Swimming Has No PPDA: The Lane and Its Data Void

Swimming Has No PPDA: The Lane and Its Data Void

On 28 July 2026, at La Défense Arena in Paris, 17,000 spectators rose to their feet in a single movement. Léon Marchand touched the wall in the men's 400-metre individual medley final. The scoreboard read 4:02.95, an Olympic record, and the arena roared as one body.

That night, my spreadsheet held five rows for that race. Total time. Two 100-metre splits. Reaction time off the blocks. Final placing. Nothing else.

The night before, I had rewatched a 0-0 draw in Serie A. That match produced no goals, and it still left me with more than 800 event data points: every pass, every duel, every off-ball run, every metre of pressing, every clearance. No goals, but dense with information.

Numbers do not lie, but they know how to hide things. In the pool, they hide more than in any sport I have ever analysed. And most fans, including professional bettors, do not realise what they are missing, because what appears on screen looks complete.

Sixteen years, two notebooks and a stopwatch

I entered the profession in 2026 at Thanh Niên Báo as a swimming reporter. Back then I had exactly two tools: a notebook and a hand stopwatch. I sat in the third row above the stands, timing each lap, then comparing my figures with the electronic board when the officials published results. My margin of error against the touchpad was usually two to three tenths of a second over 100 metres, and close to a full second over 400 metres.

Those three tenths were my first lesson. They taught me that the human eye is a poor measuring device, and that spectator feeling is an even poorer one. A swimmer who looks explosive at the finish may simply be holding speed, while the one who looks exhausted may have swum the fastest final 50 metres.

In 2026 I lost two million dong following a purely emotional tip about the V-League. Angry with myself, I built a hand-made xG tracking sheet for ten rounds of Hanoi FC and found the club had scored 13 goals from just 9.2 xG, a 40 per cent overperformance. I wrote a warning, readers told me I was wrong, and by round 16 the team went completely silent in front of goal.

In 2026, at the World Cup in Russia, I used a hand-calculated PPDA figure to predict South Korea's shock win over Germany in Group F. In 2026, when football shut down worldwide, I spent eight months archiving data from 2,400 Serie A matches played between 2026 and 2026 and regressing it against Asian handicap movements.

2,400 Serie A matches, and one evening I realised I was watching the pulse of an entire football culture. Football stopped moving, but those 2,400 matches kept whispering inside my spreadsheet. That archive revealed a systematic bias: bookmakers generally price away teams about five per cent below their true level.

That Saigon summer, I learned that data also needs watering. Nothing grows in an empty cell.

But when I returned to the pool, where I have sixteen years of genuine observation, I found a far larger paradox. Swimming is the sport where result and data are almost identical, yet it is the poorest in publicly available data of any elite sport I have touched.

What the pool actually releases

Start with what exists. When a race ends, organisers publish an official results file. It contains total time, placing, reaction time for events timed individually, and splits for races long enough to be split. A 100-metre event gives you one 50-metre marker. A 200-metre event gives you three. A 1,500-metre event gives you nearly thirty.

Reaction time is the only genuinely biomechanical metric in swimming measured by machine. At Olympic level, a swimmer leaves the blocks in roughly six to seven tenths of a second. The gap between the fastest and slowest reaction in a world final is usually about two tenths. That sounds small, but in the 50-metre freestyle, where the world record is barely over twenty seconds, two tenths is one per cent of the race.

Splits are more useful, but narrower than people assume. A 50-metre marker tells you whether a swimmer went out fast or slow, but never why. It does not show whether the turn was good or bad, whether the breakout came early or late, or how breathing changed between metre 30 and metre 45. A split is an exclamation mark with no subject.

On the equipment side, swimming was digitised very early. Omega timing has been tied to the Olympic Games since 2026, and the touchpad appeared at the 2026 Mexico City Games. For more than half a century, swimming has had a measurement chain accurate to one hundredth of a second. But that chain measures exactly one thing: time. It does not measure how a human being produces that time.

What it hides

Four categories of data sit outside every public results file.

The first is stroke rate and stroke count per length. This is the swimming equivalent of cadence in running. An elite 100-metre freestyler typically covers each 50-metre length in roughly forty to forty-five strokes in the first half, rising in the second half as fatigue sets in. The ratio between strokes and time is distance per stroke, the single most important technical efficiency indicator. It is not published.

The second is turn efficiency. Over 200 metres a race contains three turns; over 1,500 metres, twenty-nine. A suboptimal turn can cost one to three tenths of a second. Multiplied by twenty-nine, that is nearly ten seconds, which is the entire gap between gold and eighth place in a world final. The results file gives you total time and nothing about how bad turn nineteen was.

The third is underwater distance. Since the fifteen-metre underwater limit was tightened, every swimmer must choose a breakout point. Break out early and you keep rhythm but lose momentum; break out late and you keep momentum but lose oxygen. This is the biggest tactical decision of the first twenty metres of every length, and it is entirely invisible in public data.

The fourth is the velocity curve. A swim race is not a straight line. It is a sequence of oscillations: fast off the start, slowing before the turn, surging after it, then gradually decaying. Total time is merely the area under that curve. A football analyst has per-minute xG to see the rhythm of a match. A swimming analyst has one aggregate number.

There is a technical reason for part of this poverty. Swimming happens in water, and water is brutal for sensors. Underwater cameras exist, but they serve broadcast, not measurement. Three-dimensional motion tracking has been trialled at some meets, but the installation and calibration cost per lane prevents it from becoming standard. A football pitch needs one camera system to generate thousands of data points. A ten-lane pool needs ten such systems, in a humid and reflective environment.

Who holds the data

Swimming data does not disappear. It simply flows into closed pockets.

National federations hold the largest pocket. Leading national teams maintain internal databases recording stroke rate, stroke count, turn times and even cardiovascular recovery markers for every athlete across every session. This is treated as strategic property, much as football clubs guard GPS running data.

Universities hold the second pocket. The US collegiate system maintains a public, searchable times database covering millions of swims across decades. It is one of the few genuinely open swimming archives, and it exists not from goodwill but because recruitment and scholarship systems need a public yardstick to function.

Private aggregators hold the third pocket. A handful of international websites compile results from meets worldwide and allow athlete history searches. But they only recycle official results files. They create no new information; they merely rearrange the same data in a different order.

Broadcasters hold the fourth pocket, and it is the most misleading. Watching a final on television, you see graphics showing swim speed, the gap to the record, even virtual race lines. Those graphics are built from available split data plus interpolation. They are beautiful, smooth, and they make you believe you are being fully informed. That is the biggest blind spot: viewers are given a vivid visual and mistake it for depth.

Four records, four voids

Test this against the biggest moments of the recent cycle.

On 31 July 2026, Pan Zhanle swam the men's 100-metre freestyle in 46.40 seconds, breaking the 46.80 world record he had set himself in February 2026. Four tenths of a second removed in five months over the shortest distance in swimming. Where did those four tenths come from? A little from the start, a little from the single turn, and most of it from holding speed over the final twenty metres. The results file gives me 46.40. It does not tell me where the four tenths live.

On 4 August 2026, Bobby Finke swam the 1,500-metre freestyle in 14:30.67, breaking the 14:31.02 world record set back at the 2026 Olympic Games. This event contains twenty-nine turns. If the record fell by thirty-five hundredths of a second, each turn needed to improve by roughly one hundredth on average. That is a precision no device outside a national team can measure.

In May 2026, Summer McIntosh swam the 400-metre individual medley in 4:24.38, breaking the 4:26.36 world record that had stood since the 2026 Olympic Games. Two seconds across eight years, across four strokes, eight turns and four stroke transitions. How was that improvement distributed between butterfly, backstroke, breaststroke and freestyle? No public file answers it.

On the women's side, Katie Ledecky continued her dominance with 800-metre freestyle gold in 8:11.04 in Paris, her fourth consecutive Olympic title in the event. For over a decade she has been number one at the longest distances. And throughout that decade, I have never had a public file allowing me to compare her stroke rate at metre 700 in 2026 with metre 700 in 2026.

Four records, four voids. We live in an era where every touch is recorded to a hundredth of a second, yet we still do not know how the record was made.

The contrarian case: swimming is the most data-driven sport, and that is the problem

Here I must argue against myself.

The conventional view is that swimming lacks data, so swimming analysis is weak. Look closer and swimming is the sport where result and data nearly coincide. A goal in football is a discrete event, and xG exists precisely because we need a measure between goals and chances. In swimming there is no such gap. Whoever touches first wins. Time is the result, and the result is the data.

The paradox lies elsewhere. Because the result is so clean, people assume the analysis must be equally clean. And once you have an absolute figure accurate to a hundredth of a second, it is easy to believe every question can be answered by comparing one number with another.

I have seen this thinking many times in valuation work. Someone looks at a 50-metre split and concludes a swimmer will win because the first half was faster. But correlation is not causation. The fastest first half may belong to the swimmer burning reserves others are saving. In the 200-metre events, the leader at 50 metres is often not the winner.

This is where I return to PPDA. PPDA is not a number; it is a confession. It does not tell you which team presses better, it tells you what that team is forced to choose. A swimming split works the same way. It does not tell you who swims better, it tells you how someone distributed their energy.

But to read that confession you need more than one marker. You need stroke rate, stroke count, turn time, the velocity curve. You need everything the pool is hiding.

So my counter-intuitive conclusion is this: swimming's problem is not a shortage of data but a shortage of context for data. We have enough numbers to know what happened, and not enough to know why. In a sport where result and data coincide, that missing context makes every predictive model fragile.

Here I must admit the limits of my own trade. The models I built from 2,400 football matches work only because football generates hundreds of data points per match, allowing constant recalibration. Swimming gives me none of that. With five rows for a race, every model becomes a form of storytelling decorated with numbers.

The transfer window of a sport with no transfers

While the spreadsheet stays empty, money still moves. And swimming's money tells a different story from its data.

At the Paris 2026 Olympic Games, the world federation paid direct prize money to medallists for the first time, at fifty thousand US dollars per gold medal. That is a change of substance. Before, an elite swimmer's income came from personal sponsorship, appearance fees at invitational meets, and, for a very few, professional team-based leagues.

The team-based professional league was expected to transform the sport's economy by turning swimmers into contracted assets. It ceased operations after 2026, leaving a gap no institution has filled. No transfer contracts, no release clauses, no wage bill. Only invitations, prize money and sponsorship.

But one form of transfer genuinely exists, and it is systematically overlooked: coach transfers and training-group transfers. After Paris 2026, Bob Bowman left his coaching post at Arizona State University to become head coach at Texas, and Léon Marchand turned professional inside Bowman's training group there. A four-time Olympic champion and one of the most respected coaches in history moved within weeks. In football that is a blockbuster deal. In swimming it is a footnote.

In Australia, a single coach guides two of the world's leading female freestylers across different distances. A training group at a suburban club producing more Olympic medals than an entire nation is normal in this sport, and it shows where real power sits: in training centres, not in competitions.

That is why I say emotion is the most expensive thing in the transfer market. In swimming, the most expensive thing is trust between an athlete and a coach, something with no release clause and no valuation. When a swimmer changes training centre, they do not change shirt. They change an entire physiological and psychological system.

This is where data should be worth something. If Marchand's training data under Bowman in Arizona and Texas were public, we would learn more from it than from a thousand articles about four gold medals. It will never be public. It is the private property of a collegiate programme and a coach.

Three signals to track

The first lies in the next cycle. If swimming keeps commercialising through prize money without opening its data, the sport will increasingly resemble a show rather than an analytical subject. Prize money creates stars. Data creates successors. Stars alone leave a sport shallow.

The second lies in young athlete files. When a junior swimmer jumps in performance within one season, the right question is not how much they improved but where that improvement was allocated. If I lack the data to answer, I must tell clients I lack the data. That is a sentence I have had to say many times in recent years, and it never gets easier.

The third lies in the results files themselves. If major meets begin publishing richer splits, turn times or stroke counts in the coming years, that will matter more than any record. Because every record is a data point, but not every data point is a record.

Sixteen years watching the pool taught me one simple thing. This sport has been measured to a hundredth of a second for over half a century, and it is still not understood to that hundredth. There is a gap between measuring and understanding, and swimming has stood in that gap longer than any other sport.

When a world record falls, I want to know in which segment it fell, not merely that it fell. Until someone opens that locked data vault, a swimming analyst will still have five rows in a spreadsheet and a vague belief that five rows are enough. I no longer believe that. But I keep the spreadsheet, waiting for each new results file, the way a farmer waits for rain over a field that has been measured precisely to the centimetre and still has never been watered.