The Data-Layer Failure in Football Analytics: Empty Frameworks and Their Cost

Three in the morning in Madrid, and I open the analysis that has just been pushed through. Twelve sections, every heading in place, every table, every framework. And every one of them closes with the exact same sentence: insufficient information to assess.

Not one section. Twelve. Tactical analysis, club financial structure, results cycle, league landscape, regulatory compliance, dressing room, risk profile, the transmission chain of the entire industry. The skeleton was built beautifully. The flesh contained not a single grain of rice.

I rewound twelve times. That is a habit I have kept since 2026, when I spotted a camera-angle error in an El Clásico. That day I measured a 1.7-metre discrepancy between camera A and camera B, wrote it up, and the piece travelled more than 200,000 shares within 24 hours. Today I cannot measure anything, because there is nothing to measure. That is the part worth writing about.

The television screen does not lie; only the person sitting behind it lies to himself. When an empty analysis is pushed to market with a full headline, full structure, and a fully professional appearance, the reader does not receive information. They receive the shape of information. And a shape cannot play a match.

Context: the two layers of an analytical product

Every football analytical product passes through two layers. The first layer extracts events: which team, which player, which number, which source said it, at what moment. The second layer builds those events into a readable conclusion.

When the first layer returns empty, the second layer must stop. That is the basic rule of any analytical process. In football, its equivalent is a scout filing a report without ever setting foot in the stadium. No club would accept that. But at the content level, it happens every day.

The current cycle is the transfer window. This is peak season for form over substance. Every day brings hundreds of lines of news, most of which name no specific club, no transfer fee, no contract length, no release clause. But the headline still gets written. And the fans still read it, because they are hungry for information.

What I want to say here is not the fake-news argument. That one has been made enough. What is worth saying is that a technical failure at the data layer can travel straight to market without anyone stopping it, and once it arrives, it puts on the suit of a professional analysis.

How emptiness is born

I classify three kinds of emptiness, because these three lead to three different consequences.

The first kind: the source text never entered the system. The content exists, but never reaches the person processing it. In football terms, this is a match that has already been played, with footage already available, but nobody rewound it. You sit down to analyse something you have never watched.

The second kind: the system runs but generates no data. The framework is built, the cells are created, and all of them are empty. This kind is the most dangerous, because it looks like a finished product. A table with complete headings always looks more credible than an empty paragraph.

The third kind: the source never existed. No original article, no team, no player, no date. Only a headline.

All three are data-transmission failures, not conclusions about content. But by the time they reach the reader's eyes, they are indistinguishable. And that is where this industry shoots itself in the foot.

The French defence and the value of a falsifiable claim

I will use an old example of mine, because it serves as a measuring stick.

That French defence needed no prediction; you only had to watch how they stood.

World Cup 2026, before the knockout round, the whole world was talking about Paul Pogba and Antoine Griezmann. I was not talking about them. I was talking about the back four with Raphaël Varane and Samuel Umtiti in the middle, Benjamin Pavard and Lucas Hernández on the flanks. I built a fourteen-minute video in which I showed that Pavard needed to drop roughly three metres deeper to neutralise Lionel Messi in the match against Argentina in Kazan on 1 July 2026.

That is a falsifiable claim. There is a number. There is a specific player. There is a specific opponent. There is a date. There is a location. There is an action. There is a consequence.

That analysis travelled far enough that the Argentina coaching staff printed it out as meeting material. I did not need that to prove I was right. I needed it to prove something else: an analysis only has value when its author dares to place on the table a sentence that can be contradicted.

Now take that standard and hold it against today's content market. What percentage of football analysis published today contains a sentence that can be contradicted? I have watched enough to answer: very few.

Numbers do not know fear

This is where I have to speak plainly to the modern analytical establishment.

Modern football loves numbers, but numbers do not know fear.

xG is a good metric. It measures chance quality. PPDA is a good metric. It measures pressing intensity, and a lower value means more aggressive pressing. These metrics have saved analysts from a great deal of prejudice. I use them.

But xG does not know that a full-back is playing with a groin strain. xG does not know that a centre-back just lost his father. xG does not know that a team is playing its fourth match in ten days, and that their legs ran out of battery at minute 60. PPDA cannot measure panic. No metric can measure panic.

That is why I never read an analysis made up entirely of metrics. After every block of numbers, I need one human moment. One passage of play at minute 89. One glance at the bench. One player standing still, not tracking back.

Seven World Cup cycles at my desk have taught me one thing: the champion is the team that corrects its mistakes least often. Not the team with the prettiest metrics. Not the team that scores the most in the group stage. It is the team that repeats the fewest errors, and fixes them fastest within the match.

That kind of information lives in no table. It lives in the way a back four stands still before the ball is kicked.

Live data and the bookmakers

There is a side effect of the digitisation of sport that analysts rarely discuss, and it is the darkest one.

Live data, the kind collected second by second during a match, is most valuable to one specific group of customers: betting companies. Every touch, every player position, every acceleration becomes raw material for repricing odds within fractions of a second.

I do not object to data collection. I object to an industry selling its most detailed data to the only customer group with an incentive to make fans lose money.

VAR was created to fix human error, but in the end it created machine error. My 2026 story is a miniature example: two camera angles offset by 1.7 metres, and an entire VAR room looking at them to confirm a goal. The technology was not wrong. Humans chose the wrong frame. But when they were wrong, the system still stood behind its own conclusion.

Live data behaves the same way. It does not audit itself. Nobody on the other end asks: is this number correct, or merely new.

Shirts, ROI and the local community

While I was analysing the data-layer failure, another failure was unfolding in parallel at the commercial layer, and it too is a transfer-window story.

Shirt advertising is breaking the bond between clubs and their local communities. A club in a small city, where the stands hold three generations of families sitting side by side, sells the space on its chest to a betting brand on another continent. That brand does not care about the city. It cares about the exposure index, meaning the number of times its logo enters a television frame.

The exposure index is a perfect number. It can be measured. It has reports. It has charts. And it says absolutely nothing about whether the ten-year-old on that street still feels he belongs to this club.

When you read an industry report, this part is usually left out, because it cannot be turned into a pretty table. That is the fourth kind of emptiness: not empty in data, but empty in meaning.

Transfer window: ranking rumours by evidence

This is the most practical part I have drawn from observing the data-layer failure.

During the transfer window, noise always drowns out signal. The only way to filter is to rank rumours by evidence, not by how attractive they are.

A line of news with a club name, a player, remaining contract length, a date and a specific source ranks high. A line of news with a full structure but no names for either party ranks low, even when written by a major outlet.

Three things I always check first in any deal: the structure of the release clause, the wage headroom the destination club still has to absorb, and the agent's movements in the previous ten days. These three are structural. They do not depend on the reporter's inspiration. A club can lie about its level of interest, but its wage bill does not know how to lie.

Veteran transfer specialists do not report on feeling. They follow the money, the contract, the paperwork filed with the governing body. The rest is entertainment.

Contrarian angle: where I might be wrong

I have laid a fairly heavy argument on the table. Now it is time to check myself.

Possibility one: an empty framework is not necessarily a failure. Sometimes the most honest answer is "I do not know". Football has far too many people obliged to say something every day, and a system willing to return "insufficient data" may be behaving more correctly than its critics.

Possibility two: the industry's problem is not the empty framework, but the fake one. An analysis that openly declares itself empty is harmless. An analysis stuffed with unverifiable numbers is what causes damage, because it makes readers believe in something that does not exist. If so, I have aimed at the wrong target. I am attacking the framework while the one who deserves attack is the one stuffing in fake flesh.

Possibility three, and the one I think about most: I am a footage man. I was born before xG, before PPDA, before every expected metric. My eye was trained by rewinding. It is possible I undervalue data systems simply because I did not grow up with them. Every season I force myself to read at least one analytical report written by a young person, and I do not always find myself right.

I leave these three possibilities on the table. See it first, then believe it.

What is missing: a validation gate

The solution to this problem is not better writing. It lies in one very simple technical rule: reject any input with an empty information field.

If an extraction returns an empty title, an empty source, and an empty list of events, then the next analysis must not run. No exceptions. No "run it to make the deadline".

This is not a dry technical matter. This is a matter of credibility. An industry willing to publish empty analyses is an industry that has lost the ability to audit itself. And when an industry loses the ability to audit itself, the fans pay for the mistake, with their trust.

Closing

I still keep the notebook recording the failure patterns that repeat across World Cup cycles. In it are patterns about defences collapsing because they dropped too deep, about teams rotating exactly when the fire was burning, about matches where the stronger side lost because of a substitution at minute 70.

I will add a new entry to that notebook, an entry about analytical products with no flesh inside. And I will set out a falsifiable prediction: within the next transfer window, at least one major media outlet will push out an analysis with full structure, full headline, and not one primary source.

When that happens, do not ask why football is becoming harder to believe. Ask why we still accept reading a framework without ever checking what is inside it.