Every game we can price, side by side.

A line is the price we hold for that game, listed with its source on each game’s own page. Jackpot odds are exact: C(N,k) times the bonus deck’s combinations where the game has one. Jackpot now is the headline last advertised — for annuity games that is the sum of thirty years of payments, not a cheque.

Any prize comes from the last published prize table: the sum of the per-line probabilities of every tier whose rules we can read. Back per line is the whole calculation — the jackpot after cash value, tax and expected sharing, plus what the last draw actually paid out below the jackpot, divided by the lines that table implies were in play. The model and all its parameters are on the edge page.

Actually paid back is prizes actually paid over what the lines actually cost, across the whole current era — the measured twin of the modelled column beside it. Lines a draw is the median number of lines a draw sells, read from its prize table rather than announced. Next draw works the same calculation against the jackpot advertised for the next draw, which is the only figure here a reader could act on. It is below the price on every game, every time. The last three are empty until a game’s prize tables have been worked through. An empty cell means not measured, never zero.

9 of the 13 games we hold a price for are left out, because a row we cannot ground is worse than no row: a format inferred from the draws rather than verified by hand (its odds would be a lower bound dressed as a figure), a jackpot smaller than a prize the same draw paid — which is how a feed reporting a side game’s prize gives itself away — or a prize table too thin to say how many lines were in play.