Swimming and the Data Gap: What an Empty Spreadsheet Tells Us About World Aquatics

On 31 July 2026, at the Paris La Défense Arena, Pan Zhanle touched the wall at 46.40 seconds. The scoreboard flickered. A world record in the men's 100m freestyle — the event that functions as swimming's hundred-metre sprint.

I was sitting in Saigon, more than eleven thousand kilometres away, opening my tracking file.

Reaction time column: 0.62. First 50m split: 22.28. Second 50m split: 24.12.

That was it.

That is everything I have about one of the fastest performances in the history of men's swimming. Three numbers, and eleven empty columns right beside them: stroke count per lap, stroke rate, distance per stroke, turn time, underwater breakout distance, stroke amplitude, breathing pattern.

Numbers do not lie, but they know how to hide things. In swimming, they hide in an organised way.

Sixteen years in this business, and I have learned a way of measuring a sport's quality that few people use: count the number of data columns it is willing to publish.

Football is far ahead. A single Premier League match generates thousands of event data points, plus positional data for twenty-two players at dense sampling rates. From that, analysts build xG, xA, PPDA, progressive passes and hundreds of derived metrics. PPDA is not a number; it is a confession — it says this team chooses to let the opponent hold the ball, or chooses to win it back immediately.

Swimming has no equivalent. What we get is the official result sheet: final time, reaction time, and at major meets, splits every 50 metres. For a 100m race, that is five data points. For 1500m, thirty. Omega — the official Olympic timekeeper — owns the touchpad sensor system and holds far more data than it publishes. World Aquatics, the name FINA gave itself in July 2026, releases results. The rest sits in the internal files of national federations and training centres, where nobody is obliged to share anything.

I once asked a Southeast Asian national team coach why. He laughed: "If I publish my swimmer's stroke rate, next month his rivals will know exactly where he gets tired."

An honest answer. And also a confession.

The entire discipline of elite swimming analysis therefore becomes an exercise in reading what was never written down. I break it into four layers, plus a fifth for the things data cannot answer.

Layer one: pacing.

Pan Zhanle's splits of 22.28 and 24.12 tell a very specific story. He swam the first 50m 1.84 seconds faster than the second. In my spreadsheet I call this the "pacing gradient," and it varies enough to classify swimmers as types.

Kyle Chalmers, second in Paris at 47.31, swam the inverse pattern: economy on the way out, acceleration on the way home. Throughout his career Chalmers has been famous for his finish. The gap between the two men in the final was 0.91 seconds. It sounds small. But in an event where the eight fastest swimmers on the planet are separated by barely more than a second, 0.91 seconds is a generation.

This is where data earns its keep: it moves the question from "who swam faster" to "who swam faster, and how." The result sheet only answers the first. It never answers the second.

Layer two: reaction time, and the trap of ratios.

Pan Zhanle's 0.62-second reaction time is good. The average in an Olympic final tends to sit between 0.68 and 0.72. So he was roughly 0.06 to 0.10 seconds ahead of most of the field.

Media coverage often inflates this into a "blistering start." That is a confusion about ratios. Within a total of 46.40 seconds, 0.08 seconds is under two parts in a thousand. But when the gap between gold and silver is 0.91 seconds, nearly a tenth of that gap lives in the reaction. In other words: reaction time does not decide who wins, but it decides who stands on the podium and who goes home.

I learned this from a mistake. In 2026, at a regional meet, I advised a bet on the swimmer with the fastest heat reaction time. He finished fifth. For three days afterwards I asked myself what I had sold. The answer: I had sold a micro indicator as if it were a macro one.

Layer three: the over-performance index.

In February 2026, at the World Championships in Doha, Pan Zhanle set a world record of 46.80. Five months later, in Paris, he swam 46.40. He improved by 0.40 seconds, roughly 0.85 percent — in an event where world records are usually broken by fractions of a percent.

Football uses xG to separate luck from quality. Swimming has no such randomness: no posts, no referees, no deflected balls. It has something else, and it is crueller. I call it the over-performance index — the final time divided by the swimmer's personal best before the meet began.

This index does not measure luck. It measures the ability to convert form under maximum pressure. A swimmer with a personal best of 47.10 who swims 46.85 in the final posts an index of 1.005 — he swam half a percent better than himself on the most important night. Conversely, someone with a personal best of 46.50 who swims 46.95 posts 0.990 — he lost one percent on the most important night. At this level, one percent is the distance between a medal and a regretful article.

During the eight months football shut down in 2026, I rebuilt this principle on a different dataset: 2,400 Serie A matches from 2026 to 2026. 2,400 Serie A matches, and one evening I realised I was watching the heartbeat of an entire football culture. The principle holds across sports: the difference between the champion and the fourth-place finisher rarely lies in baseline ability. It lies in conversion.

Layer four: the standard deviation of pacing.

Long-distance events read differently from sprints. Katie Ledecky, with eight Olympic gold medals, is famous for something the result sheet never shows: the standard deviation between her 100m splits is close to zero. She won the 800m freestyle in Paris in 8:11.04 — and the story is not that number, but the way it was assembled from sixteen near-identical laps.

Ariarne Titmus is the opposite. She won the 400m freestyle by absorbing pressure and then surging at the end. The same gold medal, two opposing philosophies: one grinds opponents down with consistency, the other breaks them with change.

With only final times, these two are identical. With splits, they differ. With stroke rate and distance per stroke, they differ substantially. With underwater breakout data after each turn, they differ so much that one has to ask why nobody has yet sold a serious data product to this sport.

The grey zone of relay splits.

There is a technical detail few outside the sport know. In relay events, only the lead-off leg counts for record purposes, because the later legs start from a flying position — no reaction time, no starting signal. That creates a data paradox: some of the fastest legs on the planet occur in the middle of a relay, fall outside the official record system, and effectively do not exist in any aggregate table.

An analyst who reads only record tables will miss them. An analyst who reads relay leg tables will see a swimmer's true speed band — and sometimes see a personal record coming twelve months before it arrives.

Layer five: what the numbers cannot answer.

I have to be explicit here, because this is where credibility is easiest to lose.

A slow split can be caused by a concussion, illness, a congested schedule, a mistake in tapering, or simply because the swimmer emptied the tank in the semi-final. Numbers cannot distinguish between these. They record the final outcome and stay silent about the cause. Leon Marchand broke the Olympic record in the men's 400m individual medley with 4:02.95, erasing the 4:03.84 set by Michael Phelps in Beijing in 2026 — but from his four splits alone, you cannot tell how much belonged to butterfly technique and how much to breaststroke energy distribution.

Bobby Finke set a 1500m freestyle world record of 14:30.67, breaking the 14:31.02 mark that had stood since 2026. Kristof Milak still holds the 200m butterfly world record at 1:50.34. Those numbers say who is fastest. They do not say why the second-fastest is slower. In analysis, "why" is always more expensive than "how much."

The white data zone in Vietnam.

I once sat in the My Dinh Aquatic Sports Palace during the 31st SEA Games in May 2026. Nguyen Huy Hoang won gold there, and the stands roared like any football stand. But when I left, I carried away no indicator beyond the final time.

Vietnam has a paradox in swimming: the sport produces more precise measurement than almost any other, yet it generates less analytical data than any other. Nguyen Thi Anh Vien held multiple SEA Games records, and I am confident her team kept stroke-by-stroke tracking sheets. Those numbers stayed in the file. They never became open data, never became something an analyst or a journalist could independently verify.

That Saigon summer, I learned that data also needs watering. An unpublished metric is like a dry pool: it still has dimensions, but you cannot swim in it.

File verification: the process I run before trusting a name.

Whenever someone hands me a swimmer's file to assess, I run five steps, in fixed order.

One, plot the personal-best curve over time and compare it against the world record curve for the same event. Two, compute the over-performance index across the three most recent major finals. Three, check competition density: a swimmer racing too often tends to go flat mid-season and spike late. Four, cross-check splits against the coach's stated strategy — a mismatch between the two is the earliest signal of an unspoken physical problem. Five, find the anomaly and deliberately refuse to explain it immediately.

Step five matters most. An anomaly explained too quickly is usually an anomaly being hidden.

The counterintuitive angle: gaps have structure.

Swimming's data deficit is not technical. Touchpad sensors have been precise to hundredths of a second for decades. Cameras have enough resolution to reconstruct every stroke. The problem lies elsewhere: in this sport, information is a competitive asset, and nobody volunteers to price it at zero.

What is worth thinking about is that this choice does not apply only to technical data.

In April 2026, a major story surfaced: twenty-three Chinese swimmers had returned positive tests for trimetazidine in late 2026, around the Tokyo Olympics. The World Anti-Doping Agency accepted an explanation of food contamination and issued no sanctions. Three years later, the information became public.

The notable thing is not the conclusion. It is the latency.

It is easy to slide to two symmetrical positions. The first says: a positive is a positive. The second says: science explained it, case closed. Both miss the governance lesson: the same system was tight enough to record the positives and closed enough to keep them silent for three years — that is the data that matters most.

Correlation is not causation. A swimmer improving 0.85 percent in five months proves nothing. A country with multiple quietly handled positives proves nothing about any individual. But both events point in the same direction: in this sport, the most tightly controlled substance is not water. It is information.

An empty result sheet is not an analyst's failure. It is a finding.

This is precisely where the analysis report I received became its own subject. The document contained all nine required dimensions — technique, performance and data, competition system and qualification, the world swimming map, rules and anti-doping governance, athlete career and team system, risk profile, public narrative and expectations, industry ripple effects — and not a single line of data to fill them. No event name. No athlete name. No timestamps. No source.

A poor analyst stops and demands data. A decent analyst speculates. A good analyst reads the gap itself, because gaps have structure, and structure always means something.

A view on the next cycle.

The World Aquatics Championships take place in Singapore from 11 July to 3 August 2026. I will follow them differently from everyone else: not counting medals, but counting documents. How many federations publish complete split data? How many training centres open their biomechanical files? How many records fall to swimmers whose performance curves rise steadily, and how many fall to swimmers with an unexplained step change?

What I am waiting for is not another world record. I am waiting for one national federation — perhaps a small swimming nation with nothing to lose — to publish a full season of stroke data. Whoever does it first will surrender a small edge and gain something far larger: the position of the one who defines the standard.

Swimming stops moving every time a hand touches the wall, but data never stops flowing. The only question is whose hands it flows into, and whether anyone bothers to open the spreadsheet and read it.