The 24-22 Set and the 12-21 Decider: The Fragile Line Between Competing and Enduring for Vietnamese Badminton at the Asian Games

Nguyen Thuy Linh levelled the match in the second game. The scoreboard stopped at 24-22, the stands rose, and during the seventy-second interval almost everyone believed a comeback was beginning. The third game ended minutes later: 12-21.

I sat with those three lines for a long time. A badminton scoreline does not lie, but it does not speak on its own either. 10-21, 24-22 and 12-21 are all we have from the 26 September badminton session at the Asian Games. No shuttle speed. No rally-length distribution. No net-point conversion rate. No unforced-error count split by half-game. No service statistics. No movement-distance data.

Hand a bare scoreline like that to a sports writer and it will tell a story about courage. Read it properly and it tells a story about limits.

Context: one session, two draws, four lines of score

Nguyen Thuy Linh, Vietnam's number one in women's singles, reached the third round after beating Thalita Ramadhani Wiryawan. In the third round she met Chen Yu Fei, listed as world number four at the time of the event. The result was a 1-2 defeat, with games of 10-21, 24-22 and 12-21.

In the men's singles draw, Nguyen Hai Dang beat Odilzhon Abdurakhmanov 21-6, 21-6 in the previous round, then met Kunlavut Vitidsarn in the third round and lost 0-2, with games of 16-21 and 14-21.

In total, Vietnam's badminton presence in the source material comes down to four games won and four games lost. That is a sample so small that I have to state my position at the outset: everything in this article is presented as a hypothesis with a stated confidence level, not as a closed conclusion.

I have worked with datasets thousands of times larger, and I have learned something uncomfortable: the smaller the sample, the more attractive the story becomes. Four games won against two lower-ranked opponents and four games lost against two elite opponents are enough to describe a single session. They are not enough to conclude anything about a cycle, a playing style, or a generation.

In badminton, humility is not an ethical posture. It is a technical requirement.

Reading 10-21, 24-22, 12-21 as a data sequence

The first game, 10-21, was a game of domination. There is no more credible reading: an eleven-point margin at this level usually means one player could not find rhythm and the opponent was free to dictate the shuttle.

The second game, 24-22, was the only game Nguyen Thuy Linh won, and how she won it deserves more attention than the fact that she won it. A game that stretches to 24 points means both players reached 20-20, and at that threshold badminton rules require a two-point margin. Every rally there carries several times the psychological weight of a rally at 5-3. Crossing that zone to win 24-22 is not pure luck. It is a sequence of correct decisions under pressure.

The third game, 12-21, is the line that says the most, and it says what fans do not want to hear.

A three-game scoreline is not a random sequence. It has a shape. The shape here is very familiar in badminton: one game crushed, one game fought to the threshold, and one game collapsing quickly. I have seen this shape across many events, and it almost always points to the same variable. That variable is not basic technique. It is the stamina budget and the ability to preserve decision quality once the budget is exhausted.

A 24-22 game always leaves a bill

Here I have to use a principle I apply to football and badminton alike: the cost of a long game is not paid inside that game. It is paid in the next one.

Modern badminton is a sport of short but dense rallies. The physiological cost of a game that runs to 24-22 is significantly higher than one ending 21-15, because both players reached 20-20 and had to play at least four more points, each carrying an attack-or-defend decision, larger lateral and longitudinal movement, and psychological pressure that tightens muscles beyond what is needed.

The 12-21 third game fits that model. I have no cardiovascular data, no movement-distance data, no breathing-rate data for the decider. I only have the score. But the score here has a very familiar shape: after a long game won, a short game lost.

A nine-point margin in the decider, against an opponent ranked many places higher in the world, is the signature of a resource that ran out rather than a technique that was solved. If it had been a technical solution, the second game would not have ended 24-22.

That leads to a medium-confidence hypothesis: Nguyen Thuy Linh's limit in this match lay in sustaining intensity across three games against a top-tier opponent, not in producing a peak single game. She proved the second. She has not yet proved the first.

The third game is where a player is actually measured

I have written before that a player's true value lies in where he runs and when he stops. For badminton I would add: a player's true value lies in the third game.

The third game strips away decoration. The first game can be won with a plan prepared in the dressing room. The second can be won on a burst of momentum, a good service sequence, or a sudden lapse from the opponent. The third can only be won on foundations: physical foundations, technical foundations in balanced rallies, and the mental foundation to accept that the opponent now knows what you will play.

In the elite badminton data I have observed, a fairly durable pattern recurs: players in the world's top four are often not superior in basic technique to those ranked ten to twenty. The technical gap between those groups is smaller than spectators assume. What separates them is the point-winning rate in deciding games and the ability to keep unforced errors below expectation while tired.

Put differently, the elite do not win because they have more weapons. They win because they lose fewer points during the stretch when both players want to quit.

For Nguyen Thuy Linh, the 12-21 third game is a negative data point. It does not cancel the 24-22 second game. It shows that in this match her endurance threshold was lower than her opponent's at precisely the most important moment. In this sport, the most important moment is almost always the third game.

Nguyen Hai Dang and the compressed gap

Nguyen Hai Dang's two matches trace a different line.

21-6, 21-6 against Odilzhon Abdurakhmanov is basic professionalism. No game dropped, no rhythm lost, no letting a lower-ranked opponent drag him into danger. I rate this as positive data, and it matters more than it looks. At continental events, the number of players who drop a game to a weaker opponent is not small, and most of those cases lead to an extra game under a congested schedule, which can affect the following round.

Then came Kunlavut Vitidsarn. Losing 16-21, 14-21 is not a disaster. But the manner deserves analysis: both games were lost by margins held between five and seven points. This is the kind of defeat that carries information.

In sports data analysis, I distinguish three types of defeat.

The first is the open defeat, the 5-21, 8-21 kind. It says the two players are not in the same sport, and any deeper analysis is meaningless.

The second is the narrow defeat, the 19-21, 20-22 kind. It says the gap is very small and can be erased by minor tactical or psychological adjustments.

The third is the compressed defeat, the 16-21, 14-21 kind. This is the most interesting and the easiest to misread. Nguyen Hai Dang was good enough not to be blown away, good enough to hold each game near twenty points, but not good enough to create a turning point at any stage.

The gap sits at exactly seven to nine points per game. At this level, seven points in a twenty-one-point game usually come from three sources: unforced errors in the back half of the game, the win rate in rallies longer than ten shots, and the ability to break a deadlock after being forced into consecutive defence.

I have no data from any of those three sources for this match. I only have the score, and the score suggests Nguyen Hai Dang does not yet have a weapon strong enough to finish points in deadlocked rallies against a strong defensive player. That is a hypothesis, not a conclusion.

But if the hypothesis holds, it is good news. Deadlock finishing is a trainable problem, unlike a lack of physical foundation or hand speed, which are far harder to fix at twenty.

Single elimination and the draw nobody chooses

A factor overlooked in most commentary: the format.

The Asian Games badminton competition uses single elimination. Every walk onto court can end the tournament. There is no correction mechanism through aggregate points and no group stage to build rhythm. For a player still finding form or still developing, this is a far harsher environment than an event with a group stage.

It is worth adding that the current badminton scoring system is 21-point rally scoring, best of three games, each game to 21 points, with a two-point margin required at 20-20 and a hard ceiling at 30. Precisely because of the 20-20 threshold and the two-point rule, games like 24-22 become unusually valuable units of measurement. They pinpoint the exact moment when a player decides a match with nerve rather than technique.

The draws of both Vietnamese players placed them in the third round against opponents from the top of the continent. This is a parameter analysts call draw luck, and it influences a Games outcome more than people assume.

A player who reaches the third round, meets the world number four and loses carries the same nominal result as a player who reaches the third round, meets someone outside the top twenty and goes further. The same line in the record book. Two completely different meanings.

That is why I read the results of Nguyen Thuy Linh and Nguyen Hai Dang as encounters with the continent's highest standard, not as a measure of an entire development process.

Chen Yu Fei: the benchmark we need in order to measure

To know where a player stands, you must know who is standing across the net.

Chen Yu Fei was a top-four player at the time of the event. The defining trait of that group, judged on long-run data, is not the hardest smash or the prettiest net play. It is a stable low error rate across multiple games and the ability to raise quality in the deciding game.

Against such an opponent, there are three plausible scenarios for a lower-ranked player.

The first is being swept entirely, losing 0-2 with double-digit margins in both games. This is the most common scenario.

The second is producing one competitive game and then losing the decider. That is exactly this match. Its probability is lower than the first scenario but not rare.

The third is winning the match. Every analysis must rank that at low probability unless there is specific recent evidence.

What Nguyen Thuy Linh did on 26 September was move from the first scenario to the second. That is a real shift, but it is not a shift into the third scenario. The distance between the second and third scenarios is much greater than the distance between the first and second.

In data terms, the gap between the first two scenarios is usually measured by the number of long rallies the weaker player can win. The gap between the second and third is measured by points won after both players have passed fifteen in the deciding game. For this match, we know only the end result of the second measurement: 12-21.

Kunlavut Vitidsarn and the counter-attacking defensive archetype

In the men's draw, Nguyen Hai Dang's opponent was a player from the elite of Asian badminton, with a record built on defensive retrieval and patience in long rallies.

This is a particularly awkward type of opponent for a player in a development phase. A strong defender does not attack you with smashes. They attack you with time. They extend rallies until you are forced to manufacture the ending yourself, and when you manufacture an ending from an unfavourable position, the unforced-error rate rises.

The 16-21 and 14-21 games fit that model at the level of hypothesis. The margin did not explode. It accumulated. Nguyen Hai Dang lost points not through six- or seven-point runs by his opponent, but by conceding two or three points in stretches where he might have held the attack.

Read that way, this defeat was an expensive but valuable lesson. At twenty, playing near twenty points per game against a top-tier opponent is far better data than an easy win over a lower-ranked opponent in the previous round.

The statistics that do not exist

The hardest part of writing about this session is writing about what is absent.

The source material has no average shuttle speed. No count of rallies longer than ten shots. No net-point win rate. No unforced-error rate split by half-game. No count of service breaks. No movement distance. No point distribution by phase of the game.

As a data professional I must say this plainly: without those metrics, technical analysis is not possible. What we have is a scoreline. A scoreline is a good tool for answering who won and a poor tool for answering why.

If the data existed, here are three things I would measure from the Nguyen Thuy Linh match.

First, rally-length distribution by game. My hypothesis is that the first game had many short rallies because the opponent finished points early; the second shifted to medium and long rallies once Nguyen Thuy Linh gained the right to dictate; and the third returned to short rallies but in the opposite direction, meaning points ended early through the Vietnamese player's errors.

Second, the share of points conceded within the first three shots after losing the serve. I call this the break-rate. It distinguishes a player who loses because they are attacked from one who loses because they surrender position.

Third, the unforced-error count between eleven and sixteen points in the third game. That is the zone where most deciding defeats are shaped, because fatigue begins to matter but is not yet sufficient to excuse dropping points.

None of those three metrics exist here. So the only conclusion I permit myself is a conclusion about data structure, not about technical structure.

A lesson from the summer of 2026

I tell this story because it explains why I wrote the lines above.

In 2026, at thirty-six, working as a data consultant for a club in Indonesia's second tier, I built an xG model to persuade the head coach to push the defensive line high in a promotion play-off. The model projected roughly 1.8 xG for my team. That number was attractive enough to become an argument.

We lost 0-2.

The opponent sat deep and played on the counter. Their back line held a low block, and every one of our attempts became a harmless long shot from outside the box. The xG model still credited those shots, because it could not distinguish shot origin within the context of a defence that had dropped. I had ignored PPDA, the metric that measures an opponent's pressing intensity. I looked at a total and thought I was looking at a match.

The model was not wrong. I was wrong when I made it speak instead of my own eyes.

Since then I have kept a fixed ritual for every analysis: watch first, ask later. Never let the model lead. And when data and reality contradict each other, I write about the contradiction instead of choosing a side.

In this badminton match, data and reality also contradict each other. The data says: a national number one won a game against the world number four. Reality says: the third game ended 12-21. Both are true. Neither tells the whole story. This article exists because of the gap between them.

Croatia's PPDA and the art of picking up each pass

In 2026, analysing Croatia at the World Cup in Russia, I found something the standard tables did not show. Their PPDA was around 9.2 in the group stage, meaning they were not the tournament's highest-pressing team. Yet they led the tournament in ball recoveries in the opponent's half, around 12.4 per match. The difference lay in the timing of the press, not its intensity.

Croatia's PPDA is a reward for anyone patient enough to pick up each pass.

I thought about that lesson while reading this badminton scoreline. In badminton, the PPDA equivalent does not exist in general statistics, but it can be built. One candidate is the average distance between a player's position and the shuttle's landing point immediately before the opponent makes contact. That metric measures proactive occupation of space, and it is fundamentally different from points won.

A player can win 21-15 with a low value on that metric, meaning they won largely through opponent errors. A player can lose 19-21 with a high value, meaning they controlled space but failed to convert. A scoreline does not distinguish the two cases. For Nguyen Thuy Linh and Nguyen Hai Dang, we have no idea which case applies.

This is why I offer no conclusion about playing style. Drawing style conclusions from a scoreline alone is a habit I have committed not to repeat.

When data expires

In 2026, when the pandemic stopped every league, I was thirty-nine and working as a data consultant for a top-flight club. As the season prepared to resume, the board asked me to forecast form. I used data from the first fifteen rounds to build a model and advised the team to keep a possession-based approach.

We lost three straight matches when the league restarted. Opponents pressed harder, exploiting empty stadiums, and we lost the ball in our own half.

My model lacked two variables I had never considered: spectators, and spacing on the pitch. Without a crowd, the pressure on the ball carrier drops, which means pressing teams can push higher without being punished by noise and by familiar psychological reflex. Nobody taught my model that, because no data had ever existed for a world without spectators.

The pandemic taught me that data gets scared too — when the world stops, numbers are meaningless.

I repeat that here because it connects directly to this result. The context of an Asian Games is not the context of an annual professional badminton event. Delegation pressure, a schedule packed with multiple disciplines, the atmosphere of the athletes' village, the number of matches in a single day, and competing inside a framework where results count for the whole delegation are all variables that sit outside any model I have ever built.

When a player loses a third game 12-21, I do not know what percentage of that is stamina, what percentage psychology, what percentage tactics, and what percentage things that cannot be measured. The most honest thing I can do is state the number and state my limits.

Four checks I apply to every number

Because this is a data-driven article, I will set out the process I use. It is not specific to badminton; it applies to any sport.

Step one is checking the definition. A metric only means something when we know exactly what it measures. The point count in a badminton game is a very tight definition, so I trust it. Things like world number four require a specific date, because rankings change weekly.

Step two is checking the sample. How many matches support this conclusion? Here the number is two matches per player. That is far too small to speak about form, though enough to speak about one session.

Step three is checking context. The same scoreline can mean different things in round one and in a final. The same scoreline can mean different things against a world number four and against a world number forty.

Step four is checking for contradiction. Is there data that argues against my conclusion? Here there is: the 24-22 game argues against any pessimistic technical conclusion. So my conclusion has to live alongside that game rather than erase it.

These four steps are slow and produce no catchy headlines. They are also why I am still working after twenty-nine years of observing this industry.

The contrarian angle: the prettiest story is often the wrong one

This is where I have to say something many people do not want to hear.

The most natural way to tell this session is: a Vietnamese player won a game against the world number four. It sounds good. It shares well. It creates a sense of progress. And in a sense, it is honest.

But structurally, that telling has three problems.

Problem one is selection bias. We are picking the single game with a positive outcome inside a lost match to build the story. Pick the third game instead and the story inverts entirely. Which game you choose to narrate is an editorial decision, not an objective fact.

Problem two is that correlation is not causation. Winning one game does not prove the gap has narrowed. It proves something much narrower: during that window, in those rallies, the Vietnamese player performed at a level high enough to take the game. Same player, same opponent, same day, different window: 12-21. Whether the first event is repeatable has not been tested.

Problem three is the expectation consequence. When a result is framed as evidence of imminent progress, expectations for the next event are set higher. If later matches follow the same pattern — losing the decider by double digits — the story stops being a progress story. It becomes a different story, and the player under pressure in that new story is still the same person.

I am not saying the second-game win has no value. It has high diagnostic value: it shows that this player's technical ceiling reaches into a zone she can touch. But that diagnostic value only means something if we simultaneously accept the third game as an equally weighted data point.

Numbers are the prayer book, but intuition is the candle — I light both whenever I read a match. In this match, the candle illuminates the third game.

Squad depth and what the record book does not say

Another important signal sits outside these two matches.

In the session material, Vietnam had players in men's and women's singles, but there is no information about medals in team or doubles events. I do not have enough data to conclude, but the general pattern for mid-tier Asian badminton delegations is stable: talent concentrates in singles, while doubles and team events are thinner.

This has a strategic consequence. A badminton nation that relies on singles for international visibility places all its pressure on a small group of players. When that group meets top-tier opponents early in the bracket, the delegation's outcome depends on two or three matches, and every small error is amplified.

Compared with strong Asian badminton systems, the difference is not at the top player. It is at the third, fourth and fifth tier of each discipline — players who can reach the third round or quarter-finals of major events and create the depth an opponent must handle across a full week. Vietnam currently has no public data showing that tier exists, and this is a more important observation point than the results of a few individuals.

The Asian landscape and the position of a mid-tier delegation

Asian badminton has a fairly clear tier structure. At the top are systems with depth in almost every discipline. In the middle are delegations with one or two players capable of reaching deep rounds, but without the depth to sustain results across rounds.

What is notable is that the gap between the top and the middle is not located in the middle tier's number one player. A national number one from a mid-tier country can absolutely win a game against the world number four, as this match shows. The gap lies behind: at the number two, three and four players, and in the ability to organise training so that those players move closer to the international standard.

That is why I treat Vietnam having representatives in both men's and women's singles as a good signal but not a sufficient one. A healthy badminton nation should be measured by how many players can pass the third round at continental events, not by how many times a single player reaches it.

It is worth adding that Asian Games events differ from annual professional events. At a Games, value lies in medals for the delegation, in national visibility, and in creating benchmarks for domestic media. At a professional event, value lies in ranking points and income. These two pressures act on players differently, and I do not have the data to quantify the difference.

Why I do not compare two badminton nations directly

I live and work between two major badminton cultures, and I constantly have to remind myself of a trap.

Rankings, points systems, seeding criteria, and the way a national federation organises its team differ from place to place. Placing two metrics side by side without checking their original definitions is the fastest way to produce a wrong conclusion that looks highly professional.

Based on my experience tracking matches across both environments over many years, I have learned that cross-border comparisons only mean something when the definition, the reference date and the competition conditions are identical. Those three conditions are rarely satisfied at the same time.

In this case there is a more basic issue: the source material does not specify which edition of the Games it covers, and does not identify its source. Chen Yu Fei's world number four position is the position at the time of the event, but I do not have the specific ranking publication date being referenced. That means any cross-national comparison must stop at the qualitative level.

A responsible analyst states that limit rather than filling it with speculation. I choose to state it.

What I believe and what I do not

After removing what cannot be verified, here is what I permit myself to believe at medium confidence or above.

I believe Nguyen Thuy Linh has a technical ceiling high enough to win a game against a top-four player. The evidence is the 24-22 game, and that evidence is fairly strong because it came in the 20-20 zone, where error rates normally rise rather than fall.

I believe her biggest limitation in this match lay in the deciding game. The evidence is 12-21. Medium confidence, because one game cannot support a durable conclusion.

I believe Nguyen Hai Dang is at a stage where the gap to Asia's elite sits between seven and nine points per game, not fifteen. That is more positive information than the 0-2 result suggests.

I believe winning 21-6, 21-6 against a lower-ranked opponent is basic professional data and should not be dismissed. Holding concentration against a weaker opponent is a competition skill, not an inevitability.

I do not yet believe anything about playing style, tactics, in-match adjustment, service quality or defensive quality for either player. There is no data to believe.

Nor do I yet believe that the Asian Games generates world federation ranking points at a level equivalent to annual professional tour events. That is a technical question requiring independent verification against official documents, and I do not want to make an assertion I have not checked.

Two scenarios for the next twelve months

When the sample is small, the most honest way to write about the future is to present more than one scenario.

Scenario one is repetition. If the pattern of this match repeats at coming events — a close or won second game followed by a clear third-game defeat — then the problem lies in the stamina budget and in decision quality while tired. That is a problem that can be addressed through conditioning work and smarter scheduling, but it takes time.

Scenario two is conversion. If within six to twelve months Nguyen Thuy Linh begins winning deciding games against top-ten opponents, then the 24-22 game here will be reread as a starting point rather than an isolated peak. I assign this scenario a higher probability than the first, but lower than domestic media might expect.

For Nguyen Hai Dang the scenarios have a different shape. If the seven-to-nine-point gap narrows to three to five points in meetings with top-twenty opponents, he enters the group capable of causing upsets at continental events. If the gap holds or widens, the issue lies in the training environment and in how many elite matches he is exposed to each year.

I do not pick one scenario as a conclusion. I hold both and let the data decide.

Signals to track over the next six to twelve months

Analysis is only worth something if it generates future observation points. Here are four signals I will track.

First, Nguyen Thuy Linh's third-game scores against top-ten opponents. If she keeps losing deciders by eight points or more, the stamina-limit hypothesis is confirmed. If the margin narrows below five points, the hypothesis weakens significantly. It is a very clean measurement, because it requires only the score.

Second, Nguyen Hai Dang's win rate against top-twenty opponents. Against a leading Asian defender he held each game near twenty points without creating a turning point. If that pattern repeats across several matches, it points to a specific issue in finishing deadlocked rallies. And that is a trainable issue.

Third, Vietnam's results in doubles at continental events. This is a more important structural indicator than individual singles results, because it reveals the true depth of the development system. A badminton nation with only strong singles will always depend on two or three people.

Fourth, verification of the source and the Games edition behind the original data. It seems a small task, but it determines the entire historical context of the result, and therefore how every scoreline in this article should be read. If the edition is confirmed, we can compare the outcome with previous editions and know whether this was a step forward or a step back.

In place of a conclusion

That day, on the arena scoreboard, there were four lines of Vietnamese badminton scores. Two belonged to a defeat in three games. Two belonged to a defeat in two. None of them was a medal.

I maintain that a defeat can carry more information than a win, provided the reader is willing to read it without adding what they want to see. The 24-22 game is a real signal. The 12-21 game is also a real signal. Those two signals do not cancel each other; they coexist inside one player and one session.

What I want to know next time is not whether a Vietnamese player can win a game against a top-tier opponent. What I want to know is whether, in the third game, when both sides are equally tired, that player can still hold the score within five points.

The next scoreline will answer that. This scoreline answers only part of it, and I chose to write about the part it does not answer.