The Grey Zone on the Canvas: Analysing Martial Arts When the Data Goes Silent

Three in the morning in Beijing, November. I open a file called striking_data_1400rows.csv, sent over by a statistician friend, with one short line of text attached: "Use it for now, I pulled it from public sources." Fourteen hundred rows. Three filled cells. No fighter names, no event names, no dates, no units of measurement. Three indices stranded in a sea of empty cells, like three spectators seated inside a twelve-thousand-seat arena. Forty years in this trade taught me how to call fights I have insufficient data for. Someone hands you a blank sheet and you still have to talk for three hours, still have to keep the audience convinced you know something they do not. That night was different. Behind that empty file there was no carelessness on the sender's part. It was a product. An entire system had manufactured it, and I was holding it in my hands. When data starts to rebel, tactics finally open their mouth. When data disappears, what opens its mouth is power. I spent that night rereading three anonymous indices and understood that my craft is not the craft of reading numbers. It is the craft of reading the gaps between them. A SPORT SOLD ON RUMOUR A single English Premier League matchday generates hundreds of thousands of data points, collected by independent companies and resold to broadcasters, bookmakers, clubs and the smallest blogs. A night of professional combat sports generates a comparable volume, but most of it is owned by the promoter. The difference between these two worlds is not quantity. It is permission: who gets to count, what gets counted, and who it gets published to. On June 27, 2026, in Kazan, I called Germany's 0-2 defeat to South Korea. When Son Heung-min scored at 90+3, I said into the microphone that this was a textbook Zerg opening in Starcraft, that South Korea had patiently waited for an opening across ninety minutes. The former international sitting beside me went silent for ten seconds, then asked: "Are you sure you are talking about football?" That clip travelled to 2.5 million views. What I rarely retell is that on that night I held an independent dataset showing both teams' passes per minute, updated by a third-party provider. I could look up how many sideways passes Germany played in the 78th minute, and what that figure had fallen to by the 88th. Nobody owned that dataset. It was the shared property of the trade. In May 2026, when the pandemic shut every stadium, I worked with a game designer to replay the 2026 Champions League final between Manchester United and Bayern Munich in three-dimensional space, across twenty different scenarios. We put Bayern 2-0 up in the 80th minute and asked whether the comeback still happens. A UEFA data analyst contacted me and admitted the approach was irritating but thought-provoking. Place those three stories side by side and a very different reality emerges. In football, an analyst can buy independent data, cross-check it, and rebuild the match under their own scenarios. In combat sports, the analyst usually has two sources: the footage released by the promoter, and the statistics compiled by the same promoter. Both pass through the same pair of hands. THE COUNTER AND THE COUNTED In professional boxing, the four major sanctioning bodies — the WBA, founded in 2026; the WBC, 2026; the IBF, 2026; the WBO, 2026 — all publish periodic rankings. Those rankings function more as negotiating instruments than as measurement tools. A fighter can be ranked first by one body and fifteenth by another, with both lists published in the same week under two sets of criteria nobody can audit. No authority sits above those four bodies to say which list is correct. In mixed martial arts promotions the picture is tidier and more closed. The promoter sells the tickets, runs the broadcast, and operates the official statistics platform. The bout result is a legal document, but every metric describing the bout is a commercial document. When a fighter prepares to negotiate a new contract, his side will cite official data — the very data published by the counterparty across the table. Nobody in that room finds anything unusual about it. I once tried something simple: I took the official statistics from one event and re-counted from video at three different playback speeds. The deviation between my three counts was roughly eight percent. The deviation between my average and the official sheet was seventeen percent. Neither side publishes its margin of error. The official sheet is treated as a definition rather than a measurement. BEAUTIFUL, HOLLOW METRICS The term "significant strike" is the clearest example. Under most current systems, a strike is classified as significant if it is thrown at distance or from a power position. That definition gives a soft long-range jab and a fight-ending elbow the same weight. A fighter can lose 80-120 on the stat sheet while winning almost every exchange that actually mattered. In DOTA 2, nobody counts total damage. They count damage per unit of resource, damage within each teamfight, and the timing of each teamfight against the rhythm of the match. A player can deal less damage than his opponent and still win, because his damage landed inside the decisive three seconds. Combat sports work the same way, but the stat sheets still count raw totals. Between the pitch and the esports arena there is an invisible bridge, and I make my living proving it is shaking. The bridge lies in the fact that both sides understand a single action's value depends on timing, on resource state, and on what the opponent has already spent. Only one side bothers to record that in its systems. AGENTS AND THE MARKET IN RUMOUR The largest hidden cost in professional combat sports is not broadcast rights. It is the intermediary machinery: managers, agents, communications advisers, and the network of friendly journalists. That machinery manufactures a very particular commodity, and the commodity is controlled rumour. A rumour does not generate itself. Someone decides to leak it on the right day, to the right person, at the right level of vagueness. The usual aim is to pressure one negotiating party, cheapen another candidate, or inflate a fighter's commercial value before he steps into the cage. Major outlets are not naive about this game. They need traffic, and rumour is the cheapest content to produce. So when someone asks why I am not reporting a deal that is trending, I usually answer that I am waiting for a price. Price here means the specificity of the information. A name with no date, no figure, no confirmation from either side is not news. It is raw material. A contract is never wrong; only the person holding the pen fools himself. BETTING AND THE SPEED OF EROSION Betting in esports is eroding competitive integrity faster than in traditional sport, and the cause is not the morality of players. It is structural. A match is played remotely, without spectators, without officials physically present, among young people earning at the bottom of the pyramid. The monitoring system was built in roughly ten years, while traditional sports had nearly a century to learn how to fight match-fixing. Combat sports sit between those two worlds, on the more dangerous side. A fighter on the opening bouts of a card can earn less than a tier-two esports professional in many regions. When the betting handle on a regional card dwarfs the combined annual income of the people who directly determine the outcome, the pressure does not need to be spoken aloud. It only needs to exist. I once simulated crowd noise for an empty stadium and realised the loudest applause came from the data. In that space there is no crowd to distract you. What remains is money flow, time, and decisions made in less than a single breath. THE CRAFT OF READING GAPS Back to the three-cell file in Beijing. The response is not to go looking for another source. The response is to interrogate the emptiness itself, in three steps. Step one: who benefits from this metric not existing. If a promotion does not publish how many times its fighters absorb shots to the head in training, the beneficiary is the ticket seller. A fighter showing signs of accumulated brain injury keeps his commercial value as long as nobody counts. Step two: what that metric would show if it existed, and who would pay to make it disappear. This question usually leads into areas both the promoter and the agent prefer to keep blurred: recovery timelines, actual weight cuts, and contractual penalty clauses. Step three: what publication costs. A newsroom can publish anything, and will lose press-room access, sources, and accreditation in return. Many outlets know the facts and choose silence, not out of cowardice, but because they want to survive long enough to cover the next fight. When the stands are empty, I hear the fight through numbers rather than through my heart, and that was the first time I understood the sadness of a play. That sadness does not come from losing. It comes from having nothing to confirm that you lost. A BRANCHING SCENARIO Suppose that from next season, every professional combat sports bout must release full sensor data within twenty-four hours: heart rate, strike velocity, distance travelled, balance-loss events. What happens? Branch one: betting markets sharpen, bookmaker margins thin, and the power of broadcast narrative falls sharply. A fighter is no longer valued by aura but by his own athletic file. Branch two: fighter market value becomes measurable, and agents lose pricing power. This branch carries the strongest political resistance, and is the least likely to occur. Branch three: the sport loses its mystery. Fans buy tickets because they do not know what will happen, not to watch a live-updating data table. Full disclosure could clean the sport and simultaneously make it less compelling. These three branches are not mutually exclusive, and I do not hold enough data to conclude which wins. That is the boundary of the trade: the analyst must accept there are questions current data cannot answer, and pretending to answer them is worse than silence. THE CONTRARIAN ANGLE The prevailing assumption in sports analysis is that more data is better. In combat sports, the opposite holds more often. The more data a promoter supplies, the more easily the analyst is steered by the story that promoter wants to sell. A dense stat sheet creates a feeling of objectivity, and that feeling is the ideal environment for a biased conclusion to pass unexamined. In March 2026 I wrote a four-thousand-word piece on the El Clasico between Real Madrid and Barcelona, using data from thirty matches, drawing player-position maps, and comparing the concept of zone control to DOTA 2. It drew only twelve hundred views, but an editor at CCTV Sports noticed it and invited me to audition for the 2026 World Cup. The 2026 rebellion taught me one thing: fear the number that does not know how to lie. At the 2026 Qatar World Cup I declared Morocco would reach the final. Morocco beat Portugal 1-0 in the quarter-final, then lost 0-2 to France in the semi-final. The online crowd called me a clumsy prophet. I wrote a five-thousand-word self-autopsy, pinpointing exactly which data had led me to the wrong conclusion. It was shared twelve thousand times, and I lost four percent of my followers, but I earned the respect of veteran colleagues. The lesson was not to be more cautious. It was to be more public. An analyst who says "no data, no comment" is quietly serving the party with the most to hide. Silence is not neutral. In any information system, silence always tilts toward whoever controls publication. A THOUGHT LEFT OPEN I still keep that three-cell file, renamed mot_khoang_trong.csv. I never use it to call a fight. I keep it as a reminder that my trade only has value when I dare to say clearly what I do not know, and to say clearly who benefits from my not knowing. On October 24, 2026, Khabib Nurmagomedov defeated Justin Gaethje and announced his retirement at 29-0. On July 10, 2026, Conor McGregor broke his tibia late in the first round against Dustin Poirier at UFC 264. Both events were captured in thousands of frames, and both are retold through different indices depending on the teller. If a complete dataset for either bout were published tomorrow, would the story change, or only the storyteller?