A 40-Page Report With Zero Data Points: Football Is Buying False Certainty

July, Shenzhen, two in the morning. A 40-page PDF slides into my inbox titled "Transfer Market Analysis – Pre-Season Phase". I read it from the first page to the last. Section one, tactical analysis: N/A. Section two, club financial structure: N/A. Section four, league landscape: N/A. Section five, regulatory compliance: N/A. Nine sections, six tables, four risk-scenario models, and not a single data point filled in.

The sender was a consultancy I had worked with before. The covering email contained one sentence: "The input material is empty. Any conclusion issued now would be speculation, not analysis." Two other clients had already returned the document. I kept it, printed it, and pinned it to my office wall.

During a transfer window I receive an average of seven documents a week. Six of them carry enough charts, enough arrows, enough metric rankings to make a skimming reader feel informed. The seventh carries nothing but clearly labelled empty fields. That is precisely why it is the most worth reading.

Among thousands of numbers, the truth never needs to be shouted.

The trade of reading empty reports

Seventeen years ago I entered transfer market administration in Shenzhen after a small incident. In 2026 a short video spread with millions of views, claiming that Guangzhou Evergrande won because they ran 120 kilometres per match, because of "spirit". I opened the club's public GPS data and found the real figure: 98.7 kilometres. Their opponents had run 6.3 kilometres more. I wrote a rebuttal. I was scolded. Three months later an analyst at a European betting firm wrote to me, and from that point I had a second profession.

The lesson that year was not about 120 versus 98.7. It was that a wrong number spreads faster than a right one, because the wrong number is told as a story, while the right number is merely presented as a number.

A decade later, working with clubs, I realised the structure was intact — only the casing had changed. People no longer say "fighting spirit". They say "pressing metrics". They say "predictive model". But most of the content is still a story dressed in numbers, not a conclusion born from numbers. The new casing makes the old story harder to argue against, because to argue against it a reader needs the tools to argue against both the number and the story.

The transfer market is a chessboard. Others count pieces. I count moves.

Five checks before believing any analysis

Since receiving that empty report, I have systematised the vetting process I had used quietly for years. It has five layers, and any document — four pages or four hundred — must pass all five before I let it influence a decision.

The first layer is the evidence base. Every conclusion must trace back to a specific data point with a source and a date. If a report says "the team is in good form", the next question must be: good relative to themselves three months ago, good relative to pre-season expectations, or good relative to direct rivals over the same match sample? Those three answers can lead to three opposite conclusions, and all three can be right.

The second layer is sample size. Discussing a team's expected-goals figure after three matchdays is a game of randomness. Discussing it after thirty is a tactical story. The same number, two entirely different horizons of meaning. Many analyses fail not because the metric is wrong, but because it is presented at a horizon where it has not yet acquired meaning.

The third layer is source tier. In a transfer window this is the most neglected layer. One rumour does not carry the same weight as another. Tier one is a journalist with a verified track record and direct access to an agent or sporting director. Tier two is media relaying tier one. Tier three is an aggregation account relaying tier two, usually after two rounds of copying and one round of inference. When these three tiers sit side by side in the same news item, the reader loses the ability to distinguish their degrees of certainty. That is when the transfer market becomes an exchange where the buyer cannot see the labels.

The fourth layer is falsifiability. A conclusion only has value if a condition exists that would make it false. "This player will succeed in the new league" cannot be falsified, and therefore is not analysis. "This player will succeed if he sustains touches inside the box above a defined threshold, and will fail if that threshold falls below a certain level" — that is analysis, because it hands the reader a mechanism to check again in six months.

The fifth layer is motive. Who paid for this document, and what do they gain if I believe it? The agent wants the player's price to rise. The club wants its supporters calm. The betting firm wants money flowing one way. The journalist wants reads. None of those motives is wicked, but none of them is neutral.

That 40-page report passed all five layers without filling a single field. It has no evidence, so nothing to trace. No sample size, so nothing to contest. No sources, so nothing to tier. No conclusions, so nothing to falsify. And it declared its motive in the first line: insufficient data, no conclusion.

The biggest sample-size shock of the year

The summer of 2026 compresses the entire football calendar into one event: the World Cup finals run from 11 June to 19 July 2026 in the United States, Canada and Mexico. For the transfer market, this is a sample-size shock, not a financial one.

Look at the structure. A player enters the tournament with roughly 2,500 to 3,500 club minutes in a season. He leaves it with five to seven more matches. If his team goes deep, that number reaches seven; if his team exits in the group stage, it is three. Against a base of 3,000 minutes, seven matches are noise. On a transfer valuation sheet, those seven matches are frequently priced like a season.

This is why post-World Cup windows consistently produce signings with a higher failure rate than the baseline. Not because scouts are poor. Because they are placed in a situation where the only hot data point is also the smallest sample in the file.

Based on my experience watching matches across several World Cups, the pattern repeats fairly reliably: after each finals, the volume of deals signed within thirty days rises sharply while the quality of information behind those deals falls. That phase rewards speed, not diligence.

A season without crowds exposes every false idol.

The transfer market: where evidence is sold by the kilogram

A transfer window is the only period of the year when the volume of information grows faster than its quality. That is a structural rule, not a complaint about professional ethics.

Follow the flow. A club needs a left-back. That need is instantly converted into a list of six names. Six names generate eighteen news items. Eighteen news items generate three real negotiations and fifteen supposed ones. The player's price does not move throughout. His perceived value moves every day, and by the final day of the window the two figures have drifted far enough apart to be unrelated.

This is where quantitative analysis is most useful, and also where it is most misused. A valuation model built on age curve, minutes played, chance-creation metrics and position will produce a price range. That range is not truth. It is an anchor. Its real value is this: when a report says a club is prepared to pay a specific sum, you can place that figure beside the anchor and ask where the gap comes from. If the gap cannot be explained by age, form or remaining contract length, it is being explained by something else: panic.

Kylian Mbappé's move from Paris Saint-Germain to Real Madrid as a free transfer, announced on 3 June 2026, is the cleanest example of the two figures being pulled apart. A player at the peak of his age curve, out of contract, joining a club that paid no transfer fee. Every valuation model produced an enormous number; the actual accounting entry in the transfer-fee column was zero. The value sat in the wage bill, in signing fees, in contract structure. The clause structure is the real story.

In recent years the phrase "panic premium" has entered the analytical vocabulary. It describes the extra amount a club pays when the window has ten days left and the squad still lacks a position. In accounting terms, the premium is amortised into contract value across the years. In sporting terms, it does not create a single extra touch of the ball.

Sentimental media sells legends. I sell the map of what is true.

One metric, two definitions, two conclusions

There is a layer of checking I have not yet mentioned, and it only appears when you work directly with raw data: the definition layer.

Expected goals, passes-per-defensive-action, big chances — none of these has a single industry-wide definition. Two data providers can assign different values to the same passage of play because they weight shot location, defender pressure and a player's stronger foot differently. The result is that two datasets describing the same match can yield two opposite conclusions.

A reader has no way of noticing this from a chart. Charts always look the same.

This is why a credible analysis declares its data source in the first line, the way an experiment must declare its instruments. When that declaration disappears, you are reading a table of figures with no unit of measurement. It may still be correct. But you have no way to check.

The empty report and the full report: which is more dangerous

I once thought an empty report was the worst outcome a process could produce. I no longer think so.

An empty report declares that it is empty. The reader knows what he is holding. Nothing is hidden, nothing is sold. The ignorance is clearly labelled, and therefore it can be repaired by adding data.

The full report is the dangerous kind. It has a model. It has metrics. It has conclusions printed in bold at the end of each chapter. The only thing it lacks is a declaration of assumptions. Nobody states which dataset the model was calibrated on, how weights were set across different leagues, or how recently injury data was updated. When assumptions are hidden, the model's output looks like objective fact.

This is the blind spot of quantitative analysis itself. A model is only as good as its assumptions, but assumptions do not appear in charts while results do. Readers remember the final number. Nobody remembers that the number depends on a 24-year-old being assumed to keep developing for two more seasons.

There is one more paradox. The transfer window is where the most data is used, and where correlation is most often mistaken for causation. Suppose three of a club's five most successful signings this year were defenders. The hasty conclusion is: defenders are the most efficient position on the market. The better conclusion is: that club had a problem in defence, they bought defenders, and their improvement came from covering a weakness, not from the intrinsic value of the position. One dataset, two readings, and only one survives the following season.

Data cannot replace people. It only makes people's wrong judgements verifiable.

Turning 61 taught me one thing — data outlives reputation.

Signals for the next round

If I could track only one signal for the remainder of this window, I would choose the order in which the fields of the next report get filled in. Not the conclusion. Not the price. The order.

A club that fills in financial data before tactical data is a club selling players. A club that fills in medical data and minutes played before economic data is a club buying players. A club that fills in nothing is a club that has decided nothing, and will decide by instinct in the final ten days.

As for the two clients who rejected the empty report, I do not think they were wrong. They are paying for certainty, and certainty is the most expensive service in this market. The question I want them to carry into next month is not "who will this club sign". It is: what evidence would change my mind?

If the answer is that no evidence would, the problem does not lie with the transfer market.