World Swimming in the New Season: A Power Map and the Data War Beneath the Surface

The third split of the women's 400-metre freestyle — the segment analysts call the "death split" — lasts exactly 27 seconds. Within those 27 seconds, a swimmer can gain 0.6 seconds, or lose everything built over the previous 250 metres. I have spent many nights dissecting hundreds of race videos in this event, and what caught my attention was not the swimmer who touched the wall first, but the pacing structure of the one who finished third. She swam the opening 100 metres 0.35 seconds faster than the eventual winner, then collapsed over the final 50 metres. The split data had warned of that outcome. The audience did not see it coming. Numbers have no gender, but the people who read them do. I still remember an older coach telling me that swimming is "a sport for madmen" because you have to love the sensation of suffocation and the burn in your lungs. I did not argue with him about the emotion. I only asked one question: if that sensation really is data, then why do split times predict outcomes so accurately? CONTEXT The annual swimming season is entering a clear phase of differentiation. Following successive Olympic Games and world championships, leading swimming nations are rebuilding their squads: one generation of elite swimmers is weighing retirement while the next enters a cycle of physical accumulation. For the Australian market — the one I follow most closely — this is a phase in which selection depth matters more than any single medal. A country can have one star, but to sustain its standing across four consecutive years, it needs a system. Swimming is a sport that is strikingly short on public data: most of the most valuable metrics — anaerobic endurance, start reaction, underwater speed — never appear on the electronic board in front of the audience. We see the total time, but total time is a genderless number. It does not tell us how many metres a swimmer covered underwater after leaving the wall, who gained a faster wall entry after a turn, or who held a steadier stroke rate through the accumulating pain of the closing stretch. That is why, in every analysis I write, I begin by reconstructing the invisible architecture of the lane: stroke rate, distance per stroke (DPS), underwater speed after leaving the wall, and start reaction time. These four metrics form swimming's own heat map. Unlike the heat map in football — which I have warned has become "the new fortune-telling" — these four metrics still retain explanatory value, because they have not yet been inflated into a media product sold to a mass audience. CORE ANALYSIS Let us begin with the technical data. Among the four competitive strokes — freestyle, breaststroke, backstroke and butterfly — freestyle is where technique fragments most clearly. A swimmer at 46 to 47 seconds in the 100-metre freestyle typically maintains a stroke rate of 48 to 52 strokes per minute. A slightly slower swimmer often tries to push the rate to 55 to 58 strokes, and that is precisely when DPS collapses. There is a fascinating pattern I found when analysing the medal group: the leading finishers had a lower-than-average stroke rate but markedly higher DPS. In other words, they swim "with fewer strokes but more efficiency" — each stroke pushes more water, conserving energy for the closing stretch. But technique does not decide everything. When I analysed split data in the women's 200-metre freestyle, I noticed a striking pattern: between 100 and 150 metres, average speed drops by roughly 2 to 3 percent versus the opening split; between 150 and 200 metres, it can fall another 4 to 5 percent. The winner is the one who keeps that decline at the lowest threshold. In other words, at the elite level, victory does not come from the fastest opening split, but from the least collapsed closing split. This conclusion runs completely against the audience's intuition, because audiences are usually captivated by the sprint speed of the first 50 metres. Now let us turn to the global power map. Over the past decade, world swimming has operated on a fairly stable tier model: the dominant tier is the United States, with squad depth across nearly every event; the first challenger tier comprises Australia, China and France; the chasing tier includes Italy, Hungary, Japan, Britain and Canada. Names such as Katie Ledecky of the United States, or Ariarne Titmus and Kaylee McKeown of Australia, have served as reference standards for years. But this map is shifting. The rise of several young figures — especially in the men's middle-distance events, most notably Leon Marchand of France — is placing genuine pressure on the old order. This is precisely where data proves most useful. When an 18- or 19-year-old swims a time near the world-medal threshold, the question is no longer whether they have talent. The right question is: is their rate of improvement sustainable? I like to use a concept I call the "two-year improvement slope". If a swimmer improves by 1.5 to 2 seconds in a single year, that may signal a technical leap, or it may simply be a phase of natural physical growth. If they improve consistently by 0.8 to 1 second each year for three straight years, that signals a healthy development system and a durable physical foundation. But if they suddenly improve by 3 to 4 seconds in one season and then plateau the next — that is a warning sign I always flag in red, because it often accompanies injury or prematurely burning out a young athlete. Numbers have no gender, but the improvement slope tells the truth about the person behind it. I have watched too many young talents erupt and then vanish, and in most cases, the warning sign was already sitting in the data beforehand. Turning to the talent supply chain. The American university system, through the NCAA, remains the world's largest athlete-production machine, thanks to its enormous recruitment scale and dense year-round competition calendar. Australia's system relies on the national institute of sport and state-level swimming centres, with fewer athletes but higher density of quality, and one of the best conversion rates from junior talent to elite senior. China runs a centralised model, with specialised training camps beginning at a very early age. Each model has its own strengths, but I always remind readers that the centralised model carries a social cost the medal table never records: families pouring all their resources into a single child, and children eliminated early with no academic or career exit route. This is the murky data zone I always try to speak of, because if you look only at the medal table, you would think everything is fine. On rules and governance, there are three zones I always check before making any judgment. First is the 15-metre rule — the limit on underwater distance after the start and after the turn. Backstroke swimmers in particular exploit this zone to the maximum, and every rule change has produced a new wave of records. Second is the competition-swimsuit rule — after the 2026 ban on polyurethane suits, record times levelled off for a while, then climbed again through pure technique and physical conditioning. Third is the anti-doping system, where public trust is repeatedly tested by cases handled slowly and without transparency. Swimming is not exempt from the wider trust crises in international sport, and every time trust collapses, the commercial value of the entire sport is affected. On big-meet psychology, I once spent time analysing data to test whether a "pretty lane" effect exists — that is, whether a middle lane carries a real advantage or only a feeling. The results showed the lane advantage is almost zero in short events, but carries slight significance in longer events, where swimmers can observe rivals and adjust rhythm. That is a textbook example of how data can confirm an intuition, but can also destroy it. I do not trust emotion. I trust a data series longer than your emotion. CONTRARIAN ANGLE But here I must say what many in the industry do not want to hear: swimming's data model remains rudimentary compared with football or basketball. We have time, splits and stroke rate — but we lack real-time physiological data, water-propulsion data, and psychological data measured before the moment of the start. That means every conclusion we draw carries a margin of error that must absolutely never be forgotten. Moreover, I learned — in a moment I will never forget — that a 99 percent probability can still die. Kazan, at the 2026 World Cup, is where I learned it, even though that was a football pitch, not a pool. But the lesson is identical: when a team or an athlete is rated 99 percent to win, the remaining one percent does not disappear. It simply waits quietly. In swimming, that one percent could be a slip on the starting block, a pair of goggles filling with water, a splash into the eye during the turn, or simply a sleepless night under pressure. There is a professional temptation I see clearly in myself: once you own the data sheet, you want everything in the world to be explained by the data sheet. That is precisely when data becomes a religion. I refuse that path. Elite sport still holds a dark zone — where mentality, luck and day-of physical condition decide — that no tool measures in full. Emotion is also data, but we do not yet have the tools to measure it. In terms of public narrative, swimming has one strange trait: it creates stars but does not nurture them for long. A swimmer can win Olympic gold at 17 and vanish entirely by 22. Each time that happens, behind it lies a family story, an injury story and a financial story the medal table never tells. That is why I always remind readers not to value an athlete by their best time alone. I once took part in valuing a young talent in another sport, and my conclusion was fiercely opposed for allegedly "viewing a human being as a machine". Two seasons later, that transfer failed exactly as the data had forecast. I took no joy in it. I simply noted a fact: valuing a human being is not a calculation — it is a war between belief and the data sheet. And in that war, I choose to stand with the numbers, yet I never allow myself to forget that behind every line of data is a person who can hurt. LIMITS OF THE DATA I write this section to be honest with myself. What I presented above rests on public data, race footage and internal models, but five factors cannot be quantified: first, the impact of a mid-season coaching change; second, the quality of sleep and personal nutrition athletes do not disclose; third, officiating, especially in restart starts; fourth, financial pressure weighing on the families of young athletes; and fifth, luck — which every model tries to eliminate but can never fully remove. In the absence of these variables, any probability I offer should be read as a range, never a fixed point. OPEN CONCLUSION The signals I am watching in the next round are very specific. In the women's middle-distance group, the gap among leaders is narrowing to the point where energy distribution over the final 50 metres will be the deciding variable, rather than peak speed. In the men's sprint group, the battle will lie in underwater technique after the start — whoever holds speed better through the 15-metre mark wins. And at the system level, I leave an open question: can a nation sustain dominance through squad depth, or is the era of outstanding individual athletes returning? The answer will lie in the numbers of the next 18 months. And I will be there, reading them, line by line.