A women's 200m butterfly final results sheet from a national championship lists all eight lanes, all names, all finishing times, all referee signatures. The 50m split column is empty. In the stands, spectators scroll to the last row, see a champion, see a personal best, and nod. In my apartment in Brisbane, the same sheet reads differently: the final 150 metres of the race are entirely invisible. A swimmer finishing in 2:12.45 with an assumed structure of 30.2 – 33.8 – 34.5 – 33.9 tells a completely different physical story than the same 2:12.45 split as 29.1 – 32.0 – 35.4 – 35.9. The results sheet preserves only one of them, and it preserves the less useful one.

Swimming is a sport of intermediate numbers. A finishing time is the product of four decisions: the start and underwater phase, the pacing of the first 50 metres, the ability to hold stroke length as lactic acid rises, and the touch. Splits record all four. Without splits, only a total remains — a figure that cannot be decomposed into causes.

At international meets, splits are published to every 50 metres alongside reaction time and turn times. At most domestic meets, multi-point electronic timing does not cover every event, operating staff are thin, and results stop at the finishing time. The consequence is not the absence of data. The consequence is that a half-empty results sheet is still read as a complete one — the reader has no way to distinguish “not measured” from “nothing to measure”.

I call it the dangerous blank cell. In every technical report I have built for betting operators and federations, one principle has never moved: not assessed is entirely different from assessed and found clean. A table filled with “insufficient data” is not a safe conclusion. It is an empty conclusion, and empty protects no one.

A concrete test: a 400m individual medley swimmer improves by four seconds in eight months. Without splits, three hypotheses stand equal — the gain came from the 100m butterfly, from the 100m breaststroke (the chronic weak leg for most young Vietnamese swimmers), or from the underwater phase after each turn. Three hypotheses, three training plans, three cost levels. With splits, the answer usually surfaces in three minutes. Without them, all three are correct on paper and useless in the pool.

This is where I part ways with most domestic commentary. A commentary without splits can still be beautifully written about spirit, courage, aspiration. An analyst cannot. I do not trust emotion. I trust a data series longer than your emotion. But that series is only as long as the measurement points we hold.

Nguyen Huy Hoang in the 800m and 1500m freestyle is the clearest illustration. In distance events, the race is not decided in the first 100 metres but in the third 50 of each 100 — the segment where stroke rate begins to slip while rivals hold their length. A swimmer losing 1.5 seconds exactly there is swimming on willpower, and willpower does not survive four pool lengths. Look at total time and two swimmers look identical. Look at splits and they are two different professions.

Pham Thanh Bao in backstroke poses the inverse problem: backstroke technique depends on hip position and head stability, neither of which appears in a finishing time. A backstroker who loses rhythm in the final 25 metres of each length will post a total close to a swimmer who holds cadence evenly. The difference lives in repeatability — the only thing worth using for prediction.

There is a trap on the opposite side. Stare at splits long enough and it becomes easy to forget that behind every number is a person with a sex, a cycle, an injury. Numbers have no gender, but the people reading them do. In women's events, puberty can hold times flat or push them back two to three seconds while training volume doubles. A clean split table says nothing about whether that swimmer slept enough last night.

In 2026, in Kazan, I learned the lesson I still repeat whenever someone asks why I never issue absolute conclusions. Kazan is the day I learned that a 99 percent probability can still die on the betting table. The model was not wrong mathematically. It was wrong because it was missing exactly one variable: the psychological state of the people competing. In swimming, that variable is feel for the water — the thing no device measures. So when I read a results sheet with an empty split column, I do not conclude that the swimmer faded over the last 50. I conclude: unknown.

The point I want domestic coaches to weigh runs against common intuition. The common belief is that more data is always better, so teams pour money into sensors, software, and movement heat maps. Heat maps have become the new divination of analytics: they manufacture a sense of precision while obscuring a swimmer's actual role in the tactical system. A beautiful heat map can make people overlook the fact that the swimmer cannot reproduce that movement twice in a row.

The second problem is model import. Many propose transplanting the Australian or American system wholesale into Vietnam. I have lived long enough in both places to say that doing so without a comparable baseline is a form of prejudice with numbers attached. The Australian model rests on dense competition calendars, 50m pools in most cities, and a coaching class paid enough to work full time. In Vietnam, many coaches still hold second jobs. Importing the model does not create data; it creates handsome reports nobody uses.

The third problem sits on the writer's side. An analysis with a full table of contents, full charts, and a professional tone can be entirely hollow. The biggest risk is not a shortage of measurement devices. The risk is an empty report stamped “reviewed, no issues found”.

If I could upgrade only one thing in the next three years, I would not choose resistance sensors or recovery labs. I would choose publishing 50m splits for every event at the national championship, with reaction times, archived publicly by season. My own tracking experience shows most leaps by young Vietnamese swimmers do not come from training more, but from fixing one 15-metre underwater segment after the start. That segment accounts for under 15 percent of race time and decides finishing position in short events. Without a split at the 15-metre mark, no coach knows how much their swimmer is losing there.

What matters next season is not a new national record. It is whether the results sheet gains a split column. If it does, the debate over where Vietnamese youth talent goes will shift axis — from sentiment to race structure.

Limits of the data: this article uses no split or specific time from any swimmer, because that data has not been published publicly. The times cited above are simulated assumptions illustrating pacing structure, not competition results. Confirmed zone: the mechanics of splits, the role of the underwater phase, the difference between “not assessed” and “assessed and clean”. Ambiguous zone: coverage of multi-point timing systems at each domestic meet. Intuition zone: feel for the water and psychological state — things I have learned not to pretend to measure.