Zero Matches, Zero Football: The File That Reached My Desk With the Wrong Label

Hook

A file sat on my desk on Wednesday night, clearly labelled: Domain — football. Seventeen information points, one core viewpoint, one label. My job should have been simple. Roll the tape, pause, measure the half-space, draw the passing network, write. On Wednesday I pressed play again and again and got the same thing every time: there is no picture on this tape.

Not one of the seventeen points is football. No club, no player, no coach, no league table, no fixture, no scoreline — not even a wrong scoreline. What was there: Guadalajara, Jalisco, Mexico. A residential neighbourhood called Jardines del Sauz. A complaint filed on 24 September. Security-camera footage circulating on social networks. A wave of condemnation and calls for state accountability. And, as of the information timestamped 25 September, no reported arrest.

I have been inside football for thirty-four years. Radio commentator in 2026, sports magazine editor from 2026. In 2026 I spent fourteen hours re-watching Mumbai City's tape and mapped how Bengaluru's 4-3-3 pinned their back three, how Sunil Chhetri drifted into the left half-space to create a 3v2. I went back to the Mumbai tape, and the half-space was hiding in plain sight.

This file is not that tape. This file is the tape with no tape inside it.

Context: what a wrong label actually does

When I lost my freelance contracts in 2026, and the stadiums emptied, I started reading transfer fees as tactical screams. A fee stopped being gossip and became a question the pitch has to answer. That habit produced a first reflex with any raw material: what game is actually being played here? The question looks harmless. It is the foundation of everything downstream. Ask the wrong question and your analysis can be as precise as you like — the output is noise.

The first stage of an information pipeline extracts information points, a core viewpoint, and a domain label. Everything after that — indexing, entity extraction, sentiment modelling, linking — stands on that label. Get it wrong and everything below is wrong, but the error stops being visible. It is a camera on the wrong frame: your measurements are correct, the object is not.

This file was labelled football. Inside was a Mexican crime and public-safety report. The only bridge between the two is geographic: a city is named, and that city hosts professional clubs. That is all.

There is a specific false connection waiting to be made here — the mistake of reading geographic proximity as substantive proximity. Guadalajara is a city where people live, work, file complaints and go to court. It is also a football city. This article contains no football city. A city being named in an article does not make that article about the city's football club. That sounds trivial. It is not.

Core analysis: count first, explain later

In Moscow in 2026 I was live-writing Spain against Russia. Spain completed 1,005 passes; Russia 202. The scoreboard said Spain controlled the match. Spain made 1,005 passes, so I counted the ones Russia wanted them to make. Russia's 5-3-2 low block handed Spain the passes Russia was willing to concede — to the free foot, outside the pocket, away from the line of progression. That was where I learned the difference between passes made and passes permitted. I now apply the same method to everything: count what exists, then explain why.

On this file, the count returns zero. Football entities: zero. Tactical systems or formations: zero. Match context: zero. xG, PPDA, possession, set-piece design, referee tolerance, substitutions: zero. Zero means information absent, and information absent does not mean the gap must be filled with inference. The professional act is to leave the space empty and write down why it is empty.

That refusal is the analysis. Because the temptation is always within reach. Guadalajara is in the headline, and two historic Liga MX clubs play there — Chivas and Atlas FC. Join the dots and you get a punchy football opinion: club culture, discipline, sponsorship, cards, boycotts. All of it available, all of it invented.

The whiteboard gave me a shape; the tape gave me the truth between the lines. Here there is a whiteboard and no tape. Filling in the shape would not be analysis; it would be colouring inside someone else's outline.

Civil geography versus football geography

I work with spatial information, but in football spatial information means something specific. When I say a team shut the left half-space, left means the channel outside the right-footed defender's shoulder, the zone a flat three-man midfield will not cover. When an article says Jardines del Sauz, space means streets, houses, a police station, neighbours. Both are spatial words with different units. Reading one in the other's language looks creative and behaves like a wrong decision.

The geography in this file matters to a criminal investigation — which area, whose jurisdiction, where the evidence sits. Its value to football analysis is nothing. And something subtle happens here: zero value gets mistaken for analytical negativity, and the writer feels that not writing means falling behind. That is not analysis. That is content-production pressure.

The nine-dimension null result

I normally walk nine dimensions in a tactical piece. Here is how they looked, without numbers, because there are no numbers.

First, tactical and technical. Nobody played. No system, formation, style or coaching decision is described anywhere. There is no room for passivity because there is no subject.

Second, club finance and transfers. No figure, contract, wage bill or transaction exists. No entity exists against which FFP or PSR exposure could be computed.

Third, results and the public-opinion cycle. No table, form curve or points. Sample size: zero matches. No pressure on a manager or players, because there are none. The pressure that exists is civic — demands on authorities.

Fourth, league landscape and positioning. No league, tier or club to compare resources with.

Fifth, rules and governance. No FIFA, UEFA, CONCACAF, Liga MX or federation rule is engaged by the reported facts. The applicable framework is criminal and family law.

Sixth, management and dressing room. No ownership, sporting director, coach, squad or contract situation.

Seventh, risk profile. The football risk matrix is empty. The real risks here are data-governance risks inside the pipeline: mislabelling, unattributed sourcing, and a privacy exposure.

Eighth, media narrative and expectation — the one live dimension, dealt with below.

Ninth, industry transmission. There is no transmission path into academies, agents, broadcast, capital or betting, because no football node exists in the source.

The only live dimension: narrative velocity

One of the nine dimensions has analysable substance. Here I stop doing football and read information flow, where I can actually measure something.

The structure is familiar: an alleged act of personal violence caught on camera, made public, followed by demands for state accountability. A visible artefact exists — the security footage. A formal step exists — the filed complaint. The second carries more weight than the first, because the footage's authenticity, context and continuity are unverified in the raw material; we are told only that it is circulating.

The gap between informational heat and informational foundation is wide. I call this heat/fundamentals divergence. Intense emotion, demands and condemnation on social media, sitting on a thin base in which most points carry the notation 'Source: None'. The core claims stand on ground that has not been surveyed.

I want to be precise here, because misreading is easy. Weak sourcing does not mean the allegation is false. It means the chain of verification has not yet been built. In football I make this distinction daily. When a transfer rumour arrives I neither assume it true nor call it false; I look at the source tier. Tier one: official club announcement. Tier two: multiple reliable journalists reporting together. Tier three: social-media primary source, later followed by unspecified media. This file sits at tier three. Tier three is a judgement about verifiability, not about truth.

The expectation gap

Three gaps are visible between expectation and process.

Institutional response. Online reaction wants rapid arrest and protection. The recorded information as of 25 September reports no arrest, only that authorities are reviewing the matter to determine responsibilities and corresponding protection measures. Public expectation is more optimistic than procedural reality.

Evidentiary certainty. There is a tendency to treat footage as conclusive. It is one element among several requiring review. Public expectation is ahead of the legal standard.

Victim safety. Protection is framed as something authorities must determine, not an immediate guaranteed outcome. The resolution timeline is undervalued by the public.

Source tier and an ethical gap

Working the raw material I found a problem that has nothing to do with football and everything to do with information discipline. It is serious.

In gender-based violence cases, victims are normally anonymised; that is standard practice. The raw material reproduces the full name of the alleged victim, and does so via a chain in which 'media that followed the case' identified her — an identification chain, not official confirmation. The name does no analytical work. There is no reason for it to be in an analysis. In this piece the person is referred to as the complainant, without a name.

I treat this as tracking discipline, not squeamishness. When I built a set-piece model from 306 empty-stadium matches in 2026, the discipline was arithmetic. Now it is harder: which information is evidence, which is allegation, and which is simply a human being. The name belongs to the last category.

One more: the raw material states that these episodes have occurred for years. That is testimony, not established fact. The language here will be: according to the complainant's account. Small distinction, but it is the line between analysis and repetition.

Contrarian angle: the professional act here is the decision not to write

The normal machinery of content exerts pressure: a subject has arrived, a label has been attached, accept the label and ship the piece. Find the club, find the culture, build the broader context. Traffic will come. Doing that does not produce analysis. It produces an attempt to make a wrong label true.

If geographic proximity were a valid rule for entity linking, any crime anywhere would become an analysis of that city's football club. That rule serves nobody. Manufacturing a hypothetical link between a club and a private criminal allegation creates harm far larger than any analytical gain. Inventing a connection that did not occur, because content needs one, is a question of information integrity, not taste.

The genuine professional use of this file is defensive: it is a negative control. In a laboratory we deliberately feed an unsuitable sample into a pipeline to check whether the system rejects it correctly. This file does exactly that job. The label says football; the contents contain none. If the pipeline swallows this cleanly as football, there is a detectable defect. If it flags or rejects it, the system works.

Three real defects in the machinery

Domain labelling. If stage one has no mandatory confidence field — how strongly do I believe this label — there is no way to catch a mislabel. A low-confidence label and a high-confidence wrong label look identical. This is cheap to fix and returns the most.

Source specificity. Where core claims carry 'Source: None', that is not mere carelessness; it is a required field left empty at extraction. Until a point is attributed, it cannot be indexed as fact. It must stay tagged as allegation.

Privacy handling. The name of a private individual, especially an alleged victim of gender-based violence, should not enter an analytical pipeline. It does no analytical work and it causes harm. The question is not only ethical but archival: once stored downstream, the name is very hard to remove.

Takeaway

Two questions now open every new file on my desk. What game is actually being played here? And if the label is wrong, who catches it?

On this specific story I will watch four things. One, whether the Jalisco prosecutor's office takes a formal step — arrest, charges, or a protection order. Two, whether a recognised outlet independently authenticates the footage; authentication moves the item from allegation to corroborated. Three, whether any football-affiliated person is linked to the case — only then does this become legitimately football-relevant, and only then do safeguarding frameworks become engaged. Four, whether the complainant's name has been stored downstream anywhere, which would require remediation.

At fifty I have watched tactical analysis lose ground to punchy opinion. In football it happens through false praise of possession. Here it happens through geographic proximity. Same disease: pretending to measure what was never measurable.

My next piece will be about football. But I am keeping this file open, in the right corner of the desk. On the day my own checking procedure goes slack, I will open it again: seventeen points, one wrong label, and one decision — not to write.

The integrity of tactical analysis lies not in the refinement of its inferences but in the courage of its refusals. Where there is no pitch, I do not draw formations.