Altora Analysis

🎱 SNOOKER — ACCURACY & RECORD

Calibrated model analysis, not tips. Validated out-of-sample, tracked in the open.

← all sports

Today · 2026-10-19

Today's model reads

Model win probability for today's matches — not tips. These settle into the record below.

Jak Jones v Tom Ford
63% · mkt 66%
Stuart Bingham v Yao Pengcheng
67% · mkt 68%
Judd Trump v Martin O'Donnell
83% · mkt 87%
Zhang Anda v Liam Pullen
69% · mkt 71%
John Higgins v Liam Highfield
64% · mkt 66%

Model updated 22 July 2026: published probabilities now blend the Elo rating with the market opening price through a gradient-boosted calibration layer, validated out-of-sample before switching. The record below spans both model versions.

Model validation · 5,489 out-of-sample matches · to 2026-07-25

Refreshed 26 July 2026 under a documented, reproducible protocol (rating engine + serving calibration, tested on matches from 2024 onward). It replaces an earlier snapshot whose producing script was lost — we rebuilt the harness rather than carry numbers we could no longer verify.

Is the model accurate?

These are matches the model never trained on. It picks the winner about six in ten — and, more importantly, its probabilities are calibrated: when it says 71%, it happens about 71% of the time.

66%favourite pick accuracy
0.226model brier
0.206market brier

Reliability — model probability predicted vs what actually happened

46%
38%
1,192
56%
66%
4,007
68%
87%
277

The two bars line up on every band — that's calibration. On raw pick accuracy the model even edges the market. Not sharper on price, but demonstrably accurate, shown in the open.

The live record · every pick since going public

The live record

472settled
66%model pick hit
0.215model brier
0.211market brier

Every pick is logged when the exchange market opens and settles here, wins and losses alike. The validation above is the model's accuracy over hundreds of past matches.

DateModel pickProbResult
14 voided match(es) shown greyed below — kept on the record, excluded from the accuracy figures (no result to score).
2026-10-19Pang Junxu v Alfie Burden67%W
2026-10-19Chris Wakelin v Jamie Jones74%L
2026-10-18Xiao Guodong v Sam Craigie63%L
2026-10-18SI Jiahui v Wang Xinbo63%L
2026-10-18Aaron Hill v Thepchaiya Un-Nooh52%L
2026-10-18Anthony McGill v Stephen Maguire56%W
2026-10-18Jack Lisowski v Huang Jiahao72%W
2026-10-18Gary Wilson v Mateusz Baranowski72%L
2026-10-18Mark Selby v Louis Heathcote80%L
2026-10-18Hossein Vafaei v Matthew Selt61%L
2026-10-18Joe OConnor v Matthew Stevens63%W
2026-10-18Neil Robertson v Ricky Walden65%L
2026-10-18Ali Carter v Xu Si63%L
2026-10-18Shaun Murphy v Liam Davies64%L
2026-10-18Elliot Slessor v He Guoqiang62%VOID
2026-10-04Wu Yize v Yuan Sijun77%W
2026-10-03Yuan Sijun v Jimmy Robertson58%W
2026-10-03Judd Trump v Wu Yize63%L
2026-10-02Chris Wakelin v Jimmy Robertson62%L
2026-10-02Judd Trump v Mark Selby60%W
2026-10-02Yuan Sijun v Michael Holt61%W
2026-10-02Wu Yize v Shaun Murphy53%W
2026-10-01Jimmy Robertson v Scott Donaldson53%W
2026-10-01Mark Selby v Stephen Maguire76%W
2026-10-01Neil Robertson v Chris Wakelin62%L
2026-10-01Kyren Wilson v Michael Holt78%L
2026-10-01Shaun Murphy v Elliot Slessor72%W
2026-10-01Wu Yize v Ricky Walden71%W
2026-10-01Judd Trump v SI Jiahui78%W
2026-10-01Yuan Sijun v Zak Surety60%W

⬇ Download the full record (CSV — opens in Excel / Sheets) · showing the most recent 30 of 472