Vietnamese Swimming Data: The Gap Between Feeling and the Numbers

During an internal test session at the national swimming training centre, a young swimmer completed the 200m freestyle in 1 minute 52 seconds. The coaching staff nodded and recorded it. When I split the data into four 50-metre segments, the picture flipped: the opening segment was nearly two seconds faster than the optimal distribution, and the final segment collapsed. The speed at the start did not come from a fitness base. It was a loan taken from the segment behind it, and the bill came due in the last fifteen metres. I tell this story to expose a gap larger than one athlete: Vietnamese swimming is short on data infrastructure at a basic level. We measure results with a stopwatch, but we rarely measure how those results are made. A 200-metre race has four segments, each with its own stroke rate, stroke length, breathing rhythm and turn time. When only the final result is stored, the analyst is like reading a book by looking only at the cover. And without data, every tactical argument is just sentiment. This is why I begin every evaluation by reconstructing the race structure. Based on years of watching domestic meets, I keep seeing a recurring pattern: Vietnamese swimmers often start and open very well, but speed drops sharply from the third segment. That decay indicator, if recorded consistently, would tell coaches how to distribute effort instead of relying on feel. I usually use three measures for a race: average speed per segment, stroke rate, and distance per stroke. These three move against each other. Raising stroke rate can hold speed in the short term, but distance per stroke shrinks, meaning the swimmer is burning fuel faster than needed. Over 200 metres, the optimal strategy is almost always to keep distance per stroke stable and lift stroke rate gradually toward the end. The swimmer in that test did the opposite. To grasp the scale, set it against the international standard. At the Tokyo 2026 Olympics, the men's 200m freestyle champion swam with a near-linear distribution, with only about one second between the fastest and slowest segment. At our national level, that gap is usually three to four seconds. The difference lies in the ability to sustain. This is a training trait, and it can be measured and corrected. That gap is not a verdict. It is a training marker, and markers can change. A familiar trap appears here. When people see that swimmers with a high stroke rate often win, they easily conclude that raising stroke rate creates victory. The link is real, but the mechanism lies elsewhere: a high stroke rate is the result of a solid technical base and full fitness, not the direct cause. Forcing a weak swimmer to stroke faster will only make them sink sooner. Before asserting that X causes Y, I force myself to point to the physical mechanism connecting the two; otherwise I should only say they are related. In Vietnam, most training centres still record by hand. The coach clicks a stopwatch, writes in a notebook, and the data stays in the notebook. There is no shared database, no cross-centre comparison standard, no way to trace a performance slump back three weeks. This is the biggest bottleneck, and it does not require expensive technology to fix. A simple spreadsheet, capturing all four segments in every session, has already made a difference. I once pitched this model to three centres, and two of them kept it going after a year. In 2026, when pools closed due to the pandemic, I joined a group of coaches in recording dry-land data. We measured band resistance, breathing rhythm, and reaction time. When the pools reopened, the group with complete logs regained form about two weeks faster than the other group. That is a correlation, and I am careful not to call it causation. But it was enough to convince the coaching staff to keep logging. A shock in swimming rarely comes from a single race. It comes from a process. Every shock has its own probability. We call it a shock when we have not checked the tables in time. When a swimmer breaks a national record, the media calls it a surprise. But look at the training log from three months earlier, and the curve had already signalled it. Most people look at the result to understand the athlete. I look at the process to understand the result. There is one variable my model does not fully quantify: competitive psychology. A packed pool can push an athlete past their limit, or crush them. I once watched a young swimmer go two seconds faster in a heat with a crowd, then collapse in the final under pressure. Data cannot explain that share of variance. I note it as a confidence interval, rather than pretend the model covers everything. What I want to leave behind is a forward-looking proposal: start recording as a national habit. Four segments for every race, every session, every week. Within three years, we will have a database thick enough to see the signal before it becomes a headline. A tactical era dies when its data tables no longer have readers. I do not want Vietnamese swimming to walk that path.