Nine Empty Cells at the Hamburg Training Ground: The Line Between Analysis and Invention

Monday morning, the car park behind the training ground in Hamburg held nothing unusual: a few cars parked at an angle, a metal door creaking on its hinge, cheap coffee mixing with the smell of wet mud. I was standing there waiting for the session to open when a young colleague came running out with a freshly printed stack in his hands, his face pale.

The stack was a nine-part analysis. All nine were there: tactics and technique, club finance, the transfer market, results and the public-opinion cycle, the league landscape, rules and governance, the dressing room, the risk profile, the industry transmission chain. Every section had its own table, its own bold headings, its own line of assessment. From the first page to the last, every content cell said the same thing: N/A.

One line survived — the domain label: football.

He asked whether I could write a piece out of that stack. I went through it a third time and read the integrity check at the top of the report. Whoever built it had stopped exactly where stopping was required: no club named, no player, no coach, no match, no fee. Rather than filling the blanks with something that sounded reasonable, they recorded a blocked status and asked for the source data to be pulled again.

At my desk, a report like that counts as a failure. It is an honest failure, and in this trade honesty is far harder to keep than fluency.

Let me explain the mechanics, because most readers never see this stage. A football article today passes through three stations: collecting the source text, extracting information into structured fields, then analysing it. The report on that stack broke at the first station — collection returned nothing. Three failures are common: a server blocking access, an empty response body, or a schema mismatch in which the information exists but never lands in the field meant to hold it. The fingerprint here is unmistakable: the domain label survived, the entire content vanished. That kind of break belongs to the pipeline, not to the article.

The story does not stop at a technical fault. European football lives on data to the point where a morning without numbers feels like a power cut. In Hamburg I have followed this club since before everything was quantified. In 2026/18, HSV finished 17th on 31 points and dropped out of the Bundesliga for the first time in the club's history, after more than half a century of unbroken membership since 2026. Since then, every press conference here comes with an indicator sheet: expected goals, PPDA for pressing intensity, possession share, distance covered. Based on my experience following matches, what those sheets never measure is the state of a player at eight in the morning, walking into the training ground before anyone has said hello.

I have spent years in press boxes. The screen in front of me always runs about forty seconds behind the pitch. By the time the striker has finished shooting, the expected-goals figure has only just moved. In those forty seconds I learned something: the eye runs ahead of the data, and data is only useful when it confirms or breaks what the eye has already seen. Read the sheet first and then look at the pitch, and you will see exactly what the sheet wants you to see.

So I want to talk about the value of N/A.

Statistics has a principle called missing-value handling, and there are two ways to do it. The first is to record honestly that the data does not exist. The second is imputation — replacing the blank with a plausible estimate. In research, the second has a legitimate place if you state clearly that you are estimating. In journalism, the second is fabrication, differing only in that the fabricator does not name the act.

A report made entirely of N/A is worth more than a report made entirely of handsome numbers, because it forces the reader to know where they are standing. That is the whole content of the paper in my hand that morning. It said: there is nothing to analyse, do not analyse. And it listed the minimum input — title, source, publication date, at least three information points carrying named entities, one club plus one of: player, coach, competition. Dry, and correct.

I read further and found the structure worth learning from. The tactics section asked about shape, about PPDA, about build-up; without a shape there is no answer. The finance section asked about broadcast revenue, wage bill, net debt; without accounts the table stays blank. The transfer section asked about contract structure and fee against fair value; without a transaction there is nothing to dissect. The opinion section asked about pressure on the manager; without a manager's name, pressure on whom? Nine sections, nine times the same answer, and the answer was right.

My trade is pushed the other way. The content pipeline runs twenty-four hours, every hour has a golden slot, every golden slot needs a piece. When the data does not arrive, the pressure does not disappear; it moves to the person in front of the screen. And that person, without enough spine, will fill. Fill with a plausible fee. Fill with a name currently in the conversation. Fill with a source close to the situation who does not exist.

I almost did it once, and fortunately did not.

In May 2026, after the final match, I stood in the tunnel and watched the players cry. The bulletin already had every number it needed: 31 points, 17th place, a negative goal difference. Everyone knew. What nobody had was the internal reason. I asked the desk for a series called "28 people, 28 stories", interviewing everyone from the captain to the sausage seller at the stadium gate. The sausage seller had worked there thirty years, and he told me which season the crowd left earliest, and in which minute. Nobody had recorded that data. But if you want to know where a club stands in the heart of its city, asking him is faster than reading a table.

That summer we were relegated in the rain, and Russia taught me to love defeat — but Russia came later. What I learned in the Hamburg tunnel was this: once the data has said everything it can, the rest has to be said by people.

Then came Arp. In 2026 I was twenty-three, working as a contributor for HSV's in-house site, covering the youth teams. One afternoon, with the rest of the squad already gone, I saw a seventeen-year-old staying behind to practise free kicks alone. He had just scored 2 goals in 18 Bundesliga appearances. Two goals in eighteen games is a thin indicator, and with only that there is no article at all. I spent three days interviewing kids in St. Pauli, asking them about him. They knew his name the way you know a star's. The piece, "The boy from Hamburg", passed 2,000 shares and got me a full-time job covering the first team.

What I am trying to say is this: new information does not come from adding numbers, it comes from changing where you stand. Two goals in eighteen games is public data; anyone has it. The kids of St. Pauli knowing his name is information you only get by walking out of the stadium.

Three years later, during the pandemic, Bundesliga 2 stopped for more than two months and matches were played in empty stadiums. Goalkeeper Daniel Heuer Fernandes, shirt number 1, told me that the silence itself was keeping him awake. The stats sheet records clean sheets. The man talks about sleep. I proposed a podcast called "The Garage Corner", with players and a sports psychologist talking without a script. The first episode drew 15,000 listens, three times the projection, and the club sent a letter of thanks. In the garage, football does not need a big screen; it whispers through car horns and cigarette smoke.

That same summer I went to Russia, sent to write about Filip Kostić, HSV's Serbian midfielder, who had just been signed by Frankfurt and would leave straight after the tournament. A contract already signed is searchable information. What I brought back was not the paperwork; it was the sight of him still throwing himself into a challenge in the 88th minute of a match that no longer mattered to a club that had already sold him. Transfers live in a player's eyes as he leaves the training ground, not in the number on the contract.

Now the part few people want to hear.

The football analytics industry believes its problem is a shortage of data. Its problem is a surplus of confidence. A full table looks like truth, and the reader has no way to tell a reported fee from an inferred one. By the time the N/A report passes through three editing desks, the third desk sees only a completed table. Nobody goes back to ask who completed it.

The same holds inside clubs. In Hamburg I have seen scouting reports describing players in the language of a model, dense with confident adjectives, when the author had watched that player in three clipped videos. Football does not lack people who can read a table. It lacks people willing to say: I have not watched enough.

There is an occupational paradox worth naming. The demand for information gain on search platforms pushes writers to hunt novelty at any cost. But real added value sometimes comes from subtraction: drop an indicator that measures nothing, drop a claim without a second source, drop a paragraph that exists only for length. A shorter piece that dares to leave blank what it does not know is the more useful piece.

I do not write about football; I write about the people wearing the shirts. And people have no cell marked N/A — they simply fall silent, and silence is information to be recorded, not a blank to be filled.

The young colleague did not write the piece that morning. We called the data team, asked for the source to be pulled again, and lost four more hours. The training session went ahead. Nobody died.

The signal I am waiting for is not the next match. I am waiting for the next empty cell — and to see who is the first to dare leave it alone. From the training ground to the empty stand, I count the club's heartbeat in mud-caked boots, and that heartbeat has never once quickened because I filled in a cell I had no right to fill.

Next time a table reaches you with every cell completed, the question worth asking is not whether the numbers are right. It is who filled the blanks, and whether that person was there.