The Data Blind Spot of Vietnamese Swimming: When the Lane Is Not Measured by Numbers

In the final of the women's 200m individual medley at a SEA Games, the silver and bronze medals were separated by exactly nine-hundredths of a second. Up on the stands, none of the thousands of spectators sensed that fragile boundary until the electronic board flashed. But back in the technical area behind lane four, an analyst was staring at a tablet: the losing swimmer's stroke rate fell from 48 to 44 strokes per minute over the final 50 metres, while distance per stroke went up. She did not lose because she swam weakly. She lost because she chose the wrong patch of water in which to accelerate. That moment exposes what nearly two decades of watching Vietnamese swimming has taught me. We own a system for recording results, not a system for measuring process. The scoreboard prints only the visible part. The submerged part, where the race is actually decided, remains a blind spot the numbers never reach. For more than a decade, Vietnamese swimming has lived inside a paradox. At the SEA Games, we were once a serious force. The run of gold medals from a golden generation, the name most associated with it being Nguyen Thi Anh Vien, turned swimming from a minor discipline into a flagship of Vietnamese sport. Anh Vien, born in 2026, won dozens of SEA Games golds and appeared at the 2026, 2026 and 2026 Olympics. That is real achievement, inscribed in every sporting chronicle, and nobody has the right to deny it. Yet behind those achievements a question still hangs. What produced that success, and what will produce the next one? I have carried this question for years in my role reporting on swimming for the Vietnamese market. I am not a coach. I do not stand at the pool edge correcting anyone's arm recovery. I watch from the perimeter. And from the perimeter, what I see most clearly is a data gap. In football, the analytics revolution happened in public. xG, PPDA, progressive passes, pressing intensity — a complete language emerged to break a match down into countable particles. In swimming, the equivalent language exists in theory but is almost never used in Vietnam. That is the gap I mean. Not a gap in talent. A gap in method. Swimming, by its physical nature, is the most measurable of sports. A swimmer does not run on land with a ball at his feet. He moves through a medium of consistent density, over a fixed distance, along a repeatable path. Every stroke cycle can be counted. Every second can be dissected. Three foundational metrics are available to anyone willing to record them. The most common is SWOLF — swim golf — the sum of the time to complete one length plus the number of strokes taken in that length. The faster you swim with fewer strokes, the lower the score and the higher the efficiency. Then there is DPS, distance per stroke, the distance covered with each pull. And stroke rate, the number of cycles per minute. The latter two are inversely related: to swim faster you can raise the rate, raise the distance, or raise both. But raising the rate usually lowers the distance, and vice versa — that is the balance point every coach must find for each swimmer. If you record only the final result, you know who won. If you record those three metrics, you know why. That is the whole problem. Take a concrete example. Suppose a swimmer covers 100m freestyle in 52 seconds. On the board, that is a good result. But split the first and second 50, and something else emerges. If the first 50 is 24 seconds and the second is 28, he has bled four seconds after the turn. If stroke rate holds but DPS drops from two metres per cycle to 1.7, he is losing efficiency, not stamina. The two causes demand completely different cures. A coach watching only a stopwatch will say the swimmer needs conditioning. A coach with data will say the acceleration zone needs fixing. I have studied the equivalent metric set in football for years. In an earlier piece on a major tournament, I calculated PPDA for a strong national team and found that their collapse came not from bad luck but from strikers running too little. The principle carries over to swimming. Numbers never lie, but they know how to hide. The question is whether anyone bothers to look. So what does Vietnamese swimming actually record? The honest answer is mostly just the final result. In official competitions, World Aquatics timing systems record each length, and at some major meets they also record 25-metre speed. But at the daily training level, in provincial centres, most data stops at the stopwatch reading. Nobody counts strokes systematically. Nobody measures distance per cycle. Nobody splits turn times. That is the gap I call the data blind spot. The irony is that the tools to fill it are cheap. A phone camera on the pool deck, a simple stroke-counting app, a spreadsheet — that is enough to begin. I once built a historical dataset for a different sport in far harder conditions, and I know its value. During the global shutdown of grounds, while other journalists waited, I rebuilt history to understand the present. COVID closed the stadiums, so I reopened the directory of forgotten leagues. No league is meaningless, not even the ones nobody watches. With Vietnamese swimming, the problem is a little harder because data sources are scarcer, but it is also clearer. I propose an approach called the confidence framework — a set of criteria for deciding which threshold is trustworthy. Before concluding that a swimmer is improving or declining, you need at least ten controlled swims under the same conditions, the same distance, the same pool type. Below that threshold, every conclusion is statistical noise. This is a principle I always set in advance, and it protects me from mistaking a single bright day for a trend. Now, separating luck from ability. I have a tool I call the reality coefficient. For every good result, the question is what share is ability and what share is circumstance. In swimming, circumstance has many variables. A pool with standard depth, ideal water temperature and good filtration produces faster times than another. A swimmer who has just completed a taper will swim faster than usual within a short window. A high-tech suit can save a few hundredths. None of this is cheating — it is all legitimate. But if we mistake it for pure ability, we build a skewed forecasting model. This is why I distinguish between a record and a trend. A record is a single data point, exposed to every circumstantial variable. A trend is a series of data points, and only the series can tell the truth. People look at records and cheer. I look at the curve and shake my head. People look at the price tag; I look at the curve. Many deals die before they are announced. I have witnessed this in another sport. A player rose on the back of an extraordinary run of goals, and the entire media world crowned him a genius. But when I calculated expected goals, his actual output far exceeded expectation. That is a sign of luck, not sustainable ability. I wrote a warning, and the market later confirmed it. Apply the same logic to swimming: a single fast time proves nothing. What proves improvement is a steadily rising curve across many swims in many conditions. Luck is something I do not have. I have probability and data thick enough. There is one phase in swimming where data matters more than the main race itself: the start and the underwater phase. Over short distances, the first 15 metres after the start and after each turn can account for nearly a third of the race. A swimmer pushes off the wall, goes under, and surfaces. The time in this phase depends on push-off power, body angle underwater, and the number of dolphin kicks. This is the zone where a tenth of a second can be created or lost almost invisibly. A coach without underwater camera data is completely blind here. He only sees the swimmer surface late, but not why. Then there is the turn. Turn time can be measured separately from swim time. A poor turn can cost half a second, and in a 200m race with seven turns, the accumulated error approaches four seconds. Four seconds is the distance between a gold medal and a place outside the rankings. Yet in most Vietnamese training centres, turn time does not exist as a separate metric. It is folded into total time, and with it, into oblivion. I want to go deeper into the structure of a 200m race, because this is the distance where split data delivers the greatest value. A 200m swimmer must solve the problem of distributing effort across four lengths. He may choose a negative split, swimming faster toward the end. He may choose a positive split, fading toward the end. He may choose an even split, holding a steady pace. Each suits a different body and a different mentality. No strategy is absolutely right. But one thing is certain: without 50-metre splits, the swimmer and coach cannot know which strategy they are actually choosing. They think they are swimming evenly, but in reality they are fading. They think they are accelerating, but in reality they are holding steady. The feel of the water is unreliable. Only the numbers are reliable. Here I must say something plainly that many in the field do not want to hear. We are judging swimmers by the audience's feeling, not by the specialist's data. A medal is received with applause, and that applause measures nothing. After the stands go dark, what remains are numbers usable for the next cycle — or nothing at all. Vietnamese swimming, in most cases, ends a SEA Games with very little left besides the medal. This is where I want to argue against myself. Because there is a temptation that data analysts like me easily fall into: turning everything into numbers and losing the human part of the race. I have been criticised as heartless. I have been called cold for pointing out that a team would be relegated while still unbeaten. But I do not believe data and humanity exclude each other. On the contrary, correct data is a tool that protects people. It protects a swimmer from being judged on one bad swim. It protects him from being inflated and then abandoned when luck runs out. But data has real limits, and I must admit it. Vietnamese swimming lacks data not out of laziness. It lacks data because the recording system has not been built. There is no dedicated staff. No standard process. No common standard for comparing centres. A coach in one province counts strokes differently from a coach in a big city. And when data is inconsistent, it cannot be compared. This is a systemic issue, not an individual's fault. What is more worrying is that this gap affects not only the present but the next cycle. When a golden generation ends, what determines the success of the next generation is not the medals already won but the accumulated data archive. If this cycle we have Anh Vien, the next cycle we need a system thick enough to produce the successor. And that system must begin by recording the most basic things: every session, every stroke, every turn. I am not naive enough to think data will save a swimming nation by itself. Without hardware and people, data is just a pile of numbers nobody reads. But I also know the converse: a swimming nation without data will forever depend on luck. And at the elite level, luck is always the most expensive variable. A championship squad lies not in the wallet but in how time is compressed into metrics. The same principle applies to the lane. A turn that saves 0.15 seconds sounds small. But multiplied by seven turns in a 200m race, it becomes more than a second. Multiplied by four rounds in one meet, it becomes the boundary between two medals. And multiplied by an entire four-year cycle, it becomes a person's career. Data is never just surface numbers. It is always a human story, told in a different language. A team does not collapse overnight. It collapses when its metrics stop connecting to one another. Swimming is the same. A swimmer does not decline in one bad race. He declines when stroke rate, distance per cycle, turn time and reaction time begin to lose connection with one another and no one notices. When everything breaks at once, people call it fate. But fate at the elite level is usually just another name for data that was ignored. So the question I put to the next cycle is not how many SEA Games golds Vietnam will win. The better question is: by the next SEA Games, how many training sessions will we have recorded in full? If that number is still near zero, then every medal won merely confirms that talent is still carrying an entire system unprepared for what comes after. What needs watching is not the medal table. It is the small cameras placed at the pool's edge. The signal of the next cycle starts there.