The Illusion of Possession: Why England Lost to Croatia Despite Holding the Ball

In the 109th minute at Luzhniki Stadium in Moscow, Mario Mandzukic pounced on a loose ball inside the penalty area and sent Croatia into the first World Cup final in the country's history. Seven minutes before that moment, England were still the side holding more of the ball. Across the second half and both periods of extra time, they kept the edge in possession. They left the tournament with a familiar line: we played well, we were just unlucky.

I sat down after that match, opened my notebook, and counted the passes back one by one. It was the summer of 2026, when I was a first-year economics student in Shanghai. I had started hand-recording three metrics for every game: possession share, passes into the final third, and touches inside the opponent's box. What I found forced me to rewrite the way I read football. That piece drew only 37 views, but the moment itself permanently changed how I watch a match.

Context: one notebook and three columns

The semi-final of 11 July 2026 between England and Croatia in Moscow was not a shock once you looked at the game state. Gareth Southgate built his side with three centre-backs and two wing-backs, prioritising possession and set pieces. Zlatko Dalic placed Luka Modric, Ivan Rakitic and Marcelo Brozovic in midfield — a trio capable of dictating the tempo at the very highest level.

To understand why this match mattered, you have to look at the path of both teams. England scraped through the group, beat Colombia in the round of 16 on penalties — their first ever World Cup shootout win — and edged Sweden in the quarter-final through a set piece. Croatia went through two consecutive shootouts against Denmark and Russia, each dragging into extra time. Many expected Croatia to be exhausted against England. Reality moved the other way.

My tracking never relies on feeling. I divide the pitch into zones, log where every pass lands, and add it all up after the final whistle. It is slow and tiring work, but it answers a question the scoreboard never can: what is a team holding the ball for?

The raw picture looked positive for England. They had more of the ball, passed more, and created a fair number of dangerous situations. Stop there and the conclusion becomes England controlled the game but were unlucky. That is the rosy verdict, and it ignores most of the truth.

Before the details, a word on method, because this is a principle I have kept for years. For every number, I verify against at least two independent sources. When I cross-checked my handwritten figures against event-level data from an international statistics provider after the tournament, the two sets matched on almost every category; only a few minor discrepancies appeared in pass classification. If two sources diverge too widely, I choose neither and go back to the footage to count myself. That habit came from a time I nearly published a wrong conclusion because I trusted a single statistics table.

Core: where the ball goes, not how much

I broke the possession number down by zone. In the first half, most of England's ball circulated between midfield and their own half, among the three centre-backs and two central midfielders. The ball moved sideways far more than it moved forward. When they needed to break through, England pushed the ball wide to the wing-backs, but Croatia had prepared for exactly that. They covered the flanks and let England pass sideways, as long as the ball never entered the central channel.

The second column in my notebook tells the real story. Counting passes driven straight into the central channel — the area in front of the box, where passes can open up chances — Croatia recorded nearly double England's figure, roughly 12 to 6. Croatia passed less, but each pass carried more intent. In modern football this is what analysts call the quality of actions in the final third.

In the second half, Dalic adjusted. Croatia pushed Modric higher, dropped Rakitic to orchestrate, and let Brozovic sweep in front of the back line. They intensified pressure immediately after losing the ball, denying England time to build up. Croatia's pressing metric — the number of passes an opponent is allowed before being engaged, known to analysts as PPDA — dropped sharply compared with the first half. As a result, England's attacking sequences became shorter and less imaginative.

What stands out is how Croatia divided roles in midfield. Modric made the difference in the space between England's midfield and defence. Rakitic set the rhythm, adjusting the speed of the pass. Brozovic provided cover, reading counter-attacks before they began. The trio did not need to hold the ball a lot; they needed to hold it at the right moments.

Ivan Perisic equalised in the 68th minute with a finish after a cross from the left. That goal did not come from luck. Throughout the second half, Croatia repeatedly probed the channel between England's full-back and centre-back, where space opened every time a wing-back pushed high. It was the direct consequence of England holding the ball while leaving gaps behind.

As the match reached extra time, the fitness issue became obvious. Croatia's midfield trio kept running, kept passing, kept controlling the rhythm. England shifted to long balls towards Harry Kane, the tell of a side that had run out of short-play ideas. Mandzukic finished it in the 109th minute, precisely at the point where the more organised side, rather than the side with more possession, tends to win.

One detail I always remember from that tournament: most of England's goals came from set pieces. It reflects a side coached meticulously on dead-ball situations, and that was a genuine strength. It was also an admission that they lacked a way to score from open play at knock-out level. When an opponent defends set pieces well, England run short of weapons.

Using expected goals — xG, a measure of chance quality based on location and shot situation — the picture sharpens. Croatia created fewer chances by volume but of higher quality, especially in the second half and extra time. England's shots came from distance and from set-piece headers, mostly outside high-danger zones. The difference lay not in the number of chances but in the value of each one.

Contrarian: correlation is not causation

The easiest conclusion to reach after this match is that possession is useless. I do not write that, and I think anyone who does has misread the data. Possession is not useless; it becomes meaningless when detached from zone and intent. A team with 70% of the ball that only passes sideways in its own half generates less value than a team with 45% that keeps driving the ball into dangerous areas.

Here lies a classic statistical trap: correlation is not causation. A low possession share did not make Croatia win, and a high share did not make England lose. What decided it was the quality of actions in the final 30 metres. I was once wrong to believe the possession figure could predict results. It predicts only a tiny fraction, and that fraction is often blurred by match context.

Variance deserves a mention too. A single match is a small sample. Croatia winning does not mean every low-possession team wins. Replay those 100 matches with the same two teams and the same game state, and the results differ. Variance is not the enemy — it is a mirror held up to the arrogance of prediction. Data shows us tendencies, not destiny. One season is a statistical sample; a decade is evidence.

I also would not claim Modric and his teammates won purely because of superior tactics. They won partly because they had already come through several tense knock-out matches, and that experience produced composure once the game reached extra time. Behavioural data can capture that, but it resists reduction to a single number. Data does not lie, but it learns to hide the most important thing.

Variance warning

We must separate a team's true ability from the result observed in one match. Croatia were stronger in central midfield, and that is fairly certain. Their 2-1 win after extra time was the outcome of a sequence of probabilistic events, not an absolute verdict on the two teams. Had Perisic not equalised in the 68th minute, the story could have been entirely different. I place roughly 70% confidence on the claim that Croatia controlled central midfield better, and only about 55% on the claim that they deserved to advance more across the full 120 minutes. The remainder is the space for what data cannot measure.

Takeaway

The England–Croatia match of 2026 taught me something I still use in every report: do not measure control with a single figure, measure it with a map. Control is the ability to decide where the ball is played, not how many times it is touched. For the major tournaments ahead, I will keep tracking the same three metrics, but this time I add one axis: the quality of passes into the final 30 metres. Fans remember the goals; I remember the probability before the goals happened.