Nine Complete Sections, Zero Data: The Trap Inside Football Analysis

Nine Complete Sections, Zero Data: The Trap Inside Football Analysis

Three in the morning in Surabaya, and I was counting a data column that should have held 90 rows. It held 62. Twenty-eight late-game actions from a 2026 match against Persib Bandung simply did not exist in my file. Six hours later, the coaching staff still received a tidy report: possession charts, heat maps, per-player metric tables, and a bolded conclusion recommending a higher defensive line. Formally, it had no holes. The team lost 0-3, and the second goal came from exactly the space behind the two full-backs, where my dataset was silent.

I sat with it for three nights, reviewing every action, and found what I had skipped: the opponent's PPDA. Persib had deliberately conceded the ball, conceded the middle third, then countered into the very gap my 63 percent possession figure never reflected. I wrote a ten-page self-critique for the coaching staff and proposed a cross-checking routine before every match. The Surabaya mistake taught me to interrogate data, not to trust it.

Eight years later, the same disease came back in far better clothes. This time it was not a file missing rows. It was a report with all nine sections present, tables in every one, and not a single subject.

A league drowning in spreadsheets

The transfer window is open. In V.League, almost every club now has someone in charge of data, a Wyscout or InStat account, an internal group chat where clips are pushed every evening. Public stat platforms let anyone pull xG, key passes, duel win rates, heat maps. The supply of data has never been richer.

Ten years ago, a club wanting to know where an opponent shot from had to assign someone to rewatch three matches on tape. Now it takes one browser tab. That speed is real progress, and it also breeds a dangerous habit: mistaking the ability to look up a number for the ability to understand a match. Tools do not generate method. A club can pay for three data platforms at once and still have nobody sitting down to cross-check them and ask what those numbers are hiding.

Data volume rises; evidence volume does not rise with it. A table can be complete, colour-coded on a heat scale, and still fail the simplest question: how does this team actually set up once it loses the ball?

The trade has a dry name for the fix: null-value handling. When a field carries no information, the analyst must write "insufficient data to conclude" rather than infer a plausible value and fill it in. It sounds like paperwork. But the line between analysis and fabrication sits exactly there.

Three ways a gap fills itself

The most common form shows up in the transfer window. A fan page posts "a source close to the board." Within hours, forty other pages repost. By evening, the name has become "almost certain." Very few in that chain bother to check the three decisive items: the release clause in the contract, the player's current wage against the buying club's wage ceiling, and the remaining contract years. Without those three facts, every judgement is just storytelling.

The second form lives in tactical reports. A 3-5-2 or a 4-2-3-1 gets assigned from memory of the previous round, while the team has since changed personnel at the attacking tip and changed how it builds out. I once received an opponent report with an elaborate pressing scheme drawn out, then discovered the diagram came from the first half of round 12, nearly two months before the analysis date.

The third form sits at the smallest scale: a shot credited to the more famous player because the data logger misread a shirt number in a blurry sequence. One wrong name corrupts the whole chain of conclusions about form, favoured shooting positions, and angle preferences.

All three forms share one mechanism. The gap gets filled with whatever sounds most plausible instead of whatever has been verified. In the technical literature the phenomenon has a name: subject substitution. Nobody sits down intending to fabricate. The gap simply fills itself unless the writer actively blocks it.

I once read a scouting report on a foreign player arriving from another league. Every metric was striking: key passes, pass completion, successful dribbles. The report carried not a single line about pitch conditions, fixture density, or the fact that his old league had twelve teams while the new one has fourteen. Those variables were not in the table, and because they were not in the table they were treated as though they did not exist.

A complete structure creates no information

A nine-section report with tables in every section creates the feeling that someone is accountable. A non-specialist reader looks at it and sees professionalism. What that reader does not see is that all nine cells may be empty.

I have watched an internal assessment like that. Every item on payroll, injuries, and contract discipline read "insufficient information." In a meeting, the report was summarised into four words: low risk. A complete structure had been read as a positive conclusion.

This is the most dangerous point in the trade. Bad data can be fixed, because it leaves a trail. Empty data presented as full data leaves no trail at all, and it contaminates every decision downstream: recruitment, team selection, contract negotiation.

Risks are silent by default

Here is a rule I learned after years of working with clubs: the most serious risks never appear in the data on their own. They appear only when someone goes looking.

Wage arrears are the clearest example. A club with no news throughout a transfer window looks calm. On the day the registration list is published, it turns out the club cannot sign anyone new. A key player's injury works the same way: no announcement, until he is absent for the decisive fixture. Match integrity problems are quieter still.

The absence of bad news is not evidence of calm. It is only evidence that nobody has checked yet.

My process now has one mandatory step: before writing any conclusion, I list what I do not know. That list is usually longer than the list of what I do know. Early on, it made meetings uncomfortable. Later, it saved the team a few defeats.

VAR, the biggest update football has not finished reading

Every sport has rule changes that reshape how it is played. In modern football, VAR is the biggest of the past decade.

What matters sits in the intervention threshold itself: "clear and obvious error." The phrase sounds like a technical standard, but it depends on interpretation by league, by VAR crew, by matchday. When the interpretation shifts, the playing style shifts with it. Defenders must relearn how to tackle inside the box. Forwards must relearn how to go down. Nobody is permitted to assume the size of that effect without verifying it, because it appears in no statistical table.

The same holds for every other rule change: added time, semi-automated offside, substitution allowances. All of them alter match rhythm, and all of them are absent from the metric tables people still use to compare seasons.

A four-tier filter for the transfer window

I still use a four-tier scale to filter transfer news, and it has proven more useful than any advice about "reliable sources."

Tier one: the contract is published, the player has signed. Tier two: the release clause and the wage bill reconcile, meaning real money has reached the negotiating table. Tier three: negotiations confirmed by both sides. Tier four: "a source close to the situation."

Most transfer content online sits at tier four while being written in the voice of tier one. The reader's job is not to believe or disbelieve, but to know which tier they are reading.

In a transfer window, money travels along a very specific path: transfer fee, agency fee, signing-on payment, and the personal income tax of a foreign player. Those four items decide whether a deal happens, and none of them appears in the stat tables fans keep refreshing. Tracking them is harder than tracking rumours, but that is the only part of the market that can actually be verified.

The tackles nobody remembers

The 2026 World Cup was won with tackles nobody remembers.

On the night France met Argentina, the stands and the timeline talked only about Kylian Mbappe's speed. I sat with my dataset and found a different figure altogether: France's tactical fouls in the middle third ran at the highest rate in the tournament. Those were deliberate, calculated fouls used to cut a counterattack before it took shape. Not one of them appears in any player's ratings.

The piece I wrote that night travelled fast, largely because it described something the coverage had skipped. From then on I spent more time on defensive metrics: clearances, passing lanes blocked, the distance between lines once my team lost the ball. Those numbers are not glamorous, but they explain more matches than a top-scorer chart.

I was once challenged live on air with the accusation that I worshipped numbers and dismissed the emotion of the game. I put up the shot locations of a team eliminated in the knockout round, and the story flipped: the problem was not luck, it was the quality of the chances being created. The argument ran nearly two hours. It taught me that a position built on evidence does not need to shout.

My own data has holes too

If I tell others to distrust numbers, I have to start with my own file.

For the same match, two providers can produce xG values that diverge substantially, because they define a "chance" differently. PPDA shifts depending on how a duel is counted. Clearances depend on the person logging them, and that person may be tired by the 85th minute.

People ask whether I trust statistics. I trust statistics when I know how they were collected, by whom, and under what conditions. Those three things matter more than the number itself.

The contrarian read

Most debates about sports data revolve around numbers being wrong. The bigger and far less discussed risk is empty data presented as full data.

A wrong xG table will be caught when someone reviews the footage. A report with no subject will not be caught, because it asserts nothing specific. It merely presents. And the presentation alone is enough for readers to nod along.

The crowd also carries a familiar bias: it trusts the report that looks professional over the report that is right. A document with a logo, tables and a colour scale always beats a document with three lines and one unanswered question. In analytical work, the second document is usually the more honest one.

The transfer window has one more paradox: the louder the rumour, the thinner the evidence. Real evidence needs no volume, because it lives inside a contract. Big deals tend to be announced in silence, after everything is already done.

What to watch next round

Three things I ask myself before every report, and before every transfer rumour: Where did this data come from? Under what conditions was it collected? If the gaps stay unfilled, does the conclusion still hold?

In this window, what deserves watching is not the names mentioned most. It is how clubs publish information: who announces early and clearly, who lets rumours do the work, who stays silent until everything is finished. How a club handles information says a great deal about how it handles risk.

And when the season returns, I will still be sitting with my dataset at three in the morning, recounting every column. The work is not glamorous. It is simply the only way to make sure that next time, the goal conceded does not come from where I stayed silent.