Vietnamese Swimming: When Three Sources Are Not Enough to Conclude a Single Race

In August 2026, after a session at the national swimming championships, I stayed behind alone in an almost empty grandstand to cross-check three data sources. The first was the organizers' electronic scoreboard. The second was the results page published on the federation's website. The third was the paper sheet held by the timekeepers. For the same 200m individual medley swim by one male athlete, the three sources produced three figures differing by as much as 0.37 seconds. Not much. But in swimming, 0.37 seconds is the gap between a place in the final and an early trip home. I am not retelling this to blame anyone. I am retelling it because it exposes a persistent paradox in Vietnamese swimming: we produce an enormous number of measurements, but very little reusable data. And when data cannot be reused, every conclusion is only valid for a single afternoon. Numbers do not lie, but people always find ways to lie with numbers. Since 2026, when I was a swimming reporter for a major newspaper, I have made it a habit to record everything I saw poolside: split times for each 50m, the feel of the entry, the way a body rotated underwater, even the breathing of an athlete climbing out. Twenty years later I still do the same job, except now I write in spreadsheets instead of notebooks. The problem is that Vietnamese swimming has never had a centralized, verifiable, and queryable data system. A national championship takes place, we publish final times that day, print them into the official record, and most of the detailed data disappears. There is no archive of 50m splits for every swim. There is no tracking record of turn counts, underwater metres, or stroke-efficiency indices. There is no longitudinal database long enough to answer a basic question: where does a 17-year-old male swimmer today stand compared with a 17-year-old male swimmer ten years ago? This is not a problem unique to swimming. But swimming suffers more than other sports, because swimming is a sport where differences live in very small numbers. In football, a team can win through mentality, tactics, luck. In swimming, if you are 0.5 seconds slower, you lose — full stop. A swimming nation that wants to progress therefore needs data several times more precise than other sports. But we are doing the opposite. I have sat with many head coaches of major swimming centres across the country. Everyone understands the problem. But when I ask: "Are you willing to change the way you record data over the next three years?", most go silent. Because good data costs money, costs people, and more importantly — good data forces you to face truths many people do not want to see. Swimming is a sport of splits. If there is one type of data Vietnamese swimming needs most urgently, it is split times — the time of each 50m segment within a single swim. In theory, every broadcaster displays split times during major finals. But when I tried to compile the split times of all Vietnamese athletes at a recent SEA Games from three different sources, I got a surprising result: only about 40% of the data could be cross-checked against at least two sources. The rest was either lost, mismatched in format, or recorded only as photographs of scoreboards that could not be entered into a spreadsheet. This is the first blind spot. Because split times are where the technical story actually happens. The final time of a 200m freestyle swim, for example, can be identical between two athletes — but if you look at the splits, you immediately see who starts strongly and who finishes strongly. An athlete who swims the first 50m in 24.8 seconds and the last 50m in 27.1 is an explosive but fading racer. Another who swims 25.9 and 26.2 distributes endurance far better. The same final time, but a completely different outlook. Over the past six years I have built a manual split-time table for the Vietnamese swimmers I follow. It contains 1,847 rows, collected from television, competition records, and scoreboards provided by organisers after meets. I use it to answer one simple question: where does Vietnamese swimming's biggest problem lie? The answer emerges very clearly when I divide a 200m swim into four segments: 0–50m (start and dive), 50–100m (maintenance), 100–150m (turns and rhythm change), 150–200m (sprint). Compared with athletes of the same age from other Southeast Asian nations, Vietnamese athletes on average lose the most time in the 100–150m segment — that is, the segment with the highest number of turns and the greatest demand for rhythm change. The average figure in my table: Vietnamese athletes lose an extra 0.4–0.6 seconds against regional rivals in the 100–150m segment of a 200m swim. In a 400m swim, that gap doubles. In 800m or 1500m, it multiplies fourfold or nearly eightfold. This is why Vietnamese swimmers can match regional rivals over short distances, but fall away as the distance grows. But here caution is needed. I call this the "middle-segment hypothesis", not a conclusion. My sample contains only 1,847 rows, not large enough to eliminate confounding variables: pool conditions, water temperature, the competition schedule that day, and whether the athlete was swimming the event as a primary or secondary target. A 1,847-row sample is a starting point, not a final conclusion. But it is enough to pose a question Vietnamese swimming has never asked: if the problem lies in the 100–150m segment, why do our training sessions still devote most of their time to starts and sprints? There is another type of data Vietnamese swimming almost entirely lacks: tracking data on turn technique and underwater work. At international level, a 200m swimmer performs roughly five to seven turns depending on distance and direction. A good turn can save 0.2–0.3 seconds compared with an average one. Multiplied seven times, that is nearly two seconds — an enormous margin at the elite level. In Vietnam, major swimming centres all train turns. But almost none measures turn efficiency with data. Coaches assess by eye: "This turn is not good yet", "That turn is fine". Eye assessment has value, but it cannot produce a database for comparing athletes, sessions, and development phases. I once coi mô hình là a compass rather than a scripture — but without it, we are lost. I proposed to a swimming centre in southern Vietnam that it install a fixed underwater camera to record every turn of its young athletes, then encode the footage into data. The initial cost would be only a few tens of millions of dong — cheaper than a single overseas training camp. The proposal has not been implemented, not because of money, but because of a lack of people capable of reading that data. This is the second paradox of Vietnamese swimming: we lack data, and when we have the chance to create it, we lack the people to use it. The third data type worth mentioning concerns the competition and selection system. How many meets does Vietnamese swimming hold each year? The answer depends on how you count. If you count national-level meets organised by the federation, the number hovers around four to five. If you include youth meets, regional meets, and internal centre competitions, the number rises into the dozens. But if you ask: "What must a 15-year-old athlete do to be discovered?", the answer is unclear. This is a direct consequence of the data gap. Without a national database storing the results of every meet, from youth to national level, selection depends on the eye of each coach and each locality. In countries with developed swimming, a 12-year-old who swims well at a provincial meet can be seen by the national centre within weeks, simply because results are entered into a shared system. In Vietnam, that gap can be years, or never. I know cases of talented athletes in the provinces who were only discovered at 17 or 18 — an age at which, in leading swimming nations, an athlete already has five to six years of systematic training inside a professional system. That lost time cannot be recovered. A national selection data system does not require high technology. It needs only three things: a central database, a mandatory data-entry process for every meet, and a person responsible for analysing that data and sending information to the relevant coaches. But those three things require a change in governance, not just in technique. When I combine split-time data with selection data, a notable pattern emerges. Among the athletes with the best results at national meets over the past ten years, most began structured swimming at ages 8–10 and had at least three years of youth competition before entering the national level. This is not a shocking finding. But it carries an implication Vietnamese swimming has not exploited: if we know that the most important development window is from 8 to 14, then all investment should focus on that period. Yet looking at the actual allocation of resources, most of Vietnamese swimming's money and manpower flows into the later stage — when athletes are already 16 or 17 and preparing for the SEA Games. We have names such as Nguyễn Thị Ánh Viên — a former SEA Games record holder in multiple events, Nguyễn Huy Hoàng — the distance swimmer who has competed at two Olympic Games, Trần Hưng Nguyên and Phạm Thanh Bảo — the next generation now being counted on. But when I tried to chart each of their development curves with data, I realised I had only scattered points: a few major meets, a few recorded swims, and enormous gaps in between. I am not saying investment in the SEA Games is wrong. I am saying that if all resources are concentrated at the final stage, we are trying to repair a house from the roof down. Swimming is a sport of long-term accumulation: every year of correct training at age 10 multiplies into an advantage that cannot be compensated for at age 20. Longitudinal tracking data — data on an athlete's development across many years — is the most important type of data Vietnamese swimming does not have. Now comes the hardest part. Having laid out three types of data Vietnamese swimming lacks, I must say something that contradicts what I have just written. More data does not automatically mean Vietnamese swimming will improve. This is the trap many small sporting nations have fallen into: investing in a data system, generating mountains of numbers, and then no one uses them. Data only has value when it changes a decision. If you have perfect split-time data but coaches keep the same old training plans, that data is worthless. Worse, data can become a ritual. I have seen analysis meetings drag on for three hours with hundreds of charts, ending without a single decision changed. At that point data is just a way for everyone to feel they are working seriously. And correlation, however strong, is not causation. This is why I emphasise reusable data from the outset. The best data is not the prettiest data. The best data is data a coach picks up and says: "Because of this number, I am changing next week's plan." If I had to choose one single thing for Vietnamese swimming to do in the next three years, I would choose building a national split-time archive — simple, cheap, and startable at the very next national championship. No expensive software is needed. A decision is needed. When the pool stands empty, every model collapses; but a carefully recorded dataset remains standing, waiting for someone to read it.