The Empty Data Table of Vietnamese Esports: 27 Matches, Zero Columns, and the Question No Newsroom Wants to Ask

In a windowless newsroom in District 3, people merged seven data files from seven different sources to build an analytical report for a domestic tournament lasting six weeks. The result, after cleaning, was a table of 27,412 rows. Of the fourteen columns the original plan required, only five were fully populated. Three columns were completely empty. The remaining six had a fill rate below 40 percent. I sat until 2 a.m. checking each row and realised something I had not encountered in seven years of data journalism in Busan: this table was not missing data because people were lazy about recording it. It was missing because nobody had ever asked for that data to be recorded.

That is the starting point of the story I want to tell today. Not about a team winning or losing, not about a highlight play. It is about how the entire Vietnamese esports ecosystem is operating on a data layer so thin that any serious analysis must begin with the same question: what do we have to count?

I am Harper Brown, born in Poland, currently living in Busan and working as a data journalist for the Korean esports market. But for the past three months, I have spent most of my time reading spreadsheets, tournament minutes, and match reports from Vietnam. Not to pass judgement. But to understand why a market with tens of millions of players, thousands of semi-professional athletes, and one of the largest gaming communities in Southeast Asia is entering 2026 with an almost empty data table.

The absence of data is not a technical failure. It is a deliberate choice, repeated long enough to become the habit of an entire industry.

Seven years ago, I sat in a press room in K League 2 and was interrupted by an older male reporter while asking about the pressing stats of the home team's striker. He said, in front of the room, that women know nothing about tactics. The head coach ignored my question. That night I stayed behind to analyse the entire tracking dataset of the match and wrote a 2,000-word piece. The article was shared nearly a thousand times, seven times more than the official match report.

The lesson from that night was not that data is stronger than prejudice. The real lesson was: prejudice can only survive when the data it contradicts is not something anyone is forced to look at. And when the data does not exist, prejudice no longer needs to defend itself, because no one can refute it.

That is exactly what I see in Vietnam.

The good news first. Vietnamese esports has very stable basic indicators when it comes to market size. According to annual industry reports for the Southeast Asia gaming market, Vietnam sits among the five markets with the largest esports player base in the region, and domestic game revenue has maintained double-digit growth for several consecutive years. Major tournaments across League of Legends, Arena of Valor, Teamfight Tactics, Valorant, and PUBG Mobile all draw significant streaming viewership, sometimes exceeding the combined viewership of several traditional sports events in the same broadcast window.

But scale is not data. Scale is a headline. What our spreadsheets need, and do not have, is operational data: per-match individual performance indicators, map-by-map movement distance, engagement frequency per minute, composition-specific win rates, correlation between neutral objective control duration and final outcome. Those are the columns a decent data table needs.

In my 27,412-row file, the three completely empty columns were: "neutral objective control duration per match", "gold difference at the 15-minute map mark", and "number of times the opponent was forced to use a summoner spell in combat". These are not rare metrics. They are standard metrics in every professional-tier tournament in Korea, China, Europe, and North America. They are recorded automatically, published by default, and accessible to any analyst.

In Vietnam, we do not have them. Not because we do not want them. But because nobody has built the infrastructure to record them in the first place.

This is the point I want to make clear before continuing, because I know it can easily be read as criticism: this is not a question of competence. It is a question of priority order. When a tournament has a limited budget, they choose to pay the organising committee, the referees, the broadcast crew, the prize pool. They do not choose to pay a two-person team to sit and type statistics minute by minute. I fully understand that. I have also worked with a budget that could not afford a third camera.

But the price of that choice is paid in something more expensive than money: the ability to understand oneself.

Let me give a concrete example. In an online analysis session I joined with a group of semi-professional players in central Vietnam, we spent three hours arguing about whether a team should commit to a major teamfight at minute 22 or continue controlling the jungle objective. Nobody in the room had data to answer. Finally one person said: "I feel like we win more when we fight hard at minute 20." Nobody refuted him, because nobody had anything to refute him with. The debate ended in feeling, not in evidence.

That is not a debate. That is a storytelling session.

Ha Vi writes football in literary language, and I respect him because he can read the beauty of the defeated. But there is a distance between writing beautifully and writing correctly. I am the kind of person standing on the other side of that distance. Every sentence I write must have a number behind it, or I do not write it.

So the question is: if an esports scene has no numbers, what replaces numbers?

The answer I found after three months of reading documents is: when data is empty, narrative fills the gap, and narrative is never neutral.

The media will choose characters with compelling stories. Teams will be remembered for moments rather than processes. A player with a beautiful solo kill in game three will be mentioned more than someone who held lane for 40 minutes with a consistently positive resource differential. The cognitive map of fans is drawn by bright spots, not contour lines. And the most dangerous part: the players themselves begin to evaluate themselves through those bright spots.

I have followed Vietnamese tournaments over the past four months, based on my own experience watching matches and recording notes manually. Across roughly 60 matches from three different tournaments, I recorded the number of major teamfights per match and their timing. On average, each match had 11.4 major engagements, but the standard deviation reached 5.7, which means the sample is highly non-uniform. That does not necessarily say anything about tournament quality. It says something about data quality: a sample whose standard deviation is nearly half its mean is a sign that we are measuring several different things under the same label.

In other words, when people say "teamfight", each person is defining it differently. And when definitions differ, every number that follows is meaningless.

This is the kind of problem that seven years in Korea taught me to spot quickly. In the LCK, a "major teamfight" is defined by a specific threshold: three or more participants, at least one kill, lasting more than four seconds. That is a definition that can be verified and repeated. Without such a definition, two analysts can watch the same game and produce figures that differ by 30 percent.

Vietnam has not yet established such a standard definition set for most indicators. I do not say this to belittle. I say it to point out that this is one of the few problems that can be solved by an administrative decision, without huge money: it just requires a standard document defining 20 basic indicators, and an agreement among tournament organisers to share it.

So why has nobody done it?

I have a hypothesis, and I present it as a hypothesis, not a conclusion. In many fields in Vietnam, data is seen as a by-product of media coverage, rather than an operational record. That is, data is only produced when someone needs to write an article, rather than being produced continuously as an inevitable part of the competitive process. When data is a by-product, it will have the quality of a by-product: sometimes there, sometimes not, inconsistent quality, and nobody is responsible when it is missing.

Meanwhile, in a mature esports scene, data is a commodity. It has a price. It has an owner. It is bought, sold, licensed, litigated over when misused. That sounds dry, but it is precisely what creates quality. When data has a price, people record it accurately. When it is free and owned by no one, it drifts and rots.

I know someone will say: Vietnam has not reached that stage yet. Perhaps. But the annual season cycle we are now entering is exactly the right moment to talk about this, because it is the period when everything slows down, and people start looking back at the previous season.

And this is where I have to say the hard part.

Over the past two weeks, I have re-read all the analysis pieces I wrote over the past two years. I noticed something uncomfortable: I had many times used data to reinforce a view I had already decided on before looking at the numbers. That is a professional trap every data worker knows, but few admit. People call it many things, and the most comfortable name is "guided intuition". The more accurate name is: choose the winner first, then find the numbers to prove it.

I mention this because it relates directly to Vietnam. When data is complete, people find it hard to lie, because there is always a counter-figure. When data is empty, people can say anything, and what is called "data analysis" is just a form of storytelling with decorative numbers.

This is where I want to mention a concept I began using after the pandemic: after all my prediction models failed consecutively during the 2026 no-audience season, I understood that data does not exist in a vacuum. I had to ask a question I had never asked before: what conditions are governing this dataset?

Applying that question to the Vietnamese context, I noticed something interesting: most existing Vietnamese esports data is created by media, not by tournament organisers. That means it is not data about matches. It is data about how the community perceives matches. And those are two very different things.

A concrete example: peak viewership. This figure is frequently cited as evidence of the growth of Vietnamese esports. But peak viewership measures attention, not competitive quality. It rises when there is drama, when popular teams meet, when the broadcast slot is good. It does not rise when the professional level rises. Using it to talk about professional development is a basic logical error, yet it appears in almost every industry article.

Correlation is not causation. An esports scene can have viewership growth of 200 percent over two years while the average level of teams remains stagnant. Conversely, an esports scene can have viewership flat for several seasons while competitive quality rises steadily, and nobody notices, because no indicator measures that quality.

This is the biggest blind spot in esports media in developing markets, and Vietnam is no exception.

The question left unasked in the press room is the strongest signal I have ever recorded. When a team loses and nobody asks why they lost, the answer usually lies somewhere between the data we do not have and the story we chose to replace it with.

I want to offer three concrete proposals, not because I think they will be implemented immediately, but because I believe a problem only truly exists when someone writes it out as something that can be done.

First, standardise definitions. A 30-page document defining 20 basic indicators, agreed among domestic tournament organisers, could change analysis quality within a year. The cost is close to zero. The barrier is coordination, and coordination is always the hardest thing.

Second, require data recording at the referee level. Referees already have to record minutes for every game. Just add three columns: the end time of major teamfights, the number of kills in the teamfight, and the neutral objectives taken after the teamfight. Those three columns, multiplied by thousands of matches per year, create a data asset that no university could buy.

Third, separate data from media. Tournament organisers should publish raw data as part of their responsibility, independent of whether someone writes about it. When data is an obligation, it will exist. When it is an option, it will disappear by the fourth week of the season.

I know these three proposals sound very dry. No images, no drama, no one to cheer for. But in seven years of work, I have learned that real change in sports does not come from big moments. It comes from small improvements nobody wants to write about, repeated long enough to become the standard.

And this is the part I want to close with, not with a summary, but with a look forward.

The annual season is entering its middle phase. This is when teams begin to form an identity, when fitness is truly tested, when referee controversies emerge, and when the first signals of which teams will really go far begin to appear. As an outside observer, I see some interesting tactical signals that should be measured: the speed of composition rotation between games is clearly increasing, the average time to take the first tower appears to be shortening, and the gold difference at minute 15 between winners and losers appears to be narrowing.

But I say "appears" because I do not have the data to confirm it. And a data journalist without data is just a person guessing with an education.

That is the state I am in, and perhaps the state of an entire esports scene.

What I want readers to take from this article is not a conclusion about the level of Vietnamese esports. I do not have the standing for that, and the data does not allow it either. What I want is a question: in the 27 matches of this season to come, how many will we be able to genuinely analyse with numbers, and how many will we only recount from memory?

I do not predict the shock. I only read the map the rest chose to forget. The problem here is that the map is blank in exactly the squares we need it most.

When the stands are empty, I hear the sigh of data more clearly. In Vietnam right now, the stands are not empty. But the data layer beneath them is nearly silent.

And that silence, if it continues for a few more seasons, will no longer be a technical problem. It will become a cultural feature. By then, the question will no longer be how to record data, but how to persuade an industry used to storytelling that numbers do not make a story dry. They make it more naked, and therefore more true.

For now, I still sit alone each night with my spreadsheet, typing each number into empty cells, as a small way of not surrendering to the silence.