A 'Football' Tag on a Courtroom: When the Sports News Pipeline Is Not Allowed to Say 'I Don't Know'

Three in the morning in Shenzhen. In the edit room, I was clearing the last batch of files before a broadcast deadline when a single line appeared on the screen: Domain — football.

I opened the file. There was no team. No pitch, no referee, not one player's name. The eighteen information points inside described a criminal case in the United States: a trial declared void because the jury could not reach a verdict, a family destroyed after the deaths of three young children, and an interview due to air on CBS's 60 Minutes.

I sat still for a long while. My trade is listening to silences; I record the air before I record the match. This time the silence was not on the grass. It was in the fact that a single wrong label can make an entire sports news pipeline write about a match that never existed.

Context: the label is an invisible editorial decision

For more than a decade now, every sports newsroom in Vietnam and across the region has run on labels. One round of V.League 1 fixtures generates hundreds of files: video reels, match reports, statistical sheets, wire copy, social clips. Nobody reads all of it. People tag, then route — this file to the tactics desk, that one to transfers, another to the national team desk.

The label looks like a technical formality. In reality it is the most consequential editorial decision of the day, except that nobody signs their name to it.

When a file about a courtroom is routed to the football desk, the football desk will find football in it. Not because the writer is incompetent. Because the writer is paid to find.

I went back and checked the source by hand, and the result was plain as daylight: no xG, no PPDA, no possession share, no transfer fee, no contract, no club, no league, no manager, no player. Nine professional analysis dimensions — tactics and technique, club finance and the transfer market, results and the opinion cycle, league landscape, rules and governance, management and dressing room, risk profile, media narrative, industry transmission — returned one and the same answer: insufficient data to assess.

That is where I want to linger.

Core: when a system dares to say 'I don't know'

In the content industry, 'insufficient data' is the most expensive and rarest answer there is. A pipeline is built to always have an output. It is not built to stay quiet.

In that night's batch there were three breaks in the chain. The first was at intake: a justice-domain file was tagged as sport. The second was at verification: no gate compared the label against real entities inside the file. The third, and the most dangerous, was at production: the pressure to publish, to offer an angle, to deliver 'information gain' by search standards.

Those three breaks together produce a machine capable of generating a perfectly fluent analysis of a match that does not exist.

I can picture that article. It would have a team in defensive crisis. It would have a manager under pressure. It would have a winter signing at a very reasonable fee. It would have pressing intensity declining over the last three matches. All of it smooth. All of it false.

Fluency is the camouflage. A grammatically broken piece is caught in three seconds. A factually false piece with the right rhythm can live on a front page for days.

My trade taught me this fairly early. In 2026, when I was thirty-one and writing for a football site in Shenzhen, I published an analysis of the city derby showing how the coach had placed a Brazilian striker in the 'false nine' role, with seven successful dribbles and five shots from a local forward. The piece went up, and a group of male supporters responded bluntly: what does a woman know about tactics.

I did not argue. I opened a shared viewing session in a small coffee shop and sat listening to the very people who had attacked me explain how they read a match. That evening turned out to be a conversation about feeling, not about numbers. And I understood something: an argument over a piece of sports analysis was never about how much data you had. It was about whether the reader recognised their own match inside your sentences.

A fabricated article fails precisely that test. But it only fails if a real person actually reads it.

After that coffee shop evening I dropped the counting style. I moved to supporter-centred essays, and I set myself a rule: every script starts with a collective emotion, and I spend at least two weeks interviewing before I write a single line of tactical analysis.

A year later I was assigned to the 2026 World Cup in Russia. At Kazan, an Asian side beat the reigning champion and sent them out at the group stage. Standing in the mixed zone, I watched German supporters weep and fail to look away from the pitch. I did not ask them about defeat. I asked a man named Klaus about the first time he ever watched his national team. He told me about the 2026 final and the scarf he had kept since he was ten.

That day I did not record the score. I recorded how a match had touched a man's childhood.

Between two passes there is a world the camera does not capture. That world is where verification happens — and it is exactly where an automated content pipeline never sets foot, because it generates no data to count.

Back to that night's file. What stands out is that the analysis system still did the hardest part correctly: it said 'I don't know'. Eighteen information points, nine dimensions, and not one dimension was papered over with guesswork. The null-handling mechanism worked exactly as designed.

It worked because a human being was reading the file at three in the morning.

And that is the entire problem.

The sports content industry runs on a paradox: the paradox of the novelty requirement. Current search standards demand that every article deliver at least one new insight, one angle nobody has taken. That requirement is reasonable for a human-written piece published after watching a match. It becomes a disaster when applied to a machine that must publish every day regardless of whether there is anything to say. A system forced to always produce new insight will always produce new insight. Even when nothing exists to be understood.

With a criminal case, the consequence is far more serious than a wrong piece about a defence. That file concerned the deaths of three children and a mother whose defence rested on postpartum psychosis. A newsroom treating it as football material has already failed twice: once technically, once ethically. And if somebody had actually written tactical analysis from that file, the failure would be threefold.

The only forward-looking item in the entire document is a legal one: whether prosecutors will seek a retrial, and the outcome of a hearing scheduled for 29 September. That is a justice story, not a pitch story. It needs a human editor, and quite possibly it needs silence.

Contrarian angle: the fault is not in the machine

The first reflex is to blame the algorithm. The sharper reading lies elsewhere: the pipeline was running exactly as designed. It was built never to return 'insufficient data', because that answer brings no revenue, no pageviews, no metric.

The wrong label is not yet the scandal. The scandal is that the wrong label could survive without being stopped at any point along the chain. When an error passes through nine layers of processing, the problem is not the error. The problem is the nine layers.

I would push the counter-argument one step further, in a less comfortable direction. People console themselves that this only happens at small outlets running automated content without a decent editorial desk. My experience in the industry says the opposite. The heaviest volume pressure falls on serious newsrooms — those with daily output quotas, rivals to keep pace with, and a league run-in where every day without a piece is a day lost.

In football, as in film, the scene people remember most is the silent one. But the silent scene never makes it into any script if scripts are judged by page count.

What remains

I kept that file in a separate folder rather than deleting it. I added a line to our internal process: any file tagged as football must resolve at least one concrete football entity — a club, a player, a competition, a contract, a real match. If it cannot, the file leaves the sports desk. If the file touches a real tragedy, it leaves the machine altogether and reaches a human editor, who has the authority to decide that today nothing gets written.

When the lights go out, the match truly begins. And on some nights, the only match being played is the one between the writer and the urge to write at any cost.

The annual season is long, and it will be generous enough to produce real stories: a back line cracking under a congested calendar, a young forward physicalised too early at U18 level, a loan with an obligation to buy quietly chaining a small club in place. Those stories deserve to be told, and they need storytellers who are physically present at the ground.

The one thing I do not want to see again is a match written from a room with no windows.

I do not write about goals. I write about the silence before the roar. And sometimes that silence stretches so far that the correct article is the one that never gets published.