweat runs into the eyes. There was no definition of arm position. No diagram of a boundary. No standard that turned subjective judgement into something checkable. The result was that every season the same incident was called three different ways in three different penalty areas, and nobody was technically wrong.

IFAB began to amend. From the 2026/17 season, the Laws introduced for the first time an illustrative boundary between shoulder and arm, fixing the start of the arm below the shoulder. By 2026/19 that threshold was stated more sharply: the dividing marker is the bottom of the armpit. That is precisely the sentence I did not have in my head on the evening of June 16, 2026, in Valencia. Next, the 2026/20 season added a rule on goals: accidental handling in the immediate build-up to a goal was penalised. The 2026/21 season shifted the focus to the standard of an unnatural arm position, where the arm makes the body bigger. The 2026/22 season narrowed the build-up rule, covering only the goalscorer. And by 2026/24 one principle was clarified: a ball rebounding off a player's own head or body onto the arm is generally not an offence.

Seven amendments in seven seasons. A referee working in European competitions may have to apply three different versions of the law across three matches in the same month, depending on the competition and the date. That is why I never answer a question about the law on the spot.

So readers can check for themselves, here are the specific markers. The 2026/17 Laws were approved at IFAB's annual meeting in Cardiff in March 2026 and took effect on June 1, 2026. The 2026/19 edition took effect on June 1, 2026, two weeks before France against Australia in Kazan. Risdon's contact fell exactly in the transition between two versions, and that is why a man who had taught the law for thirty years could speak wrongly on air without hesitation. That note sits on the fourth line of my reference sheet. Without it, I would make the mistake again.

In August 2026 I started a project many colleagues considered eccentric. I logged every VAR intervention in La Liga and the Champions League, one line per incident: the error code from my own classification, the minute of play, the distance from the referee to the monitor, the seconds between the signal and the announcement of the decision, and for offside calls, the ball speed and the gap between hip and shoulder of the player flagged. By March 2026, when European football stopped for the pandemic, I held 523 matches. I spent six weeks recounting every line. I have said that being 67 does not give me the memory of a 27-year-old, but it does give me the habit of checking everything twice, and this dataset is how I find what is right.

The first result forced me to rewrite the opening of my report three times. Among offside decisions contested by at least one of the parties, 74 percent fell into the group announced later than 30 seconds, with an average of 47 seconds from the referee's signal to the final conclusion. In other words, the controversy is not created by the outcome of the incident. It is created by the waiting. The stands cannot see the line. They see a man pressing a hand to his ear, and then they see a stadium holding its breath. Forty-seven seconds is more than enough for trust to evaporate.

The second group is handball. Across 523 matches I logged 118 handball incidents inside the penalty area referred to VAR. Only 31 of them led to a penalty after review. In 44 incidents the referee kept the original decision not to award. In 43 incidents the referee kept the original decision to award. The striking detail lies elsewhere: in those 118 incidents, the arm of the defending player sat in what I call the grey band, higher than the bottom of the armpit but lower than the shoulder line, on 61 occasions. More than half of the most contested incidents of an entire season concentrated on a strip less than twenty centimetres tall on the human body. A line that does not exist on skin, only in a document.

At this point the data forced me to look at what I least wanted to write about.

Within the same 523 matches I separated incidents involving players returning from long injuries. I counted how often referees penalised the team of such a player, and how often they did not penalise when that player was the one fouled. In this group, the rate at which referees awarded decisions in favour of the returning player was markedly lower than that same player's own average before the injury. The plainest reading of that number is this: a player just back from injury usually has to absorb more contact before hearing a whistle. The media calls it a player having to prove he deserves to return. But the data is not about deserving. It is about the risk of re-injury, and about the fact that referees, like everyone in the stands, are shaped by the story told around a player.

And that is where I have to address the asymmetry that television debates usually skip.

Stands pressure is not distributed evenly. I measured the average time from the VAR signal to the announcement, split by whether the home side belongs to the highest or lowest revenue group in its league. The gap is not large in absolute terms, but it exists and points consistently: in matches with a full stadium and a big home club, the average review took longer, and the rate at which the original decision stood was also higher. I found no evidence of any secret instruction. I found something else, more frightening because it requires no one to give an order: a referee with fifty thousand people screaming at his back processes information differently from a referee in a half-empty stadium. This is psychology, and it can be counted.

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

Once that dataset existed, my proposal was simple. Each review should last a maximum of 30 seconds before the referee must either change or stand by the on-field decision. Not because technology cannot be faster, but because waiting is what destroys trust. In April 2026 I published a 48-page report on my personal blog, with a spreadsheet attached so anyone could recount every line themselves. The Valencia football federation then invited me to advise a group drafting a procedure. I attended three meetings, presented exactly what I had counted, and said nothing more about what I had not counted.

But stopping there would have meant writing a technical report rather than an article.

The hard part lies elsewhere. Every television debate about handball opens with the same sentence: if I were on the pitch, this is how I would see it. It is an emotional question, and it always has an answer ready. Viewers watch the replay in slow motion, from four angles, with nobody touching their back, and they have the right to freeze on the least favourable frame. Referees have no such right. That is precisely why the clear and obvious standard for VAR intervention is so hard to apply. An incident can be clear to a television viewer and ambiguous to a man standing fifteen metres away. When VAR arrived, people hoped it would erase the grey zone. In practice it only moved the grey zone from the pitch into a sealed room, where nobody can see the line, and therefore every disagreement is pinned on one person.

I once wrote a long article defending a referee using wrong numbers. I discovered the error after it had aired. I published a public apology, took the piece down, and reposted the raw data so readers could check it themselves. That time I understood that defending someone with wrong data is worse than silence.

And there is one more thing the data shows, far less comfortable than any procedural fix.

The armpit boundary is a line drawn by a technical committee. Its virtue is that it turns a subjective judgement into a measurable marker. Its flaw is that nobody on a football pitch has ever thought that way. No defender holds an arm below the bottom of the armpit on purpose. No referee in a lower division, after a day in another job, can watch a contact at real speed and determine precisely which point on a moving twenty-centimetre strip the ball struck. We tried to reduce uncertainty by drawing more lines, and the more lines we drew, the more players were penalised for standing half a foot wrong. The price of precision is absurdity.

Yet saying a law is absurd does not mean the referee was wrong. These are two different things, and I see them mixed together almost every matchday. Referees apply the text in force. If the text in force produces a bizarre outcome, the responsibility lies with those who drafted it, not with the man holding the whistle in the 94th minute in the rain.

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

My conclusion after 523 matches is not that VAR should be scrapped. Nor is it that VAR needs more power. The conclusion is humble and very hard to carry out: whenever a law cannot be applied consistently by an ordinary referee, on an ordinary pitch, on an ordinary evening, it is a badly written law, and it must be rewritten. Data will not do that for us. Data only shows us where to fix things, and it is useful only if we are willing to read it.

I said one wrong sentence on June 16, 2026, and lost something that took me forty years to build. Years later, I still open the notebook with that line on its cover every time I have to answer a question. When I counted every incident, I understood that the law does not judge anyone. It only waits to be applied correctly. If a twenty-centimetre line can bring down the broadcast of a man who studied the law for four decades, what happens to a referee who has only three seconds and a pair of eyes, while the rest of us have been given forty-seven?