The Empty Template: When Every Box in a Football Report Read N/A

The file opened just after one in the afternoon. The header was complete — domain: football. Below it, nine analytical sections, thirty-three sub-tables, and in every substantive cell the same three characters: N/A. Title N/A. Source N/A. Article type unclassified. Summary blank. Information points, an empty list. No club, no player, no coach, no competition, no date, no number. The page existed; nothing on it did.

The easiest thing to do with a file like that is the most dangerous: fill it. Nine empty templates and a competent language model are all it takes for tomorrow's report to acquire a familiar club, a plausible scoreline, a minute for the goal and a sentence beginning "sources indicate." The only hard information left in this file was that there was no information in it.

I keep records nobody asks for. In 2026, on a university placement in Kolkata, I was credentialed as a runner at the FIFA U-17 World Cup and assigned crowd colour. Six to twenty-eight October, fifty-two matches. I did not count goals. I counted cards, and I counted handball calls — referees from three confederations, one Law 12, three visibly different readings of it. The same notebook has the final: England 5–2 Spain at Salt Lake before 66,684. I stopped writing "controversial decision" that month and started writing "Law 12, deliberate play, and the competition's disciplinary annex."

A year later, Russia 2026: sixty-four matches, one spreadsheet — every review, every overturned call, the record twenty-nine penalties, and the first VAR-awarded penalty in a World Cup final, Perišić's handball, France 4–2 Croatia, Luzhniki, 15 July 2026. That file paid me for the first time. The lesson was protocol first, verdict second.

In 2026, when the stadiums shut, I did what I do with anxiety: I read. The ISL ran entirely behind closed doors across three Goa venues and I covered it from Liverpool off broadcast feeds, roughly 140 match reports at two o'clock kickoffs. Between games I annotated the league handbook and the AIFF disciplinary code — 312 pages. I read them because the rulings I could not see were the only part of the bubble nobody was reporting. That is the desk this file landed on.

The pipeline runs in two stages. Stage one extracts the article: title, source, date, information points, entities. Stage two analyses those points across nine dimensions. This time stage one succeeded structurally and failed completely substantively. An empty payload that still validates is more dangerous than an obviously broken one — a broken payload sets off an alarm; an empty one travels quietly.

Any football article of publishable length yields at least one of four things: a named entity, a competition, a date, or a figure. This payload had none of the four. That is the signature of a body-extraction blocker — a paywall, a consent wall, a JavaScript-rendered page, an empty response. Note that classification succeeded: the domain label came back correctly as football. Which means classification ran on metadata, a URL or a feed category, while the extraction of content stopped dead. One document, two stages, two different outcomes.

The second finding matters more. When the article's source field is itself N/A, the source cannot even be graded — and that is worse than a bad grade, because a bad grade lets a reader discount the claims. An absent grade lets the reader neither discount nor trust them. That is a second-order failure: a poor source is normally at least identifiable as poor.

The sharpest trap sat in the financial-compliance and risk rows. Every cell read N/A. N/A does not mean "no issue found." It is the same error football makes every week: a disciplinary committee closes a case, the wire copy says "cleared," and the file says "insufficient evidence." Those are not the same finding. An unassessable risk rating is not a low risk rating, and an unassessable compliance check is not compliance.

There was another layer. Six risk buckets — sporting, financial, personnel, regulatory, public opinion, systemic — all came back null. But one risk could be asserted with high confidence, and it was a process risk rather than a football one: any decision taken on this payload would be a decision taken on zero evidence. Alongside it, the time-sensitivity field was marked "not assessed." So even if the content were later recovered, nothing would establish whether it described yesterday's news or news from three weeks ago. Transfer stories decay in days, injury stories in hours. A payload without a timestamp is a scoresheet with no date on it.

All four information-value ratings — sporting, industry, timeliness, reference — sat at zero. And the minimum list assembled for a re-run reads like any decent disciplinary file: title, named publication, publication date, article type, at least five discrete information points, and the entities derived from them. Miss one of those and the question of analysis does not arise.

My own ledger offers the comparison. In 2026, Salt Lake recorded 131,781, and that remains the last number Indian football can actually verify. A figure that cannot be verified is not a figure; it is an estimate. This file contained no figure at all, so verification never became the question — only whether we would choose to believe something.

Here is the uncomfortable part. Writing "analysis withheld" on this file is easy, and it is faintly comforting. But refusing is not courage; it is a filing system working correctly. An industry that ran a closed franchise league for the better part of a decade without building the structure beneath it has limited standing to deliver a moral lecture about empty templates. I went looking for a villain and found a filing system instead.

The real difficulty is not that machines can lie. It is that no newsroom keeps a ledger of the cells that stayed empty. The filled row earns the credit; the N/A earns nothing. Editors ask for the full page, not the incomplete file. The pressure to fill an empty template is cultural, not technical — and it sits less on the reporter than on the process that hands the reporter a count every morning.

The projection is straightforward, on one condition. The pipeline needs a minimum-content gate — a character threshold and at least one named entity — that hard-fails stage one instead of emitting an empty-but-valid payload. Publication date and source quality should be required fields, in the way a clause number is required in a disciplinary code. That change would show up within a single pipeline cycle. Asking who erred this time is beside the point. The question is whether, the next time the scoresheet arrives blank, we will have the nerve to call it blank — or whether we will quietly type in a familiar club's name.