Empty Analysis: When the Football Data Industry Sells Belief Without Evidence

I opened the report on an October evening in Barcelona. Forty pages. Nine sections. Every section had tables, bolded headings, and clearly marked confidence notes. But by the third page, a chill ran down my spine: every data cell was empty. Every conclusion read "insufficient information." Every evidence entry pointed to an empty set. It was the most perfect analysis I had ever read, perfect about something that did not exist. Forty-four years in this trade taught me something no classroom ever did. The most dangerous enemy of a sports journalist is not false information. False information can be argued with, exposed, cross-checked line by line until the truth is pulled out. The most dangerous enemy is empty information presented as if it were full. A report with tight structure, technical terminology, and the outward appearance of science, yet nothing inside to back it up. This kind of product is more dangerous than a rumour, because everyone knows to doubt a rumour, while an empty report wears the coat of precision. I call it an "empty analysis" — or, in more technical terms, a "null return." The system ran, it returned a result, but the result contained nothing. And the terrifying part is that most readers lack the tools to tell it apart from a real analysis. Modern football runs on data. Every Champions League match generates millions of data points. Every transfer window is a maze of contracts, annexes, sell-on clauses, and performance bonuses. Clubs hire analysis teams, media outlets hire data teams, and online platforms spring up like mushrooms to translate dry numbers into language fans can swallow. That boom created a market so hungry for analysis that it will pay for anything that looks like analysis. When demand outstrips supply, the market does what every market does: it fills the gap with counterfeit goods. An empty report still sells. A piece full of nothing but "insufficient information" can still make the front page, as long as it is formatted well enough. I have watched this mechanism operate from the inside. In 2026, when I uncovered Real Betis's suspicious contract with a Brazilian winger, I did not have enough evidence to conclude anything immediately. The instinct of a young newsroom would be to push the story with dramatic speculation, filling the gaps with imagination. I chose the opposite: build the file, cross-check test results over three years, log every abnormal rise in hematocrit, and wait in silence. By 2026, when that player was banned for two years over erythropoietin, my file was thick enough to stand. Betis hid the doping inside a contract annex; I read backwards page by page to find it. Had I chosen to sell an empty analysis to the newsroom back then, I could have made the front page two years earlier. And I could have been wrong. By 2026, at fifty-one, I was forced to learn how to read social-media data. Girona sold a twenty-two-year-old defender for ten times his market valuation. I downloaded forty thousand interactions, found twelve thousand bot accounts sharing a single API key, and traced it to a contract between the club president and a media company run by his own younger brother. Girona inflated the player's value with a bot network; the real value lay in the server logs. Once again, the truth was not on the presented surface, but in the buried layer of raw data. What I learned from these two cases is not an investigative trick. It is a principle about the honesty of data. A conclusion is only trustworthy when independent evidence backs it. And when evidence does not exist, the only honest answer is to say that it does not exist. That is exactly what the forty-page report did right. It did not invent an analysis of any player, club, league, or transfer. It admitted that the input-data extraction stage had failed, that there were no information points to analyse, and that any conclusion issued now would be fabrication, not analysis. Professionally and ethically, that is a far nobler choice than filling the empty cells with fluent prose. But I am not naive enough to believe every empty dataset is handled that way. Most analysis pipelines have no automatic gate to reject an empty output. They simply keep running, and the end reader receives a product with no usable value but formatted for consumption. The failure is not in the algorithm. It is that no one was assigned the responsibility of asking: what actually is inside this data? This is the point I want those in the trade to face directly. Over three decades, I have watched football power change shirts many times without changing its nature. From the 2026 press room to the 2026 Girona bot network, power only changes shirts. In 2026, at the World Cup on American soil, they shut the press-room door in my face; three decades later, I flung the files wide open to answer. That closing in 2026 and the emptiness of an analysis today share the same nature: both are ways for a system to evade accountability. Now comes the part where I must be fair to the other side. There is a legitimate argument for producing empty analysis as an internal draft. A pipeline operating at scale cannot only produce perfect results. When thousands of articles pass through a system in a day, returning an empty output is an honest signal that the input is broken, and it lets engineers detect the fault. If every system were forced to generate content at any cost, we would end up with a sea of text that sounds smooth but bears no relation to the truth. In that case, an honest empty set is worth more than a thousand fabricated sentences. The problem is not that the system returns a zero. The problem is that the zero gets pushed to market as if it were another number. An engineer has the right to receive an empty output in order to fix a bug. Fans do not deserve to be treated as an error-checking gate. So where does responsibility lie? Not with readers, who have no duty to cross-check every report. Responsibility lies with the newsroom, the platform, anyone who puts their name on a product before verifying that it is real. Every published analysis carries an implicit promise: that behind it there is data, verification, and someone who took responsibility for reading backwards through every page. When that promise is broken — whether by negligence or intent — the loss is not just one bad piece. It is the erosion of trust in an entire industry. I do not trust transfer fees; I trust the numbers that were crossed out. And I do not trust a report merely because it is well formatted. I trust what can be read backwards, cross-checked, and rebuilt from the raw data layer. If the raw layer is empty, then everything built on top of it — however magnificent — is only a castle on sand. The football industry is entering an era where data becomes currency and analysis becomes a commodity sold to hundreds of millions of viewers every day. In that era, the greatest value a practitioner can hold is not the ability to produce more analysis. It is the courage to refuse to publish an analysis with nothing inside. Because an empty set called by its right name is not a failure. It is the highest level of honesty data can reach.