The Empty Report: When the Football Analysis Room Runs Out of Things to Say

A Tuesday morning in Osaka, I opened the scouting report my assistant had sent overnight. The filename followed the standard convention: nine encoded segments, enough for anyone in the room to know exactly which match, which round, which competition it belonged to. Fourteen kilobytes. I opened it and started reading.

Title: none. Source: none. Type: unclassified. One-sentence summary: empty. Core viewpoints: listed as headings, each one blank. Information points list: not a single line. Entities involved: one instruction remained — "identify from the information points above" — while above there was nothing to identify. Time sensitivity: not assessed. Source quality: no source field existed to assess.

Fourteen kilobytes. Enough to hold the entire skeleton of a report, not enough to hold a single fact. The tactical meeting happened sixty minutes later. And I sat in that room, coffee gone cold, realising I was facing something no coaching course ever teaches: an empty report, placed in front of a coaching staff waiting for an answer.

There are silences in football that are not tactical silence. They are the silence of the system.

Over twelve years in the trade, I have watched the football analysis room transform from pencil-notation notebooks into multi-tiered data pipelines. Brentford, FC Midtjylland, Brighton, Liverpool — each built its own architecture, but all split the work into two independent operational stages.

Stage one is deconstruction. An article, a scouting report, a video clip, a raw statistics table goes in, and the system breaks it into discrete information points — team names, player names, numbers, timestamps, verifiable claims. Without this step, everything downstream is interpretation on a void.

Stage two is deep analysis. It takes stage one's output and runs it through nine dimensions: tactics and technique, club finance and the transfer market, results and the opinion cycle, league landscape and team positioning, rules and compliance, management and dressing room, risk profile, media narrative, and industry transmission.

Nine dimensions. Not nine chapters. Nine doors, each of which opens only when at least one piece of evidence stands behind the handle.

The operating principle of that system is almost cruel in its clarity: when no information points exist, no door opens. And inversely, when the pressure to complete a template exceeds data discipline, the door opens in another direction — toward fiction.

I have seen both scenarios. What made me write this piece, after more than a decade standing on both shores of Vietnam and Japan, is a seemingly innocent remark: that empty report was not an isolated incident. It is a symptom of an ailment the global football industry is spreading at alarming speed.

What actually happens when a data pipeline returns a void?

The fault lies in stage one, not stage two. The domain label was still populated correctly — "football" — meaning the upstream classifier received something football-related. If ingestion had failed entirely, even that label would be empty. Correct label, empty content: that is the signature of a fault at the extraction step, not the collection step.

This is the most dangerous kind of pipeline fault, because it is silent. It doesn't crash, doesn't flash red, doesn't fire an alert. It leaves behind an openable file — with title, with format, with structure — missing only content. In a working environment where speed outranks accuracy, that file slides straight into the meeting room.

There, in front of a coaching staff needing an answer about the weekend's opponent, the pressure to fabricate appears. Not deliberate lying. The pressure is subtler: the pressure of a still-blank form, a skeleton waiting to be filled, a line of "N/A" repeated eighteen times on a page.

In organisational psychology, this is called the completion effect. The human brain is more uncomfortable with gaps than with errors. A blank field creates unease; a wrong number filling it restores a sense of safety. With modern analytical models, this pressure is amplified exponentially, because their ability to generate fluent text is always readier than their ability to say "I don't know".

I have seen this on both shores. In Europe, an analysis room once produced a report on a midfielder the staff had never watched on tape, containing a line-breaking pass metric so beautiful it was hard to believe. It later emerged that the metric had been interpolated from a match the previous season, when the player still operated in a completely different role. Mathematically correct, tactically wrong, and worthless as a decision.

In the J.League, where I live and work, the problem takes a different shape. Many mid-tier Japanese clubs still keep small analysis rooms, but have begun hiring third-party data services to supply automated reports before each round. Those reports are handsome, regular, with diagrams and tables. And in some cases they are written by a system that has never watched a single minute of the analysed team — only stitched raw data patterns from different leagues together.

Every gap needs filling, but not every gap needs filling with a number. Some gaps should only be filled with honesty.

What worries me most is not empty reports in themselves. It is their consequence: when an analysis room accepts processing empty inputs and still outputs conclusions, the quality of the whole system collapses downward. If anyone can say anything about any match, then deep analysis — which demands evidence, reliability, and verifiability — becomes empty ritual.

There are three variants of the problem, all stemming from one blind spot: a process that accepts processing when the input contains nothing.

The first variant is pipeline error. In any two-stage system, stage two depends absolutely on stage one's output. The architecture has a technical rationale: it allows task separation and quality assurance at each tier. But it also creates a rigid dead point — if tier one is silent, tier two has nothing to start from, and if tier two must produce, it produces from nothing.

The second variant is source error. The input exists, but its provenance cannot be verified. A transfer rumour from an anonymous account, a clip cut from a larger context, an article translated from another article no one checked. In the transfer window, this is a plague. Aggregator accounts in Vietnam and Southeast Asia often cite "a European journalist" whom nobody can identify. When a report draws conclusions about a deal from such a source, it does not analyse — it annotates rumour in the language of analysis.

The third variant is interpretation error. The data is complete and sourced, but misread through missing context. This is where expected goals becomes dangerous. The metric is built on a probability model, and that model assumes a certain context of finishing quality, defender positioning, time pressure. When the metric is thrown into a TV debate without context, it becomes a number that cannot be questioned.

xG does not explain match decisions. It describes chance quality within an assumed model. Between those two things is a gap most commentary skips.

I want to tell a small story to clarify that gap. The first match I dissected was Cerezo Osaka 3-1 Kawasaki Frontale in round 14 of the 2026 season. I was nineteen, a second-year student at Osaka University of Health and Sport Sciences. I showed how Cerezo shifted from 4-2-3-1 to 3-4-2-1 in possession, and how Kenyu Sugimoto stretched the opposition back line with lateral movement. The piece was four days late because I kept adjusting the numbers. A youth-team coach read it and invited me to watch a training session.

The lesson from that session was not tactical. It was that every number I wrote could be carried off and challenged by someone who had watched the match ten times. And if I invented a pass, someone would know.

Data discipline does not come from abstract ethics. It comes from the presence of someone who can verify.

The present environment is thinning that presence. A match article is written by an automated system, published on an aggregation platform, translated into five languages, then cited as a source. Nobody in that chain watched the match. And nobody can say exactly which match is being discussed.

This is where my empty report becomes a symbol. It is not a technical accident. It is the inevitable result of an industry operating on the assumption that there is always data to analyse, and that the silence of data cannot happen.

In 2026, at the World Cup in Russia, I was twenty, freelancing for an online football site. In the round of 16, Japan led Belgium 2-0, then lost 2-3. In my analysis, I described how Belgium shifted from 3-4-3 to 3-2-4-1 after the 60th minute, using Marouane Fellaini as a target and exploiting the space behind Japan's two full-backs. The piece, with hand-drawn diagrams, reached 120,000 views in 48 hours.

Some stoppage times teach a whole generation how to lose. That match was one of them. But the lesson I carried away was not about the score. It was that I had drawn every pressure arrow and every pass myself, because I had to watch the footage at least ten times per diagram. No step in that process let me enter a number I had not checked.

In 2026, the pandemic emptied stadiums. I was twenty-two, writing my master's thesis at Osaka University. I collected data from 180 J.League matches in the no-spectator period and compared them with 180 matches from the same clubs the previous season. The result: home teams lost 40% of their pressing advantage in the opposition third, and the home win rate fell from 48% to 41%.

An empty stadium is football's coldest laboratory. In that laboratory, I learned that much of what we call "form" is actually environment. And when the environment disappears, what remains is the real data. But even in that laboratory, I still had to face hundreds of rows of missing data — and the only honest choice was to mark them as missing, not interpolate them into a smooth number.

Another facet of the same problem sits in youth development. When former stars open youth academies, the business model is often built before the training model. Those academies shine in image, roar in publicity, but mostly lack a deep system for training grassroots coaches. Investing in grassroots coaches — the people teaching a nine-year-old how to stand in the right position — is considered less attractive than opening a centre with a famous name on the gate.

That empty report is a miniature of the problem: we invest in the surface coat, not the foundation. An expensive output pipeline is meaningless if the input extraction step is neglected.

This leads to a question I consider central to the current phase. When the transfer window opens and rumour noise drowns out signal, how do you tell a credible report from one that is merely fluent?

My answer is a four-tier filter I apply to every piece of information that passes through my hands. Tier one: who is the original source — a named journalist, an official club, a traceable agent? Tier two: which facts in the report can be independently verified — fee figures, contract length, release clauses? Tier three: where does the publication moment sit in the negotiation cycle — before, during, or after the contract is signed? Tier four: who benefits if this information spreads — the agent, the selling club, or the buying club?

A report passing all four tiers is usable. A report passing only the tier of linguistic fluency belongs back where it came from.

In the transfer market, the fool looks at value; I look at timing. And in data analysis, the fool looks at the number; I look at where the number came from.

Here I want to invert the angle once more. What if that empty report is actually the most honest thing in the entire operational chain?

A report that says "I don't know" — by leaving every field blank instead of filling in guesses — is a report giving accurate information about the state of the system. It warns that the extraction step has failed. It shows the pre-match process has a problem. It forces someone to go back and check the source.

Compared with a complete report containing three invented numbers, that empty report causes far less damage. An invented number flows into a tactical decision, then into a match result, then into a whole week's training lessons. A gap stops at itself.

This leads me to a paradox I believe is characteristic of the current football phase. Football analysis does not fail from a lack of data. It fails because too many models are forced to complete their output even when the input is empty. This is a cultural failure, not a technical one.

The cure is also cultural. There must be someone — in every analysis room, every newsroom, every coaching staff — with the authority to say "no" to an empty report, and the authority to refuse to fill it with anything unverified.

Tactics are the only thing that survives after reflexes stop working. But analysis is the only thing that survives after memory stops interfering. And if it is built on fictional data, it survives only in the meeting room.

I know some readers will say I am making too much of a small technical incident. Data also has a right to be silent, and one broken file says nothing about the quality of an entire industry. That is a reasonable reading, and I keep it beside my own, because I have been wrong in hasty judgements before and know caution is never excessive.

But I have also seen too many reports filled in just to make the meeting. I have seen numbers appear on slides with no traceable source. I have seen tactical conclusions drawn only because someone did not want to leave a field blank.

I return to my empty report. The first thing I should do, instead of calling my assistant into the meeting, is send that file back to the extraction step with an error log. The second is to request re-ingestion of the source. The third is to accept that the meeting can happen without that report.

In every subsequent match cycle, there is one simple thing to try: count the fields filled with evidence, and count the fields filled with fluent language but no source. If the ratio tilts toward the latter, the problem is not in the match. It is in the pipeline.

Every match is a maze; I only redraw the map. But when the map is blank, the first task is not to scribble in roads, but to check whether the compass still works.

And if the compass still works, an empty report is not a full stop. It is the starting point of a more honest process — one in which the greatest value is not the ability to say a lot, but the ability to know exactly when to stay silent.