The Perfect Analysis Dossier for a Match That Never Existed

The perfect football analysis document.

I opened the file at seven in the morning, before the first bulletin of the day went on air. Twenty-three pages. Nine chapters. Forty-two tables. A three-tier transmission diagram with arrows. There was a risk warning section, a glossary of professional terms, even a disclaimer. The skeleton was so neatly built that one could print it, bind it in leather, and place it on a conference table.

Across all twenty-three pages, there was not a single match.

No club. No player. Not a single minute of play. Every data cell repeated the same sentence: insufficient information. The expected-goals table was empty. The passes-allowed-per-defensive-action table was empty. The wage bill was empty. The injury risk table was empty. The only thing rated high risk was a technical footnote describing the very process that had produced the document.

A template that informed on itself.

I read it through. Then read it again. By the third pass I realised something more frightening than the emptiness: had I not been paying attention, this document could have gone to air. It had a title. It had a structure. It had the authority of a professional text. The only thing it lacked was football.

And I thought: this is the most worthwhile story in sport right now. Not a hundred-million transfer. Not a derby. But an entire industry learning how to manufacture perfectly hollow shells and label them with two words: analysis.

People look at the league table to see who is leading; I look at the bottom of the table to find who is about to no longer be there. This time I did not look at any table at all. I looked at my own industry's desk, where a machine was printing documents about matches it had never watched.

THE FAITH IN DATA

Modern football analysis runs on an almost religious belief: more data means more understanding. That belief is not wrong. It is only insufficient.

I began writing about Korean football in 2026, when match reports were still typed on machines in a few local newsrooms and the only trustworthy numbers were the scoreline and the scorers list. Eighteen years later, I sit before a screen that can summon a heat map of any player, a passing-network diagram, top sprint speed, distance covered minute by minute, and the expected-goal value of a shot from an eleven-metre angle. I have witnessed a revolution.

That revolution taught a whole generation that football can be measured. It gave us a shared language: xG, xGA, PPDA, progressive passes, model-based transfer values. In K League 1, where I follow things closest, clubs now hire dedicated data specialists, and a few have their own analysis departments with five to seven staff. That is good news.

But something came with it that few mention: once everything can be measured, there is pressure to measure everything. And when that pressure meets a media industry that needs to publish every day, including days when there is nothing to say, the result is a production line.

That line has two stations. The first breaks the source material into information points: who, where, when, which number. The second takes those points and pours them into nine analytical frames: tactics, finance, results, league context, rules, dressing room, risk, media, industry transmission. It sounds entirely reasonable. A frame wide enough to hold almost any football story.

The problem is that the frame does not create content. It only holds a place.

THE MACHINE

Picture the process as a canning factory. Station one is intake. Station two is packaging. If intake receives an empty crate, packaging still runs at full capacity. It still seals the lid, still prints the label, still stamps the date, still stacks the pallet. The finished can looks exactly like a can with fish in it. Only when you open it do you learn there is nothing inside. The label even states plainly: raw material missing.

The document I held that morning was precisely such a pallet of empty cans, and what chilled me was how honestly naive it was. It did not invent a club. It did not attach a fake number to a non-existent player. In every cell it wrote the truth: I do not have the information.

A machine that can say 'I do not know' is far more honest than a machine that fabricates.

But wait. Look at the frame again, because the frame is the real culprit.

When you design a nine-chapter template for every football analysis, you are implicitly declaring that every football story can be divided into nine equal parts. That a story about a young midfielder in the second division and a story about a governance crisis at a big club both need the same frame. That if a chapter is empty, it is the data's fault, not the question's.

That is the moment method becomes ideology.

I have seen this in many places, not just one file. I have seen it in match reports whose summary paragraphs could be swapped between clubs without anyone noticing. I have seen it in transfer opinion pieces where the opening is always a rhetorical question and the closing always an assertion distilled from three identical sources. I have seen it in the newsroom meetings I once sat in, when someone asks: there is no news today, what do we publish? And the answer is always: build the frame first, the content will come.

The content will come. Sometimes the content never comes.

FOUR KINDS OF FAILURE

After spending a morning with that empty document, I began to classify. When a football analysis frame fails, it fails in one of four ways.

The first is the right frame with an empty filling. You have all the chapters, all the subheadings, all the tables, but each chapter contains a single sentence confirming there is nothing to say. This is the most honest kind of failure, and the only one this morning's document committed. It deceives no one. The problem is that it can still be read as a finished product.

The second is the right frame with a padded filling. You have a few real data points, not enough to fill the frame, so you stretch them, repeat them, interpret them in circles, turning a small observation into a grand conclusion. This is the most common failure in the industry. A match with three notable moments becomes a four-thousand-word piece by describing those three moments in twelve different ways.

The third is an empty frame with an invented filling. There is no data, but because the frame demands a section, you make one up. You assign a number to a player you have never watched. You quote a coach without a source. You build a transfer scenario from a rumour of unknown origin. This is the most dangerous failure, because it produces false information that looks exactly like true information.

The fourth is the wrong frame. You have complete data, but you are asking the wrong question. You use nine chapters to analyse a story that needed one. You measure what is easy to measure instead of what needs measuring. You turn a story about a player's loyalty to his city into a wage comparison table. This is the failure nobody calls a failure, because every cell is filled.

The document I held belongs to the first kind. But it is a warning about the second, third and fourth. Because once you have the frame in hand, filling it becomes the goal. And once filling it becomes the goal, the origin of what you use to fill it becomes a minor detail.

That is when our profession is in danger.

KIM JIN-KYU AND FORTY-SEVEN PASSES

I tell an old story, because it is the lesson I still use to check myself whenever I sit down to work.

In 2026, at twenty-eight, I wrote about K League 2, the division nobody wants to read about. In a match between Busan IPark and Seoul E-Land, I noticed a midfielder named Kim Jin-kyu, number sixteen. He did not score. He had no highlight-reel moment. But for ninety minutes, the ball kept passing through his feet and emerging somewhere more dangerous.

I went home, pulled the tapes, and counted by hand. Forty-seven chance-creating passes that season. The highest in the league. A number the league's official statistics never put on its homepage, because the homepage only published goals and cards.

I wrote a piece with a provocative question: why do the big clubs not see Kim Jin-kyu. I called him a kind of forgotten Pirlo of Korea. Several coaches called me a troublemaker. Six months later, Jeonbuk Hyundai Motors signed him for 1.2 million dollars, a record fee for a player from K League 2.

The lesson I drew was not that I was clever. The true value of a piece of analysis lies not in the frame, but in whether you are willing to sit down and count forty-seven passes.

The frame can be written in ten minutes. The forty-seven passes took two days. Every data revolution on earth has failed to change that. Better tools do not make the work smaller. They only make it clearer, and sometimes clearer also means harder to fake.

With only the frame, I would have written a neat analysis of how poor K League 2 is. The frame always leads to the safe conclusion. It never leads to Kim Jin-kyu.

Consensus is where a story goes silent; I choose to stand where the wind blows against me.

CALLING HARRY KANE AN OPPORTUNIST

In June 2026, at the World Cup in Russia, the whole of Korea celebrated a historic two-one win over Germany. In the middle of that joy, I wrote a piece I knew would get me stoned: that Harry Kane was overrated.

What was my basis? The data itself. Kane's five group-stage goals came largely from penalties and from rebounds off other players. His expected-goal value in that phase was only about 2.1, while he scored five. A gap like that means his scoring rate was far above what is sustainable, and it would regress to the mean. That was a verifiable claim, not an opinion for entertainment.

Thousands of English and Korean fans attacked me. The outlet had to append a note that this was a personal view. I called Harry Kane an opportunist, and an entire country wanted me silenced.

Then came the semi-final against Croatia, and Kane went quiet. Many people began messaging me to apologise. I did not celebrate. I hosted a live stream analysing the expected-goal values of the whole tournament, turning the storm of criticism into a class teaching viewers to read football data.

That affair did not kill me; it only sharpened the judgements that followed.

But there is a detail I rarely tell. Before publishing, I had rewatched all five of those goals no fewer than ten times, from four different camera angles. I did not derive the conclusion from an existing table. I built that table from what my eyes saw. Had I done it the other way, had I taken a ready-made frame and poured numbers into it, I would have written a piece identical to every other piece about Kane, and it would have been worth nothing.

A controversial piece is only worth something when it is built on watching football. Built on the frame, it is only noise.

THE VIRTUAL SEASON

In March 2026, football stopped. No matches. No match reports. The whole industry fell into a content crisis. Colleagues at many outlets struggled with the question: what do we write when there is nothing to write?

Six weeks without football, I opened Football Manager and let the whole world keep running inside an old computer.

I simulated the rest of the 2026 K League 1 season using the game's data, publishing an update every day with the question: if the season continues, who wins? My simulation produced a result nobody believed: Ulsan Hyundai, then fourth, would overtake Jeonbuk by exploiting the defensive errors of their rivals in the run-in.

At first people laughed. A journalist playing a game and calling it a prediction. But when the real league returned, Ulsan won exactly as simulated. The site's readership rose three hundred percent in three months, and I was invited as an expert on a major sports channel.

I tell this not to boast. I tell it to make a point: that simulation was only valuable because I poured into it a real season of data, every round, every injury, every card, every run of form. Had I simply opened the game, pressed run, and published the default result, it would have been a con. The machine does not prophesy. It only recalculates what humans have already put in.

The nine chapters in this morning's document are such a machine. It is not wrong. It is merely empty. And an empty machine running a correct process is still an empty machine.

WHAT EACH CHAPTER NEEDS TO LIVE

I sat and listed every chapter in that frame, asking myself: what does this chapter minimally need to have content?

The tactics chapter needs a club, an opponent or a competition, and at least one description of on-pitch behaviour. Without those three, any conclusion about shape or style is invention.

The finance chapter needs a club, a specific transaction or reporting period, and a number. A transfer fee, a wage bill, a sponsorship deal, a net loss. Without a club there is no finance to discuss.

The results chapter needs a competition name, a season stage, points or position, and a form sample of five to ten matches. And to check whether results reflect process, it needs shot data. Luck does not last.

The league context chapter needs a competition, a club, and at least one comparator. Title contenders, continental places, relegation group. Without a map there is no position.

The rules chapter needs a governing body, an alleged breach, a subject, and a procedural stage. Without those four, every sanction scenario is imagination.

The dressing-room chapter needs the names of decision-makers, a coaching-power model, and at least one reported signal about internal relations or contracts. Without people there is no dressing room.

The risk chapter needs an event. An injury, a suspension, a congested schedule, an expiring contract. Without an event there is no risk to measure.

The media chapter needs an outlet name, a headline, a stance, and at least one quantitative expectation signal. Odds, a prediction, a poll. Without a signal there is no expectation gap.

The industry transmission chapter needs a trigger event and two nodes, upstream and downstream. An academy, a club, broadcast rights, derivative markets. Without a trigger there is no chain to draw.

Nine chapters, nine sets of minimum conditions. And the document I held that morning satisfied not one of them. It is not wrong in saying it lacks information. It is wrong in continuing to exist as a complete text, ready to be read as a complete text.

THE SILENT DEATH

This is the part I most want to dwell on, because it applies to the whole industry, not just one faulty file.

Silent death happens when a product fails but looks like a success. There is no explosion. No alarm. No one is sacked. There is only a document pushed to the next stage, then the next, then onto the page, then into the reader's eyes, and the reader believes it because it is neatly formatted.

In football, we are used to checking the honesty of a piece of information through the reputation of its source. A big outlet says it, we believe it. A strange account says it, we doubt it. But we are not yet used to checking the honesty of a process. And when analysis has become an industrial process, the right question is no longer who said it, but how many times the speaker actually watched football.

For three years I have had a habit when reading a football analysis. I look for how many direct observations the author put into it. A specific moment in the eighty-third minute. An off-ball movement the camera never followed. A detail about how a centre-back points and adjusts the defensive line before a corner. Things that are not in any data table, cannot be generated from a template, and cannot be invented without watching.

If a four-thousand-word analysis contains none of that, I file it as an empty can. It may not be wrong. But it has never existed as an act of understanding.

Based on my experience following matches in K League across many seasons, I can say that people who really watch football recognise those who do not within a few paragraphs. They do not need technical jargon. They only need to wait and see whether the author tells them a detail that only someone at the ground would know.

The giant in sports media today is not a relegated club. The sleeping giant is our own capacity for observation, lulled by the false comfort of pre-filled forms. And I say this with new caution: that sleeping giant, seven years after I first used the phrase, has itself become a cliché. I break it with a different question — if the giant is not asleep, but has voluntarily lain down because lying down feels safer, how do we wake it?

WHERE I MIGHT BE WRONG

At this point I must argue against myself, because a claim without self-criticism is only a dressed-up opinion.

I might be wrong to weigh one faulty file so heavily. Perhaps it is only a plumbing failure, not a disease of the industry. A broken filter does not mean the factory makes rubbish.

I might be wrong to undervalue templates. Templates save time. They force writers to answer questions they might otherwise forget. They are a safety net for newcomers. Without a frame, many analyses would be messier, not deeper.

I might be wrong to expect so much of data. I am the one who used data to talk about Kane, and I know data can be misread, distorted, used to prove the opposite. Data is not truth. It is only evidence, and evidence needs someone who can read it.

And I might be wrong to suspect motives. Perhaps that empty document is a praiseworthy failure, a sign that someone in the system refused to fabricate. If so, the real problem is not the emptiness, but that we have no mechanism to say aloud: this piece lacks data, stop.

What I do not think I am wrong about is the diagnosis of the mechanism. When a process is designed always to return a complete text, it will always return a complete text, even when there is nothing to write. The risk lies there, not in any individual.

THRESHOLDS

So what should be done?

I do not believe in vague moral appeals. I believe in specific mechanisms, because my job is reporting, and reporting must be checkable.

First, a hard stopping threshold. If an analysis names not a single club or player, it must not proceed. It sounds obvious, but this morning's document proves that the obvious does not happen automatically.

Second, a clear distinction between two states: a piece with no football content, and a broken extraction process. They lead to entirely different actions. The first is a valid conclusion. The second is an incident to fix.

Third, a mandatory column in every analysis: number of live viewings. Not to boast, but to bind oneself. If that column reads zero, the piece must be downgraded to a brief, or not published.

Fourth, accepting that some days have nothing to say. This is the hardest thing in a media industry driven by reader rhythms. But an outlet brave enough to stay silent when there is no news will be more trusted than one that speaks every day at any cost.

WHAT I WANT TO KNOW

I closed the file near noon. Twenty-three empty pages, formatted more beautifully than many real articles I have read.

I thought of all the people who read about football every day, who have no time to watch the full ninety minutes, who trust what is written because it is carefully presented. They deserve better than a frame filled with nothing.

And I thought of the question I will carry through the coming season, every time I sit before a new analysis. If a piece of football analysis can be produced without watching a single match, then what we are doing — is it analysis, or merely a ritual to fill the gap between two rounds of fixtures?

People look at the league table to see who is leading; I look at the bottom of the table to find who is about to no longer be there. This time, I looked at the middle of the table — where the most beautifully written lines conceal the fact that they have never touched the ball.