The Finalist Index: The Column Nobody Counts in Vietnamese Swimming

In the spreadsheet I have opened for seven straight years there is a column with just two letters: CK. Final. It sits between the entry-time column and the seed-rank column, and it has never appeared in a single medal table I have ever read.

I started filling that column on a heat-session morning at a SEA Games. The stands were nearly empty. On the scoreboard the lanes appeared heat by heat, eight lanes each. People only really fix their eyes on the last heat of each event, the one with the names who will race for medals that evening. Between heats, I counted something else: how many Vietnamese swimmers appeared in that last heat.

The number I counted was far lower than the impression the medal table gives. And it was stubbornly stable from one edition to the next.

That is where the CK column was born.

The SEA Games pool runs on its own logic

To read the CK column correctly, you have to grasp three quirks that make the SEA Games pool different from the Olympic pool.

First, the event programme is decided by the host nation. The number of events swings between editions. Some editions cut events where the host has nobody competitive; some add events where the host is strong. Every comparison across time therefore carries a systematic error, and that error is not small.

Second, each country's entry quota is capped. A team cannot bring forty swimmers and stuff them into every event. Allocating entries is a strategic decision: concentrate them in medal-capable events, or spread them thin to build a squad. Those two choices produce two very different medal tables from the same talent pool.

Third, a final here means top eight after heats, as at the Olympics. But the quality of the heats is not the same. In some events the last heat contains only five or six swimmers. That means the CK column does not measure the same thing in every event, and I have to keep a separate note for events with fewer than twelve entrants.

The era I call "before 2026" is stamped with one name: Nguyen Thi Anh Vien. At the 2026 SEA Games in Singapore she won eight gold medals in a single edition, a figure no Vietnamese swimmer had produced before. Her career SEA Games total passed twenty golds. That was the period when Vietnam's medal table looked so good that people forgot the rest of the pool.

The era after is stamped with another name: Nguyen Huy Hoang. At the 2026 Asian Games in Jakarta he took bronze in the men's 1500m freestyle, Vietnam's first swimming medal at continental level. That is the fact I use as the anchor for the whole analytical chain that follows.

Between those two names, my CK column tells a very different story.

Evidence chain one: squad depth

I take the number of final appearances and divide it by the number of events in which Vietnam entered swimmers. I call that ratio final density.

In the editions around 2026, Vietnam's final density was high but its distribution was extremely skewed. Most final appearances came from a very small group of swimmers, usually three or four people, each racing four to six events. A multi-event swimmer at a SEA Games can carry five final slots across three days.

In other words: squad depth back then was the depth of one person, multiplied by the schedule.

After 2026 the structure flipped. The number of Vietnamese swimmers with at least one final appearance rose. But total final appearances did not rise in step. The squad spread wider and thinner in every event, the signature of a system shifting from "concentrate resources on one individual" to "spread resources across a group" without any single event yet reaching medal threshold.

This is what the medal table cannot say. An edition can yield fewer golds but a higher final density, and vice versa. Those two curves diverge, and the second curve is the forecasting curve.

Evidence chain two: where the golds cluster

I sorted every Vietnamese swimming gold at recent SEA Games into four groups: distance freestyle (400m and up), sprint freestyle (50m to 200m), breaststroke, and everything else, meaning backstroke, butterfly, individual medley and relays.

The result shows a concentration I rate as the riskiest of the four: breaststroke and men's distance freestyle account for the bulk of the golds. Sprint freestyle, the group that decides a nation's standing at continental level, is almost empty.

That concentration is not random. It reflects a deliberate investment choice. Breaststroke and distance freestyle have lower regional competitive density, their long-term conditioning cost can be offset by training volume, and most importantly they do not depend on explosive qualities that larger-population nations tend to dominate.

But the price of that choice shows up in the relays.

Evidence chain three: relays are the real depth index

The 4x100m or 4x200m freestyle relay requires four swimmers of sufficient level at the same moment. A country can have one star. It cannot conjure four stars out of nothing.

I track the simplest measure: the gap between the split of the slowest and the fastest swimmer in Vietnam's relay teams at each SEA Games. The smaller the gap, the flatter and more durable the squad. The larger the gap, the more the team depends on one individual carrying a leg.

Across several editions, Vietnam's gap was markedly larger than that of the teams it competes against directly. The cause I usually see when I rewatch the footage: a leg loses speed in the first 15 metres, then the swimmer tries to compensate with a higher stroke rate, which produces technical collapse in the last 25 metres and costs further time at the turn.

A relay is not four good swimmers. It is four swimmers with one shared rhythm. And a shared rhythm is built at system level, not at individual level.

Evidence chain four: the seed-rank delta

This is the index I am fondest of, and the one that took me longest to clean.

The method: take a swimmer's ranking by entry time after the heats, subtract that swimmer's ranking in the final. A positive result means the swimmer raced better than their seed; a negative result means they dropped.

For Vietnam, the average seed-rank delta in the strong events was always positive. Vietnamese swimmers tended to race finals better than heats, sometimes by two or three places. I once thought that was a sign of big-meet temperament.

Then I re-tested the assumption.

The problem is that SEA Games heats are usually swum at low effort. Strong swimmers deliberately hold back, needing only to qualify, not to post a fast time. Their heat times are therefore artificially depressed while final times are real. A positive seed-rank delta, then, measures pacing tactics more than it measures courage.

That is why I added a second filter: count the seed-rank delta only among swimmers seeded in the top three. In that group nobody holds back carelessly, so the number becomes cleaner.

After filtering, Vietnam's seed-rank delta fell close to neutral. Read plainly: when rivals also unload in the heats, the gap between Vietnam and the regional leaders does not close in the final. It stays, or widens.

Numbers do not lie, but they know how to hide something. The first version of the seed-rank column hid exactly what I needed to see.

Evidence chain five: age structure

I added a birth year to every final appearance, then computed the median age of Vietnam's finalist group at each edition.

Around 2026 the median age was higher, but the standard deviation was low: the squad clustered in a narrow age band and was past its peak physical development. After 2026 the median age fell and the standard deviation rose. The squad got younger but less even.

The thing I track most closely is the number of final appearances by swimmers under 20. This is the best leading indicator I have, because a swimmer who reaches a final at 18 usually has two more games cycles ahead.

If that number rises steadily across three consecutive editions, depth is genuinely being built. If it spikes in one edition and vanishes in the next, that is the effect of one generation, not of a system.

One exception: the overseas-Vietnamese pipeline

Another block in my spreadsheet tracks training origin. A meaningful share of Vietnam's final appearances comes from swimmers born and trained abroad who then wear the national team colours. The case of Le Nguyen Paul, a butterfly and medley swimmer developed inside the United States collegiate system, is the example I use most when explaining this to colleagues.

This pipeline solves a short-term problem very well. It brings technical foundations and a training culture that a domestic base would take years to build. But it also raises a structural question: if the main source of reinforcement sits outside the domestic development system, what exactly is that system being measured by?

I have no definitive answer. I only have an observation from the data: final appearances coming from this pipeline are very unevenly distributed by event, concentrated in butterfly and medley, and almost absent in distance freestyle, the group where Vietnam has its strongest domestic tradition.

The counter-intuitive angle: the hosts write the programme

At this point I have to argue against myself, otherwise the whole chain above turns into a form of data worship.

The most serious problem with the CK index is that it is not stable across editions. The event programme is decided by the host. One edition may have 38 events, the next 40, the following 36. Entry quotas per team also change. When the denominator moves between editions, every line chart becomes a game of chance dressed up in statistics.

I tried computing the CK index only on the set of events common to consecutive editions, that is, only events present in all three successive games. That shrinks the sample but removes most of the error. The results kept the same direction, differing only in amplitude. That is the only reassurance I can offer about this index's durability.

A second counter-intuitive point: the gold medal is a decorative index in some cases. Some events have only six entrants, and the three medals are handed out almost by default. A gold in that event and a gold in an event with twenty competitors do not carry the same weight. The medal table treats them identically.

A third counter-intuitive point, and the one that forced me to rewrite most of my spreadsheet: correlation is not causation. A rising final density does not automatically produce more golds at the next edition. In my data there are at least two editions where final density rose while medal output fell. Conversely, there is one edition where density fell and golds rose, because one individual swam beyond the level of the entire region for three days.

If I concluded that final density determines medals, I would have committed exactly the error I teach new staff to avoid: reading a single column and writing onto it a story larger than the column itself.

PPDA is not a number, it is a confession. I wrote that about football years ago, and only when I moved to the pool did I realise it applies here even more. A nation's CK index is a confession about the structure of its resources: where it concentrates, where it accepts a gap, and which cycle it is preparing for.

The profile-check section

Whenever I am asked to assess a young swimmer's file, I always open three columns first: number of events raced, seed-rank delta within the top three seeds, and the time trend across three consecutive swims.

Those three columns say more than every medal that swimmer has won. A medal records the outcome of one specific race. Those three columns record the curve of a career.

That Saigon summer, I learned that data also needs watering. A spreadsheet left alone dries out and becomes a field of meaningless numbers. It only lives when you keep rereading it each games cycle, adding new columns, and accepting that some columns must be deleted because you asked the wrong question from the start.

My CK column has been deleted and rebuilt three times in seven years. The current version has an extra note column recording the reason for reaching the final: qualified on merit, qualified through sensible pacing, or qualified because the event had few entrants.

Signal for the next cycle

SEA Games 33, hosted by Thailand in December 2026, brings a new event programme. It is the first test of the fourth version of my spreadsheet.

Three things I will count before I look at any medal table: final appearances by swimmers under 20; the gap between fastest and slowest splits in Vietnam's relay teams; and the number of events where Vietnam has at least two swimmers in the final.

The third is the index I trust most. A country with two people in one final has internal strength. A country with one person in five finals has one outstanding individual. Those two situations produce identical medal tables, and very different futures.