On the night of July 28, 2026, at La Défense Arena in Paris, Léon Marchand touched the wall in lane four. The scoreboard read 4:02.95 — an Olympic record in the men's 400m individual medley, breaking Michael Phelps' previous mark set in Beijing 2026 (4:03.84). The stands nearly exploded. I sat in the twelfth row, a hastily printed split sheet in hand, and the thing I circled was not the final time.

It was the final 100m: 58.24 seconds. Nearly four seconds faster than Marchand's own opening 100m, and faster than the closing split of almost every rival that same night. A finishing kick is nothing new — Phelps, Lochte and earlier generations did the same. What is new is the energy distribution: Marchand swam his opening 100m butterfly slower than both Phelps and Lochte at the same career stage, then made it back with an even backstroke and breaststroke leg that looked almost suspiciously controlled.

If you only look at medals, you see a home-nation star. If you look at the splits, you see a structural shift in how elite teams distribute energy in the medley. That is why I never open an analysis with a medal. I open with a baseline.

Swimming: a data-rich, method-poor sport

Swimming is among the most transparent sports in terms of published data, and among the most shallowly read. Nearly every major meet publishes splits down to each 50m, alongside reaction times, average speed per segment, stroke rate and distance per stroke. At the most recent Olympics, organisers even provided per-metre tracking data in the distance freestyle events. The raw material is not scarce. The method of reading it is.

Most swimming commentary is built on a feeling about a moment, not on the structure of a race. A swimmer who posts the best time of the morning heats and then loses the evening final is often labelled "mentally weak". Put the two split sheets side by side and you usually find a different cause: a breaststroke leg that dropped 0.8 seconds after an all-out morning swim, while the rival swam the heats conservatively and preserved the closing speed.

Swimming is a sport of efficiency per metre, not of peak speed. You can swim the fastest opening 50m in the world and still lose. You can also swim a slower opening and win, as long as your energy curve is flatter. Analysts call this the "cost of speed": every second you go above threshold early in a race, you pay back several times over at the end.

There is a natural experiment I still use as a reference point in every analysis of conditions. In 2026, when the pandemic halted competition, the Bundesliga — Germany's football league — returned in May in empty stadiums. I compared nine prior seasons of data with 93 matches played without crowds. Home win rate fell from 41.3% to 34.7%; average goals per match dropped from 3.1 to 2.7. The lesson is not about football. It is about method: only when an environmental variable is removed do you see its true weight.

In swimming, the equivalent environmental variables are the crowd and the pool. A packed arena creates noise that affects breathing rhythm and the perception of speed. Pool depth, water temperature, filtration and current all affect times, even when organisers try to standardise. At major meets, conditions tend to be better controlled, making times "cleaner" — and therefore harder to compare directly with times from smaller meets. When I build a baseline for an athlete, I always separate three groups: major-meet times, regional-meet times and domestic times. Mixing those three into one chart is the most common error I see in amateur reports.

The power map: four events, four different stories

In the women's 400m freestyle, the baseline has shifted sharply over four years. Ariarne Titmus of Australia took the world record down to 3:55.38 at the 2026 World Championships in Fukuoka. In Paris, Summer McIntosh of Canada and Titmus opened a gap on the rest that I estimate — based on results from 2026 to 2026 — at roughly 2.5 to 3.5 seconds across the medal places. It sounds small, but in distance freestyle it equates to swimming about 0.6 seconds faster per 100m across four consecutive legs. That gap is not one of effort. It is a gap of coaching systems.

In the men's 200m butterfly, Kristóf Milák of Hungary still holds the world record at 1:50.34, set in 2026. What matters is not the absolute mark but the stability: Milák, Léon Marchand and the chasing group show that the spread between first and eighth in major finals has narrowed to roughly 1.2 to 1.8 seconds. The event is shifting from a one-man race to a group race. As someone who follows the data, I read that as a healthy signal: an event with a single perpetual champion is usually a sign of a methodological gap in the rest of the field.

In the men's 100m freestyle, Pan Zhanle of China broke the world record with 46.40 seconds in Paris 2026. Split it into two 50m legs and compare with the historical standard and you see something unusual: his back-half speed barely drops relative to the front half, while most rivals lose 0.5 to 1.0 seconds on the way home. Holding speed on the return leg — not peak speed — is what separates champions from finalists in sprint events. I must be clear: this is a small-sample observation. One race does not make a trend; a trend appears only once you have at least three seasons of data collected under consistent conditions.

The same holds for Adam Peaty in breaststroke. His world record of 56.88 seconds in the 100m breaststroke, set in 2026, has stood through several cycles. After injury and a period of disruption, Peaty's sequence of results shows a familiar pattern: peak speed remains, but the ability to repeat that speed across multiple rounds declines. In the data, this signals a loss of physiological reserve, not a loss of technique. It is also the point journalists often misread: they look at one defeat and conclude something about form, when the data points to a scheduling and recovery problem.

The key insight: the gap between elite swimmers is no longer created at peak speed, but at the ability to repeat near-peak performance across multiple rounds on the same day. That is why strong teams have moved toward distance-style load management, once reserved for marathon and triathlon.

The consequence is a change in how we read a meet. Heats were once a formality. Now they are part of strategy. A swimmer who goes 95% in the heats can reach the final with a good lane, while one who goes 100% chasing a personal best often pays 1.5 to 2.5 seconds in the evening final. I have seen this at many meets: the swimmer with the best morning time rarely repeats it at night. The data sheet calls it the double-swim effect.

The Vietnamese case: read the baseline, not the medal

Vietnamese swimming has produced a generation with a regional mark: Nguyễn Thị Ánh Viên, Nguyễn Huy Hoàng, Trần Hưng Nguyên, Phạm Thanh Bảo. If you look only at SEA Games medals, the picture seems stable. If you compare their baselines against Asian and world standards at the same age, the gap remains large — often three to eight seconds in middle-distance events, depending on the discipline. That gap is not about one individual. It is a systems gap: the number of swimmers meeting continental qualification standards, the number of highly competitive meets per year, and the number of internationally certified coaches.

I have spoken with several young Vietnamese coaches and found a common thread: they have the ambition but lack the tools. A stopwatch and a notebook can still do a great deal, but they cannot replace a multi-year data system. Swimming is an accumulative sport: the value of data lies not in a single swim but in a multi-year series measured by the same method. When an editor says no, I learn to listen to the data — and that lesson applies to any swimming nation building a baseline from zero.

Look at the competition structure. In developed swimming nations, a young athlete may race twelve to fifteen high-quality meets a year, each with archived split data. In many developing nations, that number is far lower, and a significant share consists of low-competition domestic meets. As a result, athletes are untested under real pressure until they reach the international stage. That is why the baseline matters more than the medal early in a career.

Rules, equipment and the fairness question

One rarely mentioned but directly relevant variable for baselines is equipment regulation. After the boom in high-tech suits around 2026, the international federation tightened rules on material, buoyancy and coverage. Since then, records set under the new framework are broadly comparable with one another, but not directly comparable with pre-2026 records. Anyone building a long-term baseline who ignores this marker is mixing two eras into one chart.

Anti-doping testing also affects how data should be read. An anomalous junior performance, without enough testing samples over time, will always carry an uncertainty band. In my analysis, I do not pass judgment on individual ethics. I simply record the level of data uncertainty and adjust the weights when building a model. That is the approach of a data person, not of a judge.

The swimming value chain: from pool to market

Swimming is not only about elite athletes. It is a value chain. Upstream is the youth training market and the pool system. In the middle are athletes and competitions. Downstream are media, sponsorship, equipment and derivative markets such as digital content.

When a country invests in a standard 50m pool, the impact does not arrive within a single cycle. It arrives roughly five to seven years later, when children who began training there reach peak competitive age. That is why short-term plans often fail in swimming, while infrastructure investment has a long but reliable lag. If you want to read the future of a swimming nation, count the standard pools built in the previous decade, not the medals at the last Games.

The counterargument: when correlation is read as causation

At this point I must argue against myself. The greatest temptation for a data person is to turn every correlation into causation.

Countries with many swimming medals tend to have high per-capita income. Does that mean money creates medals? Not exactly. That correlation is confounded by many factors: pool infrastructure, sporting culture, population, and migration — many athletes compete for one country but train in another. If you chart per-capita income against swimming medals, you will see a positive trend. But if you isolate the variable "number of 50m pools per million people", the relationship is far stronger and more meaningful for policy. Money is necessary. Infrastructure is close to sufficient.

Another example: a junior athlete's breakout is usually attributed to "raw talent". Look at the monthly improvement curve and you often see something else: a change in meet scheduling, an injury handled correctly, or simply that the athlete has exited puberty with a body better suited to the event. In swimming, puberty is an undervalued variable. A girl racing the 200m may temporarily slow as her body changes, then accelerate again a year or two later. Judge her by one season and you will conclude wrongly.

A performance dip within a single season is not evidence of career decline; it is often the signal of a temporary adjustment in the training or physiological cycle. Being right too early is its own form of rejection — many predictions that an athlete was "finished" have been overturned by long-term data eighteen months later.

I do not argue with emotion; I present a data sequence. But I admit the limits of data: it tells you what happened and how unusual it was, not what will certainly happen. Every model has error. The wise reader reads both the number and the confidence interval. And as a journalist, I try to present both.

Signals for the next cycle

So what is worth tracking in the next Olympic cycle?

First, the spread between first and eighth in the 200m events. If that spread keeps narrowing, it signals that coaching knowledge is spreading. Results will be harder to predict, and that is good news for the sport.

Second, the gap between heats and finals. The share of swimmers who repeat or improve their morning time in the evening final is an indicator of a team's load-management quality. This is data anyone can collect, yet few bother to.

Third, developing swimming nations, Vietnam among them. If the number of swimmers meeting continental standards rises, and the number of highly competitive meets per year rises, medal results will follow — not immediately, but within roughly one cycle. This is the kind of signal data can see before the media does, if you read it patiently.

Amid the noisy stands, I choose to sit with the numbers. Not because the numbers are always right, but because they are always honest about what they measure. The race is over, but the data is still swimming a few extra laps.

---

This article is based on public data from international competitions and personal tracking notes. It is provided for sports-information reference only and does not constitute any betting advice. Sports results are highly uncertain; please read the analytical conclusions rationally.