Why professional combat sports remain a blind spot in sports analytics

On a Tuesday night in Beijing, I opened an eight-dimension analytical report on a combat sports bout, sent over by the technical team. The document was immaculately formatted: eight major sections, each with tables, an assessment column, a risk column, an evidence column. By the tenth line I realised all eight sections said the same thing: insufficient information, cannot assess. No fighter names. No weight class. No sanctioning body. No fight date. Not a single figure for significant strikes landed per minute, no takedown accuracy, no gate revenue. Style matchup: N/A. Record quality: N/A. Weight-cut risk: N/A. Pay-per-view revenue: N/A. Champion's Clause: N/A. Cumulative brain trauma: N/A. Eight dimensions, eight blanks. My first instinct was to call the technical team and ask why they had sent me a blank page. Then I read the diagnostic note at the end and understood that what they had sent was not a technical fault. It was a mirror. When data starts to push back, tactics finally open their mouth. This time the data did not push back. It disappeared, and that disappearance said more than any statistical table I have ever read. I have been writing about sport since 2026, when I was a young reporter in Australia, and I have spent nearly forty years since in press rooms, studios and empty grandstands. In that time football went through three data revolutions: the event-data revolution of the 1990s, the spatial model and expected-goals revolution of the 2010s, and the real-time positional data revolution of this decade. Basketball followed the same path with motion-tracking cameras. Tennis has point-by-point analytics. Baseball had sabermetrics before I finished university. Combat sports did not. That is the last great paradox of professional sport. An industry with billions in revenue, global broadcast contracts and a betting market larger than tennis, operating on a data system thinner than the Norwegian third division. The root cause is competitive structure. A footballer plays 45 to 55 matches a season, 90 minutes each, roughly 4,000 competitive minutes a year. A top mixed martial artist fights two or three times a year, 15 to 25 minutes each, under 75 minutes total. The sample ratio between the two differs by more than fiftyfold. In statistical terms, that is the gap between a dataset large enough to build a model and a dataset only large enough to tell a story. And when data is only large enough to tell a story, the storyteller always beats the analyst. The second cause lies in the nature of the contest. Football is a game of large numbers: thousands of passes, hundreds of duels, the law of large numbers flattening luck. Combat sports is a game of the single event. One right hook at the twelfth second can invalidate the forecast model you spent six months building. That is why current machine-learning models for fight outcomes hover barely above a coin flip, while football models have crossed the 55 to 60 percent threshold in some leagues. I once simulated crowd noise for an empty stadium and discovered that the loudest applause came from the numbers. In combat sports, the numbers have never been allowed onto the floor. Start with the first and most misunderstood dimension: style matchup. Traditional martial arts call it the chain of strike, counter and counter-to-the-counter. Every style breeds a counter, every counter breeds a counter-counter, and that spiral is what tactics actually means. Projected onto mixed martial arts, you get the familiar chain: the striker beats the wrestler, the wrestler beats the submission specialist, the submission specialist beats the striker from top position. The clearest case study is Khabib Nurmagomedov against Conor McGregor at UFC 229 on 6 October 2026 in Las Vegas. The event drew roughly 2.4 million pay-per-view buys, the highest in the organisation's history at that point. Before the fight, the entire media discourse revolved around trash talk and the brawl at the press conference. But the underlying data told the ending: Khabib had one of the highest takedown accuracy rates in lightweight history, while McGregor had one of the lowest takedown defence rates among former champions. When those two numbers meet, the result stops being a forecast and becomes a consequence. Yet even that case study reveals the limits of data, because another pairing proves the opposite. Israel Adesanya and Alex Pereira met four times across two sports, under two rule sets, producing three different outcomes. In kickboxing, Pereira won twice. In mixed martial arts, Adesanya won by knockout in the second round at UFC 287 in April 2026, after losing to Pereira by technical knockout in the fifth round at UFC 281 in November 2026. Same two people, same skill set, four fights, three endings. Change the rule set and the old data becomes void. That is why the eight-dimension report could not write a single line under style matchup. No fighter names means no pairing. No pairing means no counter chain. And no counter chain means every tactical conclusion is literature. Dimension two: record quality. This is where professional combat sports deceives itself most. In boxing, a 40-0 record sounds terrifying until you check the records of those forty opponents. On average each had lost twelve fights. A few had lost thirty. Some had never won at all. That is not a record, it is a portfolio. The mechanism is simple and economic: a manager buys a cheap win streak to manufacture a sellable asset. Renting a weak opponent costs far less than the added value of a 40-0 line on a poster. And because no body compels the publication of a strength-of-schedule coefficient, the trap stays invisible until the fighter meets the first genuine opponent of his career. In mixed martial arts the phenomenon is milder but survives at regional level. Detection is not hard: sum the win rates of all opponents over the past three years and compare with the divisional average. A gap above fifteen percentage points means the record has been inflated. Dimension three: condition and weight-cut risk. This is the most dangerous dimension and the most ignored in daily commentary. A fighter competing at 70 kilograms typically walks around at 82 to 85. That 12 to 15 kilogram gap must be cut in the seven to ten days before the official weigh-in, mostly through dehydration. The medical consequences are well documented: reduced plasma volume, increased blood viscosity, impaired kidney function, electrolyte imbalance, and most importantly an elevated risk of brain injury in the 24 hours after weighing, because the brain has not rehydrated. In 2026 a fighter in Brazil died after a weight cut for a regional event. In 2026 a young ONE Championship athlete died from dehydration complications during the cut. After those two cases, ONE Championship introduced urine-based hydration testing and limited cut magnitude relative to walking weight rather than absolute weigh-in weight. That was a correct reform, and it is a rare example of an organisation fixing its own rules before the law forced it. But it applies to one organisation. Across the rest of the combat sports world, fighters still step onto the scale pale-lipped and dry-mouthed, and nobody records a number. Dimension four: organisational barriers. Professional boxing is the classic case of fragmentation. Four major sanctioning bodies coexist, each with its own rankings, mandatory challengers and fees. A genuine champion may need three fights to unify four belts, and two of those three may never happen because negotiations collapse. Fans get a ranking system nobody understands; fighters get a career carved into contract fragments. Mixed martial arts went the opposite way: extreme centralisation. One organisation holds most of the market, controls access to opportunity, and applies the Champion's Clause, under which a champion's contract automatically extends by a set number of bouts. That clause turns a title into a legal shackle rather than a passport. The inevitable result is litigation. In 2026 a group of former fighters led by Cung Le sued the largest promoter for anti-competitive conduct, alleging that exclusive contracts and the acquisition of rivals had suppressed the labour market. The case ran a decade and closed with a settlement worth hundreds of millions of dollars in 2026. Large as that figure is, it remains far smaller than the aggregate suppressed earnings of thousands of fighters across those ten years. Dimension five: business model. Revenue rests on four pillars: pay-per-view, broadcast rights, live gate and sponsorship. Mayweather against Pacquiao in May 2026 drew roughly 4.6 million pay-per-view buys. Mayweather against McGregor in August 2026 drew about 4.3 million. Those numbers sound like the peak of a healthy industry. Now look at the other side of the ledger: athlete revenue share. In the major North American team leagues it sits near 50 percent. In professional combat sports, figures disclosed in litigation and financial filings have typically ranged between 16 and 20 percent. That is the lowest split in all of professional sport, and it explains why so many mid-tier fighters hold second jobs despite having appeared on national television. Dimension six: judging and scoring. The 10-Point Must System was designed for boxing, where each round has a clear winner based on landed strikes and damage. Applied to mixed martial arts, it meets a structural problem: a round can consist of three minutes of ground control with no strikes at all, and the system has no language to describe the value of those three minutes. The biggest controversies flow from here. Canelo Alvarez against Gennady Golovkin in September 2026 ended in a draw, with one judge scoring 118-110 for Canelo, a margin almost the entire professional community considered indefensible. Mayweather against Canelo in September 2026 produced a 114-114 card. Manny Pacquiao against Jeff Horn in July 2026 triggered an independent review by the sanctioning body, which concluded the result need not change, a conclusion that only widened the argument. The problem is not individual judges. The problem is that the scoring system was not designed for the sport it serves, and nobody is responsible for redesigning it. Dimension seven: the category error between competitive and performance martial arts. This is what the eight-dimension report called domain misclassification risk, and it matters more than it appears. Wushu Taolu is a forms discipline scored on two components: technical difficulty and performance quality. No opponent. No rounds. No knockout. Judges award points based on angle, rhythm, power and completion of pre-registered high-difficulty movements. Applying win-loss logic to Taolu is as wrong as applying goal difference to gymnastics. Sanda is the opposite: full-contact opposition permitting punches, kicks and throws, scored on landed strikes and successful takedowns. Sanda sits between professional kickboxing and traditional martial arts, with a metric set entirely its own. The ambiguity has practical consequences. The International Wushu Federation has been recognised by the International Olympic Committee since 2026, and wushu appeared on the programme at the Beijing 2026 Olympics as a special tournament not counted in the official medal table. Every time wushu moves closer to a full slot, the question returns: which discipline would be included, Taolu or Sanda? The two share a name and almost nothing else. Dimension eight: health and career risk. This is where combat sports data is weakest, and where that weakness does the most damage. The three basic inputs for screening cumulative traumatic brain injury risk are cumulative head strikes, number of knockouts suffered, and sparring intensity. No professional organisation publishes all three. Knockout counts exist because they sit in the results. Cumulative head strikes do not, because nobody counts gym rounds. Sparring intensity is entirely invisible from outside. Which means a fighter can hold a record that looks safe, ten bouts, two knockouts, while absorbing thousands of head strikes across a decade of training. The file the public sees is the tip of an iceberg whose submerged mass is never measured. Here I must be explicit about method: when a dimension lacks data, the correct conclusion is not low risk. The correct conclusion is cannot assess. That is the difference between innocent and untried, and in sports medicine that confusion can kill. So if the data is this thin, why do millions of people argue about combat sports every day? My answer is that we do not lack combat sports data. We have too much fake data. Count what the industry produces daily: unsourced transfer rumours, edited videos manufacturing conflict, predictions from accounts that never publish their hit rate, and rankings constructed to sell advertising. That is not the absence of signal. It is the overproduction of fake signal, and it is far more dangerous than a blank space. A blank space is honest. It tells you that you do not yet know. A fake signal lies to you in the tone of certainty. The 2026 World Cup was not merely a tactical scandal; it was a broken mirror reflecting an entire football culture lying to itself. The same is happening in combat sports, except nobody has written it down. I once thought esports would be the lesson for combat sports. There, every input is logged, hundreds of data points per minute, actions-per-minute computed to the second. If data were the answer, esports would be the cleanest sport on earth. It is not. Esports is wrestling with match-fixing and betting, and it is wrestling faster than combat sports because its regulation lags behind its own growth rate. Complete data does not erase integrity problems. It relocates them somewhere harder to see. Between the football pitch and the esports arena there is an invisible bridge, and I make a living proving it is shaking. That bridge connects two beliefs: that more data leads to clearer truth, and that with less data the truth is still there, merely waiting to be discovered. Both beliefs are wrong. Combat sports lacks data not because people are lazy about measuring. It lacks data because the structure of the sport generates very few measurable events, and because the people who control that structure have no incentive to measure more. A fighter with a murky record is easier to sell than one with a transparent record. A fragmented belt sells four times. An endless argument generates infinite traffic. Opacity is profitable. That is the whole story. When the stands are empty, I hear the match through numbers rather than through my heart, and that was the first time I understood the sadness of a phase of play. In combat sports the stands are never empty. They are simply full of noise nobody verifies. What I want to leave behind is not a verdict on the combat sports industry but a way of reading it. When someone tells you a fighter will win because his record is better, ask who built that record. When someone tells you a fighter has fully recovered from injury, ask who examined him and by what method. When someone tells you a puncher has one-punch knockout power, ask whose chin that knockout landed on. Those questions do not require big data. They require patience, and a little courage to accept that you may not know anything at all. In forty years in this trade I have learned one thing worth keeping: the analyst's greatest enemy is not ignorance. It is unfounded certainty. That eight-dimension report returning blanks is, in the end, the most honest document I have read this year. It did not tell me who would win. It told me exactly what I did not know, and why. A sport willing to print a blank page like that will progress faster than one willing to print forty pages of numbers nobody can verify. And if there is one thing I want to see in the next decade, it is a combat sports organisation willing to publish its complete fight data publicly, including the lines that make it look worse. Until that happens, every argument about combat sports will remain an argument about belief, decorated with numbers chosen by the storyteller. And I will still be here in Beijing, reading blank reports, waiting for someone to dare fill them in.