Empty Cells, Hard Truth: The Case for an Auditable Data Chain in Football Analysis

Last week a file landed on my desk. The name was unremarkable — "match-post data sheet." I opened it. Zero rows. No team, no players, no shot map, no timestamp. Only the format was standing — nine columns, nine headers, and an infinite blank below. A note was attached: "Stage-2 deep analysis needed, urgent."

I left my chair and made tea. Because I know the most tempting thing in that moment — filling the table. I could. I have twenty-seven years of match-watching behind me, a standard xG model, a PPDA calculator. I could pull names from football history, place a plausible expected-goals figure, and write a convincing story out of fourteen shots. The reader would never know. But the first vow of the Data Monk sits exactly here — an empty cell means an empty cell. Filling it with your imagination is not analysis; it is forgery.

This piece is about that silent table. About the least discussed, least glamorous, most necessary question in football analytics — what does an analyst do when the data does not arrive? And why is this capacity to say "no" the truest evidence of an analyst's worth?

The table I was called to fill

When I began writing for the sports fortnightly in 2026, football analysis meant eyes and memory. How far a player ran, how clean a pass was — these were matters of language, not numbers. My editor would say, "Tell the story." We told stories. Numbers existed only in the scoreline.

In 2026, sitting in Chattogram, I built a standard xG and PPDA model for Abahani Limited Dhaka versus Sheikh Russel KC. I tracked fourteen shots. Abahani's xG was 2.3, Sheikh Russel's 1.7. PPDA — passes allowed per defensive action — was 8.7 for Abahani and 11.2 for Sheikh Russel. The model predicted a 1-1 draw. The match ended 1-1.

That night a habit entered my life that never left. I forced every reporter to file a post-match data sheet. No sheet, no copy. Editors were pleased, because numbers earn belief. But that same habit carried a danger — I began to trust my own framework too much. A format present feels like information present. This is one of the great false beliefs of football analytics.

A year later, at the 2026 World Cup in Russia, I worked as a freelance data consultant for a regional broadcaster. During the Croatia-England semifinal I ran a live xG dashboard. Croatia's xG was 1.4, England's 0.8. Luka Modrić covered 12.8 kilometres, completed 67 passes, and his late pressing pushed England's PPDA down to 12.9. Croatia won 2-1.

At that tournament I decided that within fifteen minutes of every match I would fill a fixed data template. Copy became faster but less lyrical. I accepted that. Because I had learned that on a live feed one thing is always detectable — a feed can fall silent. Latency grows, cells sit empty, and that is exactly when an analyst's character is tested.

The architecture of data absence

A football data pipeline is like a river. Upstream sit the event collectors — operators in the stadium, video-tracking cameras, optical systems. Then comes the cleaning layer, then the modelling layer, where shot quality becomes xG and passes and defensive actions become PPDA. Finally comes the interpretation layer — where I sit, the analyst.

The trouble is that silt can settle at any bend. An operator mis-tagged a goal event. A tracking camera read one player as two on a rainy night. The feed arrived late. Or — most commonly — the match belongs to a competition for which the data provider never sold a licence. Then you receive a flawless format with nothing inside.

I call this the architecture of data absence. The surrounding structure stands; only the content is gone. And here two kinds of analyst are born. One thinks, "The format exists, so I must fill it." He goes to outside sources, estimates, places a rough expected-goals figure, and files a confident report.

The other stops. He says, "I will not analyse, because I do not have the material to analyse with." This stopping is not easy. Workload, deadlines and editorial demand all push you toward the first path. But a professional standard emerges precisely here — an answer honestly drawn from an empty dataset is worth a thousand times more than a beautiful report filled dishonestly.

I know the line sounds dull. Who wants an analyst to say, "I couldn't"? But suppose you are a club's data consultant. The coach asks, "Is the opponent's left side weak next match?" If you do not hold the shot maps from that team's last six matches, your honest answer should be: "I don't have six matches of data, so I can't tell you that." If instead you offer a guess, and the coach builds a tactic on it, you have not merely erred — you have made the coach err.

Nine doors, each with a lock

I follow a method I call the nine-dimension checklist. It is no magic; it is a door-lock. Every match analysis must pass through nine doors — tactics and technique, club finance and the transfer market, results and the public-opinion cycle, league landscape and team positioning, rules and governance, management and the dressing room, risk profile, media narrative and expectation, and industry transmission.

Behind each door sits a question. One asks how sophisticated the tactic was. One asks the ratio of wage spend to revenue. One asks which tier of the table the team actually occupies. Answering these requires solid information. Without solid information, every door stays shut.

Our industry is ashamed of shut doors. Nobody wants to say, "I don't have the information here." So many break the door down — with guesswork, with memory, or worse, with invention. In my system a shut door is a shut door. I write, "Insufficient information, cannot assess."

That sounds like failure. The truth is the opposite. Suppose a match analysis has seven doors open and two shut. You walk through the seven, draw an honest conclusion, and separately state the two shut doors. The reader knows what is known and what is not. That is real analysis.

By contrast, imagine an analyst forcing all nine doors — no real information in any, but a guess planted in each. The result is a complete, elegant, fluent and entirely fabricated report. Readers will read it, believe it, share it. Because fabricated things tend to be beautiful — reality is messy, imagination is smooth.

Two things are therefore compulsory in my method. One, every claim carries its source. Two, every uncertainty carries its measure. Source means not just "someone said" but the source's tier. A transfer rumour from an agent's inner circle is one tier. From a club's official statement, another. From a claim spreading on social media, the lowest tier of all. Without stating the tier, the reader cannot know how far to trust it.

The lesson of latency: when a dashboard lies by telling the truth

My deepest data experience in football comes from live dashboards. In Russia in 2026 I learned this — a live xG dashboard is a strange creature. It does not lie, but it does not tell all the truth either.

Because of latency. A shot happens. The ball enters the net or goes wide. The operator tags the event. The system processes it and sends it to the dashboard. The whole journey can take seconds to minutes. During that time the number on the dashboard is old. It does not yet know about that shot.

Why does this matter? Because viewers and journalists both treat the live number as final truth. If at minute sixty the dashboard shows Team A on 0.9 xG and Team B on 1.3, a storm erupts on social media — "Incredible, B ahead of A!" Yet in reality A may just have created a big chance the dashboard has not yet caught.

On live work I keep one rule. I state the number, but I state the time with it. "At this moment, that is, at this reading in minute sixty, the xG is this. But remember, the latest event may not yet be processed." One added sentence, and the number no longer deceives.

I have a favourite line I return to again and again: start with the xG, but end with the cold Tuesday. What does it mean? It means the model's number shows you the path, but the decision must be taken on a real, tangible, cold day — a day when the player is tired, the pitch wet, and the coach's job on the scales. Data is a map. A match is a road. You walk the road with the map, but the road's mud is not on the map.

On that Croatia-England night, Modrić's 12.8 kilometres and 67 passes are beautiful. But the number alone says nothing. It does not say at which moment Modrić ran, why he ran, or how his running helped drag England's PPDA down to 12.9. The number is the frame of the story. You must write the story, and to write it you must watch the match.

This is where I use my second favourite line: the dashboard is not the match; it is the match. It sounds contradictory, doesn't it? It means — treat the dashboard as the match and you become a fool, because the match happens on the pitch. But treat the dashboard as part of the match and it becomes your eyes. The difference is subtle but vast. If you treat data as the decision, you become its slave. If you treat data as the witness, you become the judge.

The data chain: what blockchain teaches us

So far I have spoken of empty input, honest answers, latency, source tiers. They look like separate rules, but they all point at one thing — football analysis needs an auditable chain. And it is exactly here that the philosophy of blockchain becomes useful to my work.

I do not mean blockchain as cryptocurrency. I mean it as an idea — an immutable, timestamped, publicly verifiable chain of records. Football analysis feels the absence of this idea every day.

Imagine a transfer story arrives. Who said it first? Which agent? Which club official confirmed it? What fee? How many years? If these facts are not arranged in a chain, nobody can verify them later. Numbers shift, sources vanish, nobody takes responsibility.

Take my 2026 Abahani model. That day I tagged every one of fourteen shots, computed xG, derived PPDA. But I made one mistake — I assumed that once the report was printed, the numbers would last forever. In reality? By season's end they were lost. In a different spreadsheet, a different file, a different email. Nobody could verify how my 2.3 xG was actually derived.

Here blockchain-like thinking helps. If every analytical step were joined into a chain — raw event data, then cleaning, then model output, then my interpretation — and every step were timestamped and linked to the previous, then no one could later lie. The sentence "I never said that" becomes impossible, because the record is immutable.

Imagine the benefit. A coach reads my analysis. He can see that my conclusion "the opponent's left side is weak" came from three particular matches, which filter was used, which variable was dropped. He sees not only the final number but the whole journey. That is real transparency.

In my professional life I have seen one thing — the greatest harm in football comes when a number is severed from its source. The number then stands as an independent truth, though it was in fact an estimate. This severance breeds misinformation. And an auditable chain prevents the severance.

The pressing-metric trap: running and destination are different things

Football data is most abused in pressing and effort metrics. PPDA is a fine metric. A low value means a team is pressing aggressively. But a trap sits here.

I use a line that grounds my pressing analysis: a pressing metric measures the horse's speed, but it does not say where the run is going. It means a team can achieve low PPDA because it is genuinely pressing in an organised way. It can also achieve low PPDA because it is chasing in frustration, gifting space. Same number, two causes.

Covered distance and high-intensity sprints are the same. They are sold as effort metrics. "Look how many kilometres this player ran, how many sprints — what commitment!" Yet pointless running also produces pretty numbers. A team that is behind runs more, because it chases the loss. A team that is ahead keeps the ball and runs less. So you cannot judge a player's quality from the height of a covered-distance figure.

Where does this error come from? From the habit of severing a metric from its destination. Computing PPDA is easy. But writing beside it — "was this pressing organised, or desperate?" — is hard. To write that, you must watch the match. Again we return to the same place — the number is the match's witness, not its substitute.

I have a long-standing suspicion about load management, directly tied to this metric abuse. Load management is romanticised — "we must protect the player." Yet in reality it is often a polite name for clearing the path for commercial tours and friendlies. Clubs use covered-distance figures to give their decisions a scientific face, when the decision is in fact financial. Here data is not the instrument; data is the wrapping.

When a threshold becomes law

I have a big weakness, I know. I love converting every argument into a line — a threshold, a cutoff, a clear boundary. "If the xG gap exceeds 0.5, I will say the team played well." This habit keeps me disciplined, but it also builds a trap.

The trap is threshold legalism. When a limit stops being a limit and becomes a law, the analyst no longer thinks — he only checks the rule. But football is a disorderly game. A number true in one match is false in another.

So my rule now — beside every threshold I write its sensitivity. That is, "I am using a 0.5 xG gap, but from 0.3 to 0.7 the result may invert." Add that one line and the threshold is no longer law; it becomes a decision — with responsibility and with limits.

An example. Two teams. Team A's xG 1.8, Team B's 1.2. The ordinary analyst says, "A created better chances." I say, "A did well, but this gap is close to my cutoff. If one of A's shots was actually offside and the system missed it, the gap shrinks. So I am not certain about this number."

This caution looks like weakness, but it is strength. Because the analyst who admits his uncertainty stays credible. The analyst who is certain every time is either brilliant or a liar. And in football analytics the brilliant are few and the liars many.

The contrarian question: does an empty result mean "no risk"?

Now the contrarian place I fear most. When I face empty input and write, "Insufficient information, cannot assess," I believe I am doing honest work. And that is true. But a danger hides.

Imagine an automated pipeline. It reads my report. It sees that in the risk grid every cell says "insufficient information." The pipeline may conclude, "No risk identified." But the truth is the opposite — "Whether risk exists cannot be said." These two are entirely different.

This is the subtlest trap. Confusing the absence of information with the absence of risk makes an analysis produce a dangerous false confidence. So in my reports I place a clear flag: "Insufficient input, reprocessing required." Sometimes I write it in capitals, so there is no room for misreading.

And a further contrarian point. We say good data means good decisions. That is half true. Because correlation and causation are different. When two things happen together, it seems one causes the other. But in football this seductive error happens daily.

Suppose a team wins more in the matches where it presses more. The analyst says, "Pressing is the cause of winning." But it may be that the team only gained the freedom to press when it was ahead. That is, the win is not the cause of pressing; pressing is the result of the win. The reverse is also possible. Without separating these two directions, you write a beautiful but wrong story.

I believe an analyst's most necessary skill is not extracting numbers but drawing a clear line between which question a number answers and which it does not. The analyst who can draw that line is reliable. The one who cannot is merely loud.

Waiting for the cold Tuesday

Back to that silent table. I finished my tea. Then I closed the file and wrote an answer: "Fourteen shot slots, zero rows. In this state I cannot reach any tactical conclusion, because the minimum material for analysis is itself absent. Please send raw event data, match schedule, and source tiers. Then I can walk through the nine doors."

My hand did not shake as I wrote it, but there was a satisfaction. Because I know I may have lost a report today, but I saved a profession's standard.

The future of football analytics is not in the abundance of numbers but in their accountability. The clubs and broadcasters showing live xG today will take their next step toward an auditable chain — a timestamp, a source, a measure behind every number. This need not be blockchain technology, but it must be blockchain's philosophy — immutability, transparency, verifiability.

The analyst who refuses to fill an empty cell may be slow, may be boring, may be less exciting. But he is the one thing rarest in today's football world — credible. And credibility is built over a season but destroyed by a single fabricated number.

So next time a dashboard shows you a gleaming number, ask — which match did this come from, at which moment, from which source, at what latency? Ask whether the latest event has not yet entered the system. And if a cell is empty, pause before filling it. Because that empty cell is your greatest witness — it reminds you that honesty is the last and only model of a Data Monk.

Start with the xG, yes. But end with the cold Tuesday, where number, pitch and pressure become one truth. And before that truth is made, let the courage to leave an empty cell empty be yours.