The File That Came Filed as Football — Line 3, Tactile Guides, and the Ledger of a Wrong Label

The first page of the notebook carries the date, then the weather, then the drill. That habit dates from 2026, when I covered 42 training sessions and 12 district matches of the Barishal Football Academy U-18 side alone, and the coach let me stand near the tunnel for exactly one reason — I never once published something off the record. That morning I opened the machine the same way. The file in front of me carried a single word on its label: football. I opened it and stopped. I read the station names — Balderas, Juárez, Hidalgo, Mina, Guerrero. There is no club in it. No coach, no formation, no scoreline, no transfer, no registration window, no wage bill. There is a service advisory: several stations on Line 3 of the Mexico City Metrobús will be temporarily closed.

What became clear that morning was not about football. It was about how we do our own work. The label was wrong. And a wrong label, if nobody catches it in time, quietly contaminates an entire data estate.

The notebook remembers the beat before the story does. The condition is simple — the notebook has to remember to check the label, too.

The work described is an accessibility rehabilitation. Two components are explicit: refurbishment of the tactile guide strips that blind and partially sighted travellers use to navigate, and replacement of register covers. The scope is written in numbers — 1,200 linear metres of tactile guide and 114 register covers. The timeline is specific — September to October 2026. To limit disruption for users, the works will be carried out at weekends and in stages. The programme began in August and spans Lines 1, 2 and 3. The source is a named government body: Semovi, Mexico City's mobility secretariat.

Worth noting, the advisory is internally coherent. No claim drops out of the sky, no spending is described without a paper trail, no date is inserted on a guess. In the language of Bengali football reporting — the story is clean. The problem is not in the reporting. The problem is in the address.

So how did a transport advisory enter a football dataset? The answer is technical and irritably ordinary. Automated classifiers work on keywords. This advisory contains the word suspension — there, it means service suspended at a station; in football, the word means a player banned. It contains staged — there, phasing of construction works; in football, phasing of a squad. It contains lines — there, transit lines; in football, divisions or tiers. It contains regional, and it contains rehabilitation, the same word football uses for a recovery from injury.

This collision of vocabulary is not only a language problem, it is a structural one. Which word changes meaning in which context is invisible precisely when the classifier has no cross-check rule for context. A second signal surfaced here as well. The fields that should have been filled in that file — entities involved, primary source — were left empty. One carried an unexecuted instruction, another read that the source was unspecified. In other words, the upstream stage never fully ran.

A wrong label and a wrong claim differ in this: a label can be corrected, while correcting a wrong claim costs trust.

I recognise this trap because I have stood in front of a similar mistake — from the opposite direction. In 2026, alongside covering the Barishal Football Academy senior side, I covered lower-league matches behind closed doors. After six players tested positive, the club entered a 14-day quarantine camp and three matches were postponed. Around us, the rumour market was open. Who was infected where, how many had recovered, how long it would take — all of it guesses. I waited. The official test results came, and then I published the timeline covering the training loads, isolation rooms and return-to-play protocol for 23 players. The club secretary gave me first access to the medical log because in the two years before, I had not added a single sentence of my own. I learned the rhythm of empty seats in that quarantine camp — the gap between presence and absence was the actual story.

Absence is not a void; absence is itself a measurable piece of evidence — if you have the nerve to admit it.

In 2026 I followed all 64 matches of the Qatar World Cup remotely, logging Argentina's 3-3 final, Messi's seven goals, Mbappé's hat-trick. In that same year I broke, from Barishal, that 19-year-old striker Rakib Hossain — six goals in 14 matches — was joining Bashundhara Kings on a one-year loan. Before publishing I checked two club officials and the player's agent, three sources in all, then verified the registration paperwork, the salary cap and the loan fee clause. A loan scoop is a receipt with a deadline attached.

The transfer window is open now. A dozen claims circulate daily and readers want a reliability filter. That work sits exactly here. Source tier, agent motive, the club's wage structure, the remaining length of a contract — put those four side by side and half the rumours fall away on their own. A story without a named source, a story without a contract behind the number, is not news. It is noise.

The analytical framework in front of me was built from nine football dimensions — tactical analysis, club finance and the transfer market, league landscape, rules and governance, management and dressing room, risk profile, media narrative, industry transmission. Every single box was filled with one answer: insufficient information, cannot assess. A less careful analyst could have written nine beautiful paragraphs out of the same empty field — on a three-at-the-back shape, on the sustainability of a wage bill, on dressing-room politics. That would have been the real damage.

The contrarian view sits right here. The conventional assumption is that football data's biggest enemy is missing information. In my experience it is the reverse. Confidently invented information does the most harm. An empty box is visible; a reader can question it, a model can flag it. Attach a wrong label and the error goes invisible — the file looks right, the headline looks right, the date looks right, only the context is wrong. And once the context is wrong, every layer of analysis built on top arrives at the wrong address.

The lesson from empty stadiums applies here too. In 2026 the league returned behind closed doors: cameras present, crowd absent, turnstile counts stopped. That absence itself became testimony — where the market stood, who had drifted away, which club had stopped counting gate money. An empty data field is testimony of the same kind: it tells you, out loud, where the pipeline is weak.

Nobody hid the Semovi advisory, nobody distorted it, and it named its source. That part of the work was correct. But if ten years of watching from the ground teaches one thing, it is this — having information is not the same as being reliable; reliability needs its own ledger. Sources are not quotes; they are coordinates on a long map. The Line 3 file was pinned to the wrong coordinate, and read from the wrong place, Lines 1, 2 and 3 all look like something else. The press box is my metronome; the crowd is the song — and a metronome that slips out of tune makes the song unlistenable.

The question now turns toward the football dataset itself. Before the next transfer window opens, how many files are carrying a label on their cover that has nothing to do with the paper inside?