The Silence Between Two Lanes: What Remains When Swimming Data Refuses to Speak

The scoreboard above the pool showed everything: lane number, athlete name, nationality, final time. Between two of those columns sat a small box labelled 50m split. That night, the box was empty. Nobody in the stands complained, because spectators pay to see who touches the wall first. But in the press room, a coach opened his laptop, downloaded the results file, stared at the screen for three seconds and said something I heard clearly: "Without splits, how do you know how he swam?"

I remember that sentence because it matched exactly how I felt in the summer of 2026, when the world's swimming pools went dark and I lost my newsroom job. That was when I knocked on Dr Emily Chen's door at the Australian Institute of Sport to ask about co-measuring ground contact times across fifteen national hurdlers. We found that women's 100m hurdles champion Celeste Mucci held an average ground contact time of 0.088 seconds across eight hurdles — 0.012 seconds longer than the theoretical optimum. A technical hole sitting quietly inside a result everyone was praising. The COVID laboratory taught me that data knows pain — if only we agree to listen.

Then, once, an analysis file came back empty. No title, no source, no information points, no entities. Nine analytical dimensions, fully templated, and the only surviving field was a single label: "swimming". I sat looking at it for a while. An empty file is itself a diagnosis.

What we call "a result"

A number on a scoreboard looks like a fact. It is not. It is a compound, and the compound only holds if four ingredients are present: the event, the pool length, the date, and the split series. Remove one and the whole structure collapses.

Pool length is the most neglected ingredient. A 25m pool has twice as many turns as a 50m pool, and every turn is a wall push plus an underwater glide — two things faster than any arm stroke. Short-course times are therefore always lower over the same distance, and World Aquatics ratifies short-course and long-course records as two separate systems. Placing two "1:55" figures from different pool lengths side by side is a meaningless subtraction. A swim time without a pool-length tag is not a hard datum to analyse — it is a datum that cannot be analysed at all.

The second ingredient is the equipment era. Between 2026 and 2026, full-body polyurethane suits appeared and distorted the entire frame of reference for the sport. At the 2026 World Championships in Rome, forty-three world records fell in eight days. The world governing body banned the suits from 1 January 2026. Every result inside the 2026–2026 window must therefore pass through a filter before being compared with the textile era. Skipping that filter is not a lack of nuance. It is an arithmetic error.

The third ingredient is the qualification standard. The A-cut and B-cut system splits entry into two tiers: an A-cut grants direct entry, while a B-cut depends on the allocation of remaining quotas. A swimmer who hits a B-cut may or may not travel, depending on how many athletes have hit the A-cut in the same event worldwide. A swimmer going faster than their own personal standard and still staying home is ordinary inside this system. It only looks strange to a reader who has never been told how the mechanism works.

The fourth ingredient — the split series — is the one most often stripped out of regional reporting. And that is where I want to linger longest.

The back half does not lie

Splits are the per-50m times. Reading a split series reveals three things a final time never will: how effort was distributed, where technique broke, and what the tactical intent was.

An 800m freestyle swimmer who is three seconds faster over the first 400m than the second may be saving fuel for a final. Seven seconds faster usually means the fuel was burned in the heats. Same final time, two completely different structures, two opposite coaching conclusions. Without splits, both read identically.

In our 2026 hurdles study we learned exactly this lesson in another sport. Celeste Mucci won, the result was good, but a 0.088-second ground contact time showed that on every landing she lost an extra hundredth of a second against the theoretical threshold. Eight hurdles, nearly a tenth of a second accumulated — enough to change a placing in a continental final. Nobody saw it, because the scoreboard only shows total time. A beautiful result can hide a technical hole for years, and only granular data is sharp enough to cut that shell open.

In swimming, splits matter even more. It is the only sport contested in a medium with roughly eight hundred times the drag of air, and drag rises with the square of velocity. A small deficit in the back half is not merely a fitness issue. It is a technical signal: stroke length shortens, stroke rate rises to compensate, the head lifts, the hips sink. Each change leaves a trace in the split series.

And this is where I feel the loss most acutely when reading domestic results: many result sheets carry only a final time. A young swimmer races the 200m breaststroke, touches the wall, and gets a provincial record. But nobody knows her first 100m, whether she stalled at the 150m mark, or whether her turn technique at 50m matched the one at 100m. Nobody knows, and because nobody knows, nobody fixes it.

Six-tenths of a second before anyone touches water

There is a stretch of time in swimming that spectators never see and analysts stare at: the gap between the starting signal and the feet leaving the block. At world level, reaction time sits roughly between 0.6 and 0.7 seconds. The difference between a good reactor and an average one in the same final can reach a tenth of a second — in a 50m race, that is the distance between a medal and fifth place.

Then comes the 15-metre rule. After the start and after every turn, a swimmer may travel underwater for a maximum of 15 metres. The underwater dolphin kick is the fastest form of propulsion the human body produces in water, faster than any surface stroke. Those fifteen metres therefore became one of the fiercest technical battlegrounds in modern swimming: whoever holds the underwater glide longer without infringing gets a free gift on every turn.

In a 25m pool, that gift appears twice as often as in a 50m pool. This is why turn technique and underwater technique carry different weights in the two pool types — and the second reason why comparing the two record sets directly is meaningless.

The list of parameters a decent measurement system must capture is therefore short: reaction time, 50m splits, the 15m split after the start, actual underwater distance, turn splits, stroke rate, and distance per stroke. Seven parameters. A camera in the right place and reasonably good extraction software can capture all seven. The cost is far lower than one overseas training camp.

I have covered swimming at many levels for fifteen years, and what I have come to see is this: the biggest gap between elite swimming and regional swimming is not in the water. It sits in the measurement apparatus outside the pool wall.

Domestic qualifying can be harder than a world medal

There is a famous paradox in American sport: sometimes clearing national selection is harder than winning a world medal. The US Olympic Trials format awards the top two places on the day, with no regard for historical results, potential, or age. Swim the fastest time of your four-year cycle but finish third on one particular morning, and the Olympic career ends there.

That mechanism creates a very specific risk type: internal upset risk. An event with six swimmers packed within two-tenths of a second produces shocks no world championship can replicate. And the same mechanism explains why American swimming has depth: internal pressure always exceeds international pressure, so athletes must learn to peak on one designated day.

On the other side, many Southeast Asian nations — Vietnam among them — treat the SEA Games as the de facto peak of the four-year cycle. That is rational in terms of resources and public attention. But it creates a distortion in training periodisation: if peak form is scheduled for May, then an Asian or world championship in July pays the price. No athlete can hold two peaks in one year. Nobody can, whatever the training plan says.

This is the kind of distortion only data exposes. Reading a results sheet, you see an athlete whose form dropped at an international meet. Reading a split series over time, you see an athlete who peaked six weeks earlier.

The puberty barrier and the cost of a one-person programme

In forecasting a female swimmer's career, there is one variable analysts consider among the most important and most often ignored: the puberty barrier. The stage of physical change — especially in girls — frequently brings a plateau or a temporary regression in performance. Girls who broke age-group records at fourteen or fifteen sometimes need two or three seasons to rediscover the structure of their own new body.

Without longitudinal data, that phase gets read as "finished". With longitudinal data, it gets read as a curve inside a development path. Same phenomenon, two conclusions, two destinies.

Nguyen Thi Anh Vien is the most instructive case in Vietnamese swimming of the past two decades, and I say that as someone who has followed her since her junior years. She appeared early, carried the expectations of an entire sporting system on one pair of shoulders, and sustained an elite career longer than most projections allowed. But she was a one-person programme: one athlete, one coaching group, one budget line, one expectation. When a system pours itself into an individual, every fluctuation of that individual becomes a fluctuation of the system.

The American NCAA model runs the opposite way: hundreds of universities, each nurturing a few dozen swimmers, together forming an enormous supply pipeline. Nobody is irreplaceable. Pressure is distributed, and that distribution is precisely what depth means.

Nguyen Huy Hoang followed Anh Vien in the distance group, and the question facing him is not speed. The question is whether a system stands behind him or only a coach and a schedule.

Four kinds of truth that must never be mixed

Swimming has a dense governance architecture: equipment rules, start rules, turn rules, eligibility rules, and an anti-doping system with a biological passport tracking physiological markers over time. Every dispute can ultimately land at the Court of Arbitration for Sport.

Inside that environment, there is one classification principle I consider the most important and the most frequently violated in the press: the absolute separation of four kinds of truth — a confirmed violation, a contamination dispute, a procedural violation, and a pure public allegation. These carry different legal consequences, different timelines, and very different levels of reputational damage.

Collapsing them into the single word "doping" is technically wrong and humanly cruel. An A-sample positive is not a sanction. A B-sample positive is not a finding about intent. A case won on procedural grounds does not mean the athlete is biologically innocent, and a case lost on procedural grounds does not mean the athlete is morally guilty.

I write this not to defend anyone. I write it because I have seen a young athlete branded "doping" on social media because of one truncated news line, when the actual file was an unresolved supplement dispute. The price of collapsing four kinds of truth into one is a career.

An empty box is not a clean bill of health

This is where I want to argue against myself.

When an analysis file comes back empty, the natural reader reaction is "nothing to worry about". No red flags, no warnings, no risk row marked. Nine clean rows. That feeling is pleasant, and it is entirely wrong.

An empty risk matrix admits two readings: either there are no risks, or there is no subject against which to assess risk. In my professional experience, the second is far more likely. Misreading the second as the first is the most dangerous error a sports data professional can make.

More dangerous still is the third reaction: filling the empty box with general knowledge. A well-meaning analyst short of data tends to "help" by populating the gap with observations that sound entirely reasonable about real athletes who are currently competing, who have families and sponsorship contracts. Those observations have no source. They are born from intuition, presented in the language of data, and read as fact.

The greatest danger in this profession is not a wrong number. It is a number that was never measured and still got written down.

The same mechanism operates at the media layer. The labels "next Phelps" and "next Ledecky" appear far more often than they come true. A fifteen-year-old who swims fast at a junior meet receives a comparison to the greatest swimmer in history. Two years later, when the puberty barrier arrives, the label returns as a headline in the opposite direction: "prodigy loses herself". Nobody in that chain did anything professionally wrong. Data was simply missing, and the absence was filled with emotion.

I do not believe in luck; I believe in the track each athlete chooses to stand on. And that track only becomes visible when someone bothers to record it.

There is one thing I have to keep reminding myself, because my professional instinct is to turn every moment into an equation. Stand at the pool edge at five in the morning, when there is only one swimmer and one coach, and you do not hear data. You hear water breaking in rhythm at every hand entry, a short even exhale on the third stroke, the dry snap of a foot pushing off the wall. Those things cannot be measured, and they do not need to be. A sport made only of data is a sport that has lost its memory. A sport with no data at all is a sport that has lost its memory in the opposite direction.

What needs building is not another medal

Every record is a hypothesis confirmed; every defeat is an equation waiting to be solved again. Vietnam already has athletes good enough to enter that equation. What is missing is the instrument set to solve it.

Three things can be done immediately, and none requires the budget of a swimming superpower. First, make split series mandatory in the publication of every national meet result — near-zero cost, near-infinite value. Second, state pool length and competition year on every results sheet, so that suit era and pool differential stop quietly distorting comparisons. Third, build a longitudinal database for the fourteen-to-eighteen age group, so the puberty barrier is read as a curve rather than a full stop.

The Gatlin–Coleman equation taught me that speed is never a single variable. That holds on the track, holds on the pitch, and holds twice over underwater, where every hundredth of a second is the product of a complex system of forces the naked eye never fully sees.

The question I leave behind is not for coaches. It is for those who hold the pen: if tomorrow the scoreboard shows an empty box in the split column, will any of us stop and ask why — or will we keep writing about whoever touched first, and let the fifteen metres beneath the surface remain a territory nobody reads?