When Data Stays Silent: The Empty Frame and the Trap of Confidence in Football Analysis

That night, in an editing suite in Chengdu, I sat in front of six monitors replaying the same incident inside the penalty box. One screen showed a clear handball. Another showed an arm tucked tight against the body. A third was so blurry that I could not tell a shoulder from an elbow. I sat there for four hours looking for a truth, and by the time the roosters crowed on the outskirts of the city, I realised something I had never considered fifteen years earlier: sometimes, I have no data. I only have the feeling that I have data. That was the moment I began writing about a subject few in my profession want to discuss: the silence of data.

Today I want to tell you a story. Not a story about a match, a club, or a star player. This story is about the very foundation all of us — the people who analyse football — stand on every day, and about how empty that foundation can be without any of us noticing.

Context: When a Full Frame Looks Empty

Imagine a typical analytical table you see online after every big match. It has lineups, possession percentages, pass counts, shot counts, heat maps for every player, xG, PPDA, kilometres covered. This table looks complete. It has a title, charts, colours, footnotes. A machine validator would say: pass. A rushed editor would say: publishable. A busy reader would say: seems professional.

But the question I want to raise is: does that table actually contain information? Or does it only contain the structure of information — an empty frame painted over with numbers that the people who produced them do not themselves understand?

In the software industry, there is a concept called "silent failure". It occurs when a program completes, returns a structurally valid result, raises no error, shows no red warning, yet is empty inside. No real data. No real values. Only a shell. And the most dangerous thing about silent failure is that systems checking only for structure will keep accepting it. Empty results get pushed everywhere, layered onto each other, used as foundations for larger conclusions, until someone — usually at the most critical moment — discovers that everything was built on sand.

I recall a specific case from 2026, when I worked as a VAR data analyst for a sports television channel in Chengdu. Across 28 rounds of that CSL season, I reviewed 147 controversial refereeing incidents and found 12 offside decisions directly linked to camera placement. Twelve out of one hundred and forty-seven. Nearly ten percent. It sounds small, but each of those decisions could shift a match, an Asian competition slot, or a young player's career.

The match that haunted me most was round 25, Guangzhou Evergrande versus Shanghai SIPG. Wu Lei scored, the assistant referee raised the flag, VAR confirmed. The calculated deviation was fifteen centimetres. But the problem was this: not one camera in the broadcast system captured the correct horizontal plane of the last defender. We had six camera angles, but all six were positioned at heights and inclinations that made projecting a perpendicular line onto the goal line an almost unsolvable equation.

What I discovered was not whether Wu Lei was offside. What I discovered was that the VAR system had confidently delivered a conclusion for which it did not possess sufficient data. It did not say "I don't know". It said "offside, fifteen centimetres".

That was when I began building my own database of "viewing-angle error" — for every incident, I recorded how many cameras there were, where they were placed, at what height, with what angular offset, and the potential margin of error in centimetres at that distance. Later, when I wrote articles, I always attached camera-angle diagrams. Not to complain about referees. But to tell readers: here, this is the limit of what we are seeing.

Analysis: Four Forms of Data Silence

From that database, and from nearly three decades of watching professional football, I have distilled four forms of "data silence" that I encounter repeatedly in daily analytical work. These four forms apply not only to VAR. They apply to every dimension of modern football.

First form: Data severed from context. A player runs 12.4 kilometres in a match. This number is printed on every statistics sheet. BUT: running 12.4 kilometres in straight lines chasing a ball already passed away is physically different from running 11 kilometres that include 32 accelerations, 8 turns, and 14 bursts. The number 12.4 is not wrong. But it is silent about its own real meaning. Distance covered and sprint counts are packaged as effort indicators, yet ineffective running also produces beautiful numbers. And once a number is printed, no one traces back to ask: how much of this 12.4 kilometres was useful?

Second form: Data with structure but no entity. I encounter this most often in transfer analysis. A story about a deal has complete structure: club name, player name, transfer fee, contract length, wage. It looks very specific. BUT: is that club within the league's spending thresholds? Does that wage break the existing wage structure? Are there hidden clauses in the release terms? These questions often go unanswered, and when they go unanswered, administratively complete information becomes analytically empty information.

I remember once analysing a La Liga deal, when every outlet reported the fee and the contract length. No one noticed that the club sat right at the league's wage ceiling, and to register the new player they would have to sell another within ten days. Complete information. Full structure. But the context — the part that gives the story meaning — was entirely silent.

Third form: Data diluted by process. This is the most dangerous form, and the hardest to detect. In any newsroom's information chain, there are at least five pairs of hands: the original writer, the editor, the fact-checker, the headline writer, the publisher. Each pair shifts the information slightly. A 15-centimetre margin of error in the original can become "clear" in the second draft, "almost certain" in the third, and "certain" in the headline. By the time the article reaches the reader, it has been flattened into a decisive conclusion — while the original truth was an uncertainty.

I once witnessed such a process in a league I will not name. A referee was criticised for a decision to disallow a goal. In the original report, the match record stated: "the images are not sufficiently clear to determine the attacker's position". In the front-page headline, it read: "Referee disallows goal in controversial call". These two sentences do not say the same thing. The first addresses the limit of the data. The second implies a referee error. And the gap between those two sentences — the gap filled by process — is precisely what we need to talk about.

Fourth form: Data confirmed by itself. This is the trap even skilled analysts fall into. You have a hypothesis. You look for data supporting it. You do not find data refuting it — but not because refuting data does not exist, rather because you did not look where it might be. You review the incident from your familiar camera angle. You see what you expected. You conclude. And in your mind, the conclusion appears confirmed by your own search process, when in reality it was confirmed only by the limits of that search.

It took me years to realise this: a good analyst is not the one who reaches conclusions fastest, but the one who knows exactly what percentage of their conclusion rests on data and what percentage on gaps.

Contrarian Angle: Truth Is the Adversary of Justice

This is the part I want to dwell on longest, because it contradicts the instinct of most people in my profession.

In my early years, I believed my task was to find the truth. I believed every incident had one truth. There was a correct decision. There was an objective answer. And my job was to use data, cameras, every tool available to uncover that truth and present it to the world.

I was wrong about one very basic thing. I confused physical truth with measured truth. Football has physical truth — a player's toe is either on one side of the line or the other, no grey zone. But everything I have as an analyst is not that physical truth. It is a photograph of that physical truth through a camera system with error, through an algorithm with thresholds, through a pair of human eyes with limits. Three layers of noise. Three layers of silence stacked on each other. And when I printed the result, I called it truth.

This is why, at 45 — an age when many colleagues have moved into management, teaching, or retirement — I write something I likely would not have had the courage to write at 30: often we think we are seeking justice, when in fact we are only seeking a prettier camera angle.

I am not saying this to dismiss VAR. VAR has corrected many errors. VAR has exposed many things. The empty stadiums of 2026 showed me something I will always remember: VAR does not save football, it exposes football. It exposes that what we thought was objective turns out to be subjective. That what we thought was clear needs redefinition. That "clear and obvious error" — the phrase VAR teams everywhere use to decide whether to intervene — is itself an ambiguous clause. Clear to whom? Obvious in which frame? At what threshold does an error count as clear?

It took 37 reviews before I understood that the human eye is not a measuring stick. But neither is the camera. Nor the algorithm. The least reliable measuring stick of all is our confidence that we hold the measuring stick.

Psychology has a concept called the overconfidence effect. It shows that people tend to rate their judgement accuracy higher than reality, especially when given more information. But more information does not mean more accuracy. More information in the same direction can create a feeling of certainty without increasing accuracy at all. And in football — where we have millions of data points, hours of video, mountains of tables — the temptation to overconfidence is enormous.

I set myself a rule many colleagues consider extreme: when analysing a controversial incident, I begin by writing on paper, "I know nothing about this incident". Then I ask: "If I were forced to argue the opposite of my initial instinct, what would that argument be?". Then I ask: "What would have to be true for that opposite argument to be correct?". And finally: "What evidence do I have? And from how many camera angles did I look for it?".

Those four questions take about two hours per incident. Not every incident is worth two hours. But important ones are. Because a wrong conclusion here will spread across the internet within twenty minutes, and take years to correct.

Key Takeaway: Learning to Say "I Don't Know"

Among the four forms of data silence, the fourth — data confirmed by itself — is the hardest for readers to detect, and the one writers most easily conceal. Because to expose it, the writer must do something the entire football industry dislikes: say "I don't know".

Our profession has a strange fear of silence. Writers fear saying "I don't know" lest they seem unprofessional. Editors fear publishing an article with gaps lest readers leave. Readers — I have spoken to thousands over twenty-nine years — are not actually afraid of silence. They are afraid of pretended knowledge.

I think we need to relearn an old skill that technology has made us forget: the skill of tolerating uncertainty. When an incident sits right on the boundary, when a deal has only rumours and no paperwork, when a player is announced by the club's media team to return at the weekend, the right skill is not to reason toward a decisive answer. The right skill is to hold that uncertainty intact, write it honestly, and let readers judge for themselves.

On the matter of "returning at the weekend", I want to state plainly something I have observed over many years: return schedules are controlled by the club. "Waiting until the weekend" often means the injury has not healed. A club has many reasons to announce a return earlier than reality — reassuring fans, protecting market value, applying pressure to rivals. The average reader lacks the tools to distinguish a medical bulletin from a media bulletin. But the analytical writer does. And the duty of the analytical writer is to draw that line.

The strange thing is that when I began writing this way, my readership did not fall. It rose. The average reader has a very good instinct for when a writer is genuinely honest with them. What exhausts them is not uncertainty. What exhausts them is pretended certainty — the feeling that every article is trying to drag them toward a conclusion the author had decided before collecting any data.

Conclusion: The Limit Behind the Frame

I found the error not at the centre of the pitch, but at the edge of the frame. At the edge of the frame is where the light is weakest, the resolution lowest, the camera angle most skewed from truth. That is where I understood that what we lack is not just better data — but humility before what data does not say.

I am still in this profession at 45. I still sit in front of six monitors night after night. But the way I look at each monitor has changed greatly since 2026. Now, when a frame appears so clearly that I want to conclude immediately, I have learned to ask myself: is this clarity the clarity of truth, or the clarity of the frame I am being shown?

That is the question I want to leave with you today, as you watch any analytical table, any prediction piece, any headline about a transfer. Is what you are being shown the entire truth, or only the prettiest camera angle the writer chose to hand you?