Vietnamese Swimming and the Data Gap: What the Lane Has Not Told

There is a habit I have carried through many years of working with data: after every competition, I do not look at the medal table first, I go looking for the splits. At a domestic swimming meet a few seasons ago, I asked for the 50-metre split file of an athlete who had just set a personal best. The organisers replied that the timing system could only output the final number. That personal best, in data terms, was reduced to a single bright point on a time axis. A small GPS drift is enough to teach me: verification is everything. In 2026, because of a single synchronisation error, I once misread a player's sprint distance in a V.League match. The error was so small it was nearly invisible, but it was enough to make me understand that a number without a clear source cannot be trusted, no matter how beautiful it looks. When I moved to covering swimming lanes, that feeling returned intact. Swimming is a sport measured to the hundredth of a second, which at first glance makes it the discipline with the most solid data foundation of all. In Vietnam, however, most published swimming data stops at the shallowest layer: the finishing number. Deeper layers — splits, stroke rate, distance per stroke, underwater speed after the start, recovery time between heats — are almost absent from the results tables that fans can access. That gap has nothing to do with how many medals we hold. It has everything to do with how much we know about those medals. To understand why this gap matters, one has to look at the structure of Vietnamese swimming. For many years, the key training centre for the sport was a national-defence sports facility in Can Tho, tied to the name Nguyen Thi Anh Vien. It was a centralised model: a small group of athletes, one lead coach, a relatively closed physical environment. That model once produced results far beyond its scale. But it also concentrated the data on a few individuals, while the rest of the system was barely measured in any structured way. When an athlete delivered a big result, we had photographs, video, a finishing number — but very little of what lay in between. I often compare this with the field I come from: football. In professional football, a GPS unit on a player's back records distance covered, accelerations, heart rate, and even the sprints that led nowhere. In swimming, the equivalent data has to be splits and technical indicators. In Vietnam, the number of meets whose timing systems can publish 50-metre splits to the public can be counted on one hand. This creates a paradox: the sport measured by the most precise numbers is the very sport where an analyst like me struggles most at the middle stage. I believe in numbers, but only after a number has passed three rounds of checks. In swimming, those three rounds are: the data source, the consistency of the measuring system, and comparability across meets. In Vietnam, all three rounds have problems. The first round is the source. Swimming results are usually published as a flat list: name, nationality, final time. The viewer does not know whether that time was measured by an automatic system or by a human finger, or what the margin of error was. Hand timing alone can drift by several tenths of a second — enough to reshuffle a ranking, enough to turn a place in the final into a seat at home. The second round is consistency. An athlete competing at three different meets may face three different timing systems, three pool configurations with different water temperatures and depths. Swimming is a discipline extremely sensitive to environment: water one degree colder, a shallower wall, a current generated by the filtration system — all of it affects times by tenths of a second. Without recording those conditions, two results cannot be compared fairly. The third round is comparability across meets. This is the hardest round. In football, I can place an Austrian team's data next to a Portuguese team's as long as they share the same data system. In Vietnamese swimming, a result at a national championship and a result at the SEA Games may be measured by two different systems, published in two different formats, and no one maps them together. As a result, most conclusions about Vietnamese swimming today are built on the shallow layer. We know who finished fastest. We do not know why. Take a more concrete example. Nguyen Thi Anh Vien, born in 2026 in Can Tho, is the most successful swimmer in the history of Vietnamese sport. She was the first Vietnamese athlete to win a swimming medal at an Asian Games, in the 400-metre individual medley in 2026, and the first Vietnamese swimmer to reach an Olympic final, in the same event, at Rio 2026. Her national record in the 400-metre individual medley is 4 minutes 36.85 seconds, as recorded in world swimming federation files. But if you ask me where her biggest leap came from — the start, the turn technique, or the ability to distribute effort over the final 100 metres — I have to answer that I do not have enough split data across her career to say with confidence. I have her finishing times at many meets. I lack the splits. This is a deeply practical matter. If we knew Anh Vien improved mainly through turn technique, we could teach that to the next generation. If we knew she improved through training volume, we would know to invest in the programme. If we knew she improved through some psychological factor, we would know to add a psychology team. Without split data, every inference is guesswork dressed up in numbers. Anh Vien is the ideal case for analysis because she has more public data than anyone. For other athletes, the situation is far harder. At the SEA Games, Vietnamese swimming generally sits in the medal-contending group but does not lead. Across many editions, Singapore has dominated the pool with a carefully invested squad, including names who have won world and Olympic titles. Vietnam, Thailand, Malaysia and Indonesia share what remains of the medal table. That picture has been fairly stable over the years. This stability can be read two ways. First: Vietnam has maintained its position in the contending group, which is an achievement. Second: if the picture has barely changed across many editions, then perhaps the foundation producing the results has barely changed too — and that is a sign the system has not produced a leap. To distinguish between these readings, I need more than a medal count. I need the average age of the squad, the number of first-time SEA Games athletes, the number of athletes who improved their personal bests compared with the previous edition, and the size of those improvements. Most of those numbers are not published. On the talent pipeline, Vietnamese swimming has seen some male figures emerge after the Anh Vien era, including Nguyen Huy Hoang — an athlete who has won medals at continental level and competed at the Olympics in long-distance freestyle events. But in terms of depth, the number of Vietnamese swimmers who can compete at continental level remains thin. One or two figures do not make a generation. The point I want to stress: the pipeline is not a story about innate talent, it is a story about data and systems. Countries with strong swimming traditions do not merely have more pools. They run systems that track young athletes from a very early age, measure physical development, and adjust programmes based on numbers. When data is not collected, choosing whom to invest in becomes a decision based on feeling. The economics of swimming are also worth viewing through a data lens. The cost of taking an athlete to an international stage includes training camps, pool rental, travel, nutrition and medical care. Without data tracking the return on investment, budget allocation tends to fall into one of two extremes: pouring everything into a few established names, or spreading it thin across everyone. Neither is optimal. One simple indicator I would like to see: cost per second improved. If an athlete receives extra budget and improves a personal best by one second, is that investment more or less effective than spending the same money on another athlete? Without split and recovery data, this question cannot be answered. In football, I am used to the concept of expected goals — a model estimating the probability that a shot becomes a goal based on position, angle and pressure. In swimming, the equivalent concept could be "expected time": based on an athlete's splits at previous meets, we estimate a reasonable time range for each segment in a new race, then compare against reality. If an athlete finishes half a second faster than the model predicts, that may be a sign of genuine progress. If they finish a full second faster but their splits show they only accelerated over the last 15 metres — while the rest of the race was slower than predicted — then it may simply be a beautiful result on a beautiful day, not repeatable progress. This is precisely the tool I lack when covering Vietnamese swimming. I have the result. I do not have the structure inside the result. If I stopped at the call for "more data", I would not have done my job. One thing must be said clearly: more data does not automatically produce results. There is a familiar trap in how Vietnamese sport tells its stories: whenever an athlete shines, the whole system is credited; whenever that athlete declines, the individual is blamed. This is a form of confusing correlation with causation. The emergence of a rare talent does not prove that the development system is working well. It only proves that the system has produced at least one rare talent. Conversely, the fact that a successor generation has not repeated earlier results does not prove the system has broken. It may simply be that probability has not smiled again. To distinguish between these two possibilities — a weakening system, or mere random noise — we need exactly what we are missing: data on the entire population of athletes, not only the winners. Data does not tell stories; it records everything so that I can tell them myself. But if data records only the winners, the story I tell will always be one-sided. There is one more point to weigh: the cost of collecting data. An automatic timing system capable of publishing splits is not cheap. It requires sensors at every 50-metre mark, synchronisation software, and operating staff. For a sport with limited resources, the choice between buying more equipment and buying more training-camp slots for athletes is a real trade-off. I do not argue that every Vietnamese pool needs a top-tier system immediately. What is more feasible: standardising manual record-keeping. A timekeeper can record 50-metre splits by hand with reasonable accuracy, and as long as that data is stored in a shared database, we already have a foundation for analysis. The current problem is not strictly a shortage of equipment; it is a shortage of the habit of archiving. A culture of support does not live in the shouting, it lives in the frequency of patience. A sport that wants to go the long road needs fans patient enough to care about meets without medals, and perceptive enough to ask questions about process, not only about results. If I had to pick one signal to track over the next one to two years, I would not pick medals. I would pick the appearance of split data on the results sheets of national meets. A competition that publishes full splits is a sign that someone inside the system has decided that process matters as much as outcome. I could be wrong. Perhaps a new talent will emerge and render all of this analysis moot. If that happens, I will be the first to update my table.