Defence Is the Most Underpriced Variable at Every World Cup

On December 6, 2026, at Education City Stadium, Yassine Bounou saved two Spain penalties in a row and sent Morocco into the World Cup quarter-finals. Tens of thousands of red-and-green flags went up in the stands. On my screen, the most telling line of data sat at the other end: Spain completed 1,019 passes across 120 minutes and generated 0.9 xG. That figure belonged to the lowest bracket of the tournament.

Four years earlier, aged fourteen, I opened a blank spreadsheet in the middle of the 2026 World Cup summer and logged more than 1,200 shots from all 64 matches by hand, estimating chance quality from shot angle, distance and the number of defenders screening the ball. My first xG spreadsheet taught me this: every goal carries a hidden story. The press that year praised France's attacking flair; my column of numbers showed the champions allowed opponents an average of 0.7 xG per match. The claim was untimely. It was also not wrong.

Since then, before writing a single line, I open the spreadsheet.

PPDA — the number of passes an opponent is allowed before a team makes a defensive action — is the measure I use most. The index does not count tackles, and it does not count clearances. It measures aggression: the lower the PPDA, the more a team forces its opponent to pass quickly and into tight spaces. A team that defends well does not necessarily defend a lot.

At the 2026 World Cup, Walid Regragui's Morocco kept its PPDA inside the tournament's lowest bracket through the group stage and the knockout rounds, despite never exceeding 40 percent possession in the knockouts. When I extracted the data for all 32 teams and published it on my personal Substack in November, the argument was simple: Morocco owned the most aggressive defensive shield in the field, not a concrete block standing still and waiting for the ball. Their defensive block moved with the ball.

Morocco 2026: when defensive data spoke first, the world listened later.

One detail rarely gets mentioned. Morocco conceded mostly from set pieces and from sequences in which the back line had already been pulled out of its shape. An aggressive defensive block carries a structural weakness: it depends on all ten players shifting in sync. One mistimed step and a gap opens. Achraf Hakimi pushing high on the right is part of the design, and part of the risk. Sofyan Amrabat swept behind him to compensate. Yassine Bounou covered the rest.

The core point sits here: defensive data only carries predictive value when it measures structure, not merely outcomes. A team that concedes few goals because its block is good is one case. A team that concedes few goals because its goalkeeper was outstanding three matches running is an entirely different one. The same number, two opposite implications.

The 2026 World Cup expands to 48 teams, which lifts the number of matches and stretches the quality of the opposition. A larger sample sounds good for analysis, but it also dilutes the signal: six group-stage matches between teams of unequal level produce flattering defensive numbers that prove nothing. For anyone patient enough to wait a full season to prove a number right: the knockout rounds remain the only place where defensive data faces real pressure.

On the other side, my own model has failed in memorable ways.

In 2026, when the pandemic forced Europe's top leagues to play in empty stadiums, I collected data from more than 3,000 prior matches and estimated that home teams benefited by an average of 0.38 goals per match from home advantage. The first three rounds after the Bundesliga restart confirmed home win rates falling exactly as predicted. But when I extended that conclusion to other leagues, the results did not hold together. Home advantage did not vanish; it fragmented by league, by schedule, by how comfortable each referee felt.

I do not predict the future through intuition; I only read the traces numbers leave behind. And traces always carry noise.

That is also why I do not treat VAR as a tool that reduces controversy. VAR moves controversy off the pitch and into the review room. A penalty in the 88th minute is now decided by how a video referee interprets the notion of sufficient contact to go down, rather than by whether the on-field referee saw the incident at all. The boundary of argument narrows, but it deepens: it sits inside the grey zone of the law.

Every dataset is a scripture, and I am a slow reader.

What I carry into the next major tournament is not a list of strong teams. It is a question: which team is winning because of structure, and which team is winning because its luck has not yet been named? The same league table, two very different fates. The market usually only sees the table.