Seo Seung-jae, Two Finals in One Day, and the Four Empty Data Columns in Copenhagen

On August 27, 2026, at the Royal Arena in Copenhagen, Seo Seung-jae walked into two finals on the same afternoon. The first was the men's doubles final, partnered with Kang Min-hyuk. The second was the mixed doubles final, partnered with Chae Yoo-jung. Two matches, two partners, two entirely different systems of spatial geometry, and a break in between long enough only to change a shirt, drink water, sit down for a few minutes and stand back up.

I rewatched that footage at one in the morning in Beijing. Not to see who won. I wanted to look at Seo's feet in the third game of the mixed doubles final, specifically the moment he retreated to the rear backhand corner and pushed back up to the net. There is a very short beat there. Reading a scoreboard will never show you that beat.

On my second monitor I kept a spreadsheet open. Four columns: Advancement, Execution, Physical Fit, Key Data. Four columns, and all four empty. I sat with that frame for three weeks, filling it in and deleting it, filling it in and deleting it, until I realised the problem was not the data. The problem was the frame.

Men's doubles and mixed doubles are two different sports sharing one court. In mixed doubles there is a convention that is not written into any rulebook but is close to inviolable at world level: the woman covers the front court, the man covers the rear court. That structure creates a system with an axis. In men's doubles there is no axis at all. Two men rotate continuously, and every rotation is a decision about who owns which gap over the next three-quarters of a second.

Seo played two finals inside those two spatial systems on the same day, and won both. He became the first Korean player to win two world championship titles at a single edition. I raise that fact not to celebrate an individual, but because it poses a question that my four-column frame has no cell to answer.

It is worth stepping back a few paces. Copenhagen 2026 marked the coming of age of a new doubles generation. In men's doubles, Kang Min-hyuk and Seo Seung-jae took the title. In mixed doubles, Seo Seung-jae and Chae Yoo-jung took the title. In women's doubles, Chen Qingchen and Jia Yifan claimed their fourth world championship. In men's singles, Kunlavut Vitidsarn became the first Thai world champion. In women's singles, An Se-young delivered Korea's first women's singles world title.

Four of the five disciplines had champions from Asia. That is an easy statistic to read and an easy one to forget. The harder job is explaining why it happened, and the answer is not a continent. It is the way coaching schools organise space.

The Korean school builds men's doubles around a rotation with no stopping point. The principle is simple to state and very hard to execute: after every lift you play, the player who played it must move immediately into the position his partner has just vacated. Nobody stands still. The court is divided into sliding zones of responsibility, and the two players take turns filling the empty one. I call it the geometry of compensation.

The classical Chinese school chooses differently. It accepts a partly fixed structure: one player anchors at the net to intercept and finish, one player sits behind to generate pressure. The strength of that model is net interception. Its weakness is that when the lift is driven into the two corners, the rear player has to cover more ground, and the gap in mid-court opens up as a rectangle.

The Indonesian school takes a third route: it lifts the tempo of the rally with flat exchanges in mid-court. It builds no rotation and anchors no positions. It tries to deny the opponent time to rotate by keeping the shuttle below about a metre and a half and moving fast.

These three approaches produce three different shapes of gap, and this is where the data table starts to lose its bearings. A tracking system can record the positions of four players on court every hundredth of a second. It can compute distance covered, number of direction changes, peak speed. But when I opened the data table for a men's doubles match in Copenhagen, I found an empty column in exactly the most important place.

This is worth pausing on, because it explains why my four-column spreadsheet could not be filled in.

The fastest smash ever recorded in international competition is 426 km/h, struck by Mads Pieler Kolding in 2026 according to the Badminton World Federation. That is an impressive number for broadcast and a fairly useless one for tactics. At world championship level, all four players on court know the smash is coming. The question is not how fast it is, but which cell of the six square metres the defender cannot cover in time.

I spent many evenings measuring that gap by hand, frame by frame. What I logged from a handful of men's doubles matches in Copenhagen: at the moment the shuttle leaves the attacker's racket, the gap between the two defenders typically falls between five and six square metres. That figure is not stable. It stretches with the quality of the previous shot, with the foot position of the player who has just retreated, and with the eye direction of the player about to receive.

And here is the part I have to state plainly: eye direction exists in no standard dataset, yet it determines where the gap opens. Good attackers do not look at where they will smash. They look at the defender's hip, read which way the weight is tipping, and hit the opposite side. The retreat of a men's doubles player does not live in his feet. It lives in his eyes.

I do not write to persuade anyone; I write to arrange what my eyes have already seen. And across three weeks with that four-column frame, what I saw was an analytical system designed to produce rows rather than questions.

In mixed doubles the geometry is simpler but the decision pressure is denser. The woman at the net has less time to react, because the distance from her starting position to the contact point is shorter. The man at the rear has more time but must cover a larger area, usually about two-thirds of the court's length. That split creates a vertical axis, and every mixed doubles tactic is an attempt to break that axis.

China typically breaks the axis by driving the shuttle into the two corners to pull the man out of mid-court, then dropping into the space that just opened. Korea typically breaks it by having the woman intercept early, turning the man into a sweeper in mid-court rather than the primary attacker. These two routes produce two completely different kinds of women's player, which is why comparing the statistics of women playing in the two systems is a meaningless exercise.

Seo Seung-jae does both jobs. In men's doubles he is the rotator. In mixed doubles he anchors the rear court with a proactive partner at the net. He has to switch between two spatial programs, and switch within the same day.

The cost of that switching is not muscular. It is decision latency. When a player is used to every gap being shared responsibility, he reacts more slowly in a system where the space behind his partner's back is his alone. Conversely, when he is used to a fixed axis, he tends to wait in position rather than rotate. Both habits take seconds to unpick, and at this level seconds are an entire game.

That is why I watched Seo's feet in the third game. He retreated about a quarter-beat later than in the men's doubles match on the same day. But he compensated by pushing up earlier on the next shot. He could not fix the latency. He budgeted it.

In my notes from the years when I was working as a coach, there is one entry I still keep. It records a doubles session in Beijing where the coach asked two players to rally for forty minutes without being allowed to stand still for more than one beat. After twenty minutes, both began rotating without making decisions. Their bodies found the empty zone on their own. That is the point where technique becomes reflex, and also the point where tracking data becomes meaningless, because it can only record the trajectory of a decision that has already become instinct.

Every tournament has invisible pairs; I spend a fair amount of time hunting them. In Copenhagen 2026, the invisible pair for me was a Korean women's doubles pair that went out in the quarter-finals. They lost, and the scoreboard said nothing more. But on review I counted seventeen occasions across two games where one of them had to retreat to the rear right corner and the other failed to fill the mid-court gap in time. Seventeen times across two games, roughly once every thirty seconds. That is a systemic problem, not a form problem.

Now to what I consider the biggest blind spot in badminton analytics today.

The sports data industry has built infrastructure that lets anyone who pays receive near real-time positional data on players. Most of that infrastructure was built for broadcast, and most of its revenue comes from a market I would rather not name too many times in an article about badminton. But the technical consequence is clear: that infrastructure determines what gets measured. And what gets measured determines what is treated as important.

Smash speed gets measured. Distance covered gets measured. Shot contacts get measured. Eye direction does not get measured. Decision latency does not get measured. Rhythm allocation does not get measured. And so, in most modern tactical reports, the three things that decide a men's doubles match at world championship level sit outside the frame.

That is why I wrote about a spreadsheet with four empty columns. Not because I had no data. Because I had data, and that data did not answer the question I was asking.

Eighty pages of report are only the visible part; the submerged part is the nights spent asking whether I have watched enough. I know this sounds like a professional confession, and it is. But it is also a technical description. A report with no submerged part is a report with no error margin, and a report with no error margin cannot be used to make decisions.

There is another layer of the problem on the institutional side, and it connects directly to Seo's case. The Badminton World Federation's competition system obliges players in the top group to enter a minimum number of World Tour events each year, including all Super 1000 and Super 750 tournaments. That rule exists to protect the interests of hosts and audiences. It does not account for the fact that a player entered in two draws doubles his match load under the same obligation.

For a player competing in both men's doubles and mixed doubles, the cost does not double linearly. It compounds, because two sets of spatial programs have to run alternately within the same week, sometimes within the same day, and the nervous system has to keep overwriting one reflex with the other. I do not have enough biomedical data to assert this with a number, and I will not pretend otherwise.

During the Olympic qualification window that pressure is heavier still, because ranking points only count inside a limited timeframe, and skipping a tournament for physical reasons can cost a seeding position at the next Games. Players do not get old by age; they get old by wasted minutes. But there is another kind of ageing, quieter, that comes from minutes that were not wasted at all but spent converting between two systems.

For readers in Vietnam, this story has a practical layer. Nguyen Thuy Linh and Le Duc Phat are two rare cases of Vietnamese badminton reaching the main draw of World Tour level events and the last two Olympic Games. In singles, the spatial geometry is far simpler than in doubles: one player holds the court, and every gap is created by his or her own movement. But the principle of reading an opponent's eye direction is identical. A world-class singles player does not hit to where the opponent is standing. They hit to where the opponent will stand after being wrong-footed by one beat.

This is the technical layer that youth training centres tend to skip, because it cannot be taught through repetitive drills. It can only be taught by watching tape and asking yourself questions.

The blind spot lies in the assumption that everything important is measurable. Across three weeks with the four-column frame, I tried many ways to fill it. I tried converting eye direction into a binary variable: looked before hitting, or looked after hitting. I tried converting decision latency into the interval between the moment the shuttle leaves the opponent's racket and the moment the first foot leaves the floor. Both produced numbers that looked plausible. Both were wrong, because they measure the consequence of a decision that has already happened, while what I need is the decision that has not yet happened.

A good doubles coach does not ask his player how far he moved. He asks his player where he looked before the shuttle arrived. That question has no column in the spreadsheet.

There is a way to test the limits of any analytical model, and I use it fairly often. Hand the data table of a match to someone who did not watch it, and ask them to describe the match back to you. If they cannot say where the gaps opened and why, the data table has not finished its job, even if every cell is filled.

My four empty columns, by that standard, carry more value than a table full of numbers. They point precisely at what I do not yet understand.

What I want to verify at the next major tournament is not who wins. I want to see whether a pair can win a quarter-final at world level by deliberately accepting more gaps than the opponent, as long as those gaps sit where the opponent's eyes are not looking. If that happens, it will be recorded in no data column at all. And that is exactly why it is worth watching.