Methodology: Verification and Exact Probability

Foundation and registry methodology reviewed: .

Prefer primary sources

Use official operators, official lottery trustees, game rules and public authorities. A news article or search result can help locate a source but is not final proof. A blocked page stays unavailable; we do not bypass CAPTCHA, login, geography or access controls.

Verify fields, not whole pages

Each registry field has a status: VERIFIED, UNVERIFIED, UNKNOWN, NOT_APPLICABLE or REVIEW_REQUIRED. Reviewed fields identify a source, date and scope. A VERIFIED record means its base number format is established; it does not imply that every price, prize, legal or tax statement was checked. Null represents a gap and is never silently replaced with zero.

Reproduce the probability

For unordered selections without replacement, C(n,k)=n!/(k!(n−k)!). The shared engine uses exact BigInt multiplicative arithmetic. Separate independent player pools multiply their combination counts; draw-only bonus and supplementary balls do not add a second ticket pool. A different mechanic requires a different reviewed model.

Seven official jackpot denominators match the exact model. Japanese Loto 6 and Loto 7 have reviewed formats but no established numeric official jackpot odds, so their results are labelled model calculations. A mismatch is ODDS_MISMATCH_REVIEW_REQUIRED, not a reason to change the registry automatically.

Review dates and scope

A review date records the comparison performed on that date, not a promise of live freshness. Prices may include or exclude fees and optional products; timezones are explicitly scoped or left unknown. Overall odds can be rounded and are kept distinct from an exact jackpot denominator. Taxes and claim deadlines are not filled from ordinary game marketing.

Manual change detection

A local manual source checker can compare selected normalized evidence, availability and HTTP metadata against an explicitly accepted review baseline. Dynamic jackpot amounts, advertising and navigation are outside those selected excerpts. Missing excerpts or changed fingerprints require review. The tool does not parse guesses into new facts or modify published registry/pages, and no background scheduler is installed.

Random generation

Random mode uses browser Web Crypto rejection sampling when available, with a pseudorandom fallback for older environments. Balanced and personal-number modes constrain the sample and are not uniform across all combinations. No mode predicts results or improves a particular valid ticket's chance.

Responsible use and maintenance

English is the semantic source for reviewed educational content; supported language editions preserve its meaning and limits. Old unreviewed translations remain inactive audit material. We review factual discrepancies against authoritative evidence, record original and reviewed values, and update affected data, explanations and translations together. No public inquiry or correction-submission channel is provided.

Lottery Explorer · Exact combination calculator · Sources · Editorial Policy