A 48-Team World Cup and the Empty-Data Problem: Analytical Discipline in a Major Tournament Season

In the working folder on an old laptop in Guangzhou, I keep a Word file named empty-analysis.docx. Inside there is a single line: "Insufficient information to assess." The file was created at 2:40 a.m., after a phone call from an editor at a large sports outlet. He needed 2,500 words of deep analysis on a match, delivered in four hours, along with a template the newsroom had prepared. The template had nine sections: patch and meta, tournament format, roster and form, region, club finance, rules and compliance, risk profile, public narrative, forecast. Forty-one cells. I opened it and found I had exactly one thing to enter: the names of two teams. No video, no data sheet, no lineup, not even certainty that the match had already been played. I sent back one line. Three days of silence. On the fourth day he wrote: "Thank you. At least you didn't make it up."

I tell this story for a different reason. During a major tournament season, empty-analysis.docx shows up more often than any other document on my machine, and it forces a choice: a piece that reads well, or a piece that is right. My job pays for the first. My discipline lives on the second.

104 matches and a data paradox

The 2026 World Cup expands to 48 teams across 12 groups of four. That is 104 matches in total: 72 in the group stage and 32 in the knockout rounds, with the top two from each group advancing alongside the eight best third-placed teams. The figure of 104 matches is a media problem before it is a technical one.

In the Chinese market, where I have worked for eleven years, a single major match generates four to seven content products: a preview, live coverage, a post-match piece, a data sheet, a heat map, and a long analysis. Multiplied by 104, that is roughly four hundred to seven hundred pieces in six weeks. In Vietnam, where I was born and still write for a few outlets, the tempo is lower but the structure is identical: output scales arithmetically, while the underlying data does not.

Here is the paradox. The number of matches in the tournament grows, but each team still plays three group games. A team reaching the semi-finals plays at most eight matches. Eight matches is the entire corpus of elite competitive data we hold on them across four years. Meanwhile, a domestic league player appears in thirty-eight matches a season. We analyse most energetically what we know least about.

I learned this early. In 2026, as a first-year student in Guangzhou, I started a football blog and built my own data table to break down a Chinese Super League match between Guangzhou R&F and Shanghai SIPG. Striker Eran Zahavi accelerated 57 times in one match, 34 percent above the average for other forwards, and scored six goals across three consecutive rounds. I wrote a piece called "The Sprint Machine" using data on 23 under-23 players across two seasons. It drew 32,000 reads, eighteen times the site average, and brought my first collaboration offer.

The lesson was not in the number 57 or the 32,000. It was that I counted something nobody counted, and therefore held something nobody held.

Three matches cannot tell a story

A team takes roughly twelve shots per match. Three group games means thirty-six shots. At that sample size, the error band around conversion rate is wide enough that a team scoring four goals from three expected goals and a team scoring one goal from three expected goals can be described through two completely opposite narratives, while the real difference between them sits inside the noise. A team that wins twice and loses the third has not "lost form"; it has simply finished a sample far too small for anyone to conclude anything.

The same holds for clean sheets, pass completion, duels won, press escapes. Every outcome-based metric trembles at a sample of three. An analyst without discipline fills that tremble with adjectives: character, class, weakness, decline. Adjectives carry no error bars, which is precisely why they are used so heavily.

I used to fill gaps that way. In 2026, during a live World Cup commentary, I mispronounced Sadio Mané's name three times in the first half. The audience mocked me, and I deserved it. I did not deny it; I recorded the voices of 47 internationals and practised pronunciation every night. In that same tournament, during France against Argentina, I estimated Kylian Mbappé had reached a top speed of 37.2 km/h, against Gareth Bale's record of 36.2 km/h, and wrote a series predicting he would break every transfer-fee record within five years, with an estimate of 400 million euros. That series got me hired by a sports business magazine.

In 2026, I was wrong. But from that mistake I saw the value map of an entire decade: what I got wrong was not the prediction, it was predicting without a confidence label. Numbers weep, if we are willing to listen.

Morocco 2026 and the structural signal

Qatar 2026 gave me a cleaner counter-example than anything I had done before. I chose to follow Morocco, the most underrated side in the deep-run conversation. Across their first five matches they kept four clean sheets and allowed opponents an average of just 2.1 touches inside their own penalty area per half.

Looking closer, their 4-4-2 defensive block sat roughly 2.1 metres deeper than the norm. The consequence: passes played into the final third against them fell 28 percent, while counter-attacks ending in goals rose 60 percent. I wrote twelve analytical pieces during the tournament and predicted Achraf Hakimi would reach a commercial valuation of 80 million euros within two years. By the day Morocco reached the semi-finals, my communications plan had been ready for a while.

What makes this case fundamentally different from the three-match lesson is this: I was reading mechanism, not outcome. Goals are a sparse event; five matches give you a sample you can count on one hand. Player positions are a dense event. Every match generates tens of thousands of tracking frames; every defensive block repeats hundreds of times per match. Five matches may give you five outcomes, but they give you tens of thousands of positional observations. The positional sample is enormous even when the goal sample is tiny.

That is the line between an analyst and a commentator. A commentator counts goals, which have already been counted for him. An analyst counts distances, which nobody has bothered to count.

The data nobody counts

In 2026, when stadiums closed during the pandemic, my familiar data source vanished along with the crowd noise. I turned to something colleagues considered pointless: within the Bundesliga, I tracked 15 matches without spectators and counted audible player shouts. The average was 19 clearly audible shouts per match, up 34 percent on the previous season.

Nineteen shouts. A metric that appears in no official data set, no sponsorship contract, no paid measurement. Yet it plugs directly into a theme I have followed for years: the five-substitution rule deepens squads while turning the final twenty minutes into a war of attrition. With five changes available, the closing twenty minutes become a period in which defensive organisation is repeatedly reloaded, and the side that maintains better communication structure holds its shape longer. Shout data is a cheap proxy for something expensive.

The pandemic did not kill football; it stripped away the breathing so we could hear the heartbeat. Some colleagues called me romantic. But a well-known podcast invited me on regularly, and my formal journalism career began with that supposedly useless series.

Esports is teaching football to read patches

Before I moved into media, I competed in esports and organised tournaments. That experience shaped how I read football more than any tactics course.

Esports has three things football lacks. First, patches. Every publisher update resets the entire value system overnight, and the whole community has to re-read the game from scratch. Second, versioned data. Every metric is tied to a specific version, and nobody mixes data from two patches in a single chart. Third, replay databases. Thousands of matches are stored in queryable form, turning analysis from a craft into a discipline with an underlying database.

Football is walking the same road, just more slowly. A World Cup finals is a patch: the format changes, the number of teams changes, the number of matches changes, and every historical statistic loses part of its comparative value. Esports is teaching football to speak the language of a new generation.

Any football analyst who knows how to read a patch understands immediately why qualifier data cannot simply be carried into a finals. Different opponents, different pressure, different rest rhythms, and most importantly different motivations. Blending those two data sets into one chart is a beginner's error, and I still see it weekly on major sites.

Five checks before publishing

From all those wrong and right calls, I extracted a five-step routine that applies to every analytical piece during a major tournament season.

First, define the question before hunting for data. If the question is "which team is stronger", I will never get a decent answer, because it is an infinite question. If the question is "where does this team's defensive block stand when it loses the ball", I have an answer within twenty minutes.

Second, check the sample size and the unit of observation. Three matches is far too few if the unit is goals. Three matches is plenty if the unit is positions. Same data block, two opposite conclusions, simply from choosing the wrong unit.

Third, separate mechanism from outcome. Mechanism survives small samples; outcome does not. Morocco allowing 2.1 touches per half inside their box is mechanism, and it repeated. Morocco reaching the semi-finals is outcome, and it might not have happened.

Fourth, actively hunt for data that contradicts your own thesis. Before writing about Morocco, I spent a full session looking for matches in which they were stretched and structurally broken, and I found them. They did not destroy the thesis; they bounded it. A thesis without boundaries is a slogan.

Fifth, label confidence inside the text, in plain words, without hedging language. When I write that a conclusion rests on only two matches, I am giving readers the right to judge for themselves. When I stay silent about it, I am selling them a certainty I do not possess.

The price of honesty

At this point I have to argue against myself, otherwise this piece is just a slogan stretched long.

My pieces that end with "insufficient information" have markedly lower readership, often a third to a quarter of pieces that end with a firm forecast. A colleague once issued a completely wrong but supremely confident prediction; it spread across platforms and he was promoted. The market pays for certainty, and it pays steadily. That is a fact, not a complaint.

But the opposite trap is more dangerous. Some analysts become addicted to caution and never commit to anything, losing the only thing that gives this trade value: the ability to be early and right. Someone who only ever says "more data is needed" will never be wrong, and will never be useful. The strongest competitor is not the fastest runner, but the one who can read the wind of the market.

The solution is not silence. It is labelling: how much I believe this, based on how many observations, and what would change my mind. Those three sentences turn a guess into a testable hypothesis, and turn the writer from a loudspeaker into an interlocutor.

What remains after 104 matches

When 104 matches arrive within six weeks, the scarce commodity will not be opinion. Opinion is free and always in stock. The scarce commodity will be properly labelled uncertainty, because it is the only thing that lets a reader distinguish analysis from a prediction wearing analysis as a costume.

The next major tournament will have 48 teams, many of which have never played a knockout match at this level. There will be groups where three teams split six points and all three advance or all three go home. There will be teams that win three matches and exit in the next round, and teams that lose twice and go deep. Each scenario will generate hundreds of explanatory articles, and most of them will be stories written on top of a void.

In my Morocco folder in Guangzhou there are twelve published analyses, and in the margin of page seven there is a handwritten note about a Hakimi retreat in the 78th minute, when he left his position to cover the far post and immediately returned. None of the twelve pieces mentions that detail, and it remains the most valuable note in the whole folder.

As for empty-analysis.docx, it now has a second line, with a date and a confidence label. I have not deleted it.