ASEAN fan loyalty: the algorithm hands out the marks, the plane ticket gives the answer

The 6:20 a.m. flight from Tan Son Nhat to Kuala Lumpur has a feature I have logged four times in three years: roughly a third of the passengers are wearing red. They are not travelling for meetings. They carry banners rolled inside plastic tubes, they carry drums, and they carry the one document no airline ever asks for — the belief that their team needs someone standing behind the goal at a stadium 1,600 kilometres from home.

On 17 October 2026, a ranking published by ASEAN Football, a Malaysian media outlet, called those passengers the most loyal supporters in Southeast Asia, at 85 percent. Indonesia sat seventh on 65 percent. Thailand second on 80, Malaysia third on 78, Cambodia fourth on 75, Brunei fifth on 72, Singapore sixth on 70, Myanmar eighth on 60, Laos ninth on 55, Timor-Leste tenth on 50, the Philippines eleventh on 45.

I read that table three times. The first as a reader. The second as someone who sat through nine home matches of a club in Long An counting the goals conceded in the final ten minutes. The third as someone who once misnamed a Russian player three times in one half and spent 47 days putting it right.

Three readings produced three different answers. Only the third is worth writing.

The first thing to say about this ranking is how it was made. According to the article itself, the percentages were compiled by an artificial intelligence model reading social media posts, and the article states plainly that these are not official statistics of any federation. The method counts only supporters flying directly from their home country to an away venue; Vietnamese and Indonesians who have settled or work abroad fall outside the calculation.

The same article carries a second dataset, measurable and verifiable: average attendance per match at the 2026 AFF Cup. Vietnam 25,599. Indonesia 24,875. Cambodia 17,169.

Two datasets sit side by side in one piece of writing, and they do not tell the same story. The ranking puts Vietnam 20 percentage points above Indonesia. The attendance data puts them less than 3 percent apart. One is sentiment turned into an algorithm; the other is tickets sold at the gate.

The reaction the article records is heated debate among Southeast Asian football fans, with Vietnamese supporters described as capable of overwhelming fanbases long known for being large and loud. That framing — surprising, counter-intuitive — matters more than the placings themselves.

One more note on the name. ASEAN Football is a Malaysian media outlet. The AFF is the ASEAN Football Federation. One word separates them, and in a story shared across social platforms, one word is enough for an unofficial statistic to borrow the authority of a governing body.

A supporter flying from Hanoi to Bangkok for a group-stage match pays for a flight, two hotel nights, meals, and two days off work. For a young office worker in Saigon, three such trips across a season land somewhere between 15 and 20 million dong — my own estimate, built from ticket prices and the actual spending of supporter groups I have travelled with. That is a month's salary.

So when an algorithm says the Vietnamese stands are 85 percent loyal, it is describing something other than loyalty. It is describing the depth of the urban middle class, the ability to take leave, and the penetration of credit cards within a particular slice of the population. The measurement speaks to purchasing power more than to feeling, and it functions as a demand-side economic indicator rather than a psychological one.

Read that way, the ranking becomes startlingly coherent. Singapore sixth on 70 — a wealthy island state where every away trip is a flight, and flights are affordable. Brunei fifth on 72 — a small country with high average income, where the furthest journey in the region is a few hours in the air. Cambodia fourth on 75 — a smaller football economy whose distances to Bangkok, Hanoi and Vientiane are short and cheap. Those numbers match wallets, not devotion.

Then the calculation betrays itself. Indonesia has an enormous overseas community in Malaysia, Saudi Arabia, Taiwan and Hong Kong. The Philippines has more than ten million workers abroad. An Indonesian supporter living in Johor Bahru, driving thirty minutes to Shah Alam to sing for ninety minutes, is counted by the algorithm as zero. He did not fly from home. He left home ten years ago. That exclusion rule is not neutral — it builds a systematic disadvantage for any nation that exports labour.

This is where I want to linger, because it explains almost the entire Indonesian paradox. Indonesia's overseas community is one of the densest football-supporting networks in Asia. It does not appear on the ranking. It appears in the stands of regional tournaments, in red and white shirts scattered across every city with a direct flight. A model reading posts in Bahasa, English and partial Vietnamese sees the surface. It does not see what lies beneath, and it scores what lies beneath as nothing.

I do not keep time for the team; I keep time for the people who kept time for the team. And most of those people do not fly. They sit at a pavement cafe at seven in the evening, phone plugged into a power bank, watching a blurry stream. They generate too little data for an algorithm to register, yet they are the entire foundation of what we call national football.

There is a further disadvantage, purely geographic. The Philippines is an archipelago of eleven major islands and thousands of small ones; to travel away in Southeast Asia, a fan from Manila must fly, and fly further than a fan from Hanoi. From Timor-Leste every trip is a flight, because there is no road to take. Those supporters are marked at 45 and 50 percent while clearing physical barriers larger than anyone else on the list. The ranking reads like an income table wearing the costume of a passion table.

Malaysia third on 78 deserves exactly the emphasis it earns, no more. A Malaysian outlet placing Malaysian supporters among the regional elite is a mild self-favouring pattern, but the position is not absurd. Malaysia has organised football, good stadiums, and a genuine geographic advantage: land borders with Thailand and Singapore, plus cheap intra-regional flights. I do not treat that as evidence of fraud. I treat it as a reason to read the table with the eyes of someone who knows who wrote it.

Then there is the false precision. 85, 80, 78, 75, 72, 70, 65, 60, 55, 50, 45. Eleven figures, no sample size, no timeframe, no model description, no confidence interval. A sliding scale descending as neatly as if it had been arranged by hand. In my trade, a percentage without a method is a statement without a speaker. It rings, and it is hollow.

Drawing on my experience covering matches, I have a far simpler cross-check. Across nine home games at Long An in 2026 I logged one pattern: three goals conceded inside the final ten minutes. It was a small sample, but it had clear boundaries — nine matches, one club, one season, and I could name the minutes. If anyone asks my basis, I open the notebook and point. The 85 percent table opens nothing at all.

Correcting one wrong name took 47 days; keeping faith with one person lasts forever. I learned that after the 2026 World Cup, when I misnamed a Russian player three times in one half and spent 47 days rewatching every qualifying match to give him back his name, along with a biography no outlet had mentioned: his father had played for Real Madrid. Since then every name I print clears at least three sources. A ranking of eleven nations has no source but itself.

The only trustworthy dataset in that article is the attendance set, because attendance is a transaction. Someone took out money, swiped a card, scanned a code at the turnstile. Vietnam at 25,599 per match. Indonesia at 24,875. Cambodia at 17,169. Those three figures say that Southeast Asia has two large demand anchors in Vietnam and Indonesia, and a rising third market worth watching. For anyone who has spent thirty-nine years reading stand structures, that is the information you can actually work with: rights negotiations, sponsorship valuation, fixture scheduling.

The gap of under 3 percent between Vietnam and Indonesia in attendance data is a more valuable counter-intuitive finding than the whole ranking. It says the two football economies sit level on matchday demand, regardless of an algorithm separating them by 20 points. It also says that anyone citing the ranking to argue about stand strength is citing a source that contradicts its own accompanying data.

As for the twelfth-man effect — the only thread in this story that touches actual football. The ranking counts away supporters only, the group that can shape the atmosphere in an opponent's stadium. But what share of a result does an away stand genuinely decide? From what I have watched, the effect lives in atmosphere, in pressure on referees during sensitive moments, in forcing opponents to shout louder to stay organised. It does not live in goals. Goals come from recovery time, travel schedules, pitch surfaces, heat, and the ten minutes after the eightieth.

And here is the loop few notice. The algorithm measures social posts. A ranking born from social posts generates more social posts. Vietnamese media will carry it in a tone of pride. Indonesian media will carry it in a tone of rebuttal. Both extend the lifespan of a statistic with no method, and both push higher the very index that produced them. A system feeding itself, detached from anything measurable on the pitch.

Between two whistles there is a world the scoreboard cannot measure. But that world cannot be measured by a model reading posts either. It is measured by the people who were there when the team lost three in a row, and were still there for the fourth.

I saw it at Long An. After the club exited the AFC Cup, the stands were nearly empty. A few hundred stayed, and they sang. Three days later I sat with Mr Sau, the stadium gatekeeper, listening to him describe the generations of players who wore the Long An shirt from 2026. He remembered every name. He had no data at all. He had a notebook and a memory. My piece afterwards brought hundreds of supporters to gather outside the ground singing the national anthem after the final match — not because a ranking had rated them, but because they recognised they had been counted in another way.

Then came the empty-stadium season of 2026, when every gate closed. I refused match reporting because I could not reach the ground, and switched to recording wind, the sound of a dressing-room door closing, phone calls to ground staff. A mother in Da Nang called and wept for fifteen minutes about the scarf her son had left behind. The emptiest stadium still breathes; you simply have to listen with your heart. No model scores that breath, and the ranking concedes as much by staying silent about it.

The biggest misunderstanding from outside is this: people read the ranking as a contest of passion and argue over who is more fanatical. What is actually being argued is mobility and social-media visibility. Filipino supporters are not less loyal than Vietnamese supporters. They simply live in a country where a domestic flight to a home game costs more than another nation's international trip.

A second misunderstanding is subtler: the largest and loudest group is always assumed to be the most loyal. I was born in England and could reach for a comparison about how European leagues allocate away tickets, but that comparison would take me off the point, because the issue here lies elsewhere. The loudest group is usually the youngest, fittest and most available. The most loyal group is usually the quiet one in the corner of the stand, seventeen seasons of season tickets deep, through two relegations and one change of ownership.

Which is why I will not use this ranking to say anything about Vietnamese football. I use it as a mirror for how regional media now manufactures numbers: fast, handsome, shareable, unverifiable. A mirror is worth looking at, provided you look at it the right way round.

Football never owes us a result; it only owes us a story. The story here is not Vietnam's first place. It is that an openly unofficial statistic crossed borders, crossed four languages, crossed thousands of shares, and by the time it stopped, nobody remembered it was unofficial.

The signals I will track over the coming weeks are concrete. First, whether ASEAN Football publishes a method, a sample size and a timeframe — if it does, the ranking deserves a second read; if it does not, it is engagement content. Second, whether the AFF or any member federation issues a clarification, because silence will be read as tacit endorsement. Third, attendance data from the next regional tournament, the only thing that will tell the truth about who goes to watch football and how often.

And one question I leave to people in my trade. When an algorithm grades the stands, who grades the grading? If the answer is nobody, we are letting the easiest things to measure define the things most worth measuring. What I want is not a fairer ranking. I want a profession patient enough to say: we have not measured this, and we will not pretend otherwise.