New concepts, datasets, and "try this at home" experiments that extend the methods in Code the Jackpot. Same goal every time: more structure, fewer random guesses — purely as study, never a promise of winnings.
If a fair draw fills the space evenly, can the shape of a ticket — its sum, spread, clustering — still carve away combos you'll never need? We measure it in Joker and Eurojackpot, and walk into two very convincing mirages before the honest answer. Get the script.
The time-domain says the draw has no memory. But a real machine can still be crooked in space — pairing certain numbers. We audit every pair and triple in Joker and Eurojackpot against a fair-null. The verdict is clean. Get the script.
We try to predict which of the x1..x20 overlap columns will be zero on the next draw — with gap analysis, Markov chains, the "diagonal" trick, a fair-null test, and even the lab's own scoring functions. Across Joker and Eurojackpot the answer is the same, and it's clean. Grab the script.
A popular idea is to filter Tzoker combinations by keeping the historically over-represented numbers at each sorted position. It shows an out-of-sample “lift” above 1 — but when we test it against simulated fair lotteries, the edge evaporates. Here’s the honest experiment.
The draw is random but players are not. How to prove conscious selection from public winner counts, flag "popular" combinations, keep each rule honest, and use it to lose less to prize-sharing — with full, game-agnostic Python.
We rank values across st1–st5 using gap percentiles, split the ranked list into buckets, and backtest where the next draw lands.
A spacing-based method for rich-state features (like sum and OSLCS): build a contrarian candidate corridor and a forbidden “avoid” list using follower-exclusion + spacing stats. Download full docx file.
A practical approach as to test the covering with different mappings. Get the script. Download here.
We scan 5–6 x-columns and 0/1 patterns, filter tot_df + hist_df in lockstep, then rank candidates using heavy-lift (ratio_of_ratios), delay-percent due-ness, plus a support penalty so tiny subsets don’t fool us.
A compact set of 5-number lines built so every possible winning draw shares at least 4 numbers with at least one line in the set.