When the Data Pipeline Returns Zero: Vietnam's Sports Analytics Profession and the Temptation to Fabricate

A night in Da Nang, and my computer screen glowed a single shade of gray. I had just finished running the extraction code for a transfer-market report, and the result that came back was an empty frame. No team name. No player name. No patch. No date. Just a row of dashes sitting next to each other like tiny graves, each one bearing the same line: insufficient information.

A layperson looking at it would assume my system had broken. They would not be wrong. But what they would not understand is this: the moment a data pipeline returns zero is the most honest moment in the entire process. Because everything else — every beautiful chart, every smooth ranking table, every loudly confident transfer prediction — can be the product of a machine fooling itself.

The greatest lesson in sports data analysis is not how to find a number, but how to accept that sometimes there is no number to find.

That is the story I want to tell here. Not a match. Not a contract. But the story of the gap — the thing that Vietnamese sport, from football to esports, wrestles with every day yet rarely dares to name.

CONTEXT: A LOUD SPORTING NATION WITH A WHISPERING DATA FOUNDATION

Talk about Vietnamese sport and people talk about emotion. The My Dinh stands erupting. Red flags with yellow stars covering street corners after every national-team victory. V.League nights with drums, horns, and social media flooded with status updates. That is the noisy part, the surface, the part everyone sees.

Beneath that surface lies something entirely different: a data infrastructure so thin it is genuinely worrying. I am not talking about goals or yellow cards, the things already printed on the scoreboard. I am talking about a deeper layer — touches, distance covered, long-ball completion rate, xG (expected goals), PPDA (passes allowed per defensive action). These are metrics that developed football nations have treated as a minimum standard for over a decade, while in Vietnam they remain a luxury.

The same pattern repeats in both football and esports. The field I have been attached to since the early days of my career — first as a player, then as a tournament organiser — sits in the same trap. Tournaments have sponsors, have online audiences, have matches watched by hundreds of thousands, yet detailed performance data on each competitor, on the impact of each patch on the meta, on pick-ban rates, often exists only scattered across a few personal spreadsheets, a few screenshots, or the heads of veterans.

That gap between the noise and the data is the root of many problems. When people lack data, they tell stories. When people lack numbers, they use adjectives. And when a sport runs on adjectives, fabrication stops being fraud — it becomes reflex.

I lived inside that reflex for years, and I know how dangerous it is. People say a player is good, but good compared to whom, under what circumstances, under what pressure. Nobody answers. People say a team is in crisis, but a crisis of tactics, of fitness, or of psychology. Nobody distinguishes. Big words are used to fill small gaps, and the price is a sport that never quite knows where it stands.

THE NHA TRANG STAND AND THE FIRST MILESTONE

I will tell a personal story, because how I came into this profession explains how I see data today.

That year I sat in the stand at Nha Trang, notebook in hand, eyes fixed on every phase of play. I counted. Tran Bao Toan had fourteen successful tackles, twenty-three ball recoveries, and lost the ball only six times against U19 Myanmar. Those were numbers I recorded myself, with nobody asking and nobody paying. I sat among a crowd that was roaring, and I did not roar. I counted.

What was strange was that I did not need to wait for a goal to see his value. A goal is the reward of a moment; fourteen tackles are the evidence of a process. And in football, process is what predicts the future, while a moment only recounts the past.

I called an editor at a sports paper. I proposed a piece that dissected the numbers. He agreed to meet but did not promise to publish. A week later I sent the draft along with a stat table I had compiled by hand. Not the old style of commentary — no lines like he plays well, he is full of promise — only numbers tied to specific actions on the pitch.

The Nha Trang stand has no wifi, but every number there smells of real sweat.

That article did not put me on the front page. It did something bigger: it forced me to stop writing subjective lines. From that day, every quality I assigned to a player had to come with a number. I learned data entry, learned to draw charts, and began building my own database. Every later piece started from a measurement question, not from an emotion on the pitch.

From the Nha Trang stand to the transfer price sheet, the road is longer than a season.

On that road I learned the first lesson, and the most important one: data is not born from enthusiasm. It is born from discipline, from accepting that you sit still and count even when the whole stand is screaming. A data person must be able to separate themselves from the crowd — not out of contempt for the crowd, but out of a wish to serve it with a truth that emotion cannot see.

THE NIGHT GERMANY COLLAPSED AND THE TRAP OF THE EASY EXPLANATION

If the Nha Trang stand taught me how to count, the 2026 World Cup taught me how to doubt ready-packaged explanations.

The Germany versus South Korea match kept me awake all night. On television, commentators repeated one phrase: Germany has run out of luck. It sounded grand, literary, and completely useless. My stat table showed a different picture. Germany generated 2.14 xG but produced only three shots inside the box after the 60th minute. South Korea had 0.82 xG, yet scored in the 90+3rd minute from a counterattack worth just 0.18 xG.

Read those two lines side by side and the run-out-of-luck story collapses instantly. Germany did not lose to luck. Germany lost by betting on the wrong zone. They controlled the ball, they created chances on the edge of the box, but when they needed a thrust through the middle they had no ideas left. South Korea needed only one moment in the right place, in the exact zone their opponent had left empty all second half.

I sent the piece to the newsroom. Two days passed with no reply. I published it myself on my personal blog, with one emphasised line: there is no running out of luck, only betting on the wrong zone. The piece was shared ten thousand times, and it brought an invitation to collaborate from a collective specialising in tactical analysis.

The night Germany collapsed, I understood: the championship formula is always missing a variable called collapse.

But there is one thing I must make clear, because I myself nearly fell into the trap. The phrase collapse variable easily becomes a hammer for every nail. After that night, I saw many people — including a younger version of myself — start assigning the word collapse to every defeat, every missed prediction, every surprise result. That is laziness disguised as depth.

The rule I set for myself from then on: before mentioning the collapse variable, you must point to at least one metric showing the team bet on the wrong zone. No number, no collapse. No evidence, no conclusion. This is what I believe Vietnamese sports analysis must take to heart, because we have a very strong tradition of storytelling and a very weak tradition of verification.

THE 2026 PANDEMIC: WHEN THE PITCHES CLOSED, A LIBRARY OPENED

Then the 2026 pandemic hit. Every league froze. No crowds, no drums, no V.League nights. For many people it was a deadly void. For me it was a forced holiday I had never dared to ask for.

In the 2026 pandemic season, I built a valuation model for Vietnamese players out of matches played without crowds.

I collected data from 240 V.League 2026 matches via an Opta account I had obtained through a connection made at the 2026 World Cup. I built a valuation model based on age, minutes played, xG, distance covered, and long-ball rate. There was no laboratory here — only a man in front of a screen, a cold cup of coffee, and the belief that data can listen to the past in order to speak about the future.

The model surfaced something that made me read it three times: Nguyen Quang Hai was undervalued by roughly 40 percent against expectations, because he produced 0.31 xG-assisted per 90 minutes — a figure on par with leading imports. I published the report on social media. It sparked fierce debate and led to a job offer from a sports analytics company.

Covid closed every pitch, but it opened for me a data library I had never dared to dream of.

What I learned from that pandemic season was not in the model. It was in the market's reaction. Many people opposed my conclusion, but almost nobody opposed it with numbers. They opposed it with feelings. They said Quang Hai cannot be that expensive, or he has already been valued correctly. But when I asked in return — correctly according to what — the answer was usually silence.

That was the moment I realised something about the Vietnamese sports market: we are very good at arguing about conclusions and very bad at arguing about methods. People are ready to say this number is wrong without needing to know how the number was calculated. And once method is off the table, any number can be replaced by a prejudice.

I began to give that phenomenon its own name: the conclusion-first syndrome. In this syndrome, people already hold the answer in their heads, then go looking for data to defend it. Data stops being a tool for discovering truth and becomes a weapon for defending belief. And once data becomes a weapon, it can be bent to the wielder's will.

THE DONNARUMMA DEAL: WHEN THE MODEL SPOKE BEFORE THE MARKET

In 2026 I was a new employee at a transfer company. Euro 2026 took place after a year's postponement due to the pandemic. I was tracking Gianluigi Donnarumma, a goalkeeper about to run down his contract with AC Milan.

My model showed his post-shot save rate versus expectation at +4.1, top of the tournament. I told my boss that PSG would sign him before 15 July. Four weeks after the final, PSG announced the deal. Agents began sending me player files for my team to assess, because they knew I had a model.

But this is where I want to pause, because this story is often told wrongly. It is not the story of a man who predicted correctly. It is the story of a method that withstands verification. The +4.1 rate is a measurement, not a prophecy. And a measurement only has value when we admit it can be wrong.

My model is not perfect, but it listens to the past, which many experts fail to do.

From then on I wrote in throughput form: age, starting minutes, post-shot xG, distance covered, and only then a conclusion about value. I focused on the group of players nearing contract expiry, where data genuinely creates a competitive edge. The transfer market is where people sell the past, but whoever keeps a clear head buys the future with data.

THE TURNING POINT: THE GRAY SCREEN RETURNS

Everything above leads to the gray-screen moment I described at the start.

For years I built a career on the belief that data is the answer. More data, better. More metrics, more precise. But one day I realised that belief has a fatal flaw, and the flaw only shows itself when the pipeline returns zero.

Imagine the workflow of a sports data analyst. You have a source. You run extraction code. The result is an empty frame: no team name, no player name, no patch, no date. Every cell simply reads insufficient information.

There are two ways to react. The first is to hunt for the fault: check the pipeline, check the source, run it again. The second is to fill the gap with guesswork. The second is far more dangerous, because it leaves no trace. A fabricated number looks identical to a real one. A name inferred from context looks identical to a verified name.

Numbers never lie; they only wait patiently while you lie to yourself.

This is what I want to say plainly: in sports analysis, the greatest enemy is not missing data. The greatest enemy is the willingness to fill gaps with whatever sounds plausible. People do not fabricate numbers because they are bad. They fabricate numbers because they are under pressure to have an answer. And that pressure, in a sporting nation where everyone wants to appear knowledgeable, is enormous.

THE PARADOX: EMPTY DATA IS A GIFT

Here I will go against the intuition of the majority.

The popular belief in data circles is: more is better. More metrics, more sources, more charts. But in reality, a complete but dirty dataset is more dangerous than an empty but clean one. Because an empty dataset forces you to stop. A dirty dataset hands you a result that looks very convincing, and you will never know you were wrong until reality slaps you in the face.

When my pipeline returns zero, what is really happening is that the system is confessing. It is saying: I have nothing to say about this. And in a field where everyone wants something to say, that confession is a rare act of honesty.

Correlation is not causation, and an empty dataset is the most valuable reminder of that.

Take a concrete example. Suppose you discover that in the last ten matches, Team A won eight whenever Player X touched the ball more than seventy times. You could conclude: Player X is the key to victory. But look closer and you may find the opposite: Team A controlled the ball well in matches they led, and when leading, Player X had more of the ball. Player X did not create the wins; the wins created Player X. This is the trap every sports data person faces, and it only shows itself when you are brave enough to look at what you do not know.

In basketball the trap is subtler still. A player can have a very high plus-minus, but that figure depends on who he plays with, who he plays against, and for how many minutes. If you cannot control those variables, plus-minus becomes a compliment disguised as a number. I have seen reports praising a player on plus-minus without mentioning that he only played in garbage time, when the opponent had already let go.

In esports the problem is even more serious. A patch is an invisible referee with the power to decide a championship. A team can win not because it is the strongest, but because it adapts fastest to a change nobody anticipated. When you lack patch data — champion win rates, match duration, lane-priority tempo — every conclusion about true strength becomes an illusion. You are praising meta adaptability and calling it talent.

FROM LABORATORY TO MARKETPLACE: VIETNAMESE-STYLE VALUATION

I must tell one more story, because it explains why empty data is especially dangerous in Vietnam.

In developed markets, transfer prices form from a complex mix: statistical models, sponsorship contracts, age, commercial potential, and pressure from rivals. In Vietnam, many deals are still priced by a spoken formula: look at each other, negotiate, then add thirty percent for affinity.

Vietnamese-style valuation: look at each other, negotiate, then add thirty percent for affinity.

I say this not to mock. I say it to point out that when valuation relies on relationships rather than data, the data gap gets filled with sentiment. And sentiment, in Vietnamese football, is usually given a lovely name: affinity. This player has affinity with the club. That coach has affinity with the title. But affinity is a word that cannot be measured, and once it cannot be measured, it becomes a curtain hiding every decision without a basis.

This is where the role of the data person becomes more important than ever. Not to replace intuition — the intuition of veterans is a precious asset. But to give intuition a system of checks. So that when someone says this player has affinity, we can ask: affinity measured by what? By xG-assisted per 90? By chance-conversion rate? By chances created for teammates?

The transfer market is where people sell the past, but whoever keeps a clear head buys the future with data.

There is a paradox I often think about. The fiercest opponents of data are usually those who have never tried to build a model. They imagine data as a cold machine wanting to replace humans. But in reality, a good model is built by humans, runs on human assumptions, and always has places where it must bow to what it does not know. A model does not replace intuition; it forces intuition to answer.

ESPORTS: WHEN THE PATCH ACTS AS REFEREE

I want to give esports its own section, because this is where the data gap causes the clearest damage.

In traditional sport, the rules are stable. A football match in 2026 and one in 2026 still have the same number of players, the same pitch size, the same basic principles. In esports, the rules change constantly. Each patch can wipe out one strategy, revive another, and reshape the entire way the game is played within weeks.

That means data in esports has a very short lifespan. A figure correct on the previous patch can become meaningless on the next. And this is precisely where the temptation to fabricate is strongest, because people need new data, but the new data has not yet formed.

I once witnessed a textbook case. A team won several matches in a row after a new patch launched and was immediately hailed as the strongest in the region. But when I looked at pick-ban data, I saw something else: the team was simply exploiting a champion that had been over-buffed in the patch, and once that champion was nerfed in the next patch, they returned to their true position. Nobody fabricated numbers here. But everybody ignored one variable, and that variable decided everything.

That is why I believe that in esports, the skill of reading a patch matters as much as the skill of playing. A champion team may simply be the team that reacted fastest to a change. And if we have no data to distinguish true strength from adaptation, then every championship carries a little mystery that cannot be explained.

REFEREES, VAR, AND FAKE TRANSPARENCY

There is another domain where empty data has a direct impact on the emotions of millions: refereeing.

I watch a great many matches, and what I have realised is that fans do not react violently to wrong decisions. They react violently to unexplained decisions. A referee can blow wrongly, but if he can explain why, fans will accept it, even while disagreeing. What enrages them is silence.

VAR was born with a promise of transparency. But in many cases it delivered something else: fake transparency. People see the referee run to the monitor, people see the lines drawn, but people do not hear the reason. And once the reason is unspoken, data — however abundant — cannot fill the gap of trust.

This is a perfect example of my argument. Missing data is not the problem. The problem is data that is not turned into explanation. A decision without explanation is a decision that cannot be verified, and a decision that cannot be verified can always be doubted. In sport, where trust is everything, doubt spreads faster than any goal.

I do not believe VAR is wrong. I believe a transparency tool without an on-pitch explanation mechanism is forgetting the most important party: the fans. And when fans are forgotten, every argument about refereeing will never end.

WHAT READERS SHOULD DO WITH ALL OF THIS

I am not writing this to praise myself. I am writing to warn, and to share a principle I believe is central to this profession.

When you read a piece of sports analysis — any piece, including mine — ask yourself three questions. First, where does this number come from? Second, how was it measured? Third, what would make it wrong? If the writer cannot answer those three, you are reading an essay, not an analysis.

And if you are a data person, here is a private note. Do not fear the gap. The gap is your most honest friend. When the model returns insufficient information, write exactly that. Do not turn a dash into a number just because you want your piece to look fuller.

I used to think a good analyst was the one with the most answers. Now I think differently. A good analyst is the one who knows precisely what he does not know, and dares to say so. Because in a field where everyone wants to appear knowledgeable, the one who dares to admit emptiness is the only one not lying to himself.

WHAT I AM TRACKING IN THE COMING ROUNDS

The regular season is now in the phase where everything is easiest to disguise. The table says little. Form is not yet stable. And the arguments about refereeing, about VAR, about decisions left unexplained on the pitch, are simmering beneath the surface.

This is when data becomes most valuable, and also when data is easiest to fabricate. Because when nothing is clear, people need stories to fill the ambiguity. And when they need stories, they are ready to invent numbers.

Three signals I will track in the coming rounds. First, the change in PPDA at clubs under relegation pressure — when a team reduces the passes it allows before each defensive action, that usually signals a change of approach, often a deeper defensive block. Second, the gap between xG and actual goals at teams on a hot streak — if they outscore xG persistently, it is either skill or luck about to run dry. Third, how esports teams adapt to the new patch — because in esports a patch does not just change the game, it changes who the best players are.

A CLOSING, ON MY OWN TERMS

I began this piece with a gray screen and an empty data frame. I want to end it with a thought about the future, not a summary.

Vietnamese sport stands at a fork. One path is easy: keep telling stories, keep using adjectives, keep filling gaps with whatever sounds plausible. The other is hard: build data infrastructure, accept that some questions have no answers yet, and measure patiently until they do.

The second path is longer, more expensive, and less celebrated. But it is the only path to a sport that can walk onto the world stage with confidence, without leaning on inflated numbers.

I still keep the old habit: taking notes every day. Every match, a few lines. Sometimes I look back at old pages and see numbers I was once proud of, now just traces of a man learning not to lie to himself.

Numbers never lie. They only wait patiently while you lie to yourself. And the task of this profession, in the end, is not to produce as many numbers as possible. It is to build a system honest enough that, when there is nothing to say, it dares to say nothing.