The Empty Cell in Johor's Data Sheet and the Trap of the Perfect Report

On the eleventh night of March 2026, Kuala Lumpur had been raining since dusk. I sat in front of my screen with four Johor Darul Ta'zim matches, rewinding at quarter speed, my left hand on the space bar, my right hand writing into the spreadsheet. The sheet had twelve columns: minute, ball carrier, direction of pass, number of opponent passes before Johor regained the ball, position of the defensive line, and a final column - the distance between the two centre-backs - left empty. The broadcast frame cut off the right touchline, where the right-sided centre-back stood. Across twenty-two first-half sequences, I did not know where he was.

I could have filled that cell with a plausible number. Eighteen metres. Twenty metres. Nobody could check. A complete spreadsheet always looks more trustworthy than one with a hole in it. I left it blank. That empty cell taught me more than the rest of the sheet.

My first blog post was not about football. It was about the space between two Johor defenders.

Three weeks later, I finished a 2,500-word piece on how Johor Darul Ta'zim press in the Malaysia Super League. Their average PPDA was 14.2, meaning opponents were allowed fourteen passes before Johor made a defensive action. But the real story sat elsewhere: Johor's midfield moved in fragments, without any clear zonal reference, and that fragmentation opened the spaces Kedah Darul Aman exploited twice after the break. My first draft was twice as long and mostly speculation about player psychology. I cut it. The published version kept only data, diagrams and one open question at the end. A large fan page shared it - twelve thousand reads in the first week.

What I remember most, though, is that blank cell.

Southeast Asian football does not lack feeling. It lacks a data column.

In England, where I was born, every Premier League match is captured through hundreds of positional data points per second. In Malaysia, where I work, most Super League matches have one camera angle, sometimes two, and no tracking system at all. Every tactical analysis here therefore starts with manual labour: rewatch, pause, count, record. One match takes three to four hours. One season takes a month.

That sounds like a disadvantage. It is also a strange advantage. When you have to count every pass yourself, you are forced to look at the grass rather than at a summary table. You see the left centre-back stepping up half a second earlier than the man beside him. You see the holding midfielder turn his head twice before the ball is played backwards. You see the opposing striker feint toward the flank and cut inside instead. None of that ever sits in a data column.

Before the ball is circulated, I have already seen three decoy receivers and one real lane.

On the night Croatia met Denmark in the 2026 World Cup round of sixteen, I was twenty-three, working as a commentary assistant for a sports channel in Kuala Lumpur. I kept using the phrase "spatial binding" to describe how Croatia stretched the opposing shape. The content director called me in after the broadcast and said one thing: the audience does not understand what you are saying. I did not argue. The following month I rewatched all four Croatia matches and drew their attacking and defensive transition maps. I realised I could say the same thing in a simple sentence: they dragged the opposing defenders upfield to leave space behind them.

Since then I have followed one rule: do not name the concept, draw the pass. Every term has to be translated into an action on the pitch. Stretch. Squeeze. Offside trap. Run. Sentences got shorter, verbs got stronger, and the writing became readable for people who do not coach for a living.

Every formation diagram lies when you view it from the stands; the truth is on the grass, where spaces move.

A report can be formally complete and substantively empty.

In 2026 I joined the data analysis group for the expanded thirty-two-team Club World Cup in the United States. Fourteen of us shared one report template: eighteen sections, from projected lineups to expected goals, from heat maps to set-piece breakdowns. It was a beautiful template. Anyone who filled all eighteen sections felt they had done the job.

The problem is this: filling eighteen sections does not mean you have eighteen findings. Some sections were completed with fluent prose that carried no fact at all. I once saw a section that read: this team needs to improve its transition capability. Correct. But nobody who read it knew what to do, and nobody could verify where it was wrong.

That is the trap of this profession. People notice a spreadsheet with a blank cell, but they do not notice a report full of words and empty of facts - and it still gets used to make decisions.

I learned to read reports differently. First I look for the section that could be wrong. A section that cannot be wrong usually says nothing. Only a section that puts forward a specific number with a measurement condition is worth discussing. In Malaysia I once read a four-page scouting report on a player whose speed assessment rested on exactly two live viewings and a seventy-second highlight reel. Those four pages could persuade a room to sign a contract. They could also be entirely wrong, because the sample was too small.

That is why I always state the sample size in my writing. Saying a midfielder completes 89 percent of his passes without saying how many passes across how many matches is half a sentence.

The distance between evidence and inference is often one deleted footnote.

On November 22, 2026, at Lusail, Saudi Arabia beat Argentina 2-1. I stayed behind and rewatched the tape for two days. In the first half, by my own manual count, Lionel Messi was caught offside seven times. The Saudi defensive line sat an average of about fifty-two metres from goal, and they worked to one synchronised rule: when the ball entered central areas, the entire back line stepped up on the same beat.

When I wrote about that match, I opened with a question: if the back line steps up half a second late on the eighteenth sequence, what happens? That question has no certain answer. I kept it in the piece. A reader messaged me afterwards saying I should have offered a clear conclusion. I disagreed. The article was published after the match ended, when every option already had its answer. Leaving one question before a sequence is the only way readers can touch the uncertainty the players had to carry.

A match does not truly begin when the referee blows his whistle. It begins when a defender decides to leave his position.

But I have to be honest about one thing: the count of seven offsides is mine. The organisers have a different figure, the data sites have another. I stated that clearly in the piece. My definition of an offside situation is not identical to the one a semi-automated system uses, because the system measures body parts at a precision of hundredths of a second, while I measure with my eye on a frame rate of twenty-five images per second.

If I had hidden that detail, the article would still read smoothly. It simply would not have been honest.

In a transfer window, the emptiest things are usually the numbers that get published.

We are in a transfer cycle now, and I spend most of my time reading contracts rather than rumours. A deal is not decided by the headline fee but by its structure: how much is paid up front, how much is contingent, who holds the image rights, and when and for how long a release clause can be triggered.

There is a paradox I have tracked for years. The price of young players rises faster than their professional quality. A player with fewer than fifty top-flight appearances can be valued at one hundred million euros. The buyer is not paying for what the player has shown. The buyer is paying for a scenario of what he might become, plus potential resale value. When scenarios become the basis of valuation, the market stops measuring and starts speculating.

The same happens with metrics. Distance covered and sprint counts get packaged as measures of effort. They look scientific. But a player who runs twelve kilometres in a match, most of it chasing the ball ten metres behind the pass, has produced a beautiful number for a meaningless performance. Running a lot is not running well. Running well means running into the space before the ball arrives there.

In my data room we keep an unwritten rule: when a metric makes a player look better than the eye suggests, open the tape and check three random sequences. If those three do not match the metric, the metric is struck from the report. Not because it is mathematically wrong, but because it does not measure what we need measured.

I do not write to praise a goal. I write to point out every footfall that brought it there.

The environment bends tactics in ways no diagram can draw.

When the pandemic halted global football in 2026, I lost my live commentary contract. I was twenty-five, at home, and had time. I downloaded data from Europe's five major leagues across two consecutive seasons and compared them. The average home win rate fell from forty-six percent in 2026-19 to thirty-nine percent in the remainder of 2026-20, when matches were played in empty stadiums.

I called the index I built the crowd pressure index, designed to measure how noise affects referee decisions and pressing rhythm. The results showed that proactive pressing teams lost roughly eleven percent of their effectiveness without supporters. Not because they ran less, but because opponents gained time to look up and choose a pass. Noise does not create pressure. Noise shortens decision time.

I brought that lesson back to Malaysia and applied it differently. In Kuala Lumpur, matches usually kick off at nine in the evening to avoid the daytime heat. Humidity can exceed eighty percent. Grass after an afternoon storm holds the ball more slowly. If I take a defensive line sitting fifty-two metres high in Doha and paste that figure onto a night match in Kuala Lumpur, I have committed the most basic error of the trade: using one place's measuring stick to judge another.

A Malaysian side may choose a low block not because the coach lacks ambition, but because they have forty-eight hours of rest, a domestic flight, and a pitch that does not permit accurate long passes. Those decisions never appear in any dataset. To understand them you have to sit in the stands and feel the heat rising off the grass at fifteen minutes past nine.

In 2026, again at the Club World Cup, I saw more clearly than ever how environment bends plans. European sides like Real Madrid and Manchester City controlled the ball well, but their scoring efficiency dropped by around eighteen percent when they had to travel over four thousand kilometres and had fewer than three days between matches. I proposed a fatigue coefficient based on kilometres flown, consecutive matches and pitch temperature.

My model called three of four quarter-finals correctly. When a colleague said football cannot be reduced to mathematics, I did not argue. I printed the chart and pinned it to the board. But I also wrote directly beneath the chart that the model failed on the fourth match, and that the failure mattered more than the three successes, because it pointed to a variable I had not measured.

The biggest blind spot in analysis is the belief that a complete report is a trustworthy one.

I once sat through a two-hour meeting where everyone debated a tactical diagram projected on a large screen. It had arrows, coloured zones, symbols. It looked convincing. After two hours, an assistant coach asked one question: do we have data showing which of these coloured zones are real? The room went silent. That diagram had been drawn from one person's feeling, not from one match's data.

The paradox sits here. The best analysts I have worked with are not the ones with the thickest data sheets. They are the ones who state most clearly what they do not know. They open a report with a line like: I have good data on the defensive block, weak data on attacking organisation, and almost nothing on how this team reacts to conceding before the twentieth minute. Those three sentences set the credibility of everything that follows.

The opposite approach is the real trap. When you fill every blank with fluent prose, you do not increase the report's value. You merely hide the uncertainty and transfer that risk to the decision-makers, who will sign a contract or pick a lineup believing you have computed everything.

In transfer analysis, that trap is far more expensive. A club can ignore a gap in the data and pay for it with a five-year contract. An agent can supply a figure nobody verifies, and the gap turns into money. I always rank transfer information in three tiers: confirmed in writing or by the club, reported independently by multiple sources, and coming from a single source with unclear motives. The third tier is not news. The third tier is an unanswered question wearing a news jacket.

I remember a line I once wrote at the top of a client report: this model holds only if the schedule stays unchanged, the temperature stays below thirty-two degrees, and the starting eleven changes by no more than two players. If any of those three conditions fails, read the rest of this report as a hypothesis, not a conclusion. That client later told me the opening line was the most useful thing in the entire document.

A blank is not a weakness of analysis. It is a data point.

Back to the spreadsheet from that March night in 2026. Years later I realised the distance between Johor's two centre-backs, the one I could not measure across twenty-two first-half sequences, was among the most important facts of that match. Not because I knew the number. Because I knew I did not know it, and therefore built no conclusion on top of it.

Had I filled that cell with eighteen metres, I would have written a different article. Perhaps better in prose. Certainly worse in tactics.

When you read football analysis, look for what the author admits. If you find nothing, you are most likely reading a complete report. And a complete report, in most cases, is one that has hidden its own gaps.

Data does not lie. The people presenting it do, usually in sentences that read very smoothly.

Tonight I have another match to rewatch. Before I switch on the machine, I always ask myself the same question I asked in Doha, in Lusail, in Kuala Lumpur: does this back line step up together, or one man at a time? And if someone steps up half a second late, which space opens before the ball arrives there?

The answer is not in any column of my spreadsheet. But the blank cell, at least, has prepared me to know where to look.