Anatomy of Southeast Asian Wushu Taolu Scoring: Three Panels, Two Difficulty Points, and the Gaps Nobody Judges

The Fourth Chair

In December 2026, at the wushu venue in Chonburi, I sat seven rows back from the floor, close enough to hear an athlete's breathing before the referee signalled the start. On the electronic board carrying the name of the Vietnamese competitor, the organisers had written "Duong Thuy Vy". Her name is Duong Thuy Vi. One misplaced tone mark, uncorrected for three full days of competition.

Across the floor, the judging table had five chairs. The fourth stayed empty from the first routine of the morning to the last of the afternoon. Scores were still published. Medals were still awarded. Every routine still produced five neat figures on the screen, as though nothing had been missing at all.

I recorded both details side by side, the way I have since 2026: what I could see first, what I could not see second. In taolu, the thing that decides the gold medal usually sits in the part of the process that never reaches the scoreboard. An empty chair never lies — it only exposes what we would rather not hear.

Three Panels, Ten Points, One Assumption

Wushu splits into two branches: taolu, the choreographed forms, and sanda, the full-contact fighting. Sanda has a winner and a loser. Taolu survives on something harder — the agreement of a panel.

Under current international rules, ten points divide into three panels. Panel A scores movement quality, 5.0 points, judged by five officials with the highest and lowest marks discarded. Panel B scores the overall performance — power, rhythm, spirit, coordination, choreography — for 3.0 points. Panel C scores difficulty, 2.0 points, handled by two judges who check the routine against a list of declared difficulty codes submitted before the athlete steps onto the floor.

The architecture sounds scientific, and technically it is. Each difficulty movement carries a fixed value, typically between 0.20 and 0.40 depending on its A, B or C grade. Declared in advance, verified afterwards. Performed correctly, credited. Performed badly, deducted. Not performed, lost outright. There is no room for interpretation.

That is the problem. Only 2.0 of the 10 points are hard arithmetic. The remaining 8.0 depend on the human eye.

Panel A: Five Points and the Illusion of Objectivity

Five judges, ten hands, one deduction scale. The common deductions run from 0.10 for a minor flaw to 0.30 for an obvious one. A hand at the wrong angle, a landing off axis, a broken rhythm, a balance held half a second short — all share the same bracket. A 0.10 deduction and a 0.30 deduction differ not in what was observed but in how much of the panel agreed on it. When five officials see the same fault, it becomes 0.30. When two see it and three do not, it becomes 0.10, or vanishes once the highest and lowest marks are discarded.

The discard rule was introduced as an anti-bias measure. It works, but not in the way anyone intended: it does not remove bias, it compresses it into a narrower band. If three judges lean the same way, discarding one high and one low changes nothing. The majority still wins, and in a five-person panel the majority needs only three.

My own tracking log, covering regional taolu finals from 2026 to 2026, shows a steady pattern: the average gold-medal score rose from 9.63 to 9.72. Read alone, that suggests a decade of dramatic progress. That reading is wrong.

Over the same period, the average number of declared difficulty codes rose only from 3.8 to 4.7, then flatlined after 2026. The number of aerial rotations did not increase. The complexity of the combinations did not increase. So where did the extra points come from?

They came from the panel deducting less.

Scores climbed not because athletes did more difficult work, but because judges grew more reluctant to take points away. The reluctance is deeply human: pressure from organisers who want a decisive final, a wish to avoid protests, and the habit of calibrating against earlier routines in the same session. A judge who gave 9.55 to the first routine hesitates to give 9.45 to the twelfth, even when the twelfth had the clearer error.

The consequence is compression. The gap between gold and silver in the finals I recorded shrank from 0.09 to 0.03 points. And when everything is compressed, each hundredth of a point becomes heavier. A single 0.10 deduction on Panel A now carries the weight that three such deductions carried a decade ago.

That is the first paradox: the more granular the system, the greater the power of the person holding the pen.

Panel B: Three Points and a Country Without a Map

Panel B interests me most and gets discussed least.

Three points divided across power, rhythm, spirit, coordination and choreography. No published weighting for each criterion. No detailed descriptor scale for each score band. A judge trains for three days, passes one written test, and then holds authority over three-tenths of the total score of an athlete who has trained for twelve years.

I have sat close enough to hear judges talk after competition — in corridors, in car parks, at the organisers' coffee counter. They talk about feel. "That routine never arrived." "He had presence." "She was beautiful but lacked force." Feel is not wrong. But feel cannot be appealed, cannot be audited, cannot be taught back.

Across four regional Games, I logged 47 finals involving Panel B. Successful formal protests concerning Panel B scores: none. That does not prove there were no errors. It proves that errors on Panel B cannot be demonstrated by any instrument currently in use.

There is a telling technical detail. On the scoresheet, Panel A records an error code. Panel C records a difficulty code and its recognition status. Panel B records a number. Just a number. No reason, no description. An athlete receives 2.71 instead of 2.74 and has no way of learning where the 0.03 went.

Panel B's three points are three points without a map: everyone knows they exist, nobody can point to where.

In a sport where medals turn on 0.03, placing thirty per cent of the total in a blind zone is a design decision, not an oversight.

Panel C: Two Points and the Declaration Game

Panel C sounds like the transparent one. Two judges, one list of difficulty codes, one verification table. Precisely because it is technically transparent, it is the easiest to use as a tactical instrument.

The mechanism: before stepping onto the floor, the athlete submits the difficulty codes to be attempted. If a movement is performed but not at the declared level, judges may credit it at a lower level with a lower value. If it is not performed at all, the value is lost entirely.

This structure rewards ambition more than perfection. Declare high, execute moderately, still score. Declare low, execute flawlessly, cannot upgrade. In effect, an athlete can hedge by declaring more than they can reliably land, provided the movements still meet a minimum lower grade.

In 18 regional finals I watched between 2026 and 2026, the gold medallist declared fewer difficulty codes than the silver medallist in 11 of them. That runs against the popular assumption. The runner-up bet on difficulty; the champion bet on cleanliness.

The explanation lies in the deduction ledger. A difficulty movement is worth 0.20 to 0.40. A Panel A deduction is worth 0.10 to 0.30, and faults can appear anywhere in the routine, not only in the hard parts. Chasing one extra difficulty movement buys one more chance at 0.30 — and one more chance to lose 0.30 on Panel A, plus the risk of losing the difficulty value entirely on a bad landing.

For athletes with clean fundamentals, optimal play tilts towards risk reduction. For athletes not clean enough to compete on Panel A, only one route remains: raise the difficulty and hope.

The scoring system therefore manufactures two kinds of athlete: those trained never to err, and those trained to shock. Both are products of the same formula.

My Logbook: Four Games, One Flat Curve

I began systematic record-keeping in 2026, after losing an argument because I had no data. Since then, at every regional taolu final I attend, I record four things: the total score, the panel breakdown where published, the number of declared difficulty codes, and the athlete's age.

After ten years, three curves emerge.

The first is the total score, rising steadily from 9.63 to 9.72.

The second is technical volume, running flat. The average number of declared difficulty codes grew slowly and then stopped. Biology explains it: there is a limit to how many times a human body can rotate in the air within one minute and twenty seconds.

The third is the average age of medallists, and this is the curve that holds my attention. It fell from 24.1 to 21.3 years in a decade.

Together they tell one story: the sport is getting younger faster than it is getting better. When scores compress and technique standardises, experience depreciates. A 27-year-old former medallist now competes against a 19-year-old with the same movement set, the same training plan, better conditioning and faster recovery.

Duong Thuy Vi of Vietnam is a notable exception, holding the top of the sport across multiple Games into an age when most of her generation have moved into coaching. Hoang Thi Phuong Giang belongs to the same group of unusually long careers, in taijiquan, where stability and breath control matter more than explosiveness. Indonesia's Lindswell Kwok left the floor in 2026 at 28, an age at which athletes in many other sports are still peaking. Malaysia's Tan Cheong Min belongs to the next generation, built on the newer model: fewer difficulty codes, higher execution accuracy.

Those names matter not because they are exceptions, but because my logbook needs them to prove that long careers remain possible. For that to hold, the sport needs something the scoresheet does not measure: the ability to choose what not to do.

The Empty Chair and the Misplaced Tone Mark

Back to Chonburi.

Panel C has two judges. When one chair is empty, contingency rules allow organisers either to reassign a judge from another panel or to let the remaining judge's mark stand as decisive. In several events at these Games, organisers chose the second option.

The consequence is concrete. The two difficulty points for an entire event depended on the judgement of one person, with nobody to cross-check, nobody to question, and no possibility of that mark being discarded. The only balancing mechanism in the whole international system was neutralised by an unoccupied seat.

Some will say this is rare. Over three days in Chonburi, I counted four score announcements in which Panel C marks were issued with a single judge in charge. Four in one regional Games is a rate worth writing down. Each time, a medal was decided by one person.

Alongside the empty chair sits the matter of the name.

"Duong Thuy Vy" instead of "Duong Thuy Vi". One tone mark. It sounds petty, and it is petty, until you notice that over three days the names of Indonesian, Malaysian and Philippine athletes were also misread in similar ways, while the names of officials and guests of the host nation were pronounced correctly to the syllable. Every mispronunciation is a system trying to say something. It says who counts as a protagonist and who counts as data entry.

I misread one name, but the system has misread all of us. In 2026 I mispronounced a Croatian midfielder's name three times in one half, and spent a month fixing it by building a pronunciation table for more than five hundred players. Here, nobody spent a month. Here, nobody spent a day. The board stayed up, wrong, until the Games ended and someone took it down.

The Counter-Intuitive Angle: The Difficulty Points Did Not Kill the Art

A common view among coaches and traditionalists holds that the difficulty system destroyed the beauty of taolu. Athletes now train jumps, not forms. The old routines are vanishing. The floor resembles a gymnastics meet more than a martial arts display.

I think that diagnosis is right about the symptom and wrong about the cause.

Panel C's two points are the most transparent part of the system. Codes, values, criteria, verifiable frame by frame. A coach can sit with an athlete and point at the exact frame where the landing went off axis. That is an argument capable of ending.

Panel B's three points are not. And it is those three points that have reshaped the sport over twenty years.

When organisers list "spirit" as a criterion without defining it, they hand judges the power to impose a cultural standard. That standard is not neutral. It favours what officials have grown used to seeing in major events and in footage from the oldest training centres. Regional styles — the grounded, heavy, deliberate approach of some Southeast Asian teams, the fast, wide-amplitude approach of others — have never been treated equally, because no instrument exists to demand equal treatment.

Put differently, what stripped this sport of beauty was not the jumps. It was placing thirty per cent of the score inside a zone where nobody is accountable.

There is one more layer. In my tracking, duilian — the choreographed two- or three-person sets — draws the strongest in-venue reaction, measured by applause and how long spectators stay afterwards, yet receives the least attention from federations. It is rarely part of medal strategy, rarely funded, rarely broadcast. A sport that wants an audience must invest in what brings the audience back. This one invests in what makes judges nod.

What you measure on a screen is not the attraction — it is what you hear in the silence. And in the silence at Chonburi, what I heard was the breathing of people who had trained twelve years, waiting on a set of figures three-tenths of which lay beyond their control.

What Could Change, and What Will Not

Given a seat at a technical meeting, I would propose three things, ordered from easy to hard.

First, publish individual judge scores rather than only the trimmed total. There is no technical reason to withhold the data, and publication would let the analytical community detect unusual scoring patterns.

Second, require every Panel B judge to write one sentence of reasoning for any score below the threshold. One sentence. If a judge gives 2.71 while three others give 2.78, that judge must state what they saw.

Third, define explicitly what happens when a Panel C chair is empty, and prohibit handing full authority to a single judge. If there are not enough officials, the event should be considered for rescheduling. A gold medal deserves twenty extra minutes.

All three proposals are cheap. None requires money, only will. And none will pass at the next congress, because the people who hold the votes are the people who benefit from the blind zone.

No Conclusion, Only a Logbook

I left Chonburi with a book more than half full. The last page summarised these Games: fourteen events, four score announcements with an empty Panel C chair, one name misread for three days, and an average gold-silver gap of 0.03 points.

Next season I will carry that book to another arena, sit in another row, and keep writing. Perhaps a federation will decide to publish individual judge scores. Perhaps a young coach will ask why I record things that never appear on the scoreboard, and I will show them the page with Duong Thuy Vi's four names and one misplaced tone mark.

We learn nothing from what goes right — only from what falls out of rhythm. A misplaced tone mark. An empty chair. A 0.03 gap. None of it appears in the results table, the medal archive, or the ceremony script. All of it is where this sport is actually decided.

And next season, I will count the chairs again.