Difficulty 9.10, Execution 8.40: Inside the National Taolu Score Sheet

On August 8, 2026, at Tien Son Arena in Da Nang, the electronic scoreboard flickered for seven seconds before locking. Tran Thi My Linh had just finished her Truong Quyen form in the women's taolu final at the National Traditional Martial Arts Championship. Two numbers appeared: 9.10 in the difficulty column, 8.40 in the execution column. A total of 8.75 after deductions. She lost the gold medal by exactly 0.30 points, the equivalent of a single loss of balance at second 41, which I recorded in my notebook at corner three of the mat.

From my seat in the fourth row of the press section, I carried three notebooks: one for the running order of performances, one for the fault code table, one for the raw scores of every judge. The person beside me asked what there was left to record when the result was already on the screen. I record because the screen only delivers the conclusion. People remember medals. I remember what led to them.

That night I typed every raw score into my personal coding sheet and compared it against four previous seasons. By the following morning, when the organisers released the medal list, I already had the answer to a question nobody asked: where those 0.30 points came from.

The taolu scoring system at national championships has run on a stable structure since 2026. Each routine lasts between 60 and 90 seconds, depending on the event and age group. Two A judges score technique and difficulty. Five B judges score execution. For the B panel, the highest and lowest scores are dropped, the remaining three are averaged, and the permitted margin is 0.10. Athletes must submit their difficulty declaration 24 hours before competition. The fault code table contains 12 groups, from loss of balance and wrong direction of travel to broken rhythm in the closing sequence.

Once you understand that structure, the number 8.40 says nothing about Tran Thi My Linh's skill. It says only that her routine was declared at a difficulty of 9.10, and that her execution left a gap of 0.70. From 2026 to now I have coded 412 taolu routines across four national championships and two youth championships. Among them are 1,183 deductions logged with the minute they occurred, the corner of the mat, and the fault code. I applied exactly the method I used for the 242-match dataset of one mid-table football club back in 2026: code the smallest possible event first, look for the pattern second. Numbers do not lie. We simply have not asked them the right question.

The first result is easy to predict. Athletes declaring a difficulty of 8.80 or higher made up 21 per cent of all routines but took 63 per cent of the gold medals. High declared difficulty is the shortest road to the podium. Stop there and this article looks like every other one.

The second result is the one worth writing down. I calculated the gap between the A score and the B score for every routine. The average in the gold-medal group was 0.42. In the group that won nothing, it was 0.71. In other words, the winner is not the person who performs the hardest movement. The winner is the person who closes the distance between the ambition declared and the execution delivered inside 90 seconds. Difficulty is a promise, execution is the evidence, and the judging panel only pays for evidence.

In the men's event, Nguyen Dang Khoa won with a difficulty of 8.90 and an execution score of 8.62. That gap of 0.28 was the smallest across all 412 routines I have recorded. He did not declare the hardest routine of the championship. He declared exactly what he could deliver.

I then separated the data by running order. Routines from the third slot onward averaged 0.18 higher on the B column than the first two slots. The effect repeated across all four championships. Broken down by delegation, however, it appeared clearly only for teams entering four or more athletes in the same session. Small delegations, two routines and out, collected almost none of that ordering bonus. The disadvantage is not written in the regulations. It sits in the way a panel grows accustomed to the rhythm of a session.

Then comes the group I watch most closely. Seventeen cases of athletes returning from ligament or ankle injuries, competing within nine months of the date their recovery was announced. Their average B score was 0.34 lower than their own pre-injury season. Five of the seventeen suffered a re-injury within six months of the championship. What stands out is that most of them were entered in high-difficulty routines from their very first competition back. Coaching staff wanted a fast answer. Their bodies answered with a different injury. Asking an athlete who has just returned to prove himself with his hardest movement is the fastest way to lose him for another season.

In qualification, the A column decides who reaches the final. In the final, the B column decides who reaches the podium. In 68 per cent of the finals I recorded, the ranking reversed relative to qualification, and the cause almost always sat in the B column. Outsiders see the stability of the winner. I see the collapse of the runner-up.

The contest ends, but the data does not.

There is a common reading I hear often in the press area, and it repeats every season. People say taolu is less disputed than combat because the scoring has been reduced to a mechanical scale. Combat has referees, protests, disqualifications. Taolu has a score sheet, and the score sheet ends every conversation.

My notebooks say the opposite. In combat, the referee panel can review video, and a protest must be filed within five minutes of the final command. In taolu there is no review mechanism at all. A routine exists for exactly 90 seconds, is scored by seven people standing in seven different positions, and then disappears. A 0.10 margin cannot rescue a second that was misread. The mechanical shell of the scale creates a feeling of precision, while the actual result depends on whether the judge at corner five could see the ankle at second 41.

The consequence is a grey zone sitting outside the regulations, and that grey zone is shaping delegation strategy in a conservative direction. Because difficulty must be declared 24 hours in advance, and because the B column is where ranking is decided, a smart coach will choose a routine below the true ability of the athlete in order to protect the B column. I counted 47 routines declared below a difficulty the athlete had already performed at youth level, and 31 of them won medals. The scoring system is quietly teaching delegations that the safest way to win is to say less about yourself.

What worries me more is how the advantage is distributed. A delegation with eight athletes gets eight routines, across three sessions, in front of the same group of judges. A delegation with two athletes gets two routines and a seat in the stands. After four seasons, the average execution gap between large and small delegations is 0.26, enough to move two places in the overall standings. Small delegations still turn up every year, still pay their costs, still supply athletes to the national team. They simply never get the chance to grow familiar with the mat.

In the warm-up area I heard two coaches discussing their declaration sheets. One said he would keep the old difficulty and focus on the closing sequence. The other said he would push difficulty to 9.30 in case the judges scored generously. Both were right by their own arithmetic. And both were optimising for the score sheet rather than for the athlete.

Before asking who won, ask who kept discipline.

Next season there are three signals I will log from the first round. The first is the number of routines declared below an athlete's own previous personal best. If that number keeps rising, the championship is rewarding caution instead of rewarding performance. The second is the A-minus-B gap among athletes returning from injury. If coaches still enter them in difficult routines in their first competition back, the re-injury rate will stay near 29 per cent. The third is the number of routines entered by delegations with fewer than four athletes. Every extra routine for a small delegation is one more time a judging panel has to watch them closely for 90 seconds.

Nothing prevents taolu from adopting a video review mechanism, even only for finals, under a limited procedure like the one already in use for combat. A five-minute protest window does not damage the dignity of a championship. It merely forces the judge to remember that a camera can see what the eye at corner five missed.

Every season is a chapter, and I am only the person who keeps the bookmark. But a bookmark, if it is accurate enough, can force whoever writes the next chapter to reread a few lines of the one before.