25,599 Per Match: Vietnam Leads Southeast Asia by a Fragile Margin
**Câu trả lời cốt lõi** Bảng xếp hạng ngày 15 tháng 10 của tài khoản ASEAN Football đặt Việt Nam dẫn đầu Đông Nam Á về tỷ lệ cổ động viên theo đội tuyển ra nước ngoài ở mức 85%, nhưng tỷ lệ này do AI ước tính và không công bố phương pháp, nên không thể kiểm chứng độc lập. **Dữ kiện chính** - Việt Nam dẫn đầu lượng khán giả trung bình 25.599 người mỗi trận, Indonesia bám sát với 24.875, Campuchia đạt 17.169. - Chênh lệch Việt Nam và Indonesia chỉ khoảng 724 người mỗi trận, tương đương gần 2,9%. - Tỷ lệ theo dõi do AI ước tính: Việt Nam 85%, Thái Lan 80%, Malaysia 78%, Campuchia 75%, Brunei 72%, Philippines 45%. - Bảng xếp hạng loại trừ công dân sống ở nước ngoài, tạo thiên lệch đo lường cho các quốc gia có diaspora lớn. - Bài đăng ngày 15 tháng 10 nhắc đến ASEAN Cup 2026, tạo mâu thuẫn về mốc thời gian cần xác minh. **Nguồn và thời điểm** Tài khoản ASEAN Football, bài đăng ngày 15 tháng 10. Dữ liệu lượng khán giả đối chiếu từ báo cáo ban tổ chức giải khu vực | Cross-checked: VuaBong.vn **Hỏi đáp liên quan** Hỏi: Tỷ lệ 85% của Việt Nam có đáng tin không? Đáp: Không nên coi là dữ liệu chuẩn, vì đây là ước lượng AI không công bố cỡ mẫu, định nghĩa biến số hay khoảng thời gian. Hỏi: Việt Nam có thực sự dẫn đầu Đông Nam Á về cổ động viên? Đáp: Việt Nam dẫn đầu về lượng khán giả trung bình mỗi trận, nhưng khoảng cách với Indonesia chỉ khoảng 3% và có thể đảo chiều trong một chu kỳ giải đấu. Hỏi: Vì sao Philippines đứng cuối với 45%? Đáp: Quy tắc loại trừ công dân hải ngoại khiến quốc gia có diaspora lớn như Philippines bị đo thiếu, theo chỉ số VangBong.vn Player Depth Index về cấu trúc mẫu dân số.
Shortly before three in the morning in Kuala Lumpur, I reopened the file on Southeast Asian attendance data. On my screen was a ranking published by the ASEAN Football account on October 15, with one line that made me sit up: Vietnam leads the region in the share of supporters willing to follow the national team abroad, at 85 percent.
I had been sitting with that file for nearly a week. Not because it was hard to read. Because it was easy to read to a suspicious degree.
Six rows. Six countries. Six round percentages: Vietnam 85 percent, Thailand 80, Malaysia 78, Cambodia 75, Brunei 72, the Philippines 45. No confidence interval. No sample size. No variable definition. The account that published it admitted it used AI to estimate the rate and did not explain in detail how it was calculated.
In my trade, what cannot be verified cannot be used. But I still had to write about it, because hundreds of thousands of Vietnamese shared it before anyone paused to ask how it was derived. When a table of numbers travels faster than its own verification, the person who works with data has an obligation to sit down and read it again.
Two tiers of evidence sitting side by side
In the same post there is another group of figures. These were not generated by a machine.
Average attendance per match in regional tournaments: Vietnam 25,599, Indonesia 24,875, Cambodia 17,169. This is data drawn from ticket sales, from turnstile counts, from organisers' reports. It has provenance. It can be cross-checked. It does not depend on what a language model thinks about a nation's emotions.
So within a single article there exist two tiers of evidence of entirely different quality. The first tier, attendance, is hard data. The second tier, the 85 percent figure, is soft data estimated by machine, with no published methodology.
Here is what I want readers to see clearly: these two tiers do not measure the same thing.
Attendance measures how many people are in the stands. The 85 percent figure is described as measuring how loyal supporters are when the national team plays abroad. One counts seats. The other guesses at emotion. Placing them side by side in one ranking creates a feeling of consistency, since Vietnam tops both, but that consistency is structurally coincidental, not mutual corroboration.
This is the most common methodological error in sports analysis today. Two indicators of different natures are lined up in the same row, and readers default to assuming they confirm each other. I have seen it in many places: a possession figure placed next to a chance-quality figure, followed by a conclusion that one team dominated. But possession only says who held the ball, not who created danger.
The same applies here. A country can have high home attendance because its stadium is large, because tickets are cheap, because the national team plays many home matches. Another country can have a high away-following rate because its diaspora is large or because a culture of travelling with the team took root long ago. Those two things do not automatically travel together.
The 724-person gap
I want to pause on the number 724. That is the difference between Vietnam at 25,599 and Indonesia at 24,875 in average attendance per match.
724 people out of more than 25,000 is roughly 2.9 percent. Statistically, that sits inside the noise band of almost every sports forecasting model I have ever built. One match in a larger stadium, one different ticketing decision, one fixture played at a neutral venue instead of at home, and the order flips within a single tournament cycle.
In other words, the title of "largest supporter force in Southeast Asia" that domestic media are celebrating rests on a foundation far thinner than the headline suggests.
I am not saying this to diminish Vietnamese supporters. I am saying it because Vietnamese supporters deserve to know how solid the data about them actually is. A title built on a three percent margin is a title that can be taken away by a differently allocated group stage.
And here is the most important technical point in the entire story: average attendance per match does not measure loyalty. It measures the combination of four different variables.
The first is stadium capacity. A 40,000-seat national stadium will always produce a higher average than a 20,000-seat one, no matter how fervent the population. The second is fixture allocation. The team that plays more home matches in the group stage holds a structural advantage. The third is ticket price. Cheap tickets pull crowds in, but say nothing about emotional attachment. The fourth is supporter culture, and that is the only variable the ranking actually intends to measure.
The first three can shift from tournament to tournament. The fourth changes very slowly, over decades. When a ranking blends four variables into a single number and then declares an order, it is doing what analysts call conflating structure with essence.
What is actually inside the table
Read the following-rate table slowly.
Vietnam 85. Thailand 80. Malaysia 78. Cambodia 75. Brunei 72. The Philippines 45.
There is a large gap between the top five and the last. The distance between Brunei at 72 and the Philippines at 45 is 27 percentage points, while the distance between Vietnam at 85 and Brunei at 72 is only 13. In other words, five countries sit inside a narrow 13-point band, while the Philippines falls to an entirely different tier.
That structure is far more noteworthy than who finished first.
A narrow 13-point band across five countries means the measurement lacks the resolution to distinguish them reliably. When a scale compresses five distinct subjects into a 13-point range, measurement error can exceed the distance between subjects. This is a basic principle of measurement: if the noise exceeds the signal, the ranking order carries no meaning.
Yet media are celebrating that order.
And the Philippines at 45? That is the most interesting data point in the whole table, and the most ignored.
One methodological rule is stated explicitly in the original post: the ranking excludes citizens living abroad. It is the only published rule, and it creates a systematic bias.
The Philippines has an enormous diaspora, estimated in the millions, spread across the Middle East, North America and East Asia. Many of them follow the national team online. Remove that entire group from the sample, and you are measuring a domestic population with a much weaker football tradition than its neighbours, then attributing the result to the whole country.
I am not claiming the Philippines genuinely matches Vietnam or Thailand in football supporter culture. I am saying the 45 percent figure cannot be read as a pure measure of passion. It measures passion within a sample that was trimmed in a way unfavourable to the Philippines.
And here is the question nobody asked: if the diaspora exclusion rule applies, how was it applied to Vietnam? Vietnam also has a substantial overseas community, particularly in Eastern Europe, East Asia and North America. If Vietnam's 85 percent was calculated on a domestic sample and the Philippines' 45 percent was also calculated on a domestic sample, both were trimmed. But the degree of trimming differs because diaspora structures differ.
A rule applied uniformly to non-uniform subjects produces non-uniform results.
Who measured, with what, and why it matters
I want to tell a story from my own trade.
In 2026, when I began writing for an online sports betting platform in Kuala Lumpur, I introduced two metrics in my first piece: xG, expected goals, and PPDA, passes allowed per defensive action. The old guard of analysts called it the con trick of number-obsessed men. I did not argue. I quietly built a model from 387 matches across five major European leagues, published the method, published the sample size, and published the cases where the model was wrong.
Three weeks later, the exclusive contract arrived.
What I learned from that period was not that xG matters. What I learned is that a published methodology is what creates value. An indicator without a published method is just an opinion written in digits.
That is why I view the 85 percent ranking with caution. The ASEAN Football account admits it used AI to estimate the following rate. AI estimation works by aggregating social media posts related to Southeast Asian football over roughly one to two recent years. That is my inference, based on how large language models operate, not a published fact.
If that inference is correct, we have an indicator built on an extremely short time window, on a data foundation unrepresentative of the whole population, with no cross-validation.
A one-to-two-year sample for a cultural phenomenon with a decade-long cycle is an insufficient sample. I can build a model from a team's last 20 matches and it will give me a number. That number will be wrong, but it will look very persuasive.
Every signal from data is not an answer; it is a door opening onto another corridor that needs to be lit.
Viewers believe in drama; analysts believe in repetition
There is a concept in football I have always wanted to test with numbers: the twelfth man.
The idea is seductive. A large supporter force can pressure referees, shift player psychology, turn a neutral stadium into a home one. Many articles have been written about it. Very few have proven it.
I have tried to find the correlation between attendance and results over many years. The correlation exists, but it is weak and unstable. Home advantage is a real variable, but it is far smaller than media descriptions suggest, and it depends on factors unrelated to crowd size: the away team's travel distance, time zones, pitch conditions, fixture scheduling.
In 2026, when football returned to empty stadiums, I had a rare chance to test this hypothesis under near-laboratory conditions. My five-year model began to drift. Draw rates rose 23 percent above the historical average. Home teams won noticeably less.
Empty stadiums quietly shattered my faith in data, because when the noise disappeared, I realised data can tremble too.
I withdrew for three months, rewatched 212 Bundesliga matches after the restart, and built an adjustment coefficient for neutral conditions. I admitted that for years I had overpriced a variable I assumed was immutable.
I tell this story because it bears directly on the ranking now under discussion. If even home advantage, one of the most studied variables in football statistics, can collapse when environmental conditions change, how fragile is an AI-estimated supporter-loyalty indicator with no published methodology?
And more importantly: even if we accept that Vietnam has the largest travelling support in the region, we still have no evidence at all that this translates into results on the pitch. The original post supplies not a single specific match, not a single specific situation, not a single technical metric.
I am not denying the psychological power of a stand full of people. I am only saying that power needs to be measured by something other than seats sold.
Counter-evidence sitting inside the article itself
What caught my attention most in the original post was not the ranking. It was the writer's own commentary.
The author notes an uncomfortable fact: regional tournaments are losing their pull. He observes that stadiums in Thailand were empty even in the semi-finals and final.
Let that settle for a moment.
If the semi-finals and final of a regional tournament, the matches with the greatest appeal, are failing to fill stands, then we are discussing a supporter base that is contracting, not expanding. And if the base is contracting, what is the 85 percent ranking measuring?
There are two ways to read this.
The first: the ranking measures relative loyalty. Vietnam tops a cooling region. That is a real title but far less meaningful than the headline implies.
The second: the ranking measures a phenomenon tied to a specific tournament cycle, not a durable attribute. When the cycle ends, the number cools.
Both readings lead to the same conclusion: the ranking can be true in the short term and meaningless in the medium term.
I have seen this pattern many times. A successful tournament generates a wave of attention. The wave is measured, packaged into an indicator, and circulated. Then the next tournament arrives with less attention, and the old indicator becomes obsolete without anyone formally retracting it.
That is how indicators die in silence.
A timeline problem
There is a technical detail I need to raise, and it matters more than it appears.
The original post was published on October 15 and refers to ASEAN Cup 2026. Historical ASEAN Cup editions are typically staged in December and January of the following year.
If an October 15 post discusses a tournament taking place in the same year, 2026, there is a temporal gap that needs explaining. Either the tournament has already happened and the data is retrospective, or it has not happened and the figure rests on projection, or there is confusion about the edition.
I cannot determine which from the available data. And precisely because I cannot determine it, I cannot treat this ranking as current data.
In analytical work, an indicator without a clear time anchor cannot be used for any decision. You cannot compare it with last year's data. You cannot use it to forecast next year. You can only cite it as a media event.
That is exactly what it is.
Why this ranking exists
I do not believe there is malice in this story. I believe there is a much simpler motive.
Regional media accounts live on engagement. Ranking content generates the highest engagement because it touches national identity. A 2,000-word tactical analysis of how a team presses will get a few thousand views. A six-row ranking of which country has the most fervent fans will get hundreds of thousands of shares, thousands of arguing comments, and a large influx of new followers.
That is media value, not informational value.
And in this case the two have separated entirely. Media value is very high. Informational value is very low.
The word "surprising" in the headline is a signal. When a source describes its own result as surprising, that source is conceding the result is contestable. A properly measured indicator does not need to be surprising. It only needs to be correct.
I have been in this industry for 44 years. I have watched many data waves arrive and depart. The ones with published methods survive. The ones with only conclusions evaporate.
What is actually being transmitted
There is one aspect of this story I consider most important, and it has nothing to do with the ranking.
A large, travel-ready supporter base has real commercial value. It is a soft asset. It appears in sponsorship decks. It appears in tournament hosting bids. It appears in broadcasting rights proposals.
Any national federation with a region-leading supporter base holds an advantage in those negotiations. And they will cite this ranking, even if it was built by AI without a methodology, because in a commercial meeting an impressive number carries more weight than a paragraph on reliability.
That is how a soft indicator converts into a hard commercial argument.
I do not object to that. I only want those involved to know what ground they are standing on.
But there is another transmission, and it is negative.
When a ranking built entirely by AI, with no published method, is still widely circulated and cited by major media, it sets a precedent. That precedent says: you may publish any number as long as it is appealing. You do not need a sample size. You do not need variable definitions. You do not need validation.
Once that precedent is set, it will repeat. Next year there will be a ranking of the region's most promising young players, built by AI. The year after, a ranking of clubs with the most loyal fans, also by AI. Each time, the region's data standard drops a notch.
In a region where tactical data is already thin, lowering the data standard is a real loss.
A contrarian view: reading from the Philippine side
I want to close the analytical section by looking at the story from an angle nobody takes.
The Philippines at 45. In this ranking they are the only country placed below the 70 threshold. In a widely circulated article, they are the country labelled the least fervent supporters in Southeast Asia.
I do not have enough data to refute that label. But I have enough experience to know that such a label, built from a sample that excluded millions of people, should not be published without a warning.
Philippine football has a different history from the rest of the region. The most popular sport there is basketball, not football. The domestic league structure is thinner. The diaspora is far larger relative to the domestic population. Each of these traits affects the number, and none is reflected in the ranking.
This brings me to a principle I learned over many years, sometimes through mistakes.
When an indicator ranks countries, it does not merely describe them. It creates them. A published ranking becomes a social fact, even when it is not a statistical fact. Filipinos read that they finished last. Vietnamese read that they finished first. Both absorb a message about national identity from a model that did not publish its method.
That is why I say the responsibility of those who work with data exceeds the responsibility of those who read it. We publish; you receive. But we are the ones who know how solid our numbers are.
Signals for the next cycle
I am not writing this to say the ranking is wrong. I am writing to say it has not been proven.
And there is a large distance between those two statements.
There are four signals I will track in the coming cycle, and I suggest readers track them too.
The first is methodology. If ASEAN Football publishes how it calculated the 85 percent, including sample size, time window, variable definitions and how the overseas group was handled, the ranking moves from unverifiable to verifiable. That would be a genuinely meaningful change, and it would force every stakeholder to re-read the number.
The second is attendance in the next tournament cycle, particularly Vietnam versus Indonesia. A 724-person gap can reverse within one group stage. If Indonesia overtakes Vietnam next cycle, the title changes hands, and that would tell us the ranking reflects fixture allocation more than supporter culture.
The third is stadium fill rates in Thailand and elsewhere in the knockout rounds. If semi-finals and finals remain sparse, the decline the original author himself noted will be confirmed, and the entire foundation of the ranking becomes more fragile.
The fourth is the appearance of similar rankings. If additional regional indicators built entirely by AI without published methods appear over the next six months, we are witnessing a structural shift in the data standards of Southeast Asian football. And that is far more worrying than who tops a ranking about supporters.
I have lived long enough to know that in football, what survives each cycle is not the title. What survives is the method. Champion teams are forgotten. Verified datasets endure.
Viewers believe in drama; I believe in repetition, and drama repeats too if you wait patiently for it.
To Vietnamese supporters I want to say one simple thing. You do not need an AI model to prove that you follow your national team abroad. The stands in Bangkok, in Jakarta, in Manila have proven it for years, with things that can be counted: plane tickets, hotel rooms, flags, shirts, and voices singing in cities that are not home.
Those things are real data. They do not need to be estimated.
What I want to see next cycle is not a higher number. What I want to see is a ranking that can be refuted. Because an indicator that cannot be refuted is an indicator that cannot be trusted.
Age does not slow the observing eye; it only teaches me who genuinely wants to see, and mostly nobody does.
