Vietnamese Swimming: Reading the Lane Again from a Pile of Half-Burned Data

At the men's 800m freestyle final of SEA Games 31, I sat in the technical row of the My Dinh Aquatics Palace with three screens and a notebook. Not to watch who touched the wall first — most of the stands had already guessed that. I sat there to time every 50 meters and count how many times a Vietnamese swimmer's stroke rate changed over the final 200 meters.

The number I recorded: stroke rate rose from 34 to 41 cycles per minute, while distance per stroke fell from 2.1 meters to 1.7 meters. Same overall speed, but the way that speed was produced had changed completely. That swimmer was getting faster by churning more water, not by gliding farther. For a properly trained athlete, that is the signature of a race being eroded technically — or of a tactic nobody taught.

That is the kind of anomaly I hunt for: not who wins, but the technical price the winner has to pay.

I have followed Vietnamese swimming for more than two decades, since the days I wrote for the swimming desk of a major newspaper, and I learned one thing: in this sport, people remember medals, but the real data lives in the things nobody hands out medals for — 50-meter splits, turn times, the number of underwater kicks after the start.

In 2026, I began my career as a swimming reporter. Back then, the standard way to cover a meet was: who came first, what was the time, was a record broken. Three questions. Done. Nobody asked how an athlete had distributed their energy, how they turned, or where their breathing rhythm changed.

Twenty years later, I still see most Vietnamese swimming coverage stop at exactly those three questions. The problem is not the writers. The problem is the data infrastructure. At major international meets, every lane has an electronic timing system recording 50-meter splits, reaction times, and sometimes stroke frequency. At many domestic and regional meets, that data either does not exist or sits scattered in coaches' notebooks and is never published.

That creates a paradox. Vietnamese fans watch swimming through the medal table — a one-dimensional thing. Analysts like me have to work with a mosaic: a fragment of a split here, a phone video shot from the stands there, a semi-official time sheet nobody takes responsibility for. I call those half-burned data fragments.

One example. At a recent SEA Games, I tried to reconstruct the entire men's 400m individual medley lane of the Vietnamese team. I had the total time — accurate to the hundredth of a second. But I had no 100-meter splits. Without splits, I cannot tell whether the athlete was strong in the breaststroke leg or weak in the butterfly leg. I have the final answer to a problem, but I have lost the entire working.

That is why I say: in swimming, the result is the most copyable and least valuable thing. The value lies in the structure of the lane — and in Vietnam, that structure is usually a void.

Here I will dissect it across five layers: stroke rate and distance per stroke, underwater work and turns, performance measured against the regional map, the competition system and the rules, and finally the talent supply chain. This is how I work when the data is complete, and it is also how I discover where the data is missing.

Technical layer: rate and distance.

Swimming is the sport of a single equation: speed equals stroke rate times distance per stroke. Everything else — the start, the turn, the finish — is only a way of adding to or subtracting from that equation.

When I analyze a swimmer, I always begin with these two numbers and track how they change across each 50 meters. An ideal athlete holds distance per stroke steady and gradually raises stroke rate in the second half. An athlete losing control does the opposite: distance per stroke collapses first, stroke rate has to rise to compensate, and at some point stroke rate cannot rise any further. That is when the lane breaks.

In the 800m freestyle example I opened with, stroke rate rose by 7 cycles per minute but distance per stroke fell nearly 20 percent. The speed was held at the price of efficiency. If the lane were 200 meters longer, that price would collapse. This is not a hunch — it is an equation, and that equation always belongs to the swimmer who is more economical at the end.

Technical layer: underwater work and turns.

After the start and after each turn, a swimmer has about 15 meters to work underwater before being forced to surface. At the elite level, this is where races are won and lost without the crowd in the stands ever seeing it. A good dolphin-kick phase can be 1 to 1.5 seconds faster than swimming on the surface per turn. Multiplied across 15 turns in a 1500m lane, that gap reaches 15 to 20 seconds — more than the distance between gold and fifth place in many regional finals.

In Vietnam, this is the biggest blind spot. I have asked many domestic coaches whether they separately time the underwater phase after the start. The most common answer is no, or not regularly. That means we compete in a sport where one third of the time is spent below the surface, yet we barely measure that time.

Performance layer: measuring against the regional map.

Without a benchmark, every number is meaningless. A time of 1 minute 48 seconds for the men's 200m freestyle is very good nationally, mid-pack continentally, and not enough to make a world final. One number, three meanings.

Southeast Asia is a genuinely competitive swimming region, not a weak one. Singapore has long been the leading force, with a selection system tied to schools and a national training center. Thailand has depth in the medley events. The Philippines, Malaysia, and Indonesia all have individuals who crack the continental top tier in certain events. Vietnam sits in that group, but our position is unstable — we are strong in a few exceptional individuals, not strong across an even squad.

And here is what I want to state clearly: Vietnam's swimming dominance over the past decade has come mainly from a few rare individuals, not from a system that produces athletes. When the data shows results concentrated in one or two names while the gap to the next swimmer is too large, that is not the sign of a strong swimming nation. It is the sign of a dependent one.

Competition system and rules layer.

A swimmer does not only race opponents. They race the schedule. Swimming is a sport where an athlete may have to swim heats in the morning and a final that same evening, sometimes across multiple events. Every swim is a withdrawal from the account.

I once reconstructed the schedule of a regional multi-sport games and noticed something: the strongest delegations often deliberately drop a few of their own strong events to concentrate energy on events with a higher medal probability. That is a data decision, not an emotional one. A swimming delegation with nobody doing this analytical work is shooting itself in the foot at every games.

And the rules are where tactics are shaped most clearly. The 15-meter underwater limit after the start and after each turn is a number. The rule on the number of kicks in breaststroke is a number. The rule on touching the wall with both hands simultaneously in butterfly and breaststroke is a number. An athlete who optimizes their lane right up against those numbers gains an edge; one who violates them is disqualified.

I remember a case at the regional level when a swimmer was disqualified for an illegal wall touch in a medley event. Their time was good enough to make the final. But time is not what counts. This is a reminder that: numbers do not lie, but people always find ways to lie with numbers — and the rules are where numbers are not allowed to lie.

Talent supply chain layer.

This is the layer I worry about most, because it decides the next ten years. Swimming is a sport where the peak arrives much later than many people think. Female swimmers often peak younger than males, but both need a development runway of 6 to 10 years before touching the international threshold. That means the athlete we see on the podium today started a very long time ago — and the person who will be on the podium ten years from now is somewhere in a provincial pool right now.

The question for the infrastructure is: are we measuring the 12-to-15 age group? Do we have quarterly performance tables? Can we compare progress across training centers? If the answer is no, we do not have a supply chain — we only have random talent appearances, and randomness cannot be planned.

Here I have to argue against myself. There is a huge temptation for a data person like me: take a model built in Europe or America and apply it directly to Vietnam. I nearly fell into that trap once.

The first trap is the sample-size trap. When you see a Vietnamese swimmer with a turn index better than average, it is easy to declare "we have solved the turn problem." But if that index comes from three swims at one meet, you do not have a conclusion. You have a hypothesis. The difference between a conclusion and a hypothesis is the entire credibility of the analyst.

The second trap is the pool-condition trap. Swimming in a 25-meter pool and a 50-meter pool are two different sports tactically. The short course has more turns, so the edge belongs to the good turner and the good starter. The long course has fewer turns, so the edge belongs to the one who holds rhythm. A short-course result cannot be translated directly to long course, and vice versa. Anyone who does that is comparing two things that cannot be compared.

The third and most dangerous trap is the causality trap. When I see a successful swimmer with a high stroke rate, I am tempted to conclude that a high stroke rate creates success. But it may be that their superior fitness is the cause, and the high stroke rate is only the effect. Correlation is not causation. This is the sentence I have to remind myself of every time I sit in front of a data table.

I once treated the model as scripture. Now it is only a compass — but without it, we get lost. A good model is not one that predicts correctly every time. A good model is one that knows how to say "I don't know" exactly where it does not know. And in Vietnamese swimming today, the best model will have to say "I don't know" many times, because we simply have not measured enough.

That is why I do not approach Vietnamese swimming with a formula. I approach it with a list of questions. And I refuse to answer them until I have three cross-checked sources — the official time sheet, video I can time, and my own direct observation. If one of the three is missing, I write in the notebook: insufficient data, no conclusion.

Let us call this the unreadable-data section. I believe every honest swimming analysis in Vietnam today must have a section like this. Pretending it does not exist is what weakens the credibility of an entire analytical field.

So what is the signal for the next cycle?

If in the next few years I start seeing 50-meter split tables published after every domestic meet, I will know Vietnamese swimming has entered another stage. If I still only receive a total time and a medal-ceremony photo, I will know we are still reading the answer to a problem whose working nobody bothered to write down.

When the stands are empty, every model collapses. I rebuild from the half-burned data. At a pool, the stands are never empty — but the data usually is. And that is what I am waiting to be filled.

The question I leave behind: if tomorrow every Vietnamese swimming medal table disappeared, would we still hold anything to tell our own story?