Nine Dimensions of Analysis, One Empty Data Layer

Minute 83, Yanmar Nagai Stadium, summer of 2026. Hiroaki Okuno, shirt number 25, left foot, collected the ball at the edge of the penalty area and curled it into the far corner. I was standing in the media section, my left hand holding a recorder, and what I kept on tape was the sound of 25,000 people rising to their feet in one single beat. An hour later, the hashtag #Okuno passed 12,000 posts. That night I wrote without inventing a single detail: the 83rd minute, the player's name, the shirt number, the stadium capacity, the speed at which public reaction travelled. Data first, story second.

In more than thirty years in this trade, I have never once seen that order reversed. The annual season, however, always tests a writer in its own particular way.

The annual season is a news-printing machine. Every matchday, dozens of games, hundreds of incidents, thousands of lines of statistics are pushed out within hours of the final whistle. In Japan, where I follow a team, newsrooms race minute by minute. In Vietnam, where I still read the papers every morning over a hurriedly brewed coffee in an Osaka apartment, the pace is even faster: a match ends at 7pm and by 9pm there is already a three-thousand-word tactical breakdown.

I understand that pressure, because I once lived inside it. Last week, a nine-dimension analytical document was placed in front of me. The source summary section was blank. The timing field read: not assessed. The source-quality field read: cannot be assessed. And yet directly beneath, the assessment section had been written out in full, all nine parts, each with its own heading, each with its own conclusion. A nine-storey building raised on an empty foundation.

I call this the empty data layer. Every piece of sports reporting has three layers. At the bottom is the primary data: who, when, where, how many, who confirmed it. The middle layer is interpretation. The top layer is conclusion. When the bottom layer is empty, the two layers above can still be written very fluently, because language does not need a source in order to sound reasonable. It only needs correct grammar and the right tone.

In the early hours of 3 July 2026, I organised a shared viewing of Japan against Belgium in a small bar in Osaka. Japan led 2-0. In the 69th minute, Jan Vertonghen pulled one back. In the 74th, Marouane Fellaini equalised. In the 90th plus fourth, Nacer Chadli sealed a 3-2 win for Belgium. Three names, three timestamps, one result.

The next morning, a great many analyses blamed the Japan defence. Those pieces had data: minutes, player names, the scoreline. But their primary data layer stopped there. None of those writers stayed until three in the morning with roughly fifty people in that bar. None of them recorded the sentences that caught in people's throats between songs. I did. And what I heard did not match the conclusion that "the defence made individual errors". People do not remember the score; they remember why they did not leave the bar.

That is why I always carry a recorder, even when I am only going to watch football. Based on my experience of following matches, a claim only stands when it can answer four things at once: who said it, when they said it, where they said it, and where the evidence sits. Lose one of the four and the statement can still be printed, can still be shared five thousand times, but it has left the territory of journalism. It becomes a prediction wearing the clothes of an event.

Take an example closer to the annual season. Suppose that over the last three matches a team's pressing index has dropped sharply, the players are running less, the midfield is sitting deeper. That is a signal, and the signal is worth reporting. But if I do not know who measured that index, in which match, and by what criteria, then the signal is not yet enough to be called tactics. It is an ornament pinned to the top of an article to create a feeling of expertise. Readers cannot verify it, and because they cannot verify it, they believe it.

I still remember an afternoon at the training ground. The squad finished, the press had almost all gone home. I stayed behind, purely out of habit. About twenty minutes later, a young player came back out alone, placed six balls along the edge of the box and struck them until it went dark. The next day, no newspaper carried that story, because it appeared in no press release. Three weeks later, he came off the bench and scored the decisive goal. By then I already had my primary data layer: one afternoon, six balls, alone. I keep the rhythm of the story, because I have watched it fall and rise many times.

The shouting of that night still rings in me, long after Osaka has gone to sleep.

People in this trade tend to believe that modern sports journalism is a race of collection. More data, more depth. The blind spot lies elsewhere, and it is far more discreet: what most current analyses lack is the provenance of their data, not the quantity of it. A nine-dimension analysis does not fail because it lacks a tenth dimension. It fails because not one of its dimensions rests on a verifiable fact.

Here there is a paradox that forces me to question my own profession. Fear of losing the community's goodwill makes writers swap a line reading "there is no data yet" for a line reading "it could be". Readers are happier with the second. But it is precisely the second line that leads them to misunderstand their own team over the long run. The most honest article on some days is rarely the best article. It is a short one, saying that the primary data layer is still empty, that nobody has confirmed anything, that there is nothing yet to conclude.

A report with no primary data layer can still be correct in its wording. It simply has nothing to correct when it is wrong, because it has nothing to check against. An empty stadium does not make a match smaller; it only makes the memory larger. A piece of writing without sources moves in the opposite direction: it warps the memory, and nobody ever finds out.

At my age, people do not count trophies; they count the nights when a whole city stayed awake. Next matchday, before you trust some conclusion about the team you love, try to find the primary data layer sitting beneath it. If that layer is empty, what you are reading is a prediction written in the voice of something that already happened.