Vietnamese Swimming Is Missing Its Data Layer: From 50m Splits to the A-Cut

At a domestic swimming meet, the electronic scoreboard displays results to two decimal places. Anyone can read 1,234.56. But when I asked for the 50-metre split data from the very lane that had just finished, the answer was that it did not exist. The scoreboard records arrivals, not journeys. The organisers keep the final time, not the splits. Coaches remember the feeling, not the figures. A 200-metre race — decided by four lengths swum at completely different rhythms — gets compressed into a single line, and then argued about on instinct. I have followed swimming for nearly two decades, from provincial youth meets to the SEA Games and Olympic selection cycles. My professional roots are in the pool: in 2026 I started out as a swimming reporter. That was where I learned something that later became a rule — in swimming, the one thing nobody can argue with is the clock; the one thing almost nobody preserves is how that clock was made. Vietnamese swimming has an obvious contradiction. At the results layer, we have achievements: SEA Games medals counted one by one by the media, Olympic qualifications, national records set and broken. At the data layer, the gap is far wider. There is almost no public database that would let an outsider look up the 50-metre splits of a Vietnamese swimmer across three consecutive competitions. There is no standard record of reaction time, turn time, or average stroke rate per meet. There is no comparison table between the 25-metre pool and the 50-metre pool for the same swimmer within the same training cycle. This does not come from a lack of expertise. Our coaches spot technical problems with a professional eye, and they are usually right. What they lack is the infrastructure to prove what they see, and to pass it on to whoever comes next. The competition structure of Vietnamese swimming places almost all its weight on the SEA Games. That is where medals come from, where targets are assigned, where budgets are justified. The Asian Games is the real test, where the distance to Japan, China and South Korea becomes impossible to hide. The Olympics is a selection puzzle: the A-cut and B-cut standards operated by World Aquatics, plus universality places and national federation quotas allocated by their own criteria. In such a system, results are worth more than process, and results are recorded far more completely than process. A swimmer who finishes third gets a name. But the fact that they swam the first 50 metres 0.4 seconds faster than their personal best and then collapsed over the final 50 — a detail worth an entire training cycle — usually vanishes from every report. And when the detail vanishes, the debate about that swimmer returns to exactly where it usually lives: the feelings of someone in the stands. Every strong swimming nation has two things Vietnam does not yet have enough of. The first is a competition calendar dense enough to generate samples. The second is a habit of record-keeping disciplined enough to make those samples mean something. Nguyễn Thị Ánh Viên left behind a SEA Games medal collection running into the twenties, along with appearances on the Olympic stage. Nguyễn Huy Hoàng is credited with Asian Games medals in the distance freestyle events, and Olympic appearances in his specialist events. Those landmarks are real; nobody disputes them. But they exist as isolated peaks, not as a continuous series of data that would let anyone see a slope. To know whether a swimming nation is rising or flat, you need the slope. A peak only tells you someone once got up there. The analytical framework I use at work has nine dimensions: technique, performance and data, competition system and qualification mechanics, the global landscape, rules and anti-doping, athlete career, risk profile, media and expectations, and finally industry ripple effects. Running this framework over Vietnamese swimming produces the same result on almost every attempt: most dimensions return empty. Only two have enough material to speak with any confidence — the competition system, and the global landscape. The technical dimension returns a blank: no 15-metre post-start segment, no separate turn times, no underwater dolphin-kick count per length. The performance dimension is the same: final times exist, but comparison coordinates — world record, all-time list, seasonal ranking — do not, so a swimmer cannot be positioned on the improvement axis, and the magnitude of their progress cannot be measured against the standard for that event. The career dimension lacks year-on-year growth data, so no age-versus-performance curve can be drawn. The risk dimension lacks systematically recorded injury history, so nothing can be said about safe training loads. This is where a principle gets tested. When data does not exist, people tend to fill the gap with assumption. I have seen analyses that looked highly professional, full of charts and clean colour coding, where every data point was inferred from a single race or from one person's account. That kind of analysis is more harmful than nothing, because it manufactures certainty precisely where a question mark belongs. The only approach I accept is to state it plainly: insufficient information, cannot assess. A blank field honestly marked is worth more than a blank field filled with guesswork, because the honest blank tells you exactly what still needs to be collected. Without that step, every layer above it — forecasting, ranking, valuing potential — is built on sand. If I had to choose one starting point for a Vietnamese swimming data layer, I would choose the 50-metre split. It is the smallest unit that still carries diagnostic value. The reason lies in the structure of the sport. Swimming is a discipline where the smallest error changes the ranking: one hundredth of a second can be a medal, or it can be fourth place. Meanwhile, a 200-metre individual medley consists of four lengths in four different strokes — butterfly, backstroke, breaststroke, freestyle — with four different biological mechanisms and four different ways of distributing energy. The final time tells you the total, not which segment took what, and not where a swimmer lost the medal. Only from splits can you reach the metrics that actually mean something. Stroke rate — cycles per minute — measures how hurried the action is. Distance per stroke measures how efficient it is. Multiply the two and you get stroke index, a composite measure showing whether a swimmer is compensating for inefficiency with frequency, or compensating for low frequency with efficiency. Two swimmers who finish in the same time may be doing entirely different things in the water, and the training direction for each of them must differ. Looking only at the scoreboard, they are identical. There is more. The first 15 metres after the start is where many medals are made, because it combines three elements: reaction time to the signal, the quality of the underwater dolphin kick, and the moment of breaking the surface. The 15-metre underwater limit is a specific technical rule, and the best swimmers in the world exploit it centimetre by centimetre. In Vietnam, these metrics are barely recorded, so the debate about a "weak start" has run for years without data to close it. Everyone is half right, and nobody can prove the other half. Then there are the turns. In a 50-metre pool, a 1,500-metre freestyle race contains twenty-nine turns. If each turn saves two tenths of a second, that is nearly six seconds in total — an enormous margin at continental level. But to know who turns well, who turns badly, who turns well in the first half and fades in the second, you need turn data separated out for each instance. Without it, every comment is a prediction wearing a data costume. Three years after I built a PPDA index for a football club in Bình Dương and named it Bình Dương pressing, I realised the swimming pool was missing something far more basic. In football, I had to struggle to define an abstract concept and then go hunting for data to match. In swimming, the data is right there, created at every wall touch, and abandoned the moment the scoreboard goes dark. World Aquatics operates the Olympic qualification threshold system on two levels: the A-cut and the B-cut. Meeting the A-cut generally means an official place, with a maximum of two swimmers per event per country. Meeting the B-cut only opens a waiting door: a place is granted only if slots remain after A-cut places are filled, and the order is decided by time ranking. Alongside these are universality places and quotas for countries without a qualifier. For Vietnam, this system means the cleanest road to an Olympics is meeting the A-cut in a specialist event. Every other road depends on variables beyond the swimmer's control and the coach's control. And to know whether a swimmer can reach the threshold, one thing is minimally required: a continuous series of competitions in a 50-metre pool, under controlled conditions, spanning at least one full training cycle. Without that series, any Olympic projection is a belief presented in the form of a table. This is where a technical trap appears that is rarely discussed. Swimming in a 25-metre pool means twice as many turns, and every turn produces a push off the wall. As a result, short-course times are always faster than long-course times in the same event, with differences that can reach several seconds over distance events. So when a Vietnamese swimmer races in a 25-metre pool at home and then steps onto a 50-metre stage, the times cannot be compared directly. Conversion requires parallel data on both pool types for the same person, in the same physical state, in the same training phase. We effectively do not have that dataset, so every cross-pool comparison silently carries an undeclared margin of error. The calendar matters too. Domestic meet density is not enough to generate samples, and the number of international meets Vietnamese swimmers can access is limited. A swimming nation improves through the number of times it is tested under real pressure. Without that number, every judgement about big-meet temperament is inference, even when the person making it has a very good eye. Not everything can be reduced to numbers. That is what I learned after years of building models. xG in football is not wrong; football simply contains an irrational element, and a good model is one that maps the boundaries of that irrationality instead of denying it. Swimming has its own irrational element, but it is much smaller, because the lane is measured by touchpad systems at the wall and no referee awards or denies points for a beautiful or ugly splash. Precisely because the irrational element is small, swimming is the most data-amenable sport in the Olympic family. And yet it is the one Vietnam records least. That gap does not come from technical difficulty: a stopwatch and one person standing at the end of the pool would be enough to produce splits for every lane at a national meet. What is missing is a decision to treat data as an asset rather than as administrative paperwork. I once treated models as scripture. Now they are only a compass — but without one, you get lost. There are three traps an analyst is most likely to fall into when the material is thin. The first is applying foreign models wholesale. Training benchmarks are built in swimming nations with thousands of athletes across every age group, competing in dozens of meets a year. Applying their thresholds unchanged to a Vietnamese swimmer with a calendar many times thinner is a sample error at the root. The same threshold, placed in two different data densities, produces two contradictory conclusions — and both look plausible if the reader does not know the sample size behind them. The second trap is inferring from small samples. A swimmer suddenly going 1.5 seconds faster than a personal best at a domestic meet might be genuine progress, or a new pool, or better recovery, or weaker opponents meaning no need to spend energy in the heats, or simply a favourable timing touch. Separating those possibilities requires at least three repetitions under equivalent conditions. One fast swim is news. Three fast swims are a trend. Ten are a capability. The third trap is declaring a rebuild when there is nothing yet to rebuild. When the stadium is empty, every model collapses. I rebuild from the half-burnt data. But if the half-burnt pile is genuinely only ash, constructing a new model is theatre. There are moments when the correct answer is to publish a map of the unreadable data: which zones are known, which are unknown, and which cannot be known with current resources. That honesty does not cost anyone credibility. It gives credibility somewhere to stand. Swimming carries two occupation-specific risks, and both are data problems. Shoulder injury from repeated high-amplitude loading is the most common in freestyle and butterfly swimmers; knee injury is common in breaststrokers. Both accumulate quietly over years. Without weekly load records, they are discovered only when it is late — and by then the story is usually told wrong, as an accident rather than as the consequence of an untracked process. The second risk belongs to the rulebook. The anti-doping system operates through whereabouts obligations and unannounced testing. For young athletes, the greatest danger zone is usually not deliberate doping but unverified supplements and therapeutic medication without an exemption. This is an area where a properly managed medical record can protect a swimmer, rather than merely documenting them. There is a third kind of risk few people calculate: media risk. After every SEA Games, the mood rises and then settles, and while it rises, conclusions are issued as though the sample were sufficient. Four years later, when an Olympic cycle closes without the expected result, the search begins for someone to blame — usually the coach or the swimmer, people who were never given the tools to know where they stood on the track. Reputation is only a name. What remains is always the way you read the race. The notable thing is that Vietnamese swimming does not lack skilled people. Experienced coaches still identify most technical problems correctly, with a professional eye and a long memory. What is missing is the infrastructure to prove what they say, and to pass it to a successor without starting the story again from scratch. A good coach can spot a swimmer with poor turns. A system with data will know which turn number they were poor on, after how many metres, by what percentage against their own previous season, and from which week of the training cycle. The distance between those two levels is the distance between a generation and a swimming nation. The three signals I will watch in the next cycle all sit at the infrastructure layer, not the medal layer. The appearance of split data in the official results of national meets would be the clearest sign: once organisers publish splits by default, the entire analytical layer above automatically has raw material, and fans gain the right to argue with data instead of with feeling. Next is the continuity of athlete records year on year, rather than bundled into a single meet and closed. And finally, the number of times Vietnamese swimmers race in a 50-metre pool under international conditions — the only measure that can be placed beside the qualification thresholds without conversion. I do not expect these to appear simultaneously. But if even one of them appears and is sustained over several seasons, the structure of Vietnamese swimming analysis will change. At that point, answering the question "why is this swimmer slower than last season" will no longer be a debate between two people with different memories. Based on my experience following swimming competitions from domestic youth meets to regional games, I believe the largest gap in Vietnamese swimming lies where nobody looks: the data layer. It does not produce medals immediately. It only makes every medal that follows explicable. If a swimming nation can produce individuals who exceed their own limits without a decent record-keeping system, then the more frightening question is this: how many others did that nation lose, simply because nobody managed to see them at the right moment?