Difficulty Points, the 0.01 Crack, and the Fight for the Soul of Competitive Taolu

I pressed the stopwatch three times for the same jump. The foot leaves the mat, the body rotates twice in the air, the landing comes on the front half of the left foot. All three times: 1.04 seconds. From the seventh row of a wushu arena, sitting between two coaches I had never met, all three of us held our breath at the 0.9-second mark — the moment when everything can collapse.

The scoreboard appears a few seconds later. One athlete: 9.706. The other: 9.696. A gap of 0.01 points, one hundredth of a point, something the camera in the corner of the stands cannot capture, and something that decides who gets called at the medal ceremony.

More than twenty years in those rows taught me an uncomfortable conclusion: most spectators leave the arena without knowing what they just watched. They see beauty, speed, shock. They do not see the system that produced those numbers. When data begins to push back, tactics finally speak.

Three scoring groups and a decision in 2026

Competitive taolu is scored out of 10, split into three groups. Group A, movement quality, 5.0 points: correct amplitude, direction, rotation axis, hand position. Group B, performance, 3.0 points: rhythm, spirit, the coordination of hand and eye, of breath and step. Group C, difficulty, 2.0 points: the flips, jumps and spins whose values are fixed in the International Wushu Federation's table.

That split was not there from the start. In 2026, the International Wushu Federation brought the difficulty group into the scoring system. Through later revisions, Group C values were standardised into concrete numbers: each difficulty element is assigned 0.2 to 0.4 points, capped at 2.0 in total. A men's changquan routine runs from roughly 1 minute 20 seconds upward, and inside that window an international-level athlete typically registers three to four Group C elements.

The original intent was technical. Without quantification, judges are forced to score by feel, and feel cannot be audited, replayed or explained. But every time a sport quantifies itself, it also changes itself.

I once simulated the roar of a crowd for an empty stadium, and found that the loudest applause came from the data. In taolu, that applause is worth 0.3 points.

The expected-value problem on the mat

A coach once showed me the spreadsheet he uses to assemble a routine for his athlete. The first column lists Group C elements. The second holds their point value. The third holds the success rate recorded over the last six months of training. The fourth holds the deduction if the landing fails. Group C turns a performance into an expected-value problem: element value times success probability, minus deduction cost times failure probability.

Take concrete numbers. An element worth 0.4 points, an 85% success rate in training, a failure cost of roughly 0.3 points in deductions plus the lost 0.4. The expectation is clearly positive, close to 0.3 points. Drop the success rate to 60% and the expectation falls to about 0.12 — still positive, but with enough variance that one fall pushes an athlete out of the top five. In some events, the gap between gold and fourth place is under 0.05 points. Which means the decision to register or cut a single Group C element can cost more than a full year of training.

The part few people mention sits here: difficulty earns points in Group C and pays the debt in Group B. After a 720-degree spin, an athlete needs roughly 0.2 to 0.3 seconds to recover the torso axis and the breathing rhythm. In that window, hand and eye fall out of phase, the sword line loses tension, and three Group B judges record it in numbers nobody can replay.

I followed one female athlete in the jianshu event across four seasons. She gained 0.5 points in Group C and lost nearly 0.3 in Group B. The net balance was only 0.2. That is why many teams now hire strength coaches instead of forms specialists.

The bridge to esport is clearer here than in any other sport. Between the football pitch and the esport arena there is an invisible bridge, and I make a living proving it is wobbling — but in taolu that bridge does not wobble, it sits inside the spreadsheet. A professional gamer makes a decision in 0.5 seconds and calls it actions per minute. A taolu athlete has 1.04 seconds in the air and must finish every calculation before take-off. The recovery between two difficulty elements in the same routine is their cooldown.

On the Vietnamese side, the taolu school was built on aesthetics. Figures such as Duong Thuy Vi in jianshu and qiangshu, Hoang Thi Phuong Giang in daoshu and jianshu, Nguyen Manh Quyen in the forms events — they grew up in a system where Group B was the sharpest weapon. As Group C expanded its tactical weight, that advantage was compressed. They did not get worse. The equation changed.

The suspicion is not in Group C

The most obvious reaction is to demand the removal of the difficulty group and a return to pure feel-based judging. I have heard that proposal on at least four forums.

Simulate it. Remove the 2.0 Group C points and the scale drops to 8.0 shared by A and B. Nothing is restored; Group B's weight merely rises from 30% to nearly 38%. Group B is the group with no audit trail: no value table, no measuring stick, no replay. We would trade something countable for something uncountable and call it fairness.

In the transfer market, agents are the largest hidden cost precisely because the noise they generate is never priced. In taolu, Group B plays exactly that role. It is not wrong. It simply has no books.

There is another blind spot, less discussed. International Wushu Federation rules update roughly one cycle behind the technical evolution of athletes. The same thing is happening with esports betting, where the regulatory frame lags behind and erodes competitive integrity faster than in any traditional sport. An outdated scoring table does not produce corruption; it produces grey zones. And grey zones are always more crowded than bright ones.

World Cup 2026 was a tactical scandal, but it was also a broken mirror reflecting an entire football culture lying to itself. At a smaller scale, the scoring arguments at wushu championships reflect exactly that: a discipline telling itself that any score of 9.706 can be explained.

Closing

I want the federation to publish every judge's raw score with a timestamp, every difficulty element's value and every deduction, the way a match API publishes events. Spectators do not need to understand the whole scoring table. They need to know why 9.706 differs from 9.696.

If all of those numbers were pushed out live tomorrow, would a 25-year-old athlete still register a 720-degree jump for 0.02 points, or would he spend those 1.04 seconds on a cleaner sword line?