Swimming and the Discipline of an Empty Data Sheet: An Analyst Must Not Invent Miracles

At two in the morning, I reopened the analysis file of a national swimming championship on my computer. The time column was empty. The stroke-rate column was empty. The split column, the number of strokes per pool length, the reaction time off the starting block — all empty. A young colleague stopped by, looked at the screen and asked: “So what is the conclusion, brother?” That is a beautiful question, and also the most beautiful trap the trade of sports data analysis sets for the person holding the pen. When a data sheet is empty, there is always a very human temptation: to fill it with a story. A swimming miracle. An exploding young talent. A comeback on the final leg. I sat for a long time in front of that empty screen, and what I learned that night was not a conclusion about swimming, but a discipline about how an analyst must stand before emptiness.

Swimming is the most quantified sport of all. No other sport lays truth so bare: one hundredth of a second is enough to separate gold from silver, one extra stroke is enough to trade away a whole medal. In Vietnam, the water has produced names that made the whole Southeast Asian region look twice: Nguyễn Thị Ánh Viên with her collection of SEA Games medals, Nguyễn Huy Hoàng with his Asian medals in long-distance freestyle. Behind each of those names lies a mountain of data: split times per lap, 50-metre segments, stroke rate, distance per stroke, and the physical trajectory year by year.

When I hold an empty dataset — no athlete names, no meet, no times, no stroke events — every professional analytical framework becomes a shell. In my trade, a proper swimming report must pass through nine dimensions: technique, performance and data, competition system, the world landscape map, rules and anti-doping, athlete career, risk profile, media narrative, and the industry ripple effect. With not a single data point, all nine dimensions say the same thing: insufficient information to assess. That sounds like a failure. But to a data person, it is the most honest thing that can be said.

On the technical dimension, an analyst cannot judge without split times. You cannot call a swimmer’s start “good” without knowing their reaction time off the block. You cannot praise an “efficient stroke” without distance-per-stroke and stroke-rate figures. Swimming efficiency lies in the combination of tempo and stroke length — two variables that always trade off against each other. Add tempo and lose stroke length, and you are faster over the first 25 metres and dead over the last 25. Without those two numbers, all technical praise is just sentiment dressed up in jargon.

On the performance dimension, swimming has a strict coordinate system: world record, continental record, national record, personal best. A number only means something when set beside another number. A time of 1:45 in the 200-metre freestyle might be a national record for a small country, yet only a heat time on the world stage. Remove the coordinate system, and the number becomes meaningless — and readers will fill it with a wrong meaning of their own.

On the competition-system and world-landscape dimension, swimming runs on a four-year Olympic cycle. Each meet carries a different weight: a national championship is not in the same class as a world championship. The dominance map of the world’s water always has its “kings” in each event, with a tier of challengers in pursuit. But to draw that map, you need to know who holds the throne, who is rising, who is stalling. With no names and no events, the map is just a blank sheet with grid lines.

On the rules and anti-doping dimension, swimming is among the most tightly monitored sports. An abnormal result, a performance leap too large in too short a time, must always be set beside the question of validity. That is part of the verification process, not an accusation. A serious analyst never labels an athlete, but never ignores an outlier either.

On the athlete-career dimension, every swimmer moves along an age curve. There are explosive phases, plateau phases, and technical-restructuring phases. For young female swimmers, puberty is a milestone that can flatten an entire performance curve. Without age and competition history, you cannot place an athlete at the right point on that curve.

On the risk-profile dimension, there is a risk I always stress: emptiness does not mean safety. When information is missing, people easily assume everything is fine. But a risk surface that has not been drawn is not a clean risk surface. It has simply not been looked at.

On the media-narrative dimension, this is the most dangerous place of all. An empty result can be turned into an attractive story with a few editorial moves. And on the industry-ripple dimension, a swimming star can pull an entire value chain: from the youth-training market, the equipment industry, to meets and sponsorship deals. But all of that can only be analysed when there is a triggering event.

Here I want to speak plainly about another temptation, far subtler than inventing numbers. It is the temptation to equate correlation with causation. An athlete changes coach and then swims faster — many will immediately conclude the new coach is the cause. But data only shows you that two things happened at the same time; it does not give you the right to conclude which produced which. Numbers do not know how to lie, but the people who read numbers do. Between a correlation and a causation lies a distance that the lazy analyst always wants to leap across.

I once stood under pressure to write a “miracle” story in exchange for page views. At a major meet, an underrated swimming team unexpectedly beat a stronger opponent thanks to one relay leg. The editor urged me to write about the wonder. I refused, and instead rebuilt each split: starting speed, tempo between legs, transition times. When the numbers were laid side by side, the “wonder” dissolved and a very mundane structure appeared: a sensible pacing strategy and a complacent opponent on the final leg. A miracle is just a data point that has not been regressed. When you regress it carefully enough, it always returns to some ordinary variable.

For a person in the trade, the hardest thing is not finding a number, but accepting an empty number. The analyst’s ego always wants a conclusion to perform. But swimming taught me something I have carried through my whole career: on the scoreboard, only the performance counts, not how loudly you splashed when you jumped in. An honest report that says “not enough data” is worth more than a glossy report built on numbers that do not exist.

That night, I told the young colleague that the only correct conclusion was: no conclusion. But the full stop had not arrived. This was a starting point. Data only dies when we stop asking questions. What needed to be done was not to invent a story for the empty sheet, but to go back, find the right source, collect each split again, and rebuild the truth from the beginning. An empty result is a signal pointing exactly at where the process broke.

In an era when every sports article wants a “miracle” to tell, the data person has an unglamorous but essential task: to keep the number honest. What readers need is not a beautiful story, but a verifiable truth. And truth, in swimming, always begins with a column of data that is not allowed to be left blank.