Empty Data in the V-League: The Data Monk and the Price of False Confidence

On the night of 12 July 2026, I reopened the folder called VLEAGUE_2017_XG on an old computer.

Inside were 112 V-League 2026 matches, from round one to round fourteen. One spreadsheet per match. Each sheet held hundreds of rows I had typed by hand: shot count, shot location, shooter, the phase that produced the shot, and an xG column I had calculated myself using my own model.

I clicked the source column to cross-check it.

Forty-one links returned a blank page. Nineteen more opened an article whose content had been deleted. The rest were still reachable, but the tables inside had been reformatted, and many of the numbers I had copied no longer matched.

Seven years earlier, I believed I was building a data archive. That night I understood I had built a house without foundations, and the basement was flooding.

I sat still for a long time. Then I did the only thing left to do: I wrote a report in which most of the text read "insufficient information to assess".

It sounds like a failure. It was the most honest report I had produced in years of practice.

The xG shock at Hang Day turned me from a spectator into a reader of data. But it took that night in July 2026 for me to grasp the second half of the sentence: a reader of data must also learn to read the emptiness inside his own data.

In 2026, I lost 180 million dong on a single match at Hang Day Stadium. Hanoi FC took 17 shots and generated 2.87 xG, and the game finished 1-1 against Quang Nam FC, a side with two shots and 0.94 xG. I was furious, and that anger sent me on a seven-month journey: reviewing 112 matches from round one to round fourteen, calculating xG by hand for every shot, standardising a data-collection protocol governed by rules that never changed.

The result stopped me cold. Hanoi FC created more chances than most opponents, but their finishing efficiency ran roughly 23 percent below the league average. My three-thousand-word analysis was mocked by the media. A month later, that same data correctly predicted a run of four consecutive defeats.

Since then, every V-League piece I write carries a table I built myself. I no longer write from highlight reels. I write from rows nobody else bothers to type.

For years I believed my profession was accumulation. Collect more, calculate more carefully, archive longer. That night in July 2026 taught me something else: my profession is maintaining reproducibility.

A number only has value when someone else can come back and check it.

Vietnamese football lives in an environment where most numbers cannot be checked again. The V-League does not publish xG for the public. Clubs do not publish financial statements. Internal transfer fees are almost never recorded in citable documents. Distance covered, pressing volume, duel success — the things a European league puts on its homepage every round — exist here mostly as spoken narrative.

Fans still watch. Journalists still write. Clubs still sign contracts. But when someone asks "which number proves this", the most common answer is a shrug.

Such an environment does not prevent analysis. It changes the final product of analysis.

In 2026, at the World Cup in Russia, I published a forecast that Germany would be eliminated in the group stage. The basis was specific: their average distance covered had fallen 12.3 percent against the 2026 title-winning squad, and their PPDA had risen from 8.2 to 11.7, meaning they allowed opponents more passes before making a first defensive action. I received hundreds of mocking replies. On 27 June 2026, in Kazan, Germany lost 0-2 to South Korea with a total xG of 0.41, and six of their late shots hit defenders.

Kazan does not seek revenge; Kazan simply keeps the ledger and waits for me to miscalculate. I was lucky that time. But I have always known that a model being right does not prove the model is always right. It only proves that, on that particular run, the input data was real.

In 2026, global football stopped for COVID-19. The Bundesliga returned on 16 May 2026 in empty stadiums. I examined 28 matches after the restart: home teams won only five, or 17.8 percent, against a historical home-win rate of 42 percent. My betting model multiplied home advantage by 1.32, and within one week I lost 40 million dong.

I did not fix the model by loosening parameters. I reviewed 200 Bundesliga matches from that season and found a clear pattern: without crowds, home sides still pushed forward out of habit, but real xG fell by an average of 0.45 per match. Within 72 hours I published "Home advantage is gone" and rebuilt my entire context-coefficient system, adjusting for empty stadiums, weather and travel distance.

The crowd leaves, the model breaks, and I learn to listen to the breathing of an empty stand.

But no empty stand is harder to read than an empty dataset.

When I audited my own system, I noticed something anyone working in this profession in Vietnam has seen: most claims about Vietnamese football exist in an untraceable state. An article says team A is stronger than team B because they "controlled the game better". A transfer report says player C will join club D. A comment says coach E has lost the dressing room.

Three claims, three kinds of evidence, and none of them carries a dated source.

So I began running a process I call the nine-dimension audit. It sounds grand, but the substance is simple: for every claim, I ask nine families of questions — tactical, financial, results-based, league landscape, rules and governance, management and dressing room, risk, media narrative, and industry transmission.

If a claim cannot answer any of them, it gets flagged. Not flagged as false. Flagged as not yet assessable.

What I found after months of running that process forced me to rewrite how I work: most Vietnamese football content I read falls into the "not yet assessable" box, and the danger lies in the fact that the writer does not know he is standing inside it.

Take one concrete case. A V-League club changes head coach mid-season. Media write about a "dressing-room bounce". To test that claim, what do I need? Distance covered before and after, PPDA before and after, chances created per match. In a European league those numbers sit in a public database. In the V-League I must sit down and re-code every match by hand, and even then the sample is usually three to five games — far too small to separate signal from noise.

The honest conclusion is: not yet assessable.

Another case. An internal transfer rumour appears saying club X paid a large sum for player Y. The first question is the source. Where did the report come from? Who confirmed it? How much was the fee, and did it include an agent commission or a signing bonus? How long is the contract, is there an automatic extension, is there a sell-on percentage for the previous club?

In Vietnamese football, almost that entire chain of questions collapses into a single answer: neither party disclosed it.

Without contract structure, an analyst cannot distinguish a sound deal from a mispriced one. He can only read the name.

Reading the name is a fan's job. Finding the mispriced probability inside the name is the analyst's job.

There is no such thing as a bargain; there is only probability that is mispriced and probability that is priced correctly.

The same repeats at the financial layer. A club is rumoured to be in trouble. To assess it, I need revenue, wage structure, net debt, payment cycles. Without public accounts, every conclusion is inference from indirect signs — players leaving, contracts not renewed, internal gossip. Inference from indirect signs is not analysis. It is well-founded storytelling.

At the governance layer the picture is slightly clearer, because rules are written documents. Vietnamese professional football has regulations on player registration, club licensing standards and disciplinary procedure. But when a specific case occurs, outside readers usually reach only the final ruling, never the case file. I can read the verdict; I cannot read the investigation. That confines my conclusion to description, not analysis.

At the management layer, everything blurs further. Who actually decides the matchday squad? The head coach, the technical director, or the board? That power model determines how every subsequent result should be read, and it is almost never described publicly.

So what is an analyst to do in such an environment?

Based on my experience tracking matches, the answer is this: bet only on what can be reproduced, and publish whatever cannot.

I sort every claim about Vietnamese football into three tiers.

Tier one covers what I can count myself. Shots, shot locations, minutes played, fouls, substitution timings, long passes, aerial win rate. This is the most solid tier, because I record it and I answer for it.

Tier two covers what I can derive from tier one using a transparent model. xG, xGA, pressing indices, ball progression. This tier is arguable, and the argument is healthy.

Tier three covers what cannot be verified from outside. Transfer fees, wages, internal conflict, boardroom pressure, contract clauses. Every statement in this tier must be explicitly labelled: unverified.

The most common error in Vietnamese sports media is blending tier three into tier one. An internal rumour is written in exactly the same tone as a countable statistic. The reader receives a coherent text, but the reliability inside it is uneven.

Coherence is not evidence of reliability.

I call this phenomenon false confidence. It does not come from a writer deliberately lying. It comes from a writer failing to mark the boundary between what he knows and what he believes.

Over seven years I have built myself an uncomfortable habit: every time I write a declarative sentence, I ask where it comes from, and if the source sits in tier three, I rewrite the sentence.

Many sentences were deleted. Some long articles shrank by half.

But reread years later, the sentences that survived still stand.

I do not predict the future; I only pre-read the way the past keeps operating. That is why I return to the old lesson: the nine-dimension audit is not a ritual, it is a brake. It does not make me go faster. It stops me driving off a cliff.

At 59, looking back at seven years of hand-coding V-League matches, I see a paradox.

The more I love data, the less data I am allowed to print.

I once thought my value lay in the volume of tables I produced. Now I believe it lies in knowing which table should never be drawn.

A dense table can conceal an empty dataset. An empty table, with its reason stated, is itself real data.

The day the model breaks is the day the data monk must burn the original scripture and start again. That night in July 2026, my original scripture had one line left: never pretend to hold something you do not hold.

Now look forward seriously.

Vietnamese football is entering a major tournament cycle with the weight of national-team expectations, and that weight will press down on every club in the system. As expectations rise, demand for answers rises with them. And when demand for answers grows faster than the supply of evidence, the market manufactures cheap answers.

That is when rumour becomes a currency.

It is also the moment the serious analyst holds the greatest advantage, because his edge lies in saying "not enough data" while everyone else must say something.

Here is a high-probability judgement for the next cycle: over the coming 12 to 18 months, the value of a Vietnamese football analyst will not lie in being right, but in the share of forecasts he has written down. Whoever records his predictions with dates owns the material to learn from. Whoever does not will be right forever.

A second judgement, lower probability but worth tracking: the first V-League club to publish internal performance metrics on a regular schedule will gain a recruitment and player-valuation edge within three seasons. Football is an information market with imperfect flow, and anyone who narrows the asymmetry sells product.

If that never happens, I keep doing the old work. I still type every match by hand. I still keep every spreadsheet in two places. I still date every page.

That evening I saved the empty report into the folder called DOCUMENTS. Before closing the machine, I typed one extra warning line at the top of the file, in capitals: EMPTY INPUT — DO NOT GENERATE CONCLUSIONS.

That line solved nothing. It only guaranteed that the next person who opened the file would know exactly what he was opening.

In a football culture where most numbers cannot be traced, that is probably the most honest gift an analyst can leave behind.