When Billiards Data Comes Back Empty: A Big Season Needs an Identification Gate

On a Friday night, the billiards hall on Lach Tray Street was full, a pool match was replaying on the wall-mounted screen, and I opened my spreadsheet. It has seventeen columns: event name, discipline, player, pot success rate, winning margin in racks, average time per visit to the table, table condition. That night, all seventeen columns returned a single value: empty. No event name, no player name, not one row to cross-check. I closed the laptop and did not write the piece. An analysis of billiards born from an empty spreadsheet is only an exercise in phrasing. Data never lies, but I have misheard it before — and this time what I heard was silence.

Why “billiards” is a dangerous word

In Vietnam, “bida” covers at least four different rule systems: carom (one-cushion and three-cushion), pool (8-ball, 9-ball, 10-ball), snooker, and Chinese-style pocket billiards (Heyball). These four systems do not share a metric set. Snooker measures century breaks, 147 maximums, long-pot success. Pool 9-ball measures break quality, pot rate on the opening shot, and how often a player pushes an opponent into a safety exchange. Three-cushion carom measures the moyenne — average points per inning — and the longest series. A dataset without a discipline label is not data yet; it is a list of words placed side by side.

A big season accelerates everything. Halls open, amateur tours expand, viewers demand a prediction before the first rack. The analyst faces a very specific pressure: to say something. Based on my experience tracking matches over the past four years, most of the error in billiards prediction pieces does not come from the model. It comes from the writer assigning the wrong discipline in the very first line, then spending the rest of the article defending that wrong label.

The empty record is not an exception in Vietnamese billiards. It is the default state. Results from an amateur event usually exist only as a photo of the bracket posted on a fan page, with a caption naming the winner. Forty-eight hours later the post scrolls out of the feed and the record disappears with it. Nobody stores the format, nobody stores the rack-by-rack score, nobody stores the table conditions. To reconstruct one domestic season, I have to message referees, hall owners, and whoever was holding the phone.

Four things billiards data must prove

When a record has no event name, no player name, and no description of the table and balls, the first action is to stop. A discussion of the break shot in pool 9-ball, applied to snooker, produces a concept that does not exist in the rules. An analysis engine does not know that on its own; it only knows how to fill the gap with whatever sounds most plausible. The discipline gate is the condition that gives every conclusion behind it meaning; skip it, and the rest of the article is only phrasing.

In June 2026, the WPBSA published its findings in the case involving ten Chinese players; Liang Wenbo and Li Hang received lifetime bans. In 2026, John Higgins was suspended for six months and fined 75,000 pounds, even though the investigation found he had not fixed a match. Two cases thirteen years apart, one lesson: a clean dataset does not prove cleanliness, and an empty dataset proves even less. In every billiards piece I separate two sentences: “no integrity signal” and “no violation” are different sentences, and only the second may be published with a full case file behind it. In a market where betting runs alongside small, low-visibility prize events, the distance between those two sentences is the distance between a career and an accident.

The generation born in 2026 — Ronnie O'Sullivan, John Higgins, Mark Williams — still shapes the world snooker ranking after more than three decades. Three men, three different trajectories, one starting point. Looking at a single event, I would get all three wrong. Three thousand matches taught me that one match can teach more than all of them — but only when I place that match inside its proper sequence.

Jimmy White lost six World Championship finals in snooker, five of them in consecutive years from 2026 to 2026. No column in my spreadsheet measures his hand in a deciding frame. That is why I keep a separate section called “exception notes”: the rhythm of getting down on the shot, the length of the pause between visits, the way a player chalks the tip after missing an easy ball. None of this enters the model, but it explains exactly where the model goes wrong.

Format is the biggest lever in pricing upset risk, and it is where Vietnamese data is weakest. A long event, the kind played to 33 frames, filters for true strength because it forces a player to survive hours of lost concentration. An event played to four racks compresses the skill gap down to roughly one lucky break. Most domestic amateur tours run in the short format, so a “champion” label there carries less information than a quarter-final in a long event. Putting the two labels side by side in one ranking is wrong from the counting method onward.

The prodigy filter belongs in this group too. A young player wins a short event, and social media immediately calls him a phenomenon. To test that, I need at least one of three things: a ranking, a title count across several seasons, or repeatable single-visit data. A sample of one event is noise, and noise is always louder than the truth.

The billiards transmission chain runs from upstream — halls, tables, cues, accessories — through the midstream of players, events, and broadcast, down to downstream sponsorship and the collectibles market. Two flows are reshaping that chain: Chinese-style pocket billiards events are raising prize funds and pulling mid-ranked snooker players, while Middle East capital is re-pricing the entire event map. For Vietnamese billiards, the bottleneck sits in the middle: more events, more players, but almost no records with enough fields to cross-check.

Every source sets a confidence ceiling. Official WST or WPBSA releases sit at the top. Specialist billiards media sits in the middle. Self-media sits at the bottom, and it is usually the only source present immediately after a match ends. I do not reject weak sources; I label them weak, and I do not let them carry a conclusion.

The counterintuitive part

What billiards analysis lacks is definition, before it lacks numbers. A spreadsheet can hold thousands of rows and still be useless if the first row does not state whether this is snooker or pool. Conversely, an empty spreadsheet can still generate a report that sounds very confident, because language is always ready to fill the gap. The biggest risk in this profession is epistemic: writing a confident conclusion about something you never defined. I do not write to persuade anyone. I write so that the data has a witness.

There is one more assumption I want to remove. A crowded venue is not automatically an advantage. In billiards, noise, people walking around the table, and shifting light when the hall opens another door all act on the same thing: the rhythm of getting down on the shot. I have watched players strike the ball far better in an empty hall and shrink when the next table started clapping. That variable sits in no model I have ever built, and it is why I always note the crowd condition on the same row as the score.

Next-cycle signals

In the coming cycle I am tracking four signals: a specific event name, a specific player name, a clear discipline label, and a verifiable source. When all four appear together, the spreadsheet starts speaking. The model knew in October. I only had the courage to believe it in May.