Vietnamese Badminton and the Data Blind Spot: What Nobody Measures After Every Rally

In January 2026, the Impact Arena in Bangkok opened its doors for three consecutive badminton tournaments inside a health bubble, and not a single spectator was allowed in. I sat in front of a screen in Nha Trang, muted the commentary, put on headphones, kept a stopwatch in my left hand and a pen in my right. In an empty arena, sound turns sharp as a blade: the racket striking the shuttle, shoes grinding the floor, a coach calling out to his player in the corner. I began counting the rallies, noting how each one ended, and marking it down on a blank spreadsheet.

Three days later I had 1,480 rallies in my notebook. There was nothing to compare them with, because I had never done this before. But one detail stopped me: most rallies that went past the fifteenth shot ended in an unforced error rather than a winning smash. That rate was well above what I had imagined about elite badminton. A belief I had carried for years, that class in this sport is decided by powerful smashes, began to crack right there.

Then I tried to do the same thing with a domestic badminton tournament in Vietnam. There was nothing to click. No rally log. No positional data. No table showing how many points a player won in the back half of each game. All that remained was a paper scoresheet with the score of each game, photographed and posted on social media, along with a few emotional remarks.

That gap is the subject of this article.

Vietnamese badminton has a well-documented record. Nguyen Tien Minh reached No. 5 in the world in men's singles in August 2026 according to the Badminton World Federation rankings. Three years later, at the World Championships in Guangzhou, he won bronze, Vietnam's first medal at that event. Nguyen Thuy Linh has been the leading women's player for many years, appearing at the Olympics and at numerous World Tour events. Le Duc Phat has followed in men's singles. In mixed doubles, Do Tuan Duc and Pham Nhu Thao once carried Vietnamese badminton onto the Olympic stage.

That is the record. The data behind the record is nearly empty.

Picture a coach trying to answer the simplest question: where does my player lose the most points in the third game? To answer it, he needs a rally-by-rally record with the server, the type of stroke that ended the rally, and the score at that moment. At major international events, this data exists in the form of match statistics tables and technology-assisted replays. At most domestic events and low-tier Asian tournaments, it does not exist. No stroke-tracking system. No multi-angle cameras. Nobody sitting there counting.

That is why I describe Vietnamese badminton as having a data blind spot, and the blind spot sits on three different levels.

The first level is the rally. A rally is the smallest unit of analysis in this sport, the equivalent of a possession in football. It contains the decision chain of two players: where the serve goes, where the opponent returns it, who seizes the initiative first, and how. Without a rally log, all that remains is perception. Perception is not wrong, but it has no memory.

The second level is space. Badminton is a sport where where you stand matters almost as much as how you hit. A player standing half a step too far left may pay for it with three extra strides on the next shot. That kind of data only exists when a court is fitted with tracking cameras, and in Vietnam that happens almost exclusively at international events hosted at home.

The third level is physical load. At major training centres around the world, coaches measure distance covered, jump count and sprint efforts within a game. Here, we count with a coach's gut feeling and with the number of recovery days after an injury.

Together, those three levels create a gap large enough to change how an entire national badminton programme makes decisions. Whether to play singles or doubles, whether to build an attacking or defensive style, whether to send an eighteen-year-old abroad for training or keep them at home: all of it happens on a foundation of observations nobody recorded.

Between the ages of twelve and fifteen, selection into youth squads still relies almost entirely on live observation at tournaments. A tall kid with a heavy smash gets noticed. A smaller kid who moves half a beat earlier, who reads the shuttle before the opponent strikes it, is easily overlooked, because the thing they do best appears in no scoresheet. I have no statistical proof for this claim. I only have an observation repeated many times at youth events I have sat through: players who win by reading the game are remembered less often than players who win by force.

At regional level, Vietnamese badminton regularly faces Thailand, Indonesia and Malaysia, badminton nations with denser domestic calendars and longer record-keeping traditions. The gap here is not only technical. It is also a gap in the number of matches that get recorded, and therefore a gap in the speed of learning.

The crux lies in the habit of record-keeping, not in the equipment. A sheet of paper and a pen, in the right hands and the right place, can produce data more valuable than an expensive camera system that nobody bothers to review.

I entered this trade through an Excel spreadsheet, but I stayed because of the stories inside it.

The cost of doing this is unbelievably low. A phone, a tripod, a notebook, and one person willing to stay behind after the session. The problem is not money. It is that nobody is assigned the task, and nobody is paid to do it.

Between March 2026 and December 2026, I hand-logged 63 matches featuring Vietnamese players for which I could find a full broadcast, a total of 4,117 rallies. The method was thoroughly manual: a spreadsheet with columns for rally length, server, serve type, the player who struck the final shot, the type of final stroke, and the score at the moment the rally ended. Each match took about forty minutes, plus replay time for rallies I could not log live.

I must state the limits of this sample before reporting results. It is small, and it includes only broadcast matches, which skews it toward big tournaments and deep rounds. Counting rally length by eye introduces error, especially in rallies with many changes of direction. No second person logged independently for comparison. The figures below should be read as hypotheses to be tested, not conclusions.

The first metric I call the rally attrition index. The calculation: divide the number of rallies beyond the fifteenth shot that ended in an unforced error by the total number of rallies beyond the fifteenth shot. In other words, when a rally drags on, who breaks it: the attacker, or the defender who self-destructs?

In my sample, this ratio sat around 0.58 in matches featuring Vietnamese players. In the group of matches between top-twenty players I logged over the same period, the corresponding figure was around 0.44. The gap is not in the hardest smashes, because smash speed is measurable and the gap there is far smaller. It sits from the sixteenth shot onward, when a rally has passed the threshold where technique begins to give way to stamina and decision-making.

This is where ordinary statistics tables fall silent. A scoresheet tells you someone won 21-18 and 21-16. It does not tell you that those two games contained eleven rallies beyond the twentieth shot, and that the losing player won seven of them.

The second metric I call the initiative conversion index. The calculation: among rallies in which player A produced the stroke that forced the opponent into the most movement during the first half of the rally, what percentage ended in a point for A. This is the badminton version of a question I once asked in football: when a team creates a good chance, how often does it convert?

In my sample, Vietnamese players sat around 0.61 on this metric, while top-ranked international players sat around 0.72. The eleven-point gap is not in the ability to create initiative, which was roughly equivalent in the sample. It sits in the final three shots of the rally, when initiative must be closed out with a decisive decision.

People remember the rally that ends with a cross-court smash. I remember the three shots before it, when the player already had the chance and did not take it.

The third metric is the landing-zone index. I divided the back court into nine zones and counted where winning points landed. The results in my sample clustered tightly: most points landed in three zones, the two corners near the sidelines and the rear central area. This sounds obvious, but it has a concrete coaching consequence. If most points land in three zones, defensive drills should be redesigned around those three zones rather than spread evenly across the court.

It must be stressed that these three metrics are my own constructions. They are not international standards, they have not been validated, and they are meaningless when detached from their calculation method, sample size and stated limits. I raise them to show one thing: with a single person sitting down to record, Vietnamese badminton gains three measurements that did not previously exist.

Look across to football, where I started, and a similar story unfolded in reverse. The 2026 World Cup taught me that possession is an illusion dressed up as truth. Germany touched the ball three times as often as South Korea in that fateful match, and still went home. People learned that lesson not because someone suddenly became clever, but because the data had existed long enough for people to know how to ask a different question.

Vietnamese badminton has not had that opportunity. The reason is not talent, because Nguyen Tien Minh spent fifteen years proving the opposite. The reason is that nobody was recording.

When the stands fall silent, every team strips off its mask. During those three days in Bangkok, I could hear a player breathing after a long rally, and that taught me more than any tournament summary ever has.

Data does not lie; it simply stays silent until you know how to listen.

Now let me argue the other side, because I do not want this article to become a simple call that data will fix everything.

First: correlation is not causation. If you see that player A smashes more and wins more, do not rush to conclude that smashing is the cause of winning. It may be that A only smashes when already ahead, and chooses a safer pattern when trailing. The hidden variable is the score, not the smash. In football, I once spent two days realising that a discovery of mine was merely a by-product of strong teams usually playing at home.

Second: homemade metrics very easily become lucky charms. When you build your own measurement and it looks logical in your head, you tend to use it to explain everything. The antidote is to always publish the calculation method, the sample size, and a clear statement of where the metric is useful. The three metrics above are worth exactly as much as the transparency I give them.

Third: data is not a sufficient condition. A badminton nation with excellent analytics can still decline if it lacks a next generation, lacks tournaments, lacks money. Data does not produce players. It only helps players move faster along a path they have already chosen.

Fourth, and perhaps most important: data only answers the questions people are willing to ask. If you measure smash speed because smash speed is easy to measure, you will optimise smash speed. If you measure rally length because it is harder to measure but closer to winning, you will optimise something else. The biggest blind spot in any sport is not a shortage of machinery, but measuring the wrong thing while believing you are measuring the right one.

I should also say something about the fans, because there was a time when I looked down on them. Fan emotion is a form of raw data. When thousands of people independently feel that a player fades at the end of a game, that is a hypothesis worth testing, not a complaint. The analyst's job is to turn that hypothesis into a number, or to refute it with a number. Dismissing fan emotion means throwing away the cheapest and most abundant data source available.

So what signals would show that things are changing?

First, a national championship or national youth championship publishes a rally log for every match, even as a simple file. Second, a training centre hires a full-time data analyst rather than relying on a coach who logs notes in the evening. Third, an article about Vietnamese badminton cites a metric of its own instead of citing a scoreline.

Conversely, if twelve months from now all we have are more emotional commentaries about a beautiful smash, then nothing has moved.

Every season is a lifetime of practice; every error is a session of meditation. I have sat and logged four thousand rallies without being sure where it leads. But if one day a young coach in Nha Trang, in Da Nang or in Bac Giang opens a spreadsheet and sees something nobody has seen before, then the logging has already paid for itself.

The question I leave behind: if you could keep only one measurement for a badminton player, which would you choose, and would you be willing to count it for two years to find out whether you were right?