The pots worth watching, and the warning that belongs beside them
Eight jackpots whose counter stands above the average of everything that same game has already paid, read against the interval since each level last fell. The ratio is arithmetic on two published figures, and it is not a recommendation to play any of them.
Ad disclosure: some casino links are affiliate links; we may earn a commission at no extra cost to you. It never affects the order of our list or what a page says. 18+.
Two houses whose catalogues still carry these titles
Position here is decided by the register we could read for each house, and never by a commission. Both buttons are tracked affiliate links, and the fee they pay us plays no part in the order you see. The cards show a regulator and a list of pools.
- 1
Grand Hotel Casino
Malta Gaming Authority · Kahnawake
Jackpots: Mega Moolah, WowPot!
- Sterling accounts only, on the Apricot platform with the Games Global library
- In the Malta register and on the Kahnawake list of permit holders
- One of the Casino Rewards addresses feeding the global Mega Moolah pool
Visit Grand Hotel Casino18+ · Sponsored link · Terms on the operator's site
- 2
Villento Casino
Malta Gaming Authority · Kahnawake
Jackpots: Mega Moolah, WowPot!
- Sterling only, with the Games Global library on the Apricot platform
- The Malta licence and the Kahnawake permit were both in the lists when we read them
- Trading since 2006 on the same global pools
Visit Villento Casino18+ · Sponsored link · Terms on the operator's site
The pots standing above their own history, and why that is not a tip
There is one comparison in this subject that can be made honestly, and it is not the one every site makes. Instead of putting a "counter" (the live figure a casino shows for a pot) next to another counter, we put it next to the average of everything that same game has already paid, and the ratio between the two says whether the pot in front of you is large by the standards of the game itself.
Eight jackpots stood above that line when we read the trackers (18 October 2026), and the spread between them is enormous, because one of them is 22 per cent above its own average and another 990 per cent above it (a factor of eleven). Both figures come from the same two published numbers (a counter and an average); neither of them forecasts anything at all.
What follows is the mechanism of a "progressive" (a pot fed by a share of every stake, which resets to a "seed" when somebody takes it) seen from the outside, and it is worth understanding for its own sake. It is not a list of games to go and play, and the box below gives the three documented reasons why.
Three documented reasons say that a pot above its own average is not a bargain. The "break-even" point cannot be calculated without the probability of the jackpot, and no supplier publishes that probability for any of these games. The base return of these titles runs from 89.37 to 91.62 per cent, so the deficit a player has to recover is enormous, and arriving at a profit with "optimal play", in the rare cases where it can be worked out at all, "may take tens of thousands of plays".
The two cases that make the point on their own
LotsaLoot is the extreme of the set, since its counter stands 990 per cent above the average of its own past wins and the level has not fallen for three years and twenty-three weeks. The game is still inside the Games Global portfolio declared for 2026 (the supplier's own list), so we cannot describe it as withdrawn or "retired".
It is "technically in the catalogue" and "not played", which is the distinction almost every comparison table in this trade manages to miss; the two figures above are what that condition looks like from outside.
King Cashalot is the more remarkable of the pair, because its pot has passed the largest prize the game has ever paid (by 45 per cent) and nobody has won it for nine years and eight weeks, against an "average interval" (the mean gap between two wins of the same level) of about two and a half years.
The "base game" underneath returns 91 per cent, and that is where the money has actually gone for those nine years; the jackpot has had nothing to do with it.
Read together, those two rows describe a condition rather than an opportunity. A progressive grows out of the stakes placed on it, so a "pot" that nobody is feeding grows slowly, and the "drought" (the time elapsed since a level last fell) is at once the symptom of an abandoned game and the cause of the number that makes it look attractive.
The rest of the set says the same thing more quietly, and we would rather write it out than summarise it: Jackpot Deuces sits 394 per cent above its own average after three years and eleven weeks, SupaJax 205 per cent after three years and seven weeks, and Fruit Fiesta 337 per cent after four years and fourteen weeks.
Two of the eight are different in kind. Roulette Royale is only 38 per cent above its average, which looks modest until you notice that the level has gone eleven years and eighteen weeks unwon (and that its counter, too, has passed its own largest prize ever).
The Arabian Nights pool on the NetEnt side is 22 per cent above its average (after three years and forty-one weeks), and it is the only entry here belonging to a supplier that still sells new jackpot titles, which makes its silence the hardest to explain.
Why the games with the best theory are the games nobody plays
These are, in principle, the pots where a player gets most for his stake, and that is exactly the paradox of this page. The theory says that a large pot improves the "expected value" (the average return of a bet repeated indefinitely); the practice says that the pot is large because the game has been left behind, and almost everything else about these titles is worse than the modern alternative.
Start with the base game (the reels before any "jackpot round" opens), which is where the money actually goes. LotsaLoot returns 89.95 per cent, Tunzamunni 90 per cent, Fruit Fiesta 90.09 and King Cashalot 91, against 96 per cent or better on an ordinary slot of the current generation (and 89.37 per cent on Major Millions, the lowest figure we read anywhere).
A player chasing one of these counters is therefore accepting five or six points of additional "house edge" (the share of each stake the operator keeps over time) on every spin that is not the winning one, and the winning one is the spin that has not arrived for years.
Then there is the "qualifying stake" (the bet below which the jackpot cannot be won at all), which is a trap peculiar to this generation of games, and we have found no modern network that still works in that way. Fruit Fiesta demands that every one of its fifteen lines be covered, King Cashalot that its five king wilds land on line nine, and Treasure Nile that the spin cost £4.50 before the jackpot is reachable.
The "contribution" (the share of each stake that feeds the pot) is deducted whether a stake qualifies or not, so somebody under the threshold on one of those titles is funding a prize beyond his reach, which is the oldest unfairness in this business and is still running today!
These "counters" also move slowly in both directions, which is the last thing worth saying about them. A pot that has taken three years to climb 990 per cent above its game's average will not double again next month, so none of these figures is a reason for anybody to hurry!
The address at the top of this page, and why it ends the way it does
This page is served from an address ending in .htm rather than .html, and that is deliberate rather than an oversight. The address has been declared in this site's own sitemap since the page was last rebuilt (November 2014), and an address that search engines and other sites have had for that long is worth more than a tidy one, so we kept it exactly as it stood and wrote new material behind it.
The duplicate ending in .html, which was never in the sitemap at all, now sends readers here instead of competing with this page for the same words. That is the whole of the explanation, and it is the sort of thing a reader is entitled to be told rather than left to work out.
What would actually make one of these a good bet
Three things would have to be known, and only one of them is. The probability of the top level is the first, and no supplier of any size has ever published it for a game in this trade; the full paytable of the title is the second, and it is published for almost none of them, while the stake that qualifies is the third, and that one at least you can read for yourself in the game rules.
Without the first two, the "break-even" counter cannot be worked out, and the term is not a figure of speech here: it is the exact value above which a progressive becomes a "positive expectation" bet, with a return above 100 per cent, and below which the house holds an edge over everybody playing it.
So we publish the ratio and not a recommendation, and we would rather lose the click than pretend the arithmetic is available. The eight counters and the averages behind them are tabulated on what a jackpot is worth, the frequencies behind these intervals sit on how often they fall, and the sums these games paid while they were alive are in the records, which is a better guide than any counter.
- The records
€38.4M
The largest ever paid
Every record worth the name, in the currency it was paid in, with the date and how solid the source is. Mixing currencies is how the other rankings go wrong.
Read more - Where to play
21
The casinos
Every house in our list, with the regulator we read for it and the jackpot networks its catalogue carries.
Read more
Questions we are asked about this list
Questions
Does a jackpot above its average pay out sooner?
No. A progressive has no memory of how long it has gone unwon, and the interval in the table is a record of the past rather than a countdown. What a long interval does tell you is how little the game is being played, since a pot grows only out of the stakes placed on it.
Which jackpot is furthest above its own average?
LotsaLoot, at 990 per cent above the mean of its own past wins, which is about eleven times that figure. King Cashalot is the stranger case, because its pot has gone 45 per cent beyond the largest prize the game has ever paid and has been unwon for nine years and eight weeks.
Can a big pot make a slot worth playing?
Only when the probability of the jackpot is known, and no major supplier publishes it for any of these games. Without that figure the break-even value cannot be worked out at all, and the base return of these titles, between 89 and 92 per cent, is a deficit that a theoretical edge would have to overcome first.
Why does this page end in .htm instead of .html?
Because the address has been declared in this site's own sitemap since the page was last rebuilt in November 2014, and an address that old is worth keeping. The duplicate ending in .html now sends readers to this page instead of competing with it.
Where do the counters and averages come from?
From independent jackpot trackers, read on 18 October 2026 and dated in the table itself. The suppliers publish neither the counter nor the history of past wins, so these are the best figures available, and we label them as what they are rather than as supplier data.
