The Blind Spot of Combat Sports Data: What a Fight Record Never Says

On a year-end fight night at Saitama Super Arena, I opened three tabs on my tablet: one was an international database, one was the promotion's own statistics page, and one was a record published by the Japanese organiser. Same fighter, three different records. The first tab read 14 wins, 4 losses. The second read 13 wins, 3 losses, 1 draw. The third read 12 wins, 3 losses, 1 draw, 1 no contest. Nobody in the arena knew. On the big screen, he walked out with the most flattering of the three. I sat amid the roar, opened my notebook and wrote one line: professional combat sports, after more than a century of existence, still has no unified ledger.

That is why I am writing this.

Context: an industry without a ledger

Boxing is the luckiest discipline when it comes to record-keeping. BoxRec exists as an enormous library, logging professional bouts back to the late 19th century, with opponents, results and round counts. Yet even BoxRec acknowledges gaps in small shows across Asia and Africa, and in bouts with no video. An Indonesian fighter who fought twenty times in four years may only be recorded seventeen times. The other three vanish, and nobody is accountable for the disappearance.

Mixed martial arts is more fragmented. Tapology, Sherdog, each promotion's official statistics page — every one has its own standard, its own definition of a "no contest", its own way of handling small-show bouts before a fighter becomes famous. Japan has its own history. Shooto was founded in 2026 under Satoru Sayama. Pancrase followed in 2026. PRIDE ran from 2026 to 2026. RIZIN picked up the thread in 2026, led by Nobuyuki Sakakibara. Four generations of promotions, four record systems, four different understandings of a single fight. When a fighter moves from Shooto to Pancrase and then to PRIDE, his record is sliced by organisation, and the analyst has to stitch it back together.

Traditional martial arts sit in another universe. Taolu — form demonstration, non-contact sparring — has no opponent in the competitive sense. No opponent means no win rate, no striking efficiency, no defensive index. The entire framework I use to analyse a fighter collapses immediately.

The label "martial arts" in industry reports lumps together four different things: boxing, mixed martial arts, traditional combat sports such as Muay Thai and kickboxing, and demonstration arts. Those four do not share a single yardstick. On that foundation, every combat sports data report begins with an unwritten confession: we are measuring a very small part of what happened.

Core: what the numbers measure and what they leave behind

UFC publishes its statistics through FightMetric, the dataset built by Rami Genauer that became the promotion's official standard. From it, fans get significant strikes, accuracy rates, takedowns, control time. But the definition of a "significant strike" is itself a subjective choice: a light jab brushing an opponent's forearm counts the same as a hook into the ribs.

That is the point I want to sit with longest.

Significant strikes do not measure damage. They measure the number of contacts recognised by the statistician. A fighter who lands 180 strikes over three rounds can walk out of the cage with an unmarked face, while his opponent lands only 60 — but 20 of those are kicks to the lead thigh. That thigh will not carry him into a fourth round if the fight goes long. The stat sheet says the first man won. The second man's body will say the opposite for the next three weeks.

In Japan, I have tracked hundreds of fights manually, logging each round into a columned notebook. That method cannot replace a promotion's data, but it exposes something the data conceals: the distribution of energy. One fighter throws 40 strikes in round one, 30 in round two, 12 in round three. Another throws 25, 27, 26. The totals are nearly identical. But the first is repaying himself, while the second is accumulating. A three-round fight may not expose the difference. A five-round fight will.

There is another problem the record hides even more thoroughly: opponent quality. A fighter at 20 wins and 2 losses sounds frightening, until you read the names of the twenty men he beat. If fifteen of them were late-notice novices, those 20 wins say nothing about ability and everything about his manager's media strategy. Conversely, a fighter at 12 wins and 4 losses, with all four defeats coming against top-ten opponents, has actually been tested.

I still remember how year-end cards in Japan are assembled. Matchmaking here carries a gambler's spirit: one big name, one underrated opponent with a lethal punch, and a sold-out arena. The appeal comes from the uncertainty. But data is not permitted to be vague in that way. The more a fight is staged for drama, the more its record needs independent verification.

Career longevity in combat sports does not decline in a straight line. It declines in steps. A 33-year-old fighter can still throw a spinning strike as cleanly as he did at 26, but his recovery between rounds ran ahead of his body two years ago. The promotion's data has no column for "recovery between rounds". It has only a column for "rounds fought". One column measures the past; none measures the future.

Injuries do not detonate in a single fight; they quietly accrue debt across seasons. I have written this line for seven years, and it holds truer in combat sports than in football.

In 2026, when stadiums closed because of the pandemic and I went fourteen months without hosting a single event, I sat down and built an injury dataset: 500 footballers from five European leagues and 300 track and field athletes. The result surfaced somewhere few people look. The group competing in more than 30 meets per year showed a 28 percent higher rate of hamstring tears than the group competing in fewer than 20. The under-20 group is not a lazy group. It is a group that rests at the right time.

In combat sports the equation is heavier, because there are two variables football does not have: weight cutting and year-round high-intensity sparring. A weight cut is a loan. A fighter withdraws 8 to 12 kilograms of water in the final week, then replenishes within 24 hours before the weigh-in. The body keeps accounts. The first loan is repaid in performance. The fifth is repaid in kidneys. The tenth is repaid in an injury nobody can name.

Thirty years in this trade, I believe people repeat themselves. A fighter who collapses in the fourth round usually already collapsed in the fourth round two years earlier, in a fight nobody remembers. Which means the best predictive data is not in the win-loss column, but in the bouts nobody bothers to look up.

Contrarian angle: more data is not the answer

The industry's solution is more data. More sensors, more cameras, more metrics. I do not believe that is the answer.

The problem is not volume. The problem is that the industry cannot even agree on a basic question: measure for what, and measure for whom. When three databases still list three different records for the same fighter, one more metric only makes the error more detailed. That is why I always require three independent sources, and at least one must come from a different market than the other two. Three sources pulled from the same page are just one source multiplied by three.

A more uncomfortable counterpoint: taolu has no opponent, so it generates no combat data. If data is the measure of value, the entire demonstration tradition gets pushed to the margin. Yet thousands of people still train every morning in Osaka, in Hanoi, in Tokyo for a reason the stat sheet never touches. Combat sports analysis is confusing the expansion of the yardstick with the resetting of the question.

There is one more paradox. The people who believe most fervently in data are usually the ones who have never stood in a gym at five in the morning. They read charts without knowing that a fighter can win a bout by absorbing punishment better than his opponent absorbs it. Durability has no index. It only has outcomes, and outcomes always arrive late.

Open point

Every number tells the truth, but a fight never tells the whole truth. The map is not the territory; data is not the fight.

When a fighter steps onto the scale and the screen shows 14 wins and 4 losses, the fan is reading a map redrawn many times, by many hands, for many purposes. There is nothing wrong with that map. It has simply never been the ground.

The question I carry back to Osaka after every fight night is not who won. It is this: if tomorrow an eighteen-year-old walks into a gym with a handwritten notebook, logging every round, every breath, every kick to the thigh — will this industry have the courage to open that notebook and read it?