The Empty Analysis File: Verification Standards from a K League Press Room

At 11 p.m. in Incheon, I opened my inbox and found the match analysis file the data desk had sent over. Empty title. Empty source. Empty list of information points. The only thing still alive in that file was a single label: football. My editor texted one line: "Can we run it?" I answered: no.

A file like that, pushed through a machine, becomes a very smooth article. It will describe a high press, a PPDA that has fallen for three straight matches, a left-back drifting into central midfield whenever his side builds out. The prose will flow, the numbers will look round, and not one detail in it will ever have existed. I have spent long enough at the Incheon Munhak training ground to know one thing: an article with no data is more dangerous than an article with wrong data. Wrong data can be corrected. Fabrication cannot.

Football reporting has changed. Ten years ago, a beat reporter's notebook opened with the kick-off list and closed with what the kit man said in the dressing room. Today every bulletin runs through four stages: source collection, deconstruction, analysis, publication. The K League runs its own match-data system. Newsrooms in Hanoi and Ho Chi Minh City receive data through APIs, through aggregated tables, through pages rendered in JavaScript. When one stage breaks, the whole chain goes blind.

I used to think collection was the easy part. It isn't. The easy part is writing when the data is full. The hard part is staying sober when the data is empty.

That night's file contained no player, no club, no coach, no match. It did not even contain a headline to hold on to. Four causes are plausible: the source page sat behind a paywall; the content was image or video with no text; the article was misrouted by domain; or the deconstruction stage returned a template without executing. The evidence leans toward the last one, because even mandatory fields such as article type and domain were left at their defaults. A system that returns a blank template has not done the work.

The more interesting point sits elsewhere. When a data field is left blank, it does not mean "low risk". It means "not assessable". In a relegation fight those two readings are worlds apart. No evidence of an injury to a first-choice centre-back does not mean he is fit; no information about a financial breach does not mean the books are clean.

An analysis deserves print only when every conclusion is anchored to a document that can be opened and re-checked. On 17 November 2026, Everton were docked 10 points for breaching the Premier League's Profit and Sustainability Rules; on appeal the sanction was reduced to 6. On 18 March 2026, Nottingham Forest were docked 4 points. Those lines are publishable because a written decision sits behind them, with a date and an issuing body. By contrast, "PPDA has fallen for three straight matches" is publishable only if someone can open the raw data table for those exact three matches.

My name was called wrong over the training-ground loudspeaker. Perhaps that is why I always spell every name correctly. The loudspeaker got my name wrong that morning. But the pitch never calls wrong the people who belong to it. An analysis that names nobody also misspells nobody, and that is precisely the problem. It cannot be right and it cannot be wrong. It is unimplicated.

The biggest risk in this trade is not bad data. It is the filling of blanks. An empty file pushed through a language model becomes the most confident article of the day, because a machine does not know how to hesitate. Readers do not reward hesitation either: a piece with twelve statistics is always shared more widely than a piece with three statistics and a note saying they are unverified. That incentive pushes writers toward filling.

I know what it feels like to be stoned for standing against the crowd. Stoned on live television, I still stood by the goalkeeper. Truth needs someone to defend it, not someone to pile on. But standing by a goalkeeper is not the same as inventing a save he never made.

There is a reverse failure, less often named: hiding behind "not enough data" so as never to conclude. A writer who never concludes is a useless writer. The midnight call from Park Yong-woo's mother taught me that football never ends at the whistle. She did not ask me about the fourth goal conceded. She asked whether people were calling her son by his right name. A week later Park scored his first K League goal, and I wrote about it using only what I could verify: the minute, the opponent, the scoreline, the man who assisted.

At 42 I am old enough to know everything changes, and young enough to still believe in one perfect pass. This trade will keep receiving empty files, and there will always be someone willing to fill them with fine prose. The line between a reporter and a text-generating machine is that the reporter knows when to stop and say there is nothing yet. Next time an analysis of the club you follow appears with beautiful numbers attached, who is going to open the raw table and check?