The Data Void in Vietnamese Billiards: Where Error Signs Its Name

At 3:12 a.m. on February 9, 2026, in Liverpool, I opened a spreadsheet with fourteen columns and not a single row of data. The first column read “innings.” The fifth read “cue-ball coordinates at first cushion contact.” The ninth read “time between shots, in seconds.” The last read “outcome.” Every cell beneath was empty, and I looked at it longer than the moment required. The match that file was built to record was not empty. It happened, it ran thirty-four innings, it ended 40-37, it was streamed to tens of thousands of viewers, and at least three shots made people in the room stand up. I rewatched it four times and took notes by hand. Then I understood that the thing I had been trying to build — a dataset for Vietnamese billiards — had not collapsed because I was lazy. It collapsed because there was no raw material. That is where billiards differs from every sport I have analysed. In football, I open a laptop and there are hundreds of thousands of coded events, with coordinates, with actors, with outcomes. In billiards, I start from zero, and zero here is not a starting point. It is the entire picture. Error is where reality signs its name, but to see the signature there must first be paper. Vietnam has thousands of billiard clubs and millions of people who have held a cue at least once. Three-cushion carom is the discipline with the densest international calendar. Two dates matter to anyone who follows it: the 2026 world three-cushion title in Randers, Denmark, won by Tran Quyet Chien — Vietnam’s first world title under the UMB system — and the 2026 world title in Ankara, Turkey, won by Bao Phuong Vinh, then in his early twenties. Between those two dates sits a generation long present in main draws: Nguyen Quoc Nguyen, Ngo Dinh Nai, Ma Minh Cam. What the ecosystem publishes, though, is thin: brackets, match results, a cumulative ranking, and four classic fields — innings, total points, average per inning, highest run. Main-draw World Cup matches generally finish at 40 points, some finals at 50. Everything else — the position of the balls as an inning begins, the cushion-contact sequence, cue-ball speed, the silence between shots — exists in no file I have ever opened. The largest blind spot in this sport is the position of the three balls at the moment an inning starts. One player steps in with the cue ball free and both object balls sitting near a corner, able to score with a soft touch. Another steps in with the cue ball frozen to a cushion and any route requiring three cushion contacts with a tolerance of millimetres. If both succeed, both are recorded as one point. On the scoreboard they are identical. Tactically, they are playing two different sports. The consequence is structural: every statistic in billiards begins from a unit of unequal difficulty. Football lived with the same flaw before expected goals taught the industry to separate a goal from the quality of the chance. Billiards has not taken that step, and in Vietnam nobody has framed it as a technical question. The average per inning — total points divided by innings — has three structural weaknesses. The denominator is contaminated by the opponent, so blowouts distort both players’ numbers precisely where conclusions are drawn. The division erases variance: two players can finish a tournament on 1.25 with sequences of 0,0,0,4,0,3,0,2,5 and 1,1,1,1,2,1,1,1,1,2 — the same mean, two different games. And scoring is non-stationary, because early innings are reconnaissance and late innings carry match-point pressure; collapsing them into one fraction is like judging a footballer on pass completion without caring when the passes happened. In my own small sample, the standard deviation of points per inning separated strong players better than the mean. The ratio of scoreless innings to innings of three points or more produced a stylistic fingerprint: high ratio means a player who lives on runs, low ratio means a player who lives on survival. They require different coaching, different scheduling, different matchups. Vietnam does not lack footage. Every national event has at least one phone recording it. But video that has not been annotated is memory with a hard drive. Nobody can query how a player has scored over three seasons when an inning began with the cue ball outside the centre of the table. In the sports-data economy, the invisible middle layer is annotation — rooms of people turning footage into fields. Football did not spontaneously acquire event data; it bought it. Vietnamese three-cushion has no such layer, not for want of technology but for want of a payer. A rough cost estimate: one properly recorded match to 40 points needs at least six position or description fields per inning. Over thirty-four innings that is more than two hundred data points, each requiring a scrub to the correct frame. A skilled annotator needs three to four hours per match. A national event of sixty matches needs over two hundred hours. For an international federation that is affordable; for a tournament with modest prize money it is indefensible — data here benefits coaches, players, broadcasters and sponsors while costing the collector. Public-goods economics explains exactly why the gap persists. The cushion sequence is the structure of a shot. Two players can score from the same position through different routes; both are legal and both are recorded as one point, but they differ in difficulty, in repeatability, and in the position left behind. With sequence data, coaches would have the equivalent of a shot map: where a player scores, by which route, and how much performance drops outside that zone. My own log of roughly sixty matches — a small sample I do not treat as conclusive — suggests the decisive innings cluster around short shots near the cushion and long opening cushions that loop the cue ball behind the object balls. Even a weak observation beats a belief with nothing to check it against. If I could add one metric tomorrow with no budget, it would be time between shots. Every stream carries timestamps. Rhythm signals internal state: a confident player keeps a stable interval, a wavering player shifts it within the same match, rushing to escape a decision or slowing to postpone it. Watching Vietnamese players internationally, I noticed a pattern often enough to write it down: their leading players tend to open innings faster than the field average, and their results deteriorate in matches where that rhythm is dragged down by a slow opponent or unfamiliar table conditions. Not enough data for a conclusion — but enough to know that with rhythm recorded, arguments about form would stop being arguments about feeling. Every high-level match contains dead innings, where the only goal is survival and the quality of the position left to the opponent. This is the least valued part of the sport, because it creates no points. A well-judged safety that forces three scoreless innings and a self-inflicted break is invisible outside a blank line. I call the missing metric left-position quality: after every non-scoring shot, how hard was the position handed to the opponent. Football learned to measure chance quality to judge attacks; measuring left positions is how you judge defence. Without it, every claim about a good defensive player is a claim about a commentator’s memory. Market value in billiards — invitations, appearance fees, sponsorship, headline space — is not set by granular data, because granular data does not exist. It is set by narrative. A player’s value is a story the market repeats until nobody bothers to check. The mechanism produces two distortions: it overvalues the moment, so one viral shot can build a reputation far larger than its statistical weight, and it undervalues everything that cannot be packaged into a clip — resilience in narrow defeats, holding rhythm against a slow opponent, performing in a crowded room. A deep run in a major is often remembered as proof of special potential when the draw itself did much of the work; bracket luck proves no system. Meanwhile the real knowledge sits in the clubs, held by veterans and coaches who know exactly who scores from which zone and who fails in the third inning after a long run. It is transmitted by voice, by standing beside a table, by collective gasps. That is both its strength and its ceiling: extraordinary resolution, zero transferability, no correction by evidence, and it disappears when the holder stops showing up. It also explains why international scouting of Vietnamese players fails in two directions — undervaluation from reading only the scoreboard, and misclassification from reading an unfamiliar technique as a quirk rather than a solution optimised for different table conditions. Ranking points compound through attendance. A player with a sponsor flies; a player without one does not, regardless of technique. The ranking measures the ability to travel and only afterwards measures ability. In Vietnam the gap between a sponsored player and an unsponsored one is not measured in practice hours but in flights, and every missed qualifier narrows the visible group while widening the invisible one. For the national federation, publishing per-inning data in a machine-readable format rather than a photo of a scoresheet is a low-cost step with long-term leverage: it turns one tournament’s memory into ten years of raw material. In 2026, when world sport stopped, I rewatched an entire club season at university in Manchester to build an imagined dataset on attacking behaviour without crowd pressure. When sport returned to empty stadiums, long-pass completion fell across English teams, against my prediction. It took six months to rewrite the conclusion: teams passed sideways more but penetrated less, because a psychological driver had been removed. The no-crowd season deleted a variable no model could encode — noise. Vietnamese billiards ran a version of that experiment without calling it one. Club noise in Vietnam is not stadium noise: balls colliding on adjacent tables, commentary behind your back, ceiling fans, a group standing up together as a run begins. In a sport decided by millimetres, that is part of the playing conditions, and it vanished with the audience. In my log, streamed matches in closed rooms showed shorter intervals between shots and more dead innings than comparable matches with a live crowd. Not enough data to assign a cause — but if rhythm had been recorded before 2026, we would own the cleanest comparison in the sport’s history between environment and decision-making. We lost it to the habit of not writing things down. More data is not automatically the answer. Vietnamese three-cushion has a competitive advantage built on oral transmission, and imposing a metric changes the behaviour it measures: publish rhythm and players will practise fast rather than well. Importing football-style models without the club ecosystem produces decoration — dashboards shown on broadcasts to look modern, used for no decision. And there is the risk of belief. A shot rarely fails because of technique; it fails because of absolute belief in technique, to the point where the player stops observing the table and simply re-executes what he trusts. Data in the hands of the unverified behaves the same way. Every inning is a hypothesis waiting to be refuted by the table. If I had to restart, I would begin with four fields recordable on a phone: player, innings, points per inning, and time between shots. Four fields, logged consistently for one season, would change how Vietnamese billiards sees itself. The final lesson from that empty spreadsheet is unremarkable: data does not emerge from popularity. A sport with millions of players can still have zero rows. A country with two world champions can still have no way to demonstrate how they won. And that emptiness is nothing to be ashamed of. It is a gap waiting for the first person willing to sit down and start a stopwatch. Next time you watch a three-cushion match, try counting only the innings in which a player scores nothing, and watch what they do with the cue ball on the final shot of that inning. Nobody publishes that number. Perhaps because nobody does, the sport still has room for people rather than models.