The Empty Data Sheet and the Discipline of Not Concluding: Notes from an Analytics Room in Seoul

At 1:47 a.m. on November 24, 2026, in a ninth-floor office in Gangnam, Seoul, the Japan against Germany match had finished seventeen minutes earlier. An editor messaged: eight hundred words needed by 2:30. On my screen sat a spreadsheet with four empty columns — sprint count, distance covered after the 60th minute, substitution timings, and pressing actions. The tracking-data provider had not yet delivered the file. I stared at that empty sheet for ten minutes. Then I wrote the piece, but I wrote it differently: two cells with no numbers were flagged as missing, three cells carried their sources, and the headline stated plainly that this was a preliminary report awaiting complete data.

The next morning, that piece drew fewer readers than any comeback-match report I had ever written. A colleague said I had shot myself in the foot. I kept my position, because that empty sheet taught me something twelve years of watching sport had not fully taught me: an empty data sheet and a data sheet that reads zero are two entirely different things, and an analyst lives or dies by telling them apart.

Context: when the data has not arrived yet

Professional sports analytics runs on three data layers. The event layer covers goals, cards, substitutions and shots — this arrives almost instantly. The tracking layer covers distance covered, sprint counts, top speed and positional zones — this needs anywhere from thirty minutes to several hours to sync and calibrate. The context layer covers actual starting shapes, movement patterns, pitch condition, weather and accumulated fixture load — this requires someone to sit down and read it back.

In Vietnamese sports media today, speed pressure pushes most content onto the first layer. News goes out in fifteen minutes, analysis in two hours. The gap between layer one and layer three is where wrong conclusions are born, and where a writer's credibility erodes without anyone noticing.

There is a paradox I encounter constantly: when there are no numbers, people tend to write more forcefully, not more cautiously. With no figures present, nothing restrains the sentence. Player reputations, shirt colours, head-to-head history, the general feeling around a team — all of it becomes substitute material. Such an analysis reads smoothly, sounds reasonable, and is nearly impossible to verify.

The core: five checklist items and four times the data saved me

My pre-match checklist has five items, and I only allow myself a conclusion when at least four carry numbers: total sprints for both teams, distance covered after the 60th minute, substitution timings and count, pressing actions in the opponent's final third, and cumulative xG across the last three matches. These five items are not truth. They are the minimum filter that keeps a piece honest.

Cumulative xG across the last three matches is the item I learned on a June night in 2026, when it broke everything I had believed before.

On June 27, 2026, I stayed up for Germany against South Korea in the World Cup group stage. Everyone remembers only Kim Young-gwon's strike in the 90+3rd minute and Son Heung-min's finish in the 90+6th. That night I opened the data page the moment the referee blew the whistle: Germany's xG was just 0.76, while South Korea's reached 0.92. The reigning world champions took more shots but generated chances of lower quality. The final score was 0-2, and Germany were eliminated in the group stage.

I spent the following month rewatching all 36 group-stage matches, logging xG, passing numbers and ball positions to test a hypothesis: does data reflect reality accurately even when drama obscures it? My conclusion was not that Germany lost. The conclusion was that expected goals had foretold a match in which Germany's attack could not create a single clear chance, while the press saw only possession share. From that night on, every pre-match analysis I write begins with xG, shots on target and chance quality.

Distance covered after the 60th minute is the item that came from the season without spectators.

In 2026, when K League 1 returned in empty stadiums, I realised ten years of historical data on home advantage had been neutralised. I collected figures from 42 matches played without crowds in South Korea and found two numbers that changed how I read every table afterwards: the home win rate fell from 42.3 percent to 29.8 percent, while the draw rate rose to 31.5 percent. I built a separate model, removed the crowd variable entirely, and tested it across the Jeonbuk Hyundai against Ulsan Hyundai fixtures. In the first month, the model hit 8 of 10 handicap lines.

That was the first money I ever earned from sports data analysis, but its real value was not the money. It was being forced to admit that an environmental variable can outweigh form itself. I counted every empty space on the pitch once the crowds disappeared. When spectators returned, I did not go back to the old formula; I adjusted the expected home advantage by period instead.

Pressing actions in the opponent's final third is the item that came from Euro 2026, staged in 2026.

Before the round of sixteen, I presented a report to the tactics desk at the sportsbook where I worked. France were the tournament favourites, yet their PPDA stood at just 9.1. Switzerland pressed far more aggressively at 12.8, plus a superior total distance covered of 6.2 kilometres. I recommended the Switzerland not-to-lose line. Two colleagues pushed back directly, arguing I was reading pressure data while ignoring squad pedigree. The match ended 3-3 after 120 minutes, and Switzerland won on penalties to eliminate the reigning world champions.

Since that day, every call I make must carry PPDA and ball recoveries in the opponent's final third. Without those two metrics, I do not write about pressing.

Sprint counts and substitution timings are the two items that came from the 2026 World Cup.

Japan against Germany on November 23, 2026 stunned the world as Japan came back to win 2-1. Korean media focused on the German coach's tactical errors. I read the numbers right after the match: Japan recorded 247 sprints, Germany 201; and all five Japanese substitutions came before the 74th minute. The 1,500-word analysis I wrote for my personal blog that night concluded that the ability to sustain running intensity after the 60th minute was the decisive variable, and the piece reached 120,000 views in a single night.

These four stories do not say data is always right. They say data is always useful when you know what it is missing. When the numbers do not lie, my heart starts listening. Every goal is a piece of a puzzle; I do not watch football, I decode it.

The contrarian angle: a blank sheet is the most honest document

There is one professional habit I consider the most dangerous in sports analytics: filling empty cells with reputation.

When a big club meets a small club, writers rarely start from data — they start from memory. The big club won many times before, so the big club will win now. That argument sounds reasonable, but it ignores where the model has drifted. In my world, luck is only the residual I have not yet explained. A result that defies prediction is not a shock; it is a signal that an environmental variable was dropped from the equation.

An empty data sheet is therefore the most honest document an analyst can publish. It states that there is not yet enough basis for a conclusion. The problem is that in today's media environment, publishing such a document is treated as failure. So people fill it in. And when they fill it in, they reach for the easiest material available: crowd emotion.

The three areas where I see this filling-in happen most often are youth development, injury and the transfer market.

In youth development, big-club academies are praised as talent factories. But when I recounted how many academy players actually got first-team minutes over the past three seasons, the figure sat below 10 percent at most leading academies. The rest is talent hoarding, with players sent out on loan, sold on, or simply vanishing at twenty-one. Academy reports rarely put that column in the table.

On injuries, a player returning from a long layoff is asked to prove himself in his very first start. Framing it that way is cruel and unscientific. Recovery data shows it takes a substantial number of competitive minutes to restore decision-making under pressure, while re-injury risk peaks precisely in that window. The demand raises pressure, and pressure raises risk.

In the transfer market, the young-player price bubble is bursting. A fee of 100 million euros for a player with fewer than fifty top-flight appearances is raw gambling dressed in financial language. The transfer data contains enough numbers to see it, but nobody wants to read them because the story is more entertaining.

What I take into the next round of fixtures

After that blank-sheet night in Gangnam, I set myself a publishing code. No conclusion may appear when more than two checklist items are missing; if they are missing, the piece must flag which cells are empty. Every article must cite a source plus a publication date for at least one figure, so readers can verify it themselves. Based on my experience following matches across many seasons, source tracing is the only skill that separates an analyst from a commentator. And every piece must leave behind a tool usable in the next match — a filter, a metric, a question answerable with numbers. If a reader closes the article and carries nothing away, I have written the wrong profession.

In the current regular season, the signals worth tracking are not on the league table but in secondary metrics: distance covered after the 60th minute among teams chasing continental qualification, PPDA among teams fighting relegation, and accumulated minutes for players returning from injury. These three metric groups tend to foreshadow dips in the standings before the standings actually move.

I do not believe in inspiration — I believe in standard error. When a team unexpectedly draws at home against a weaker opponent, I do not ask why they lost morale. I ask how far their PPDA has dropped across the last three matches, and how many minutes later they substituted than their opponent did. Those numbers do not tell a good story. They answer exactly one question: which variable changed.

And when the data sheet for the next match still has an empty cell, I will leave it empty. That is the only way the numbers I do have keep their value.