Three Yardsticks, One Label: Why Combat Sports Data Keeps Going Wrong

The reporter in the third row asked the coach why his athlete had lost. The coach answered with two words: difficulty score. The reporter wrote it down, and the next morning's bulletin was about a defeat. In that arena that afternoon, nobody was struck, nobody was knocked out, and there was no opponent on the mat. It was a taolu rotation: one athlete, one form, one scorecard split into three parts. The bulletin still ran under the “martial arts” heading, beside an MMA bout that ended by knockout in eight seconds. Two things of entirely different nature, sharing one label. The empty chair never lies — it only exposes what we would rather not hear. The problem starts with the label In most newsrooms, “martial arts” is a bucket. Into it go MMA, boxing, Muay Thai, taekwondo, judo, karate, wushu taolu, sanda, wrestling. That bucket is useful for allocating pages, broadcast hours and budgets. It is harmless until analysis begins, because at that point the label decides the yardstick. An empty label picks the wrong analytical framework on the writer's behalf, and a wrong framework cannot be fixed by writing better prose. In the current market phase, money in combat sports moves through contracts, purses, exclusivity clauses and pre-fight weight-check windows. Once the money is that layered, a classification error stops being academic. It becomes a pricing error: a taolu athlete judged by finish rate will always look inefficient, and an MMA fighter judged by technical score will always look imperfect. Three yardsticks that do not interchange The head-to-head group covers MMA, boxing, Muay Thai and kickboxing. Here the base unit is win-loss. On top of that sit finish rate, significant strikes per minute, takedown defence, cumulative damage absorbed. Every metric means something because there is always another person across the mat, and a binary result at the end. Sanda sits in its own group. Wins and losses exist, but the scoring logic is different. Points come from legal strikes, from throws, and from forcing the opponent out of the contest area. A sanda fighter can win while throwing fewer total strikes, because a throw is worth more than three punches. Reading a sanda bout with a boxing metrics table misreads the entire fight, even if every individual strike was logged correctly. Taolu differs at the root: there is no opponent. No opponent means no win-loss in the combat sense, only a total score drawn from three parts — movement quality, overall performance, and difficulty. A clean, rounded routine with no difficult elements can still lose to a rougher routine that pushed difficulty higher. The word “win” in taolu is already a loan word. Muay Thai shows this at the organisational level. Rajadamnern opened in 2026; Lumpinee opened on 8 February 2026; the two stadiums run two ranking systems, two ways of pricing a belt, two betting cultures. A Lumpinee champion is not automatically read with a Rajadamnern yardstick, even though both are Muay Thai and both count wins and losses. If standards split by stadium inside a single discipline, then folding every discipline into one label is an editorial decision, not a fact. When the wrong framework runs on correct data I tried something a data person would do: keep two tabs in one spreadsheet, one for combat, one for performance. Based on my experience covering bouts at arenas in Manila, Bangkok and Hanoi, the same athlete can appear in both tabs if they compete in two disciplines. The result column in the first tab always holds a binary value. The result column in the second never does — it holds a score, and that score splits into three parts that do not share a unit. That is where it collapses. A headline like “this fighter lost because he lacked a finish” is meaningless for a taolu athlete. An analysis like “high difficulty score, therefore victory” is meaningless for an MMA bout. It sounds obvious, yet both errors appear regularly, because the “martial arts” label lets the writer skip the classification step — and skipping it saves ten minutes while costing the whole piece. There is a further layer where classification error stops being a matter of wording. On 11 December 2026, in Manila, ONE Championship flyweight Yang Jian Bing died after cutting water to make weight for the Spirit of Champions event. His death forced the industry to re-examine the weigh-in process, and ONE Championship subsequently added urine-hydration testing and multiple weight checks during fight week to its official protocol. That entire logic — dehydration, rehydration, health risk — exists only in weight-classed sport. Taolu has no weight class, no water cut, no pre-fight scale. Applying a weight-class health framework to taolu generates warnings that do not exist, and non-existent warnings blur the ones that do. Every mispronunciation is a system trying to say something. Classification error does the same, except it makes no sound. An empty cell is not a neutral one When the industry meets an empty data cell, its reflex is to fill it with narrative. No numbers, so write about fighting spirit. No injury record, so write about character. That reflex is not emotionally wrong, but it converts a fixable gap into an unfalsifiable conclusion. The correct state of an empty cell is unknown. That is entirely different from neutral. Neutral is a conclusion; unknown is an obligation. We learn nothing from what goes right — only from what falls out of rhythm. The “martial arts” label is a rhythm error that has run long enough to look normal. It sits in classification systems, in platform databases, in section names, and in the way an editor assigns a story. What the system buried is not a scandal. It is a skipped step. What you measure on the screen is not the appeal — it is what you hear in the silence. One field, three values, and everything after The fix is cheap to the point of implausibility: before any analysis, lock one mandatory field naming the subject class — professional combat, performance martial arts, or sanda. One field, three values. That field decides the entire framework behind it: whether finish rate applies, whether difficulty score applies, whether weight-class screening applies. Remove the field, and the rest of the analysis becomes a string of grammatically correct, factually wrong sentences. What keeps me uneasy is that the bad label is not confined to martial arts. Wherever many things are pooled for the convenience of shelving, a framework is running wrong and nobody is checking it. Sport is simply where it surfaces earliest, because results are published in the open and scorecards cannot hide from anyone.