Empty Cells in Football Data: PPDA, xG and the Discipline of Triple Verification

At minute 64 of France against Argentina, Mbappé took the ball on the right flank and went past three players in sky blue and white inside seven seconds. Four minutes later he scored his second goal from almost the same patch of grass behind Argentina's back line. I was seventeen, sitting in front of a television with a squared notebook open at the middle page, and what I wrote down was the position of the men who did not score.

I wrote down where four France players stood in midfield while the ball was still thirty metres away. Deschamps had dropped his block, ceded the pitch, and waited for the single beat that would release Mbappé into that space. France finished the match with 39 per cent of possession and won 4-3.

Had I only owned a four-line data sheet that night — possession, shots, passes, pass completion — I would have written a completely different piece. The sheet would have said Argentina controlled the game. The sheet would have been wrong.

My tactical map was drawn on a France–Argentina night, where two shirts dissolved into one intention. It took me several more years to understand that what I was mapping was the football missing from the data sheet.

Eight years on, the public toolkit has thickened. xG, xA, xGA, PPDA, heat maps and passing networks now sit within reach of anyone with a free evening. In Vietnam, readers have grown used to numbers that once existed only inside a club's analysis room.

Public data has a property few mention: it can only fill the cells that can be filled. PPDA measures how many passes a team allows per defensive action. It describes the intensity of the contest, not the intention behind it. A side with a PPDA of 6.8 may be pressing furiously, or may be so pinned back that retreat is no longer an option. Two opposite scenarios, one line of numbers.

Go back to that night in 2026. France were never passive. They deliberately lowered their block, pulled Argentina forward, and turned the ground behind the opposing defence into goals. The same act of sitting deep can be coercion, or it can be a trap. A data sheet records only the act.

Squad depth sits at the root of it. A team that wants to press for 38 rounds needs two or three options in every position, and those options are paid for in cash. Financial rules such as UEFA's FFP and England's PSR cap spending, which caps the minutes of pressing a club can buy. Transfers are where clubs buy players; coaching staffs buy time. That time, on the pitch, shows up as PPDA.

Four frozen months, and I sat with PSG 57 times to hear them speak through empty space. In March 2026 European football stopped. I split the pitch into 12 zones, counted how often Verratti left the central axis to close down in each match, and rebuilt his movement using two separate video sources.

The answer was not in the average. Verratti pressed most in the first 15 minutes of each half and the last 15 of the match, least in the middle of the second half, when PSG chose to keep the ball and slow the rhythm. Take the match average and he becomes a steady pressing midfielder. On tape he is a midfielder trained to choose his moment. An average describes a different player from the real one.

Every time I find a repeating behavioural pattern in footage, I force myself to check it in at least two other matches, across two different video sources, before calling it a tactical habit. Highlight has value only when it is a sample of evidence, not a pretty clip.

Across those 57 PSG matches, the only thing they never rewatched was their own fear. In 2026-20 PSG pressed extremely high in the group stage and gradually dropped deeper in the knockouts. The numbers show a team changing style. The tape shows a team changing its tolerance for risk, and those are not the same thing.

World Cup 2026 handed me the inverse problem. Morocco built a wall, and I kept the diary of every brick. Regragui set up 4-1-4-1, with the 34-year-old Saïss commanding the back line and an average distance of 28 metres between units. Across the tournament they conceded one goal, an own goal by their own defender against Canada.

The press called it negative defending. I wrote the opposite, and named it a moving wall — a block that travels with the ball, counter-attacks by design, and never collapses onto its own goal. A veteran analyst shared the piece. I declined an interview because I still needed to re-verify the distance measurements before anyone quoted me.

Some called that football ugly. I do not argue about aesthetics. I record that a team with far fewer resources than its opponents moved the match from a technical contest into a contest of time, and that it worked for six consecutive games.

Here the decade's biggest trend surfaces. Gegenpressing was once an asymmetric advantage; now it is the minimum standard. Mid-table teams no longer try to press better than the giants, they try to run more. Football at that level is turning into athletics with a ball, where the key metrics are kilometres and sprints.

And this is the blind spot I see most clearly in my own trade. When a data sheet has empty cells, the writer's reflex is to fill them with a story. A single action becomes a trend, one misplaced pass becomes evidence of a failing press, one win becomes a systemic lesson. Empty cells are uncomfortable, and a story is the most comfortable thing to pour into them.

Market pressure pushes the same way. A decisive conclusion travels further than a line reading "not enough data". But the boundary between hypothesis and conclusion is what separates an analyst from a storyteller. I write "likely" when there is one sample, and "conclusion" when there are three. The conditions that would falsify my hypothesis sit inside the piece: if this team presses 20 metres less than its baseline in the next two matches, my model is wrong.

There is a symmetric trap. Some hide behind "insufficient data" and never commit to a judgement. That caution sounds scientific, but it produces no information. Readers do not need a second data sheet; they need someone who checked the sheet three times and dares to say what he saw, along with the condition under which he is wrong.

Next round I will watch one very specific thing: the first 15 minutes after a team loses the ball in the opponent's final third. That is the window where PPDA spikes, where empty cells appear most often, and where a coach's intention becomes visible. If the team presses immediately, that is a choice. If it falls back into an exact 28-metre block, that is a choice too.

Football does not speak through filled cells. It speaks through the empty ones, and the reader of numbers has to be patient enough not to fill them with his own guesswork.