The Pool Has No Legends: The Truth Lives in Every Stroke

The Moment in Lane 4

I still remember that July evening clearly, when the scoreboard jumped to 1 minute 55.42 seconds in the women's 200m individual medley. The winner was a swimmer who took six fewer strokes per length than her direct rival. The stands erupted because the girl who finished second was the most beautiful swimmer. I sat in the corner of the newsroom, hands cold, eyes glued to the split times on the screen, and a familiar feeling returned: the feeling of that Hang Day afternoon in 2026 when I was sixteen, believing that more possession meant victory.

I was deceived again. Not by a football team, but by my own eyes.

The human eye cannot measure stroke rate. The human eye cannot count the number of turns underwater. The human eye cannot see that the smoothly gliding machine was losing precisely in the final two lengths. Six weeks after that event, I sat down, reopened the entire season's data, and realized I had nearly written a tribute for the runner-up — the kind of piece I had sworn nine years earlier I would never write.

This article was born from that vow. It has no intention of disrespecting any athlete. It is simply a cross-examination using three data sources, to see what truly creates speed in the lane, and what is merely an illusion of a perfect stroke.

Context: Why Swimming Lags Football by a Decade

Football has xG, PPDA, heat maps, transfer valuations. What does swimming have?

When I started working, the only thing popularized in Vietnamese swimming analysis circles was final time. You have a number at the end of the lane. You call it truth. You don't ask how it was produced, from which segments, at what rhythm, in what bodily state.

That is like evaluating a match by only looking at the final score. It sounds right, but it ignores the entire process that produced the score. A team that wins 1-0 via an own goal isn't stronger than a team that loses 0-1 with three shots off the woodwork. But if you only have the score and no xG, you'll attribute a quality to the winner that doesn't exist.

Swimming is at exactly that point. We are evaluating an entire sport by final numbers.

Fortunately, swimming has a huge advantage football doesn't: everything can be measured absolutely. There is no luck, no linesman arguments, no goal disallowed because a finger was misplaced. Every millisecond is recorded. Every turn, every flip, every breath can generate data.

But we haven't exploited it.

Since 2026, when I worked as a swimming reporter for Thanh Nien, I began building my own tracking table. At first it was just a simple Excel file recording 50m splits. Later it ballooned into a small model with about forty variables, from stroke rate, stroke length, efficiency, to risk adjustment coefficients.

Three data sources I always use for cross-verification:

• First source is official split data from the organizing committee. This is the hard source, indisputable, but it is also the poorest — it only tells you who swam faster in each segment, not how.
• Second source is stroke rate and stroke length data, usually collected by independent analysis teams on-site. I once manually counted stroke rates of eight swimmers in a 100m freestyle final, and the error reached two strokes per length — a lesson on why three sources are necessary.
• Third source is indirect physiological data: peak heart range, recovery time between races, injury history. This is the source I always added after the Eriksen incident in 2026.

These three sources don't speak the same language. If they match, I believe. If they don't, I stop and write clearly the part the data can't explain. That is my entire working principle.

Core Analysis: Decoding the Lane with Three Layers of Data

Layer One — Splits: Who Swims Strongest in the Back Half

In swimming, there's a rule I call the "back-half law." At distances of 200m and up, and especially in medleys, the winner is almost always the one with a better back half — or at least the smallest decline.

Take a men's 400m IM final I tracked. The four strongest swimmers differed in the front half by less than 1.2 seconds. But in the back half, the gap widened to 4.5 seconds. The winner swam the back half faster than the front half by exactly 0.8 seconds per 100m — nearly unthinkable for such a brutal aerobic distance. Fourth place swam the back half 3.4 seconds slower than the front half.

| Swimmer | Front half | Back half | Difference | Rank | |---|---|---|---|---| | A | 2:01.4 | 2:00.6 | −0.8 | 1 | | B | 2:01.9 | 2:02.7 | +0.8 | 2 | | C | 2:02.1 | 2:04.2 | +2.1 | 3 | | D | 2:02.6 | 2:06.0 | +3.4 | 4 |

Notably: the swimmer in last place had the most beautiful butterfly technique of the meet. I watched his video over and over in the first 100m, and I can say this was one of the most perfect butterfly movements the eye could see that season: shoulders opened wide enough, hips undulating correctly, head not lifting too high, hands entering the water nearly perpendicular.

And he finished fourth. Why?

Because beautiful technique cannot pay an energy debt. The beautiful often pay more energy per meter, and that debt comes due exactly in the final 100m, when the muscles have exhausted their glycogen reserves.

Layer Two — Stroke Rate and Stroke Length

This is the data layer almost nobody in Vietnam pays attention to. Each swimmer has a "speed signature" of two parameters: stroke rate and stroke length. Speed is the product of these two.

But not every combination produces the same physiological result.

There are two distinct types:

• Rate-increasing type: increases stroke rate to accelerate. Advantage is quick acceleration, reacting to opponents, suited to sprint distances. Disadvantage is non-linear oxygen consumption, and easy collapse in the final 25m.
• Length-increasing type: increases distance per stroke. Advantage is high energy efficiency, sustainable over long distances. Disadvantage is difficulty accelerating suddenly, easy to fall behind when opponents accelerate.

In the meet I tracked, the men's 100m freestyle champion had a stroke rate 4 strokes per minute lower than his direct rival, but a stroke length 8cm longer. Result: he swam slower in the first 50m, then overtook by exactly two strokes in the final 15m.

This is what the eye cannot see. The eye only sees someone "swimming stronger" in the sprint, when in reality the winner managed energy better throughout the previous 85m.

Layer Three — Efficiency

This is the hardest layer, and the one where I often fight myself.

Efficiency in swimming is the ratio between output speed and input energy. It can't be measured directly by the eye, but must be inferred from speed decomposition over time.

A swimmer with high efficiency swims at the same speed with less energy. He may not stand out in the first 100m, but he conserves resources for the final 100m.

The data I collected that season showed a fairly strong correlation — but I must emphasize, this is correlation, not causation. Swimmers with better mid-race efficiency had a higher probability of reaching the final — but that doesn't mean efficiency creates a final spot.

I know this well because I once made the opposite mistake. In 2026, I believed this metric could predict results, and I lost twelve million dong on a parlay. Denmark with a pre-tournament average xG of only 0.9, I claimed they'd exit early. Christian Eriksen collapsed on the pitch in the opening match, and Denmark played with an emotion my model had no variable to measure.

I deleted the "average points" variable from the model that day and the model demanded an explanation. It forced me to add a category I call "non-quantifiable variables": injuries, psychology, schedule density, unexpected events, and a risk adjustment coefficient ranging from 0.8 to 1.2.

In swimming, non-quantifiable variables are even more brutal. An underwater collision. A disqualifying start. A pair of goggles filling with water on the fourth turn. None of these are in any split table.

Schedule Density — the Forgotten Culprit

This is the part I want to say louder in every analysis.

Schedule density is the biggest cause of injuries in elite swimming. No medical team can save a swimmer forced to race two heats and two finals within two days, in two different aerobic events.

At a recent national championship, I counted one male swimmer competing in seven events over four days. He still won medals in three, but in the final event — one where he was defending champion — his time dropped nearly three seconds from himself a year earlier.

No ranking records that. No line in a career profile says: "in April that year, athlete X missed an Olympic qualifying standard due to over-racing."

The number in the results line doesn't lie, and it also doesn't indict the schedule.

The Contrarian Angle: Why Swimming Can't Have an "xG"

Here I must be honest with myself.

I once tried to build an xG-equivalent for swimming. Something I called "xT" — expected Time, based on stroke rate, stroke length, efficiency, schedule, and a risk coefficient.

It failed. And that failure is the most important lesson I have since Eriksen.

Football can have xG because football is a probabilistic sport. Every shot is a draw from a distribution. You can shoot less and score more, and that's not a flaw — it's the nature of a sport with luck.

Swimming is different. Swimming is nearly deterministic.

In swimming, if you swim 100m in 47 seconds and your opponent swims 48, you win — unless there's a start error, disqualification, or accident. There's no "swam better but lost." The "better" exists only in the spectator's eye, not on the scoreboard.

So every attempt to apply a probabilistic model to swimming has a philosophical flaw: it models uncertainty that doesn't exist.

What does that mean for us?

It means the swimming analyst's job isn't to predict results — the scoreboard will answer that. Our job is to explain why that number was born, and predict what the next number will be based on variables that have changed.

This is entirely different from betting. In betting, you need probability. In analysis, you need causation.

I once confused the two. I once thought a good model was one that predicts correctly. Now I understand a good model is one that explains correctly.

And here's the irony: in swimming, a correctly explaining model often predicts correctly — but only because results are nearly deterministic, not because the model is brilliant. My model is right not because it's gifted, but because it's built on a decent sport that doesn't play tricks on the analyst.

The Biggest Blind Spot: Beauty

The swimming analyst's biggest blind spot is beauty.

When a swimmer swims beautifully, we tend to believe they're more efficient. The eye is deceived by smoothness, by perfect water flow around the shoulders, by grace in the turn.

But beauty in swimming is often a sign of minimizing drag, not maximizing propulsion. These two are not synonymous.

A swimmer can minimize drag to the absolute level, swim like a knife cutting water, and still lose to a swimmer with cruder technique but significantly greater propulsion per stroke.

I verified this in a silly way: I sat and counted how many times the beautiful swimmer was overtaken on the final turn across five different meets. The result was six out of ten cases. Six times out of ten, the smoother swimmer was overtaken in the final minute by an uglier but stronger one.

Of course, this is a small sample, and I don't want to assign statistical significance. But it points in the same direction as the speed decomposition data.

Beauty doesn't pay the energy bill.

The Illusion of "Controlling the Race" Underwater

There's another concept I want to demolish: "controlling the race" in swimming.

Commentators love this phrase. "Swimmer X is controlling the lane." "Swimmer Y is controlling the pace."

But in swimming, how can there be control? Each lane is its own universe. You can't pass the ball to a teammate. You can't set up a defensive line. You can't force your opponent to swim your way.

The only thing you can control is yourself.

So "controlling the race" in swimming is just a polite way of saying "swimming slightly faster than the rest and not being overtaken." That's not a tactic. That's a state.

When analysts use "control" for swimming, they're borrowing football's language — a team sport that can impose will on opponents — and applying it to an individual sport that doesn't allow it.

I deleted "control" from my swimming analysis vocabulary in 2026. It doesn't accurately describe anything in the pool.

Signals I'm Watching for the Next Cycle

After disassembling and reassembling the entire season's data, here's what I'm watching.

First, the "slow but steady" generation. They don't explode at eighteen. They improve slowly each season. But by twenty-two, they peak and sustain it longer than prodigies who burn bright and fade. I'm building a separate profile for this group, and I predict that over the next two Olympic cycles, they'll take most medal spots at distances of 400m and up.

Second, the shift from stroke rate to stroke length at short distances. Evidence shows high stroke rate is no longer optimal even at 50m. Top swimmers are swimming 50m at lower stroke rates than a decade ago, but with longer stroke length. If this trend continues, traditional sprint swimmers will face trouble.

Third, and most importantly, schedule pressure. I'm tracking how many races each swimmer competes in per season, and cross-referencing it with injury counts or performance declines. My hypothesis is there's a threshold — perhaps around five to six individual events per season — beyond which every other metric becomes meaningless.

Fourth, the value of injury data in athlete valuation. Transfer markets in other sports have begun pricing by injury history. Swimming hasn't, partly because the swimming market isn't as active, partly because managers still evaluate athletes by recent results. The young-talent bubble is bursting in football, and I believe it will spread to other sports. Young swimmers who hit peak performances — but only once — are being overvalued.

Every match sends a signal. The analyst doesn't decode, but listens.

The Data I Still Can't Explain

There's one thing I still don't understand.

In all the datasets I collected, there's a phenomenon my model can't catch: swimmers perform better at meets with larger crowds.

When I studied crowd effects on football during the COVID season, I found home win rates dropped from 44.4% to 36.2%, and away average points rose by 0.3. A clear effect.

But in swimming, this effect is much stronger, and it's asymmetric. Large crowds seem to help some swimmers swim nearly a second faster at 200m, but do nothing for others — even slow them down.

I don't know what variable creates the difference. Maybe psychology. Maybe something physiological related to adrenaline and heart rate. Maybe something I don't yet have language to measure.

Empty stadiums don't erase football. They only strip away one layer of the game's clothing. And in swimming, I'm trying to find what that layer is.

I write this part not to show honesty. I write because the analyst's duty isn't to be right, but to say what the data wants to say — even when the data wants to say I don't know.

About the Runner-Ups

Back to that July evening.

The women's 200m IM runner-up had a stroke I'll remember for my entire career. Each time her hand touched the water, it felt like the whole pool went silent for a beat. She swam butterfly soft as silk, backstroke like rest, breaststroke like a declaration, and freestyle like a sigh.

She finished second. Less than seven-tenths of a second behind the winner.

That night, I sat writing and asked myself: if I wrote about her, what would I write?

If I wrote emotionally, I'd call her a "queen without a crown," talk about "class" and "character," and ignore the fact that in the final 50m, her stroke rate dropped 3.4 per minute while stroke length didn't increase correspondingly — the classic sign of energy depletion.

If I wrote by data, I'd call her a "case study in the limits of beauty" — and that's also unfair, because she's still one of the finest swimmers of her generation.

Both ways are traps.

The right way, I think, is what I'm trying to do in this article: name the truth, without embellishment, without denigration, and let readers decide for themselves whether the number is the whole story.

Because the number isn't the whole story. But it's the only part that can't be disputed.

Possession is a beautiful lie; the score is a blinding truth. In swimming, no lie is more beautiful than a perfect stroke. And no truth is more blinding than two numbers separated by a colon on the scoreboard.

What I Believe Ahead

I don't write this to conclude that beautiful technique is meaningless. I write to warn myself — and anyone reading — that in swimming, as in football, there's a permanent gap between what the eye sees and what the scoreboard records.

That gap isn't an error. It's the nature of observing sport with human senses.

But it can be narrowed. By counting stroke rate. By measuring stroke length. By tracking schedule density. By accepting that some non-quantifiable variables will forever lie outside the model.

And by remembering something I learned with twelve million dong and a scar that never fully healed: the analyst's duty isn't to be right.

It's to say what the data wants to say.

The scoreboard doesn't know emotion. But it records the truth of emotion — as a number that sometimes can't be explained.

Predicting the beautiful swimmer will win isn't courage. It's a number that can't find its place. And I, after nine years following the blue lane, finally understand that my job isn't to find a place for that number, but to point out it has none — calmly, coldly, and as honestly as possible.