When a Football Report Has Everything Except Facts: An Audit of the Sourceless Analysis

At about one in the morning a file opened on the screen. The top line read: Domain — Football. Below it sat nine sections: tactical and technical, club finance and transfer market, results and public-opinion cycle, league landscape, rules and governance, management and dressing room, risk profile, media narrative, industry transmission. Every section was filled in. Every cell carried the same entry — N/A. Title N/A. Source N/A. One-sentence summary empty. Information points: zero. Entities: zero. Roughly two thousand words of document containing not a single fact that could be attached to a match, a club or a player.

A document like that looks impressive at first pass. It has tables, ratings, confidence markers, and a glossary at the end defining xG, PPDA, FFP and PSR. A reader skimming the surface files it as a finished report and moves on. A reader who stops at the second line notices that all nine sections arrive at the same verdict — insufficient information, cannot assess.

I pulled every clip before I said a single word, and this file is the photographic negative of that habit. In 2026, at a Bangalore school final, I cut 42 clips off a borrowed camera because a 78th-minute penalty appeal had produced two different stories from two different people. The one-page timeline carried a minute and second against every incident. The district referee coordinator used it in the weekly review, and that was the first time I understood that an incident is worth what its frame is worth, not what its echo is worth.

In 2026 I turned that single page into a 64-match VAR database for the Russia World Cup — 1,247 clips, 38 reusable review templates. Antoine Griezmann's 38th-minute penalty in the final, after the ball struck Ivan Perisic's hand, exists there as one row: incident, camera angle, law, threshold, outcome. The database said 64 matches. The narrative said something else, and the gap always appeared in the same place — wherever a claim had no timecode behind it.

During the 2026-21 ISL bio-bubble in Goa I logged 115 matches remotely and added an audio-cue column for whistles and shouts, tagging 1,046 set-piece delays and 78 on-field consultations. Empty stadiums settle one argument permanently: when the crowd is gone, the record is all you have. When the record is also empty, you have nothing.

The first thing broken in this file is not in the nine sections. It is in the top line. The domain label reads Football while the content reads zero. That order is inverted. A document is not football because its metadata says so; it is football because of the formations, fouls, scorelines or contracts inside it. When the label arrives before the parsing, this is the result — a report that knows its subject and not its event.

Refereeing has a name for this. When an assistant referee commits to a position before the ball's path commits, we do not call it a bad decision. We call it bad positioning. The decision may still be right or wrong, but the positioning error repeats for ninety minutes. A metadata-first pipeline is that same positioning error, applied to a document instead of a clip. It parks the entire analysis at an angle from which the incident is never visible.

The most expensive absence in this file is the source. Title N/A, publisher N/A, byline N/A, publication date unavailable. The media-narrative section calls this the single most consequential omission, and that line is correct — because source is not a decorative field. Source is the filter.

VAR calls the equivalent threshold clear and obvious error. An on-field decision is overturned only when the available evidence visibly contradicts it; "I think" does not cross that line. Journalism needs the same bar. A claim from an unnamed publisher cannot cross the clear-and-obvious threshold, however cleanly it is written. And when the article title itself is unavailable, the claim cannot be graded at all — not authoritative, not general, not low-quality.

That is why Griezmann's 38th minute is not simply a penalty in my notes. It is an anchor. Date, match, law citation and camera angle are welded to it. If someone tells me tomorrow the ball never touched the hand, I have frames to show. This file contains no anchor of that grade. Every claim floats, and a floating claim stays afloat five years later for exactly as long as someone repeats it.

The third failure is the quietest and the most dangerous. The document does not look incomplete. It has nine sections, a risk matrix, a transmission diagram, confidence ratings, and a glossary that reads professional. A reader who passes through it carries away an analysis that never existed.

Refereeing has a plain example. It is the match report written from the fourth official's board. The board shows the correct time and the correct number, so the report's skeleton looks clean. But whoever did not watch the play cannot tell you how the goal was built. Structure and evidence are two different objects, and making them look alike is the easiest mistake in this trade.

One point needs stating plainly, because the argument usually turns emotional here. N/A is not a defeat. "No clear error found" is a decision, and it is as legitimate as an overturned one. When 42 clips yield 14 handball and offside incidents and some of them support no conclusion, the correct report says exactly that — these frames do not decide it. A report that extracts a verdict from every frame loses its credibility within a season.

Which forces me to question my own habit. I build my own video queue, and the clips that surface fastest enter my story first. That is clip-selection bias, and it is no less dangerous than the empty file. The difference is that the empty file does not lie. It stays silent. I do not manage to stay silent with my queue. So when someone tells me the referee ruined the game, my first question is whether the full match is logged or only the part that trended. I do not argue with the crowd. I argue with the timecode.

The most useful part of this file is not what it says. It is what it refused to do. It declined to invent a story out of absent information. This pipeline failed safely rather than silently. And the central problem in football media is that most failures there are silent — in fluent prose, in confident idiom, with no timecode attached.

There is a counter-case, and it is uncomfortable. The market wants clean sentences. An editor wants 800 words by noon. Nobody commissions a piece headlined "insufficient information, cannot assess." That pressure is the root of the entire process that turns empty input into full output. The incentive is not technical. It is commercial. And when the incentive is commercial, technical fixes alone do not hold.

The objection follows that a null result given this much space is waste. The waste is in the document that is confident and unverifiable. The broadcast cut away. That is usually where the story was. Analysis begins where the source appears — not before.

So where does the repair start? Source becomes a hard precondition, not optional metadata. Every claim carries its origin — outlet, date, link. The domain label gets assigned after content parsing, not before. And every report ships with an audit trail, so a reader can tell whether what they read was extracted or invented.

I have a small precedent I keep repeating. After Mumbai City's 2-1 final win over ATK Mohun Bagan in the 2026-21 bubble, I built a 12-page crisis-review template for broadcasters covering disputed calls — incident, available audio, missing angles, likely referee communication. One producer wanted speculation. I kept the template evidence-only. Neither of us produced a better story that night. Only one of us produced a story that could be checked.

The lesson coming out of this is not new, and it comes straight from officiating: when there is no frame, there is no decision. That is discipline, not weakness. The people who argue daily about Indian refereeing standards usually skip the first step of the argument, which is evidence — clip, timestamp, law, outcome. A report built in that order can stand up to audit whether it goes for the referee or against him. Everything else is a day's echo.

The question is not how many analyses were published this week. The question is how many of them survive a Stage-1 re-run.