The Empty Report: How Vietnamese Esports Analytics Is Learning to Say 'I Don't Know

In a small room in District 3, Ho Chi Minh City, a spreadsheet opens with 47 rows. The metric column is empty. The source column is empty. The conclusion column is empty. The analyst sitting in front of the screen for two hours writes exactly one short line in the notes: insufficient data to conclude. The report is sent out as-is, not a single number dressed up.

To most people in the industry, that is a failure to be hidden. To me, it is the first time in years that a Vietnamese esports analysis process has dared to admit it does not know. When the crowd looks up at the bright screen, I dig beneath the dust of old data — and sometimes that dust is so thick that the most honest thing I can do is tell the reader I have found nothing.

An empty report is not a failed report. It is proof that the process was honest enough to refuse to fabricate a conclusion.

To understand why this matters so much, one has to look at the backdrop against which it happens: a Vietnamese esports scene growing faster than the data infrastructure it is building for itself.

The Vietnam Championship Series — Vietnam's largest League of Legends competition — has passed through many seasons with a franchised structure, a promotion system, and an online audience among the highest in Southeast Asia. GAM Esports has twice reached the group stage of the World Championship, while Team Flash and SBTC Esports have contested the domestic title for years. In other titles, Vietnam's national teams in Valorant, PUBG Mobile and Arena of Valor have all left their mark. But behind each of those matches, the volume of data recorded, standardized and publicly released remains thin.

Based on my years of watching matches, a high-level VCS game is usually captured only through basic metrics: kills, gold, damage, win rate. The metrics that describe the process — the jungler's path in the first 10 minutes, the support's off-ball movement, the number of roaming pressure plays — are hardly ever stored systematically. Meanwhile, in leagues such as the LPL or LDL in China, a single match can generate tens of thousands of raw data points, labeled, cross-checked and published to a standard any analyst can retrieve.

That gap is not about people. It is about infrastructure. And weak data infrastructure creates a very particular kind of pressure: the pressure to fill the void with anything that can fill it, even if that thing does not exist.

In the analysis trade, this has a dry technical name: null-value handling. The basic principle is that when a data field has no value, the analyst must mark it as insufficient information rather than infer. It sounds obvious. In practice, it is one of the hardest tests of the profession, because the entire market incentive — from newsrooms to sponsors to search algorithms — rewards the person who delivers a decisive conclusion, not the person who admits a lack of data.

An analysis concluding that Team A will win gets shared more than an analysis saying there is not enough data to assess Team A. A punchy headline gets more clicks than a sourced table. And that pressure is the perfect breeding ground for an occupational disease: unconscious fabrication.

I first saw it when I was very young. In 2026, at 16, I sat in the stands of a Shenzhen FC training pitch to watch an internal U16 match. Midfielder Lin Chen did not score. But over 60 minutes I counted 47 accurate passes and 11 ball recoveries in his own half. I wrote it by hand in a black notebook and did not rush to conclude. Instead, I built an evaluation framework of six metrics: off-ball movement, situational reading, recovery pressure, long-pass accuracy, processing speed and risk-avoidance index.

What I did not do mattered as much as what I did. I did not write that Lin Chen was a generational talent. I did not write that he would shine. I had six metrics and a notebook. Two months later, Lin Chen was sold by the club to a second-tier side. I was not surprised, because my data had never said he was a star — it only said he had a skill structure good enough not to be forgotten.

The lesson here is not about predicting correctly. The lesson is that the distance between raw data and conclusion must always be kept intact, never flattened. When writers flatten that distance — when they fill the empty cell with a feeling — they are no longer analyzing, they are selling a belief.

In 2026, when the pandemic froze all youth competitions, I lost my source of direct observation. No more matches to watch. But I still had one thing: a historical database. I spent months excavating 9,212 player records from 14 academies across Asia. In them, I found a very specific correlation: players with more than 1,800 minutes played at U19 level before turning 18 had a success rate after three years 2.3 times higher than the rest.

The 1,800-minute figure is not a prophecy. It is an observation threshold — and like every observation threshold, it is only true within the data that produced it.

I built a model and called it the excavation score. But that model only had value when I wrote two things alongside it: data limits and confidence. I learned this after realizing that a model without a limits section is a model that cannot be wrong — and anything that cannot be wrong cannot be right either.

That is also why I always look for a counter-arguer. In 2026, I found a data analyst in Beijing — someone who does not enjoy watching football, only numbers. We refined the model remotely. Having someone willing to say "you are inferring too far here" matters more than any algorithm.

But even with a counter-arguer, I still made mistakes. In December 2026, while interning at a sports data center in Shenzhen, I followed the smaller teams in the World Cup group stage. I noticed that young Uruguayan defender Enzo Martínez, a product of the Defensor Sporting academy, had an abnormal running gait: his left-leg push-off force was about 18% lower than his right. In injury analysis, that is a sign of a latent, unexposed hamstring issue.

I wrote a report predicting he had a high injury risk over the next six months, with a proposed recovery plan. But because I wanted it perfect, I held the draft for two weeks to re-check the charts. During those two weeks, a colleague spotted a similar issue and put it on the club's website. By the time my report was finally done, it was no longer exclusive. It was just a late confirmation.

I drew a lesson I still repeat to myself: being right but late is still wrong. And from then on, I broke every article into easily updatable parts, always published a preview with a "pending confirmation" note, and set a closing deadline for every project.

But there is one thing I have never changed, even under time pressure: I never fill an empty cell with a number that has no source. My perfectionism has a deadline — but my data honesty does not.

This is the point where I think Vietnamese esports stands at a fork. On one hand, growth pressure demands speed: audiences want content now, sponsors want reports now, platforms want numbers now. On the other, the data infrastructure is not yet thick enough to meet that speed honestly. When those two pressures collide, the first thing usually sacrificed is the source note.

I have seen this in many markets. When an emerging esports scene starts, people begin with simple numbers: kills, gold, wins and losses. Then, as audiences get used to data, they want more. Then, when there is no more to give, they start to infer. And inference, if unchecked, slides gradually into fabrication. This slide does not happen overnight. It happens one data cell at a time.

In China, where I work now, the esports data ecosystem has been through this phase and partly moved past it. Not because the people there are smarter, but because the infrastructure there is thick enough that analysts do not have to fabricate. When you have 300 minutes of labeled video per match, you do not need to imagine a trend — you can count it.

That is why I always tell young people in Vietnam who want to do analysis: do not start with a conclusion. Start by counting. Watch the 300 minutes of footage nobody watches. Record the things that do not appear in the scoreboard. And when you find nothing, write that you found nothing.

There are no miracles on the pitch, only fragments assembled before anyone else sees them. And sometimes the most important fragment is the missing one.

Here, I want to speak plainly about an issue few in the industry want to touch: the relationship between public analysis and betting data. Live data supplied to betting companies is the darkest side effect of the digitization of sport. When an analyst publishes a prediction without clearly citing data sources, what he is really supplying is not knowledge — it is a signal for others to bet on. This is why source-citing is not just a professional-ethics rule. It is a defensive wall.

A sourced prediction can be checked, challenged, corrected. An unsourced prediction can only be believed or doubted — and neither produces knowledge.

In football, I once analyzed how the five-substitution rule changes match structure. The right to five subs gives deep squads a clear advantage, but at the same time turns the final 20 minutes into a war of attrition in fitness and squad management. The interesting thing is that both conclusions are true — and they are only true when measured with the specific data of a given league, a given team. If someone just says "the five-sub rule changes football" without offering a single number, that is not analysis. It is an exclamation dressed up in jargon.

That style of writing is spreading in Vietnamese esports. People talk about "meta", "systems", "tactical philosophy" without pointing to a single metric. Big words are used to cover the emptiness of the data beneath. And as readers get used to it, they gradually lose the ability to distinguish a real analysis from a piece of prose arranged to sound plausible.

This is where the counterintuitive angle comes in.

The esports analysis industry, in Vietnam as elsewhere, is often built on a silent assumption: an analyst's value lies in delivering conclusions. More conclusions, more decisive ones, more value. But that assumption has reversed the truth. An analyst's real value does not lie in the conclusions he delivers, but in his ability to distinguish what is data from what is inference — and to dare to state that boundary.

By this standard, an empty report is worth more than a report stuffed with conclusions but no sources. Because the empty report protects the reader from a false belief. The source-less, conclusion-filled report protects no one — it only protects the writer's ego.

An empty pitch is not a stopping point, but a new stratum to excavate. A league without data is not a league with nothing to say. It is a league for which we have not yet built the shovel to dig. And admitting that is the first step to building the shovel.

In the years ahead, I believe Vietnamese esports will face a clearer choice than ever. Either it invests in data infrastructure — recording more, standardizing better, publishing more openly — so analysts have real material to work with. Or it keeps letting growth pressure fill the void with unsourced conclusions, and gradually turns the entire analysis trade into a contest of who shouts louder.

I do not know for sure which path Vietnamese esports will choose. And I will not pretend that I do. But I know one thing: if it chooses the second path, the young people now learning to count every pass and log every minute of play will be the first to walk away. Because no one wants to do archaeology in a museum someone has filled with fake statues.

Every prophecy lies in the stratum the crowd rushes past. But a stratum only has value when it is real. And keeping it real — that is the work of those who choose to count instead of to shout.

I do not drill into the moment; I drill into the sedimentation process of a talent. Likewise, I do not judge an esports scene by its dazzling moments on the big screen, but by what it dares to record when no one is watching. A mature esports scene is not measured by the number of viewers in a final. It is measured by the number of data cells it dares to leave empty, instead of filling them with a fabricated number.