Four Tables, Six Rule Systems: Why the Word “Billiards” Is Not Enough

In a pool hall in Beijing's Chaoyang district, two tables sit less than two metres apart. The one on the left has no pockets. Green cloth, four cushions, three balls, and a single turn that can run to forty contacts. The one on the right has six pockets, a triangle of balls racked at the head, and a shot clock counting every stroke. The customers call both of them billiards, and nobody finds that strange.

A few months ago I received a data extract from our internal tracking system. Every field was empty: no title, no source, no player, no event, no timestamp. One label survived — billiards. The label was linguistically correct and analytically useless. It worked exactly the way those two tables work when we file them under one sport.

In Vietnamese, “bi-a” is a very wide umbrella. Most of the time that breadth is harmless: readers want to know who won, who lost, who set a record. But when the goal shifts from retelling a match to building a dataset — form rankings, win-rate comparisons, prize-money tracking, outcome forecasting — the umbrella becomes a structural hole.

Under it sit at least six rule systems.

Snooker uses a twelve-foot table, six tight rounded pockets, fifteen reds and six colours. Its professional structure is run by the WPBSA and WST, with three Triple Crown events and a ranking list calculated on two-season rolling prize money.

American pool covers 9-ball and 10-ball: nine-foot tables, wider pockets, scoring by rack and turn. Commercial systems such as Matchroom stage finals that last a single hour.

Chinese 8-ball uses a nine-foot table with tighter pockets than the American standard, fifteen balls split into two groups, and a black eight to finish. It holds the largest domestic market in the sport, concentrated in China, with a near year-round calendar.

Three-cushion carom has no pockets at all — only cushions and geometry. The UMB governs it, and its central metric is the average.

Then there is American 8-ball, English billiards and Russian pyramid: three more systems, each with its own community and its own rulebook.

Each system has its own governing body, ranking list, calendar, prize structure and technical vocabulary. A dataset that merges them into one column loses comparability at the root, long before the question of right or wrong arises.

The most dangerous error in billiards analysis lies elsewhere: missing data is at least visible, while data that has been filled in plausibly but under the wrong discipline is not.

In snooker, a player is measured by centuries, the frequency of 50-plus breaks, the quality of safety play, and the ability to hold that quality across sessions. A world final runs two days, four sessions, up to thirty-five frames. Form here means holding accuracy into the eighth hour, when the body is tired and the scoreboard has swung several times.

In 9-ball the measurement sits somewhere else: the rate of running a rack after the break, the quality of the break itself, the ability to choose a push-out when no pot is available. A final lasts sixty to ninety minutes. Form here is the conversion rate from chance to rack inside a short window.

Chinese 8-ball changes the story again. A bigger table, tighter pockets, slower cloth, and every error in cue-ball control paid for with an entire rack. Based on my experience tracking matches at domestic Chinese events over seven years, the break-and-run rate on the main television table runs noticeably below the side tables, a gap large enough that any ranking mixing the two must publish its adjustment factor. That data needs further verification across other seasons before it becomes a standard.

In three-cushion carom there are no pockets, no break shot, no racks. The central metric is the average — balls scored per innings. Elite players at major events hold averages between 1.5 and 2.0, and a match can run forty innings, each one a three-dimensional geometry problem. In Vietnam, players such as Tran Quyet Chien and Nguyen Quoc Nguyen have carried the discipline into the world's top group, while Duong Quoc Hoang belongs to the pool line — two different currents, two different sets of metrics, sitting under one word.

Put those four metric sets in one table and average them, and the result describes nobody, no discipline, no season.

In 2026, while rebuilding a database on break-and-run rates at Chinese billiards events, I let 9-ball data sit in the Chinese 8-ball column for two days. The combined rate came out at 41 percent, a figure that exists in no system: 9-ball has an open break, a smaller table, wider pockets; Chinese 8-ball demands far tighter cue-ball control. The fault was not in the algorithm. It was in the column header. The intern on my team asked one short question: “So which sport are we talking about?”

Peak age does not follow one curve either. Data I track at three-cushion events shows the top group holding peak form into the mid and late forties. In American pool, champions cluster between twenty and the early thirties. In snooker, the stable band falls between twenty-seven and thirty-five. Average those three curves into a single “billiards peak age” and you have a parameter that discards the wrong people.

Compliance risk follows the same logic. In December 2026 the WPBSA charged ten players, mostly Chinese, over conduct linked to betting and match manipulation. In June 2026 the sanctions were published, with Liang Wenbo and Li Hang receiving lifetime bans. That case belongs to the snooker rule system, handled by a specific authority under a specific rulebook.

File it in your database under the label “billiards” and three years later someone queries “match-fixing in billiards” and receives a pattern that looks like a sport-wide crisis, when in reality it was an institutional failure inside one system. The John Higgins case of 2026 sits in the same frame: a reporting-duty breach judged by the sport's own body, not by a general court for every kind of cue sport.

Every layout is a confession; my job is to listen to it — but only when I know which rulebook I am listening to.

On one side, people argue that a wider label is more convenient: general readers only need to know it is billiards, and granularity just adds noise. That argument holds for a news brief and fails for data. An empty field makes an editor stop and ask. A field filled in wrongly but plausibly travels onward in silence, and every report built on it inherits the original error. In data work, leaving a field empty is a form of honesty.

The other side — those demanding absolute separation — is wrong in the opposite direction. Transfer between systems is real: cue action, positional discipline, table reading, mental endurance across long matches. Chinese 8-ball is itself a hybrid product: it takes the group structure of American 8-ball and places it on a table scale and cloth speed closer to snooker. That hybridisation built the sport's strongest growth market of the past two decades.

The transfer market is not a lottery; it is an unsolved equation. In the Chinese 8-ball system, players sign with teams, local sponsors and training academies, and their value is set by knockout-stage wins rather than reputation. Merge that equation with the snooker or pool version and take an average, and you get a meaningless price list.

Esports taught me that reflexes are also a form of tactics; and a professional system with no safety net after retirement pushes players toward very different decisions than a system with one. That is why rule systems cannot be separated from data analysis: rules set career length, career length sets the form curve, and the form curve sets how every remaining metric should be read.

The biggest blind spot is not the merging of labels. It is that the merge leaves no trace. A blended column produces a plausible result, and because it looks plausible, nobody goes back to check the header.

Next time a report opens with the word billiards, the first question is not who won. The question is: which table, which rulebook, which format, and can this dataset still be compared with itself last season? If the answer is unclear, the right move is not to keep writing but to stop and find the raw data. When the stands are empty, data becomes the only applause I trust — and an empty label is not data.