The Data Gap: Vietnamese Esports and the Problem of Counting from Scratch

Late Saturday night, I sat in my rented apartment in Binh Duong, reopening the dataset of a match I had watched live from the stands two days earlier. The pressure index column was empty. The duel count column was empty. The zone-control duration column was empty. I rewound the footage, counted by eye, and realized I was doing exactly the thing I had sworn eighteen years ago I would never do again: counting from zero.

People assume my job is to sit in front of a screen and let the algorithm do everything. The reality is harsher. Most of a data journalist's time in Vietnam is spent searching for data, and frequently failing to find it. That night's match was only a small example. But it represents a much larger problem, one the entire Vietnamese sports scene faces without many people naming it correctly: missing data.

Every time a data column is empty, there are two ways to handle it. The first is to fabricate. The second is to admit it. I choose the second, which is precisely why I am writing this article.

Where the gap begins

When I moved from South Korea to Vietnam to work, the first thing I had to learn was not Vietnamese, but how to work without data. In Seoul, a professional esports match leaves behind thousands of log lines: kill timestamps, gold differentials by the minute, ward placements, cooldown timers. In Vietnam, the same match, the same tournament, but what I receive after the final whistle is usually just a scoreboard and a few lines of commentary.

That gap is not about player skill. It is about recording infrastructure. A sport that wants to be analyzed must first be recorded. Recording is a decision, not an accident. Someone must pay for the person typing out every play, pay for the wide-angle camera, pay for the synchronization software. When there is no money, there is no data. When there is no data, analysis becomes nothing more than gut feeling dressed up with a few numbers.

I cover esports, but I learned my trade from football. In 2026, at twenty-five, I was a reporter for a new football site in Binh Duong. I manually logged data from 182 V-League matches off video, because no source would sell me clean data. I sat from eight in the evening until three in the morning, hitting pause, writing in a notebook, hitting play. That is how I learned that data in this market does not fall from the sky. You have to pick it up by hand.

Among those 182 matches, I found an anomaly. Long An had the lowest PPDA in the league, just 7.8. PPDA is the number of passes the opponent is allowed before your team makes a defensive action. The lower the number, the less pressure you apply. Long An let opponents hold the ball comfortably, yet conceded only 0.7 goals per match, thanks to lightning counterattacks. I wrote an article titled "Low pressing is not cowardice."

A veteran coach called my piece soulless statistics. He said football is a matter of the heart, not of spreadsheets. I did not argue. But weeks later, a young assistant at Binh Duong club invited me to build a pressing map for the team. That debate taught me something I have carried for eighteen years: numbers never lie, we simply have not asked the right question.

Vietnam's problem is not a shortage of intelligent people. It is a shortage of well-posed questions, and a shortage of data to answer them. The two feed each other. Without data, no one dares ask bold questions. Without bold questions, no one sees the need to invest in data. This loop repeats across every discipline, from football to esports.

The lesson of a model named Croatia

In 2026, thanks to my series on V-League data, I was sent as an analysis reporter to the World Cup in Russia. It was the turning point of my life. After the quarterfinals, I sat in a Moscow hotel and built a simple probability model based on xG, expected goals. Croatia averaged 2.3 xG per match. England averaged 1.1. I predicted Croatia would win, even though they had played three consecutive matches into extra time and penalties.

Colleagues laughed. They said football is not mathematics, that Croatia was exhausted, that England had a golden generation. I kept my prediction. Croatia won 2-1 after extra time. My article "Goals from probability" was shared over 10,000 times, and my editor gave me my own column called "Seeing by numbers."

Many people called it a miracle. I called it a well-managed variance. Croatia was not a miracle, but a well-managed variance. They did not score more goals than opponents by some supernatural force. They created more quality chances, and they accepted that in football, the team creating more chances wins more often over time. A single match can deviate from the model. A tournament cannot.

In 2026, I bet my entire career on a probability model named Croatia. I did not dare tell anyone, not even my editor, that if Croatia lost, I would lose all the credibility I had built over a year. But I rechecked the model seventeen times. Each time, the result was the same. I decided to trust the number over my own fear.

The biggest lesson from Russia was not that Croatia won. It was that I had asked the right question: not "which team is stronger," but "which team creates more quality chances in the same number of minutes." When the question is right, the data answers on its own. When the question is wrong, no amount of data saves you from noise.

After that tournament, I shifted from writing narratives to writing with models. I used xG as my main language. But I also recognized a bad habit of mine: I get bored quickly. I started three or four research projects at once, and most never got finished. This is another kind of "missing data," a more dangerous one: missing data caused by myself, for lack of patience to collect it to the end.

PPDA 7.8 and the rebellion in the V-League

Back to Long An's PPDA of 7.8. What made me trust it was not the number, but the event it described. When I reviewed the footage, I saw Long An deliberately ceding possession. They dropped deep, built a low block, then waited for opponents to push up in order to counter. This was a deliberate strategy, not cowardice.

The problem is that data cannot show intent. A stat sheet tells us Long An controlled little of the ball. It does not tell us why. Read only the number, and you conclude Long An played poorly. Read the footage alongside the number, and you see a team that knows its strengths and exploits them with discipline.

This is why I always remind young people in the trade: data does not replace watching the match. Data points to where you should look. It is like a map. A map tells you there is a mountain to the east. But a map does not climb the mountain for you. You still have to walk.

The V-League is a mess, but every mess has its own rules. In this league, I learned that sometimes the underrated teams are playing a smarter game than the celebrated ones. They lack the money to buy stars, so they buy ideas. And ideas, unlike stars, do not decline with age.

The veteran coach's suspicion was not entirely wrong. He worried that statistics would kill intuition. He worried players would play to beautify their numbers rather than to win. That is a legitimate concern, and I have seen it come true elsewhere. When data becomes the goal instead of the tool, it betrays its own user.

Yet I still defend my choice. Between a coach with intuition but no data, and a coach with intuition plus data, I always bet on the second. Intuition is a model not yet written down. Data is the way to write that model down on paper so you can test how right it is.

Heat maps and the new fortune-telling

There is a trend that worries me more and more: heat maps have become the new fortune-telling. Every time a match ends, social media floods with images of red and blue cells showing where a player moved most. People look at them and draw conclusions about form, role, and class.

The problem is that heat maps hide a player's real role in the tactical system. A player roaming across the pitch might simply be chasing the ball, not controlling the game. A player hugging the touchline might be following the coach's instruction to stretch the opponent's shape, not playing inefficiently.

A heat map is a product of positional data, but it is not data. It is a visualization simplified for easy consumption. Like a snapshot, it is pretty, easy to share, but it cannot tell the full story. To understand a player, we need to know where he moved, when, and why.

I once spent three weeks proving this to a group of young analysts. I took positional data for a central midfielder, built a heat map, and asked the group to guess his role. All of them guessed wrong. They thought he was an attacking player because his heat map clustered in the opponent's half. The truth was he was a defensive midfielder, tasked with man-marking the opponent's most creative midfielder, who often operated in that zone.

The lesson here: data without context is more dangerous than no data at all. It creates an illusion of understanding. Readers of heat maps believe they understand the match, when in fact they are looking at a pretty picture and fooling themselves.

This is a variant of the "missing data" problem. Instead of being empty, the data here is misunderstood. Empty is at least honest. Misunderstood is worse, because it leads to wrong conclusions presented with confidence. We think we understand the game, until the data sheet opens our eyes.

342 penalties and the dive to the right

In 2026, at the EURO, I published a study of 342 penalties across five European leagues. I logged the direction of the shooter and the direction of the goalkeeper's dive, classified by dominant foot. The results showed that Gianluigi Donnarumma, Italy's goalkeeper, dove to the right in 72% of cases against right-footed shooters.

It was a suspicious number. 72% is far higher than the 50% we would expect from a keeper choosing a random direction. I rechecked the data three times. The number held. I predicted Italy would beat Spain on penalties. My article was mocked as fortune-telling.

The semifinal came. Italy won 4-2 in the shootout. Donnarumma saved two shots, both to his right. My article reached 1.2 million views. An international sports television channel invited me to be a data expert for the 2026 World Cup.

But I want to tell you the back side of this story. My model did not predict Donnarumma would save two. It only said that if he maintained his rightward tendency, his save probability would be higher than average. There is a large gap between "tendency" and "result." I was right, but I was right because probability leaned my way, not because I saw the future.

This is what I want to stress to anyone reading this: prediction can be verified through research design, but it is not magic. A good model wins more than it loses over time. It does not win every time. Those who misread this turn themselves into fortune-tellers, and then collapse when an adverse result arrives.

After the EURO, I realized I was trapped in another snare. I began writing to seek praise. I wrote articles that looked confident on the outside, while inside I knew my model had many unfilled holes. I forgot to develop a complete model, and instead chased applause. This is when I understood that an analyst can also be a victim of himself.

Empty stands, numbers that are not empty

When the pandemic paralyzed leagues in 2026, I had more free time than at any point in my career. I used it to analyze 252 Bundesliga matches from May to June 2026, matches without spectators. The result stunned me: home win rate fell from 43% to 29%, while away teams ran 6% more.

I tweeted the comparison table. The Analyst shared it, treating it as scientific evidence of home advantage. Thanks to that, I was invited to collaborate with a European data platform. It was one of the proudest moments of my life.

But that pride had a flip side. When the stands are empty, what is lost is not only cheering. What is lost is the invisible pressure on referees, the confidence of home players, the habit of a final sprint when cheered on. All those immeasurable things add up, and they leave traces in the numbers. The applause in empty stands records a truth no one wants to hear.

The deeper lesson was methodological. I learned to use time-series data to understand systemic shocks. When a familiar variable suddenly disappears, the data reveals its true role. This is a technique I brought into esports analysis: when a familiar factor vanishes, look at what changes, because that is the true value of that factor.

I also recognized my inherent weakness. When a topic stops being hot, I abandon it to jump to a new one. I once had four unfinished drafts in a single month, each in a different field. This keeps my expertise from truly deepening. I am good at spotting problems, but poor at staying with them until they are fully solved.

Injury: a gap by design

There is one field I care about deeply, and am especially powerless in: injury and comeback. This is where the data gap is designed deliberately, not from a lack of resources.

When a player is injured, clubs usually announce very little. They say "muscle injury," "ankle injury," or simply "not fit enough." They do not state severity, exact recovery time, or long-term prognosis. This leaves fans and media in the dark.

I understand the reason. Medical confidentiality is a player's right. But I also notice that clubs announce injuries only when announcing benefits them. When a player is being courted, the injury is hidden to preserve value. When a player is about to be sold, the injury is inflated to lower the price. Injury data, when published, often serves a commercial purpose.

This puts analysts like me in an awkward position. I want to predict the impact of a player's return, but I have no data on his recovery level. I only have past matches, and those matches may no longer represent the present.

My way of handling this is to use indirect data. I look at minutes played, substitution frequency, position after returning. A player returning from a serious injury is usually used cautiously for a few matches. If he plays 90 minutes in his first match back, that is a signal. If he comes on in the 70th minute for five straight matches, that is a different signal. These signals do not tell me what the injury was, but they tell me what the club is thinking.

Vietnamese esports: a mess with rules

Now I return to esports, the field I cover for the Vietnamese market. This is where the data gap is both an obstacle and an opportunity. An obstacle because we lack tools. An opportunity because whoever goes first will shape how the whole industry looks at numbers.

Vietnamese esports is booming. Tournaments spring up everywhere, from national to community level. Viewership is rising fast. But data infrastructure lags far behind. While international tournaments already have automated systems logging every in-game event, many domestic tournaments still rely on manual typing, or worse, record nothing at all.

This creates a paradox. Vietnamese fans are increasingly knowledgeable. They watch international tournaments, they read analyses, they know what xG is, they know what KDA is. But when they want to understand their favorite team, they cannot find data. The gap between audience expectations and tournament infrastructure is widening.

I believe this is the industry's biggest opportunity over the next five years. Whoever builds data infrastructure for Vietnamese esports will own a strategic asset. Not because such data is hard to create, but because it requires patience few possess. Building data is like planting a tree. For the first three years, no one sees anything. In the fourth year, you have a forest.

But I also want to warn of a trap. When data becomes an asset, it will be controlled. Teams will keep data to themselves. Tournaments will sell data to the highest bidder. Fans will be pushed outside. I have seen this script in Korea and in Europe. It does not lead to a more transparent sport. It leads to a sport where information is a privilege.

The solution, to me, is to build an open data standard for Vietnamese esports. A standard that every tournament, every team, every analyst can read and write. This is hard work, requiring cooperation among many parties. But if it succeeds, it will be the foundation for everything downstream: analysis, prediction, training, and commerce too.

The economics of data scarcity

Why is data scarce? The simple answer is money. But the real answer is more complex.

The cost of collecting data is an upfront cost. You must pay today to receive benefits over many years. Meanwhile, the cost of a transfer star is also an upfront cost, but its benefit arrives immediately, in ticket and jersey sales. So when forced to choose between two investments, clubs always choose the star. It is a financially rational short-term decision, and a strategically wrong long-term one.

But there is another aspect few mention: data does not just cost money to create, it costs money to maintain. An old dataset, not updated, quickly becomes useless. This means investing in data is not a one-time expense but a long-term commitment. And a long-term commitment is the hardest thing to sell to a board that sits for only two to three years.

This is why I believe change will not come from clubs. It will come from outside: from analytics companies, from media platforms, from people like me. When data becomes a product with buyers, clubs will be forced into the supply chain. They will not invest because they believe in science. They will invest because someone is paying for their data.

But this is also an ethical trap. If data is created mainly to be sold, it will be optimized for selling, not for truth. We will have datasets that are pretty and easy to sell, but lacking in depth. We will have metrics designed to impress, not to explain. I have seen this elsewhere, and I do not want it repeated in Vietnam.

The counter-intuitive angle

After being right with the Croatia model, the line between counter-intuition and arrogance blurred for me. I began to believe I could see what others could not. It is a subtle trap.

Counter-intuition has value only when it explains something intuition missed. It has no value when it merely opposes the majority. A contrarian prediction without a data foundation is not counter-intuition. It is impulsiveness dressed as wisdom.

I learned this through a failure. Once I predicted a weak team would win, simply because I wanted to look different. I had no data, only a feeling that "it would be interesting if it happened." That weak team lost heavily. I wrote an apology to readers, and it was one of the hardest articles of my career.

The lesson: correlation is not causation, and counter-intuition is not truth. A good model must show the mechanism, not just the outcome. If I say Croatia won because its xG was higher, I must explain why higher xG leads to more wins. If I cannot explain it, I am only guessing, and a lucky guess does not make me a prophet.

Signals for the next round

So what happens next? I watch three signals.

First, Vietnamese esports data infrastructure will be shaped over the next two years. Whoever acts first and well will set the standard for the industry. I am closely tracking moves from major domestic tournaments.

Second, the notion of "miracle" will gradually be removed from analytical language. As data becomes common, people will stop calling an upset destiny. They will ask where it came from.

Third, the gap between fans and data will narrow, but unevenly. Some people will understand more deeply than ever. Others will be left behind with pretty but empty pictures.

For me, the work ahead remains the same: ask the right question, collect clean data, and admit when I do not know. A match can be told as an emotional story. But to understand it, we must count. And to count, first, we must have something to count.

That night in Binh Duong, after counting from scratch, I had a dataset logged by my own hand. It was not pretty. It was not complete. But it was honest. And in a market where data is still a luxury, honesty is the most valuable asset a data journalist can own. If tomorrow you open a dataset and see an empty column, do not rush to fill it with imagination. Ask why it is empty. The answer to that question is usually the real story.