The Empty Report: When Football Is Written by Machines That Never Watched a Match

The Empty Report: When Football Is Written by Machines That Never Watched a Match

Two in the morning in Shenzhen. I open a six-page document and read it from the first line to the last. It has a title. It has nine numbered sections. It has tables, a transmission diagram, a risk matrix, and a glossary of professional terms at the end. Every cell is filled. And every cell says the same thing: insufficient information to assess.

There is no team in it. No player. No scoreline, no date, no shot on goal anywhere. Nine analytical dimensions — tactics, club finance, results and opinion cycles, league landscape, rules and governance, dressing room, risk profile, media narrative, industry transmission — presented in full formal completeness and entirely empty of content.

I stare at it for ten minutes, and the only thing I think about is a match. A team completes seven hundred passes, holds sixty percent possession, and finishes ninety minutes without a single shot on target. On the stat sheet it looks like a football team. On the pitch it is not playing football.

That is why I am writing this.

Context: a document with no facts in it

The file came out of a two-stage analytical pipeline. Stage one extracts from the source article: title, source, genre, one-sentence summary, author stance, information points, entities named, time sensitivity. Stage two takes that output and expands it into nine deep analytical sections.

Stage one returned a fully instantiated template with empty values. No title. No source. Genre marked unclassified. The one-sentence summary left blank. The information-points list empty — and this is the fatal part, because that is the only field capable of carrying football content. The only populated value was a single word: football. No competition, no geography, no code.

Stage two received that empty frame and did what any system designed to always complete the task will do: it filled every remaining cell with a neutral answer. Insufficient information. Cannot assess.

Technically, that is honest behaviour. In media terms, it is a time bomb. A fully formatted document can be read as a conclusion even when it contains none. Nine sections, three tables, one diagram, a risk matrix, a glossary — all of it produces a strong visual signal: this work is finished.

The notable thing is how the default values appear. Unclassified sits where a genre should be. Not assessed sits where a judgement about time sensitivity should be. A default value is not a finding. It is the trace of a check that never happened. But placed in the right cell, in the right font, in the right order, it reads exactly like a finding.

Football has a version of this error and we meet it every week. It is when a defence is called solid after a match in which the opponent did not manage a shot. That defence was never tested. It is not solid; it was never asked a question.

Forty-two pressing sequences and one forgotten gap

In 2026 I left the newsroom to start my own tactical blog in Shenzhen. The first match I dissected was Guangzhou Evergrande against Shanghai SIPG in the Chinese top flight. I spent eight hours drawing forty-two pressing sequences by Evergrande, rebuilding their trap on SIPG's central midfield, and concluding SIPG could not build through the middle.

My conclusion was wrong. I ignored the space behind right-back Wang Shenchao. Hulk exploited exactly that space and scored in the seventy-first minute. Readers called the piece complicated and meaningless — four words I still remember.

I rewatched the footage fourteen times. On the fourteenth viewing I finally saw what should have been obvious from the start: Wu Lei had made a diagonal run that stretched the centre-backs, and that run without the ball created the space for Hulk. After rewriting the piece with six still frames, readership tripled.

When I fell in 2026, I learned that readers do not need me to be right; they need me to be convincing. And conviction does not come from volume of analysis. It comes from locating the one moment everyone else walked past.

That lesson turned out not to be exclusive to humans. My forty-two pressing sequences had exactly the structure of a formally complete document: lots of data, lots of arrows, one conclusion, and one lethal gap nobody checked. The only difference between me then and the machine now is that I was present where I was analysing. It was not.

Six frames and a three-question test

After that fall I set myself a survival rule: never publish a tactical term without a supporting image. That rule produced the six-frame method — the six most important spatial moments of any match, each one required to show who is where, how many metres apart, and what changes immediately afterwards.

The method accidentally became a reading test. Apply it to any analysis, human or machine, and you only have to answer three questions. Does this name specific people? Does it carry specific timestamps? Does it contain at least one spatial description you can picture and verify?

An analysis that fails all three can still be very long, very full of tables, and very confident. It can still make readers nod. It simply contains no information.

Every dead ball is a riddle, and I am only the man reading the pieces on the pitch. But readers have the right to know where the pieces are before they trust the solution.

Twelve point three metres at Luzhniki

In 2026 I paid my own way to Moscow to watch in person. Sitting in the stands for Croatia against Nigeria, I noticed right-back Šime Vrsaljko. Every time Luka Modrić dropped between the centre-backs to receive, Vrsaljko advanced an average of twelve point three metres beyond the back line. Croatia shifted from a 4-2-3-1 out of possession into a 3-4-3 in possession, and they broke Nigeria's press with exactly that geometric trap.

The decisive frame was minute thirty-two: the distance between Nigeria's two central midfielders stretched to twenty-eight metres. The opening goal came moments later.

From the Luzhniki stands I learned that a formation is only paper while the match lives in people. On the data sheet, Croatia were simply a team that passed well. From the stands, Croatia were a team that had found the exact seam their opponent had no time to close.

Had I watched only the replay at home, I would have seen the goal and missed the twenty-eight metres. That distance appears in no statistical table. It exists only in the third frame of a passage I had to count myself, amid the noise of seventy thousand people.

Six point seven and eight point two

In 2026, when competitions shut down, I opened a channel and streamed twice a week with a series decoding Europe's title-winning dynasties. I rebuilt ten finals from 2026 to 2026 and counted every counterattack.

Champions averaged six point seven counterattacks per match, fewer than the runners-up at eight point two. On the surface that is absurd — winners counterattacking less than losers. But the champions' conversion rate was one in five, against one in twelve for the beaten finalists.

A number detached from its situation is decoration; a number attached to its situation is evidence. Six point seven counterattacks does not say the champions attacked less. It says they chose their moments. They let opponents counter more and died less. That is a conclusion about selection, not volume.

A data analyst from Liverpool saw the series and messaged me about the counting method. The channel passed five thousand subscribers in its first month. What I learned was not in the subscriber number. It was that readers came back because they were pulled into a sequence with a beginning, a conflict, and a conclusion — not because they needed another table.

Ninety-one percent and plan B

In 2026 I predicted Italy would win the European Championship, based on the form of left-back Leonardo Spinazzola. My analysis argued that if Spinazzola sustained his touchline-hugging, number-ten style of play, Italy would reach the final.

In the quarter-final, Spinazzola ruptured his Achilles tendon. The online backlash was fierce: I had staked an entire career on one name. But I did not delete the old piece and did not dig in. Within twenty-four hours I analysed the replacement: Emerson Palmieri's advancing and pressing metrics matched Spinazzola's at ninety-one percent over the same minutes. The new piece, Italy Did Not Die With Spinazzola, became the most shared article before the final.

Tactics are not a formula; they are a chess game in which the opponent changes the rules mid-match. An analysis that was correct at publication can be wrong at reading. The only thing that preserves a reader's trust is the speed of admission and the speed of replacement.

My 2026 fall taught me I can be wrong. 2026 taught me I am allowed to be wrong, as long as I do not go quiet after learning it. That is the difference between a person and an empty machine: I can admit error. It can only fill in one more default cell.

When the risk matrix is all cannot-assess

The six-page document had a risk matrix. Six categories: sporting, financial, personnel, rules, public opinion, systemic. All six blank, each with the same explanation: cannot be enumerated because no entity, no fixture list, no squad was supplied.

The document was not wrong. It was misread. Six empty risk categories do not mean six risk categories were checked and cleared. They are an institutionalised false negative: the conclusion that nothing is wrong, when in truth nobody ever went to look.

In my trade the familiar version is a scouting report on a player nobody watched. It has every box: height, weight, preferred foot, appearances, goals. Not one line on how he moves when his team loses the ball. And that missing line is exactly what decides whether a transfer succeeds.

The machine did not create the false negative. It only made it easier to read.

The economics of capsules

This season the consumption of sports content changed again. Readers now read articles, and they also ask machines. For an answer to be cited, it needs structure: a core answer under sixty words, three to five key facts with figures, absolute dates, named sources, and full entity names.

That format is, by accident, the natural enemy of the empty report. You cannot package an information capsule out of a document with no names, no dates, and no numbers. A capsule forces the writer to pick one fact and be accountable for it. It leaves no room to hide behind insufficient information while still counting the job as done.

This is the most hopeful part of the story. Citation pressure is turning sourced attribution from a polite ritual into a technical condition. An answer without an absolute date and a named source removes itself from the game.

The contrarian angle: we built this machine long ago

There is a comfortable reading available to all of us: treat that empty document as a machine failure, a disease of the automation era. That reading is wrong.

Football readers in Vietnam have been fed template-shaped analysis for twenty years. Team A lines up 4-3-3, Team B answers with 4-2-3-1, Team A dominates midfield, Team B plays on the counter. Not one timestamp. Not one distance in metres. Not one player tied to a specific action. Those pieces were full of words, and they were exactly as empty as the six-page document.

The machine merely inherited our habits and executed them at larger scale, faster, and a little more honestly: it wrote down that it had nothing.

More frightening than a fully empty report is one that is ninety percent full, with a single invented number nestled among real ones. Emptiness is spotted immediately. The most dangerous failure is always the silent one — the error that raises no alarm.

And here I have to examine myself. If readers cannot tell a real analysis from a filled-in template, the fault is not with the machine. The fault lies with writers like me, who taught readers that a good tactical piece is a piece with a lot of arrows.

What I will do next round

From the next round on, before sharing any analysis, I will answer my three questions. Does it name specific people? Does it carry specific timestamps? Does it contain one frame I can picture and verify? Fail all three and I will read it as advertising, not as analysis.

The stands have their own language; listen to it before opening the laptop to look at the numbers. And if you write, leave in the piece at least one thing only someone who was there could have seen. That is the only thing the machine has not taken from us yet. The day it does, this profession will have nothing left to say.