Nine Sections, Zero Data Points: The Silent Failure Inside Football Analytics

THE document ran to nine sections. Full headings, full tables, full comparison frameworks, complete with a risk-warning block and a glossary of professional terms. At every position where a figure belonged — pressing minutes per match, wage structure, squad market-value gap, transfer cash flow — sat the same sentence, stamped identically again and again: “Insufficient information to assess.” On the final line the document declared its own status: complete.

No formatting errors. No syntax errors. Not one exclamation mark. A product so clean that an archive could file it under “processed” and no one would ever need to open it a second time.

I read it at 23:47, Incheon time, after a day spent cross-checking transfer records. And I understood I was holding something more dangerous than a false report. A false report can still be argued with. An empty report that is nonetheless valid leaves nothing left to argue about.

Data does not lie, but the people who write the report do.

Nine blank cells, framed as nine conclusions

Football analytics today sells reports, not conclusions. That is a significant shift over the past decade. When I began cross-checking the accounts of Korean clubs against federation registration records, what I received was usually a stack of smudged photocopies, a signature at the page margin, and a great deal of white space. Today what I receive is a structured file: nine sections, each divided into sub-sections, each sub-section carrying a “conclusion” cell, an “evidence” cell, a “risk” cell.

That structure was built to serve a real need. A club cannot decide to sign a player on the strength of a scout’s gut feeling. A federation cannot sanction anyone on the strength of a rumour. A framework is required: tactics, finance, results, league positioning, regulatory compliance, dressing-room health, risk, media, and industry transmission. Nine dimensions, nine questions, nine occasions on which the analyst is forced to answer with evidence rather than instinct.

The trouble is that the framework exists independently of its content. Anyone can print a perfect template and leave the inside hollow. And once the template looks good enough, very few people check what is inside it.

In the document I read, the tactical dimension contained no formation, no system, no expected-goals figure, no passes allowed per defensive action. The financial dimension contained no broadcast revenue, no commercial revenue, no wage bill, no net debt. The results dimension contained no table, no form, no time marker. The league-positioning dimension contained no league. The compliance dimension contained no governing body and no charge. The dressing-room dimension contained no human name. The risk dimension contained no subject capable of being at risk.

Nine sections. Not one data point.

When the machine knows it is football, but not which football

The detail that held me longest was a single line, small enough to be missed: the only field in the entire document carrying real content was the domain label — football.

The classification layer did its job correctly. It recognised that this document belonged to football. But by the extraction layer — the layer that must pull out team names, player names, competitions, events, figures — nothing was pulled out at all. The machine knew this was football. It did not know which match, which club, which season.

This failure mode is more familiar than people assume. Based on my experience of watching matches, in particular Russia against Spain in Moscow in 2026, I was drawn to an anomalous figure: the host side ran roughly 12 percent more than the tournament average. To verify that figure I had to trace backwards to the GPS units strapped to players’ backs, to the processing software, to the person entering the data. A figure produced by running on a pitch does not generate itself. It passes through at least four layers of people and machines before it reaches print.

If the extraction layer breaks while the formatting layer stays intact, the result is not a blank report. The result is a report that looks finished.

I found a contract buried under three layers of appendix and one layer of silence. That experience taught me something about document systems: the most dangerous place is not where paperwork is missing, but where paperwork is present, correctly placed, correctly templated, and semantically empty. A missing appendix prompts an accountant to phone and ask. An appendix that is templated but contains no figures gets signed and passed on.

When the upper layer collapses, the lower layers absorb it

There is a causal logic in this system that outsiders rarely see. Entities — team names, people, competitions — are generated from information points, not entered in advance. Information points are the root. Entities are the branch.

When the root is empty, the branch must be empty. Those are not two separate faults; they are one fault flowing down the pipeline, and along the way it puts on the appearance of a valid result. The tactical layer is empty because there are no information points. The financial layer is empty because there is no entity to assign money to. The results layer is empty because there is no time marker. The compliance layer is empty because there is no alleged offender. The dressing-room layer is empty because no name has been uttered.

In the risk layer, one line is not blank. It does not speak about a club. It speaks about the process itself: the risk of generating false conclusions from an empty data set.

That is the most important line in the entire document. A system lucid enough to diagnose its own illness, but not empowered to stop itself.

In 2026, cross-checking Busan IPark’s accounts against federation registration records, I found a discrepancy of 2.3 billion won linked to the transfer of striker Kim Hyun-sung, which led in turn to a shell company registered on Jeju Island. The club was forced to explain itself. The tax authority opened an inquiry. What I remember most is not the money but the method of concealment: no deletion, no alteration, merely a push into an agency-fee line that nobody read carefully.

The defensible part of a blank report

Fairness requires one admission: that report did not fabricate.

In a market that pays for conclusions, refusing to conclude is the most expensive and rarest act available. It is always possible to fill a blank cell with something like “the team is improving its pressing” or “the club needs to clear its wage bill” — sentences that are neither wrong nor right, merely unverifiable. A report like that would be applauded, because it gives the reader the sensation of having understood something. A report with nine blank cells gets sent back.

That honesty has genuine technical value. It demonstrates that the analytical framework survives the shock of emptiness: it degrades gracefully, keeps its shape, and states plainly that assessment is impossible, rather than stuffing in a plausible-sounding answer. As a piece of design, that deserves credit.

But that same honesty opens another door.

A line reading “insufficient information to assess” reads like a harmless confession. In practice it is a shield. It protects the producer from all responsibility: we stated clearly that we did not know. It protects the system from all complaint: every position was answered in accordance with procedure. And it protects the report from being reopened, because nobody wants to read a document in which every sentence is identical.

In football this genre has a large family. Minutes of a meeting that contain no minutes. A transfer announcement that contains no fee. An order to stay silent that never needs to be written down.

The pandemic exposed what the image-rights contract was built to conceal: unpaid wages are fact, reputation is a project. In 2026, when the Korean league shut down, I combed through 48 clubs and found a pattern: teams whose chairman simultaneously led a local government tended to assign unpaid wages to undeclared image-consultancy contracts. Seongnam FC was one example, at 4.7 billion won. The point lies in how such sums disappear: they are not deleted, merely moved to a line nobody checks. Exactly like the nine-section document.

Football is not clean, but financial reporting taught me to find the stain line by line.

Nobody checks the inside, only the shell

In every review process there is a question that is skipped almost by default: is this document valid, and is this document complete. Two different questions. A valid document can be entirely empty. A complete document must contain at least one real information point and at least one named entity.

Football has not yet learned to separate the two. When a club files accounts in the correct template, the regulator checks the template. When a scout files a report in the correct framework, the board checks the framework. Nobody counts how many cells actually carry data. And because nobody counts, the number of blank cells can rise cycle after cycle without a sound.

Silence is not a neutral state; it is a verifiable result. A report that is eighty percent blank is a statement. It states that the process broke at some layer, or that someone allowed it to break. Either way, it is evidence.

In 2026 I received an anonymous file describing 8.2 million US dollars flowing from a Qatari construction company to a private account, split into eleven transactions, each exactly one third of a tournament licensing fee. What let me trace that money was not a large sum but the regular repetition of small ones. The anomaly lay in the rhythm. The same applies to a document: what incriminates it is not what it says, but the rhythm of what it does not say.

If the extraction layer of a football analytics system can fail without emitting a signal, then the same fault may be running inside club data rooms, inside scouting systems, inside the medical reports a federation uses to decide whether a player takes the pitch. A test sample recorded with a single wrong digit. A fitness metric typed by hand. A blank line in the notes section that nobody turns to. Hidden transfers are not in the news bulletin; they are in the footnote nobody turns to.

The completeness threshold must come before the validity threshold

I still keep that nine-section file on an encrypted drive, beside everything else I have dug up over twenty-six years. It contains no event about football. It contains an event about the way football tells stories about itself.

A file should be treated as processed only when it contains at least one data point and one name. Without those two things it is a failure, and must be marked as a failure, rather than filed neatly under “done”.

What remains for me is to open each “done” drawer at each club, and count the blank cells inside.