The Discipline of the Empty Spreadsheet: Why the Best Football Analyst Is the One Who Can Say 'Insufficient Data'

The Discipline of the Empty Spreadsheet: Why the Best Football Analyst Is the One Who Can Say 'Insufficient Data'

On the night of June 27, 2026, at Kazan Arena, in the 90+3rd minute, Kim Young-gwon put the ball into Germany's net. I was sitting in front of a screen in Saigon, beside an open spreadsheet containing 112 V-League matches I had broken down by hand, shot by shot. There was no cheer. Only the steady turning of the ceiling fan and the sound of the keyboard. I typed one more line into the model log: the prediction was right, but being right brought none of the feeling of victory.

Three weeks earlier, I had published an analysis arguing that Germany risked elimination in the group stage. I received hundreds of jeers. An acquaintance messaged me: "You're selling paranoia to gamblers again." I did not reply. I kept Germany's pressing data unchanged: average distance covered down 12.3% from the 2026 title-winning side, and PPDA rising from 8.2 to 11.7. A rising PPDA means a team allows its opponent more passes before contesting. It is the signature of an engine slowing down, not of a champion hibernating.

But the story I want to tell today is not about being right in Kazan. It is about something far less discussed, and far harder to sell: the discipline of the empty spreadsheet.

A market that prefers answers to questions

Vietnamese fans follow football with unusual intensity. A qualifier can change the rhythm of breathing across an entire city. When the national team enters a major tournament, demand for information grows exponentially, and supply cannot keep up. That gap is always filled — the only question is with what.

I have worked as a sports betting analyst for a long time, long enough to remember when people had only newsprint and a television bulletin. Today, hundreds of new headlines appear every minute. Most of them are true. A significant minority are not.

Based on my experience watching matches and transfer cycles, I have noticed a worrying pattern: the stories that spread fastest are rarely the ones that were verified most carefully. A transfer report with murky origins, an injury claim with no traceable source, a statistic nobody can trace back — all share one feature: they answer the question before anyone has had time to ask it.

The xG shock at Hang Day turned me from a spectator into a reader of data. But it took years more before I learned the second half of this craft: reading data includes knowing when there is not yet enough of it to conclude.

In a major-tournament season, that pressure grows heavier. Fans want to know the starting eleven before the coach does. Bookmakers post odds before injury news is confirmed. An editor needs the piece before kickoff. Nobody in that chain is allowed to say "I don't know yet," because saying so means handing the space to someone else.

That moment is where the craft degrades. Not in the arithmetic. In the choice: when forced to pick between a wrong conclusion and a blank space, people usually pick the wrong conclusion, because a wrong conclusion still gets read, while a blank space does not.

Three times the data taught me to stay silent

I want to recount three occasions when the data forced me to change how I work. None of them is a story about a correct prediction. They are stories about rebuilding the frame I use to see.

The first time: Hang Day, 2026.

In 2026, during a match between Hanoi FC and Quang Nam FC at Hang Day Stadium, I lost 180 million dong in bets. Hanoi FC took 17 shots and generated 2.87 xG, but the match ended 1-1 against an opponent with only 2 shots and 0.94 xG. Watching the scoreline, I was angry the way a spectator is angry: I blamed luck.

Then I did what a spectator does not do. I audited 112 V-League matches from round 1 to round 14, computing xG by hand for every shot. The result: Hanoi FC created plenty of chances but finished 23% less efficiently than the league average. The 3,000-word analysis I wrote afterward was mocked by the media. A month later, that same data correctly predicted their four-match losing streak.

The lesson was not that I was right. The lesson was that a scoreline told me the match was over, while xG told me the match was still unfolding in another way. From then on, I launched a dedicated xG column and ended writing based on highlights and gut feeling. Every V-League piece came with a self-built data table, with a standardized metric-collection process for each match. My rigidity in presenting data became my personal brand.

But there is one detail from the Hang Day affair that I keep, because it is the seed of this article. When I set out to compute xG for those 112 matches, I realized some shots could not be classified with certainty. The angle was blurred, the defender's position unclear, the moment of contact only guessable. I considered assigning them an average value so the spreadsheet would look complete. Then I stopped. I flagged them as "insufficient data" and left the cell empty.

A spreadsheet with empty cells is harder to read than a full one. But a spreadsheet made complete with fabricated numbers is more dangerous than one with empty cells. That was the first thing I learned past fifty, later than I would like to admit.

The second time: Kazan, 2026.

The 2026 World Cup in Russia was the first time I dared to publish predictions throughout a tournament, without dodging controversy. Before the group stage, I audited Germany's pressing data and saw two signals. First, their average distance covered fell 12.3% from the 2026 title-winning side. Second, PPDA rose from 8.2 to 11.7. For a team that once defined the high press, allowing opponents nearly four extra passes before contesting was a structural decline, not a slow start.

I published a prediction that Germany would be eliminated in the group stage. The reaction was so fierce I had to turn off notifications. On June 27, in Kazan, Germany lost 0-2 to South Korea. They finished the match with just 0.41 xG, and six of their late shots went straight into defenders. It was the first time Germany had exited a World Cup in the group stage since 2026.

Kazan did not take revenge; Kazan simply kept the books and waited for me to miscalculate. And that time, I did not miscalculate. But I also refused to turn that result into a personal legend. Because I knew exactly what had happened underneath: the xG model I built from the V-League held up in Kazan only because it rested on one dry principle — conclude only when the evidence is thick, and stay silent when it is thin.

It was also from Kazan that I began publishing my own uncertainty. Before each matchday, I wrote the "Pre-match Numbers" series, presenting data with dramatic pacing to hold a mainstream audience, but every piece ended with a line stating the confidence level of the prediction. Some matches I wrote plainly: the probability leans one way, but the margin of error is too wide to bet. Nobody likes reading that. But it is honest.

The third time: Empty stands, 2026.

In 2026, COVID-19 halted global football. The Bundesliga returned on May 16 in silent stadiums. I checked the 28 matches after the restart and found something odd: home teams won only 5, or 17.8%, while the league's historical home-win rate was 42%. My betting model multiplied by a home factor of 1.32, so in a single week I lost 40 million dong.

I immediately audited 200 Bundesliga matches from that season and found that home teams still pushed forward as usual, but their actual xG fell by 0.45 per match without a crowd. Within 72 hours, I wrote the piece "Home Is No Longer an Advantage" and recalibrated the entire system.

The crowd left, the model broke, and I learned to hear the breathing of an empty stand. It was the first time I understood that part of a match lies neither on the pitch nor in the spreadsheet. It lies in the space between the stands and the touchline, where cheering usually costs the away defender half a second of focus.

From then on, I designed a "context coefficient" — adjusting xG, PPDA and result predictions for empty stadiums, weather and travel distance. My writing shifted from "absolute data" to "data placed in context." That was the first crack in my innate rigidity, while keeping my own standard of logic.

The mechanics of a conclusion built from nothing

Those three episodes taught me the same thing: error is less frightening than emptiness disguised as data.

Picture a beautifully presented analysis. A clear title. Complete sections. Every cell filled. A reader skims it and feels reassured, because a fully structured document creates the impression that someone understood the problem. But if at the source layer there is no match, no player, no club, no financial figure — then the document is not analysis. It is a form of presentation. And a form of presentation with no data underneath is more dangerous than a lie, because it cannot be caught by comparison with reality.

In my trade, this phenomenon has a name: the phantom transfer story. A player is said to be about to join a club. The story spreads. Nobody can trace the original source. When the deal collapses, nobody is held responsible, and the next story begins again from zero.

Its transmission mechanics are simple and effective:

| Stage | What happens | Who verifies | |---|---|---| | 1. Origin | An anonymous account posts information with no stated source | Nobody | | 2. Amplification | Aggregators repost it with a more enticing headline | Nobody | | 3. Legitimization | A major outlet cites the aggregator | Someone, but too late | | 4. Settlement | Fans treat it as established fact | No one left |

What is notable is that no stage in that chain requires evidence. The chain runs itself, and by stage four, denying the information is far harder than accepting it.

I once watched an injury story spread for a whole day, moving odds directly, only for the club itself to deny it that evening. Those who bet on it lost money over information that never existed.

With the experience of someone who has reported on football since 2026, when I graduated and began my career at a football newspaper, I can see the difference between two eras. In the past, slow speed was an accidental safeguard: a false story needed time to travel from one mouth to another, and that time was enough for someone to verify it. Today, that safeguard is gone. Speed has become the only standard, and speed cannot tell true from false.

Three questions before publishing any figure

Over the years I have distilled a three-step process, and I apply it to everything I write, even the shortest pieces.

First, provenance. Where did this information come from, and what does that source gain from it spreading? An agent pushing a client's value has a different motive from a club trying to reassure fans. If I cannot identify the source, I have no basis for judging reliability.

Second, the timestamp. Every piece of information must attach to a specific date. "Yesterday," "this week," "recently" are ways of writing that erase verifiability. A fact without a date cannot be checked, and a fact that cannot be checked is not a fact.

Third, the entity. I must name it precisely: which player, which club, which league, which figure with which unit. Vagueness in naming is where wrong conclusions live.

When one of those three questions has no answer, the right choice is not to guess to fill the gap. The right choice is to leave the gap empty, and to say plainly that it is empty.

This runs against my professional instinct. I am the organizing type, fond of order, fond of everything in its proper cell. An empty cell in a spreadsheet irritates me visually. But I have learned that this irritation is not a signal of a problem. It is a signal that I am being honest.

Belief is a noise variable; run the emotional regression before you bet.

I repeat this to myself whenever I read a piece of news too appealing to ignore. Because most wrong conclusions are not born from wrong data. They are born from correct emotions: the eagerness of fans, the pressure on the writer, and the craving for a clear answer in an unclear world.

A major-tournament season compresses emotion and releases it at once. People are swept up by flags and stories. In that state, a strongly asserted headline always beats a note saying the data is insufficient. That is why the discipline of the empty spreadsheet is so hard to keep: it is not rewarded. Nobody shares a piece that says "I don't know yet."

But the market, over time, rewards precisely that honesty. Not with fame, but with accuracy. The person who always gives an answer will be right about half the time. The person who gives an answer only when the data permits will be right more often, and silent the rest of the time.

The counterintuitive angle: silence is a skill, not a failure

In this industry, it is assumed that a good analyst has an opinion on everything. I think the opposite is true. A good analyst has opinions on very few things, but those opinions withstand verification pressure.

Consider an uncomfortable fact: a 3,000-word report with every section filled feels more professional than a report saying there is not enough data to conclude. Readers reward completeness, not truth. And that reward creates the incentive to build conclusions out of nothing.

I once called such documents by a name: analysis with a shape but no spine. They are handsome. They are coherent. They are sealed so tightly that not a single gap is left for a question. And precisely because of that, they are useless. A closed spreadsheet is one that cannot learn. A model that never errs is a model that is never tested.

Football taught me this in the most painful way: my model collapsed at Hang Day, collapsed in the Bundesliga's empty-stand season, and I was forced to rewrite myself after each time. The day a model breaks is the day the data monk must burn his scripture and start from the original text. Had I never accepted that collapse, I would have become a seller of certainties the data never provided.

A good bet does not exist; there is only probability that is mispriced and probability that is priced correctly. And the only way to recognize mispricing is to keep the ability to say "I don't know" at the exact moment when everyone around you is pretending to be certain.

I do not predict the future; I only read ahead the way the past still operates.

That is why an empty spreadsheet does not frighten me. What frightens me is a spreadsheet full of values nobody can trace back to a source. In this major-tournament season, as every match becomes a national event, the pressure to build conclusions will only grow. There will be more phantom transfers. More unsourced injury diagnoses. More beautifully formatted but hollow analyses.

My task, and that of anyone doing this work seriously, is not to answer every question. It is to build a habit: check provenance, state the date, name the thing correctly, and when the data is insufficient, leave it empty.

Age 59 gives me a perspective I did not have at 30: every cycle is a loop with a remainder. That remainder is where football exceeds the spreadsheet. But to see the remainder, there must first be an honest spreadsheet. A spreadsheet fabricated to look complete will hide the very remainder it thinks it is explaining.

I still keep the first empty cell from the 2026 spreadsheet, from the Hang Day affair. I never fill it in. Every time I reopen the file, I see it and remember that my craft does not begin with answers. It begins with admitting there are things I do not know, and that admitting this is the first step of any credible analysis.

If a spreadsheet does not dare to stay empty, it is lying about itself. And in a season when millions are waiting for an answer, the real question is not who will be right. The question is who dares to say "insufficient data" — and whether the reader has the patience to hear it.