When Football Data Falls Silent: Nine Dimensions of Analysis and the Trap of Fabrication

The data room of a Ligue 1 club on an ordinary Tuesday morning had nothing special about it: three screens, a running dashboard, and a twenty-four-year-old analyst staring at nine columns of figures. The first column, tactical and technical: empty. The second, club finance and the transfer market: empty. The third, results and the opinion cycle: empty. He turned the dashboard off, turned it back on, and the nine columns were still empty. He turned to me and asked in a voice close to panic: "What am I supposed to write?"

I recognized that feeling. It was identical to the feeling in the Olympique Marseille press room in 2026, when I sat there as the only female researcher and a male journalist sneered, asking whether women watched football with their emotions. The only difference was that in 2026 I had a movement diagram of twenty-two players that I had drawn myself from videotape, whereas today the analyst had a void. And a void, in our trade, is the most dangerous kind of data — because it always invites people to fill it with something.

Fate is not decided in the press room — but it begins to be written there. What few say is that the fate of an analysis works the same way: it is decided at the data-entry stage, before a single word is written. When the input is empty, every conclusion at the output is organized fabrication. That is the greatest lesson thirty-seven years of watching this industry has taught me, and it is the subject of this piece: the nine dimensions of football analysis, and the trap of fabrication when the data falls silent.

Context: from notebook to data pipeline

When I began writing for La Provence under the nickname "Madame Tactique", my tools were a VHS tape, a squared notebook, and a pencil. I watched a single match eleven times over three days, tracing every phase, recording every time a full-back was left open. That was football analysis in its manual form — slow, laborious, but with one virtue I have always wanted to preserve: every number was counted by my own hand, so I knew exactly where it came from.

Three decades later, the analysis industry has become an assembly line. Companies such as Opta and StatsBomb log thousands of events per match, each event tagged with coordinates, time, player, pressure. Metrics such as xG — expected goals, a measure of chance quality — or PPDA — passes allowed per defensive action, a measure of pressing intensity — have become a common language. A coach in the Swiss second division today can access the same dataset that a Champions League club is using.

Alongside this, the nine dimensions we call the industry's standard frame have taken shape: tactical and technical; club finance and the transfer market; results and the opinion cycle; league landscape and team positioning; rules and governance compliance; management and the dressing room; risk profile; media narrative and expectation; and finally, transmission across the football industry. These nine dimensions are not a decorative list. They are nine lenses through which to examine an event, and they share one feature: each lens only works when data enters it.

The problem lies exactly there. As the industry industrialized analysis, it also industrialized expectation. An analysis is supposed to always have a conclusion, always have an "insight", always fill every cell of the nine-dimension frame. And when the data does not arrive — because of a collection error, because a source could not be reached, because the original article genuinely contained no information — the pressure to substitute interpretation for data becomes enormous. That is precisely the moment when our trade is most likely to lose itself.

I once tracked a textbook case at club level. A mid-table French club received a scouting report on a young striker from the second division, in which most of the data cells were blank due to a data-transmission error, yet the report still concluded that the player "has great potential". When I traced it back, the author admitted he had not watched more than three matches, and those three were two-minute highlight reels. Three two-minute reels had been upgraded into a scouting conclusion. That player eventually signed with a third-division team and never rose above that level.

The frightening thing is not the mistake. The frightening thing is that the mistake was produced automatically, steadily, and no one in the production chain felt anything was abnormal. When the input is empty and the output is still full, the system stopped being honest long before anyone noticed.

Core analysis: nine dimensions and the price of an empty cell

To see the problem clearly, let us walk through the nine dimensions and examine what actually happens when a cell is left blank. Because each dimension has a different failure mechanism, and each failure mechanism produces a different kind of fabrication.

The first dimension is tactical and technical. This is where I live most of the time. An honest tactical analysis must answer four questions: what structure did the team play, at what level did they execute it, do the personnel fit that structure, and what do the key data say. When all four questions lack data, the writer has two choices: either say "insufficient information to assess", or fall into the familiar trap — describing a match by feeling. "They won because they wanted it more." I have heard that sentence thousands of times, and every time I hear it, I think of Porto 2026. When people look at Porto 2026 and see a miracle, I see an equation waiting to be solved. That year, José Mourinho's side had only about forty-three percent of possession in the final against Monaco, yet created five goal chances to the opponent's one, and won three-nil. If you look only at possession, you will say Porto were lucky. If you look at the geometry of passes forced into a pre-set trap, you will see a machine. "Miracle" here is just the name people give to what they have not bothered to explain.

The second dimension is club finance and the transfer market. Here, data has a dangerous property: it is both scarce and alluring. Figures for broadcasting revenue, commercial revenue, wage bill, and net debt are usually only partially disclosed, and the hidden part is the most important part. When a financial cell is empty, people tend to fill it with transfer rumors — and transfer rumors are the highest-noise data in the entire sports industry. I always remind younger colleagues of one thing: The transfer market is a market of hope, and hope rarely obeys valuation. A player valued at twenty million euros is not worth twenty million euros because of his ability, but because a club needed him exactly when he was most talked about. When there is no data on contract structure, wages, or release clauses, every transfer analysis is merely a storytelling exercise in luxurious settings.

I am especially interested in a phenomenon now spreading: leagues in the Gulf are attracting stars who are past their peak in Europe. The common reading is that "football is developing there". My reading is different. When a thirty-five-year-old player leaves Europe for a league with a far lower level of competitive intensity, what is being bought is not top-level playing ability, but media value. I do not say this as a moral criticism — it is a reasonable transaction for both sides. But if you call it "football development", you are conflating two different things. Football development needs academies, youth leagues, a competition system dense enough to train local players over ten years. A launch event with ten thousand spectators and a famous player does not create that system. It creates a tourism campaign with a celebrity face. This is not wrong. It is simply a different thing from what is being claimed.

The third dimension is results and the opinion cycle. This is where data and emotion fight hardest. A team can win five in a row while its expected-goals figure is lower than its opponent's in four of those five. When this happens, there are two explanations: either the team has some special quality that does not appear in the data, or they are living on luck that will vanish. The way to distinguish the two is the denominator. One match says nothing. Five matches begin to signal. Fifteen matches are usually enough to speak. When the sample is insufficient, praise and criticism are both unfounded judgments. I once watched a coach sacked after four straight defeats, even though his team's process metrics were better than the opponent's in three of those four. The board did not read process data. They read the table. And the table, at that point in the season, was too small a sample to conclude anything.

The fourth dimension is league landscape and team positioning. Here, what is usually missing is comparison. To know where a club stands, you need to know where its direct rivals stand — squad value, financial power, academy output. When these numbers are absent, people tend to assess a club in a vacuum. And assessment in a vacuum is the fastest route to a wrong conclusion. I always remember a pattern I call "the surprise team's nightmare". A small club unexpectedly succeeds — they reach a major competition, they make noise. Immediately, the big clubs begin taking away each core player. Within two or three transfer windows, their frame is dismantled entirely. Their success, seen from a systems perspective, is not a reward. It is the opening form of another talent raid. There is nothing structurally unfair about this — it is how the talent market operates. But it means an analysis of that club will be wrong if it reads the success as a destination rather than a point of departure.

The fifth dimension is rules and governance compliance. Here, what is usually missing relates to financial regulations such as financial fair play, player-registration limits, disciplinary sanctions, and competition eligibility. This is the dimension fans care about least yet which carries the greatest weight over a club's fate. When data is absent, no one can model the worst-case scenario. And when no one can model the worst-case scenario, the club will have no plan for it. I have seen clubs docked points, banned from transfers, or excluded from a European competition over issues a decent analysis should have foreseen — if it had bothered to read the documents it did not want to read. That is not a lack of intelligence. It is a lack of discipline.

The sixth dimension is management and the dressing room. This is the murkiest dimension of all, because it lies behind a closed door. The owner's investment and patience, the quality of recruitment decisions, structural stability, the leadership structure in the dressing room, the manager-player relationship, the generational transition — none of this is available as clean data. When this dimension is empty, people tend to fill it with stories from "sources close to". This is the most persuasive kind of fake data, because it cannot be verified and always appears to come from inside. I learned to treat such stories by one principle: if it cannot be wrong, it is not data. A claim that can be wrong is a claim that can be verified. A claim that cannot be wrong is a story.

The seventh dimension is the risk profile. This is the synthetic dimension, where all the others meet. Sporting risk, financial risk, personnel risk, rules risk, opinion risk, systemic risk. Each risk must be assessed by level, likelihood, impact, and mitigation. When the data is empty in the previous six dimensions, the seventh is automatically empty — you cannot assess risk without an object to assess. And this is the point I want to stress most in this entire piece. An empty risk profile is not a low risk profile. It is an unmeasured risk profile. In most systems, unmeasured risk is the most dangerous kind, because it does not appear on any dashboard until it explodes.

The eighth dimension is media narrative and expectation. Here, the data is not match figures but the gap between market expectation and objective assessment. When a player is over-praised after three matches, expectation has far outpaced the base. When a club is savaged after two defeats, expectation has sunk below the base. The good analyst is not the one who predicts results. The good analyst is the one who measures the distance between the story and the data, and knows that this distance is usually greater than the story. The media heat cycle has a memorable rule: it rises faster than the data and falls slower than the data. That is why buying on rumor is always a bad decision, in football as in any other market.

The ninth dimension is transmission across the football industry. An event at one club does not stop at that club. It travels through the talent-development chain, through the agent ecosystem, through broadcasting and commerce, through capital networks, through derivative markets, and finally through the national-team ecosystem. A transfer at Marseille can change the value of a player in South America, alter a decision at an academy in Africa, and reshape the plans of a national team in Asia. When data in this dimension is empty, people usually stop at the direct event and ignore the entire submerged part of the iceberg. And the submerged part, as always, is the larger part.

What I mean by these nine dimensions is not that they are difficult. What I mean is that they share one feature: each needs data, and when data does not arrive, each has its own way of deceiving itself. Tactics deceives itself with emotion. Finance deceives itself with rumor. Results deceive itself with small samples. Positioning deceives itself with a vacuum. Rules deceives itself with avoidance. The dressing room deceives itself with "sources close to". Risk deceives itself with silence. Media deceives itself with the heat cycle. And transmission deceives itself by stopping too soon. Nine ways of deceiving oneself, nine traps, and a profession standing in the middle of them.

The contrarian angle: the void is the strongest signal

Now I will say something I know will irritate some colleagues.

The empty cell is not a failure of analysis. It is the most important analytical result you can produce. In an industry where everyone is paid to say yes, the person who says "I don't know" owns something rarer than money: credibility. Because an analysis is only worth something if it can be wrong. If every conclusion you write is framed so that it cannot be refuted, you are not analyzing. You are decorating.

This is why I hold that a data void is the strongest signal in the entire process. It tells you exactly where the system is failing. A nine-dimension table with nine empty cells does not say "there is nothing to say". It says "your data pipeline broke somewhere between the source and the desk, and everything you write next will be inference". The difference between an analyst and a storyteller lies precisely here: the analyst sees the empty cell and stops to find the cause; the storyteller sees the empty cell and starts writing.

I learned this lesson through a personal shock. There was a period in my career when I lost nearly the entire source network I had built over many years. A series of upheavals cost me access to press rooms, my relationships with coaching staffs, even the people who used to send me internal data. For several months, every article I wrote rested on what I could observe myself. It was the hardest period, and also the period that taught me the most. Collapse is not the end of the tunnel. It is the biggest data the world provides. When every external source vanished, I was forced back to the one thing that could not be taken away: the ability to count for myself, draw for myself, verify for myself. And I realized that much of what I had written when sources were plentiful had not actually needed them. I had used sources as a substitute for thinking, not as a tool for thinking.

From then on, I built a principle I apply to every analysis: if I cannot write a decent piece using only what I have observed myself, I do not understand the problem deeply enough to write about it. This sounds extreme, but it has one very clear practical consequence: it eliminates any possibility of fabrication. You cannot fabricate a phase you have watched eleven times yourself. You cannot fabricate a number you have counted with your own hand. You can only fabricate what you have never touched.

And this is where I must speak about the darkest side of the whole story.

Sports data, in raw form, is a dual-value asset. It serves fans, and it serves another, not-small market. The same dataset on passes, on touch positions, on shot frequency, can be used to explain a match to an audience, and can also be used to price a bet. When the analysis industry is swept into a spiral of supplying data ever faster, ever more detailed, ever closer to real time, it is inadvertently building infrastructure for an industry whose goal is not to understand football, but to exploit the information asymmetry of fans.

This is the darkest side effect of the digitization of sport, and it is rarely stated plainly. Data companies sell to bookmakers. Bookmakers use the data to price. Analysts use the data to write. Fans read the analysis and believe they are understanding football better, while in fact they are being handed one part of a system whose remaining parts are designed to take their money. I am not saying every analysis serves betting. I am saying the same stream of data flows into two places, and only one of them is spoken aloud.

So when I see a nine-dimension analysis filled in perfectly, with not a single empty cell, not a single place that says "insufficient information", I do not see a good analysis. I see an analysis that stopped being honest somewhere. Because reality is never that complete. Reality always has holes, and the decent writer is the one who points them out instead of covering them up.

I know some will say this approach makes the writing dry, hard to read, unappealing. I have heard that throughout my career. In 2026, when I wrote a long analysis of Porto's geometry of proactive defending, many colleagues called it dry. But a university lecturer in Lyon used it as teaching material. That taught me that depth and accuracy matter more than easy digestibility. The space on the pitch is wider than any great figure who ever stood on it — and also wider than any appealing story told about it. If you must choose between a piece many people read and a piece that is right, choose the piece that is right. The right readers will come. They always come, only more slowly.

There is a misunderstanding about accuracy I want to correct. Many think accuracy means dryness, means no emotion. I disagree. Emotion is a variable, and a variable can be measured like any other. The psychological pressure of a penalty in the eighty-eighth minute is not something mystical. It can be observed through breathing rhythm, through how long the player stands over the ball, through the choice of angle. It can be compared between players, between matches, between career stages. The question is not whether there is emotion. The question is whether emotion is measured or inflated. The bad writer turns emotion into an unverifiable mystery. The decent writer turns it into a verifiable variable. And when that variable cannot be verified — when there is not enough data — the decent writer says "I don't know", instead of saying "it was fate".

Closing: what I want to verify in the next match

I do not believe data discipline will win in the short term. This industry runs on attention, and attention rewards what is appealing, not what is right. But I believe that in the long term, what survives is not the most-read piece, but the piece people must return to when they want to understand something real. A decent analysis is one that can still be used for verification ten years later. An appealing story is one nobody remembers ten years later.

So when the twenty-four-year-old analyst asked me what to write when all nine columns are empty, I told him he should write about those nine empty columns themselves. Write about where the data pipeline broke, why it broke, and what that says about the system. A void honestly recorded is worth more than a perfectly fabricated conclusion. And if you must choose between appearing competent and actually being right, choose the second — because the first, like everything built on a void, will soon collapse, and when it collapses, it will leave behind a larger void.

What I want to verify in the next match is not who wins. What I want to verify is whether those who write about that match have enough courage to say "I don't know" when they truly do not know. If they do, this industry is heading in the right direction. If not, we will keep producing perfect analyses of things no one actually understands.