The Empty Report: The Data Gap Reshaping How Football Clubs Decide

Seven forty on a Monday morning, I open the report file on my second screen. The first line holds a single word: football. The second line, the title field: empty. The third line, the source field: empty. The information field — the only one that actually matters — is a list without a single bullet point.

Fifteen minutes from now, a sporting director will walk into the third-floor meeting room carrying a wage bill, three names and a deadline. He will ask me which holding midfielder to sign in the winter window. In my hands is an empty file tagged with the correct sport.

The temptation does not arrive as a large lie. It arrives as nine tables waiting to be filled, each with column headers already in place: tactical sophistication, financial structure, opinion cycle, risk matrix, dressing-room relations, compliance capacity. A sufficiently skilled writer could fill all nine in forty minutes, and nobody in that meeting would notice that no event stands behind any of them.

I have sat on the other side of that table. In 2026 I presented a forty-seven-page report to the FC Seoul coaching staff. They read the summary page, nodded, and walked out to training. That night I sat alone in the office and understood something it would take me years to dare write as a sentence: nobody pays an expert to hear „I don't know“. But when the data does not exist, that is the only honest sentence available.

The day I realised data does not judge, it only exposes. And in recent seasons, what it exposes most clearly sits inside the very system producing the numbers we trust, not inside the weaknesses of any back line.

The data supply chain and where it breaks

Professional football has built a complete data supply chain over roughly fifteen years. At the front end, stadium camera systems record around twenty-five frames per second for twenty-two players plus the referee and the ball — millions of coordinates across ninety minutes. In the middle, event data providers label roughly three thousand on-ball actions per match: who passed, to whom, from where, under what pressure. At the back end, models convert those labels into xG, xA, PPDA, field tilt and packing rates.

That chain has four joints: collection, extraction, analysis, decision. People usually inspect the first and the fourth. The second — extraction — is where it breaks most often and where fewest people look, because when it breaks the final product still looks intact. A file with a sport tag, column headers and the correct format will pass every automated completeness check. It fails exactly one test: whether any event stands behind it.

I learned this from one specific season. In 2026, as new sports channels exploded, I began a tactical deconstruction project for FC Seoul. I rewatched all thirty-eight matches of that K League Classic season, built a database of the gaps between lines, and logged every action that entered the central corridor. The result: Hwang Sun-hong's side generated an average of 1.7 shots per match from the central zone, the lowest in the league. That was a real finding, and it existed only because every number traced back to an action with a date, a minute and a player.

When I presented the forty-seven-page report, the coaching staff read only the summary page. I sat down that night and compressed the entire analysis into a five-box geometric diagram, each box annotated with touch counts. From then on, every analysis I wrote had to contain at least one spatial model — not for beauty, but because space is the only thing that cannot be invented without being caught.

A tactical system only lives until it meets a larger system. I wrote that about formations, but it holds for data systems too: an analytical process only lives until it meets a larger process — and the larger process here is the head coach's question on a Monday morning.

A formation without a name has no tactics

Tactical analysis requires a subject: a shape, a pressing scheme, a build-up pattern, a set-piece design. Without a team name and a line-up, there is nothing to take apart. Yet an empty table on tactical sophistication still has three rows waiting, and every one of them can be filled with a sentence that sounds highly professional.

A real reading looks entirely different. When a team's PPDA drops below eight, I know they are pressing rather than waiting. When a side plays 4-2-3-1 but the full-backs invert, I can draw the three-plus-one rest-defence structure and identify where it will crack when the opponent switches play from left to right in two passes. All of that demands a name, a line-up, and at least six matches of video.

In June 2026 I travelled to Nizhny Novgorod to watch South Korea lose 0-1 to Sweden. Shin Tae-yong set up in a 3-4-3, and Son Heung-min was completely isolated at the top of the pitch, receiving only nine passes across ninety minutes. After the match I rewatched all six Asian qualifiers and found the problem sat not in the game plan but in the average forty-eight-metre gap between midfield and attack whenever the team had to press. I wrote two hundred pages of notes and published one short piece. That short piece still stands, because it rested on nine passes anyone could count again.

An empty report cannot do that work. It can only produce a sentence like this team tends to press high — a sentence nobody can verify and nobody can refute. In my trade, sentences that cannot be refuted are the most dangerous ones.

Without a transfer fee there is no valuation

A financial structure table has four rows: broadcast revenue, commercial revenue, wage bill, net debt. Without published accounts, a wage bill or an amortisation schedule, all four rows sit empty. And an empty row in a financial table is a far more dangerous invitation than an empty cell in a tactical one, because nobody rechecks an amortisation figure until a league commission does.

A real reading requires a specific transaction. Moisés Caicedo left Independiente del Valle for Brighton for around 4.5 million pounds in 2026, then left Brighton for Chelsea for 115 million pounds in 2026. That curve does not say Brighton buy well. It says Brighton bought a success probability at a price where the downside stayed within tolerance. To conclude that, I need the fee, the contract length, the instalment structure and the sell-on clause.

For the same reason, the deals that have interested me most over the years sit not in the transfer fee column but in signing-on fees for free agents — money flowing through the back door of financial oversight, never appearing in any amortisation line. To point that out, I need specific figures. Without figures, I have only a prejudice.

An empty table and the trap of expectation

Assessing results requires a league position, a form string and a sample size. Missing all three, any judgement about pressure on a manager is a guess wearing the clothes of a report.

There is one check I always run: points against xG. A team that has out-earned its xG over six rounds may be living off a goalkeeper, and goalkeepers do not save forever. But six rounds is the sample size of a coin toss, not of a trend. I have seen enough sides declared in crisis after four matches and finish in the top four to never issue a verdict on a sample under ten games.

An empty file has neither a league position nor a form string. It has one row waiting, labelled public pressure. A hurried writer fills it: under heavy pressure. Those three words take five seconds to write and can cost a manager his job.

A league map cannot be drawn from a label

League analysis requires knowing who is chasing the title, who is chasing European places, who sits mid-table, who is fighting relegation. It requires squad value, financial power, academy output. Without a club name, no tier boundary exists.

The models I admire most all begin by redrawing that map. Brighton built a recruitment system based on valuing players before the market valued them. Brentford abandoned their academy to pour money into data and recruitment. Midtjylland turned a Danish club into a set-piece laboratory. None of the three could have done any of it without knowing exactly which tier they occupied and how the tier above them bought and sold.

A football label says nothing about tiers. It says that somebody in a classification step saw the word football somewhere and applied a tag. That is information about the tagger, not about the league.

Rules do not exist without an accused party

Financial compliance is the field where empty analysis does the most damage, because here conclusions convert directly into decisions worth tens of millions of pounds. UEFA's financial fair play rules and the Premier League's profit and sustainability rules rest on the same mechanism: an allowable loss limit across a cycle, calculated on transfer amortisation and wages.

The 2026-24 season offered two clear examples. Everton were docked ten points for a breach, later reduced to six. Nottingham Forest were docked four. Both rulings turned on how a fee is allocated across years. To assess whether a club carries risk, I need the fee, the contract length and the financial year of recognition.

No transaction, no accused party. No accused party, and every sanction model becomes literature. I have refused to write that literature many times, and every time somebody told me that a report without a conclusion is a useless report.

A dressing room with no names in it

Management and dressing-room analysis is the hardest part to quantify and the most frequently fabricated. To assess it, I need to know who owns the club and how patient they are, who makes recruitment decisions, and which coaching model the head coach belongs to.

The difference between a full-control manager and a coaching-only head coach is not a matter of job title. It determines whose system the club is buying players for. When a full-control manager leaves, the club loses both the squad and the system. When a coaching-only head coach leaves, the club loses one person. The risk profiles are worlds apart, and neither can be assessed if the file contains no human names.

Captaincy structures, dressing-room factions, generational handovers — all of these I only write about after hearing at least three independent sources, because this is the zone where rumour carries more weight than data. A table with three empty rows has no sources at all.

The biggest risk is process risk

This is the part I want the reader to remember.

A standard risk matrix has six rows: sporting, financial, personnel, rules, public opinion, systemic. Against an empty file, all six are unassessable. But there is a seventh row that is always assessable and always rated high: process risk.

The mechanism is concrete. The sport tag survives the failure. Every content field dies. A naive completeness check sees the tag present and marks the record valid. The record then enters the system, is forwarded to the recruitment department, and a buy-or-sell decision is taken on the foundation of a file containing no events.

The most dangerous failure in modern football analysis sits in empty data that looks like full data, not in data that is wrong. Error can be distributed, estimated, placed inside a confidence interval. Emptiness cannot. Emptiness has exactly one correct handling: saying it is empty.

What I fear most is not measurement error, it is model error. A broken model does not raise an alarm. It returns a result, and the result is correctly formatted.

Media narratives do not generate themselves

Every piece of analysis sits inside a larger story: a coronation, a dynasty, a revenge arc, or a critique of money football. To place an event in the right story, I need to know what the original piece said, who wrote it, and when.

The heat cycle of opinion works the same way. A transfer rumour ignites, peaks, then dies. To know which phase it is in, I need a publication date. An empty file has no date.

And there is one test I apply before believing any transfer rumour: source grading. Who broke it first, which tier that outlet occupies, and what the player's agent gains if the story spreads. Those three questions filter most of the noise. Without a named source, there is no tier to grade, and every rumour is worth the same — which is nothing.

Industry transmission

Industry analysis follows a line: academies and talent supply upstream, clubs and competitions in the middle, broadcast rights and derivative markets downstream. An upstream event — an academy graduation, a capital injection, a competition reform — transmits downward with a lag of several seasons.

No node on that line is identified in an empty file, so there is no path to draw. But I want to say one thing about the downstream end, because it is where few people in this trade are willing to look directly: live data sold to betting companies is the darkest consequence of sport's digitisation, and it is the reason extraction speed is priced above extraction accuracy in many places.

When speed is priced above truth, a broken extraction process stops being a technical incident. It becomes a product.

The execution blind spot

There is a paradox in this trade that I have not seen anyone state plainly. Performance evaluation for analysis departments rewards completeness, not honesty. A report with nine filled boxes gets approved. A report with nine empty boxes gets its author questioned about competence.

That incentive structure produces a very specific kind of product: confident prose, standard terminology, and no event underneath. It escapes detection because it is not wrong. It is merely empty.

Gegenpressing was a good idea until mid-table sides used physicality to turn football into athletics, and then it was decoded by long passes over the pressing line. The same thing is now happening to data analysis: mid-table sides use metric volume to turn analysis into reporting, and it will be decoded by exactly one question — which action does this number trace back to.

The most valuable analyst at a club is the one who dares submit an empty file with a single line of explanation: extraction failed, rerun before anyone decides.

What to verify next matchweek

The next time a club announces a data-driven decision — a new contract, a system change, an academy promotion — the question worth asking sits at the model's input, not at the model's quality.

In an empty stadium, I hear the breath of a defender and the cracking of a tactical plan. Outside the stadium, in a room with nine blank tables, I hear something else: the sound of a system waiting for someone to fill it with confidence. The question for next season is whether anyone in that meeting room has the nerve to say the file has told us nothing yet.