The Data Gap in Vietnamese Football: Lessons from an Empty Dataset

Late on a Saturday night, after the V.League round had closed, I sat in front of a screen in Shenzhen doing my usual work: loading data, building context, and only then allowing myself to form a judgement. Twelve matches, twelve dossiers. I ran the query, waited a few seconds, and got back a nearly blank screen. No team names, no passes-per-defensive-action figures, no pressing minutes, no distance covered. Just empty cells lined up like an empty stand on a late afternoon.

An empty result. In my line of work, an empty result is rarely meaningless. When the model is wrong, the data starts telling the truth. But this time, what spoke was the silence of the data — and that silence told me more about Vietnamese football than a full dashboard ever could.

I have followed Vietnamese football for years, from local radio stations back when I was a journalism student to the transfer-data platform where I work today. The more I watch, the more I see a paradox: Vietnamese football is improving faster than its own data infrastructure can keep up with.

A football nation running faster than its own data

Over the past decade, Vietnamese football has gone through a notable cycle of achievement. The national team has won the regional championship twice — in 2026 under Park Hang-seo and in 2026 under Kim Sang-sik — alongside repeated appearances at continental finals. At club level, the V.League has stabilised at 14 teams, forming a title-contending group of Ha Noi FC, Cong An Ha Noi, Thep Xanh Nam Dinh, Viettel and Becamex Binh Duong. In youth development, academies such as HAGL-JMG, PVF, Viettel and Song Lam Nghe An have become frequently cited cradles of talent.

Behind those achievements, however, sits a thin data infrastructure. Most match data is still collected manually, by note-takers in the stands with pen and paper, or by small teams working with limited resources. Advanced metrics such as xG or PPDA appear in only a small share of matches, mainly national-team games or V.League fixtures with a data sponsor. The rest — the games that decide a relegation fate, the midweek rounds played on poor pitches after rain — have almost no reliable data to analyse.

That is why my screen was empty. It was not empty because there were no matches. It was empty because the matches took place without leaving behind any numerical trace clean enough to use.

What happens when a European model meets V.League reality

I was born in France and work in China, so I have a habit of taking European models and testing them against Asian data. That habit taught me a simple lesson: a model is only valid under the conditions in which it was built.

Take home advantage. In Europe's big leagues, the home-win rate hovers around 44 to 46 percent, and models treat this as a constant. But that constant does not exist in Vietnamese football in exactly the same way. Travel distances between V.League teams are far shorter than in Europe — an away trip from Ha Noi to Nam Dinh takes under two hours, whereas a journey from London to Newcastle eats up nearly half a day. Home is not sacred ground, only a variable that has been frozen — and when travel conditions change, that variable has to change too.

I once witnessed a vivid demonstration of this point. In 2026, when German stadiums stood empty because of the pandemic, I collected data from nine Bundesliga rounds after football returned. The home-win rate fell from 44.2 percent in the 2026-19 season to 36.7 percent, and average goals per match dropped from 3.1 to 2.8. The moment the crowd disappeared, what every model treated as fixed collapsed instantly. In Vietnamese football, that variable is even more fragile, because the crowd at many grounds is still part of the match in ways raw data cannot quantify.

Then there is pressing. PPDA is the signature; distance covered is the confession. When Vietnam played under Park Hang-seo, I tried to reconstruct their pressing signature. In big matches, the team accepted ceding the ball and waited for its moment, with a PPDA above the Asian average, yet it could switch states extremely quickly. Under Kim Sang-sik, that signature changed: the team pressed higher, won the ball earlier, and leaned more on wide attacks. The same national team, two different signatures, within just a few years. Anyone reading Vietnam's PPDA without noting the time frame is fooling themselves.

The same holds for transfer data. I once oversaw a major deal while working at a transfer-data platform: Enzo Fernandez moving from Benfica to Chelsea for 121 million euros. I used the player's World Cup data — a high pass-completion rate, an impressive number of successful tackles — to build a valuation report. But the deal ultimately depended on agents, payment terms and the buyer's haste. Data explains the past; it does not predict the future. Transfers do not pick the best player, they pick the one you mis-measure the least. In Vietnam, where financial disclosure is limited and agents play a large role, that lesson holds even more strongly.

On financial structure, the V.League picture is quite specific. Most clubs survive on one or a few corporate sponsors tied to their owners, rather than on broadcast or commercial revenue. The league's broadcast revenue is modest compared with other Southeast Asian competitions, and revenue-sharing among clubs remains flawed. This produces a data consequence: when financial resources concentrate in a few clubs, the competitive gap widens, and statistical measures of league competitiveness become distorted. A league where two or three clubs hold most of the budget will produce seasons in which the title is less surprising than the audience's feeling suggests.

The evidence lies in what the table does not tell you

If you look only at the V.League table, it is easy to think everything is clear. But the table is the end product of a process, and that process is usually hidden.

Take the relegation story. In many seasons, the gap between tenth place and bottom is only a few points, and fates are decided by matches in which both teams play under heavy psychological weight. Data shows that teams in the danger zone often see their attacking metrics drop sharply over the final ten rounds, not because they become weaker in personnel, but because pressure changes how they play. They pass shorter, run less over long distances, and lean increasingly on set pieces. A model that reads only goals scored will miss this entire psychological story.

I trust variance more than I trust the champion. In a league with few matches and high volatility like the V.League, a champion may finish only a few points ahead of the runner-up, while a relegated team may finish just one point behind a surviving one. That fragile margin makes drawing conclusions from a single season dangerous. Variance — the degree to which results swing around the mean — is what tells me which team is genuinely stable and which is merely lucky.

Then there are the academies. HAGL-JMG was once praised as a miracle, producing a generation of players tied to the image of Vietnam's national team, including Nguyen Cong Phuong, Nguyen Tuan Anh and Luong Xuan Truong. But place the data in context and you see something else: the share of academy players who stay at the top is not as high as the media narrative suggests. PVF and Viettel are similar — they train well, but their output is limited by a small domestic transfer market and modest wages. Training well does not automatically become results without an ecosystem that retains talent.

This is where I want to be clear about something the press often overlooks. Data does not get emotional, but it remembers everything the press forgets. When a young player shines in one match, the media writes about the moment. When that player fades over the next ten matches, nobody writes. But the data records both. That is why I always demand a minimum sample before concluding anything about a player: one match is an anecdote, ten matches start to be data.

At national-team level, the data is fuller — and precisely because of that, easier to misuse. The victory at the 2026 regional championship, with a 5-3 aggregate score over Thailand across two legs in January 2026, brought fans great joy. But reading only the result overlooks an important detail: in the first half of the second leg, the team trailed and had to rely on tactical adjustments and individual character to turn the tide. The PPDA data and ball-recovery positions in those two halves differed markedly, and that difference is the real lesson, not the aggregate score.

Then there is refereeing controversy, an always-hot topic in the V.League. When VAR technology was introduced, the expectation was that decisions would become more transparent. But technology is only as good as the data and process behind it. If camera angles are insufficient, if intervention standards are inconsistent between matches, VAR can become a new source of controversy instead of settling old ones. I have rewatched several incidents and realised that what frustrates fans is often not the final decision, but the sense that standards are applied differently in different matches. That is a process problem, not a technology problem.

The contrarian angle: emptiness is a signal, not a failure

The easiest mistake when analysing a rising football nation is to credit it with every achievement and ignore every limitation. I once made that mistake in a different way.

In 2026, when I was nineteen and still a journalism student, I built a World Cup prediction model based on the xG and xA of five European leagues over three consecutive seasons. The model gave Germany a 78 percent probability of reaching the semi-finals. Germany lost 0-2 to South Korea in their final group game and went out in the first round. The model correctly predicted 12 of the 16 knockout qualifiers, but it was wrong about the team I believed in most. I had discarded non-data variables: internal conflict, complacency, declining fitness. Germany 2026 was a gift, because it proved that a model also needs to fail in order to grow.

Since then, I never write absolute statements. In every analysis I add a section on data limitations and always ask myself: which non-data variables are being left out?

With Vietnamese football, that question is more urgent. A blank screen does not mean there is nothing to say. It means that what is worth saying has not been recorded well enough. When a V.League match takes place without pressing data, without a pass map, without distance-covered metrics, then every judgement about that match — whether from experts or fans — rests on memory and feeling, not on verifiable evidence. Memory can be beautiful, but it is not as reliable as a clean dataset.

There is another temptation: importing European models wholesale and applying them to the V.League. I have seen many reports do this. They take Premier League home-advantage figures, Bundesliga pressing figures, La Liga transfer figures, and paste them onto a league with entirely different conditions. The result is conclusions that sound highly professional but are wrong in essence. An imported model without recalibration is just a bias written in numbers.

And I want to offer a warning. Where does football data usually flow when it is thoroughly commercialised? Much of it flows to betting companies, where it is used to price risk for punters rather than to nurture fans' understanding. That is the darkest side effect of the digitisation of sport. A football nation that wants sustainable growth needs to keep its data serving coaching, analysis and education, instead of letting it become raw material for a betting market alien to local fans.

What I carry away from the silence of the data

Vietnamese football stands at an interesting intersection. Results on the pitch have surged ahead, but the data infrastructure behind them is still catching up. That gap is not bad news — it is an opportunity. A football nation that starts investing seriously in data will hold a big advantage, because it can learn from the mistakes of those who went before without repeating them.

But to do that requires a change in how things are seen. Data is not a tool for showing off, but a tool for self-questioning. It does not replace the observer's eye, it complements it. It does not offer the final answer, it points to better questions.

On my screen, the empty cells are still there. I could delete them and write an analysis that sounds complete. But I choose to keep them, as a reminder that data is a foundation, not an absolute truth. The next round will come, and with it new numbers — if someone is willing to record them.