How often these jackpots actually fall
No major supplier has ever published the probability of its top level, so we have replaced it with the one quantity that can be checked: how long each level has really gone between payments.
The number that does not exist, and the one that does
Ask how likely a top jackpot is and the trade will answer with a percentage, which is remarkable in itself. Not one of the four largest suppliers of these games (NetEnt, Games Global, Playtech and Relax Gaming) has ever published such a figure for any product it sells.
We looked for it in the places where it would have to be if it existed: the official game pages of NetEnt, the press releases of Games Global, the product pages of Relax Gaming, the technical notes of Blueprint and the jackpot pages of Playtech (all of them read again on 18 October 2026).
The absence is not an accident of our search, and it is worth stating plainly once before the table arrives. What those pages give is the return to player and the "hit frequency" of the base game (a measure of how often a win of any size lands at all), which is a wholly different quantity and which we have set beside the returns on reading the return of a jackpot slot.
One of the databases we read (a slot data site rather than an affiliate) puts the position more usefully than the suppliers manage to, observing that there is "no defensible universal jackpot-odds range" at all, which is not scepticism on its part but a statement about where the information actually lives: inside a "mathematics certificate" that the operator holds and the player never sees.
So we have replaced the missing percentage with a quantity that can be read by anybody, checked by anybody and still checked in a year's time, and which has the further merit of being a measurement rather than an estimate: the interval each level has actually taken between one payment and the next, which the trackers call its "average interval".
It is a humbler number than a probability, and it answers a slightly different question; it is also the only one on the table.
How long each level has really taken
| Level | Average interval between wins | Which means |
|---|---|---|
| Age of the Gods Extra Power | 2 hours 38 minutes | several times a day |
| Mega Moolah Major | 1 day 11 hours | most days |
| Cash Splash | 1 day 21 hours | most days |
| Everybody's Jackpot (Playtech) | 2 weeks 2 days | twice a month |
| Age of the Gods Ultimate Power | 7 weeks | about seven times a year |
| Mega Moolah Mega | 9 weeks 4 days | about five and a half times a year |
| Major Millions | 9 weeks 5 days | about five times a year |
| Mega Fortune Mega | 14 weeks 5 days | about three and a half times a year |
| Joker Millions | 34 weeks 2 days | about once and a half a year |
| WowPot Mega | 34 weeks 5 days | about once and a half a year |
| Hall of Gods Mega | 39 weeks 2 days | about once and a third a year |
| Arabian Nights, Betsson pool | 1 year 26 weeks | about every eighteen months |
| King Cashalot | 2 years 21 weeks | about every two and a half years |
| Thai Temple (Playtech) | 11 years 7 weeks | once a decade |
A spread of four orders of magnitude
The table opens at 2 hours 38 minutes and closes at 11 years 7 weeks, which is a range of roughly four orders of magnitude inside a single product category, and in our view it is the most useful thing a reader can be told before deciding which pot to chase. The ordering is by interval and by nothing else.
At the top of it the "Extra Power" level of the Playtech pool turns over several times a day, and Age of the Gods carries two further levels above that one ("Super Power" at 1 day 9 hours, "Ultimate Power" at 7 weeks, with "Power" below them falling too often for the trackers to record it at all). A reader who wants to watch a jackpot round resolve inside his own session is looking at a game of that family, where the amounts are correspondingly small.
In the middle sit the levels the trade advertises, and they are slower by three orders of magnitude. The Mega Moolah "Mega" has averaged 9 weeks 4 days between wins (its "Major", one tier down, averages 1 day 11 hours), which works out at something near five and a half payments a year across every casino on the network at once, while Major Millions is almost identical at 9 weeks 5 days.
Below them the arithmetic turns against the player quickly: Mega Fortune at 14 weeks 5 days (three or four payments in a year), WowPot! at 34 weeks 5 days (one or two), Hall of Gods at 39 weeks 2 days (the slowest of the NetEnt pools, and the one with the largest average prize behind it).
At the bottom of the table are two games which are no longer "networks" in any useful sense of the word. King Cashalot has averaged 2 years 21 weeks between wins (and has now gone 9 years 8 weeks without one), Thai Temple 11 years 7 weeks, and a pool which pays once a decade is a counter with a number on it rather than a product anybody plays. Playtech has not withdrawn either of them, and the trackers still list both as live, which is the gap between a catalogue and a market.
The regulator behind the pot
Malta Gaming Authority · Kahnawake
18+ · Sponsored link · Terms on the operator's site
What the suppliers do count, when they count anything
Suppliers publish no probabilities, but several of them publish "volumes", and a volume is a frequency seen from the other end. The figures are promotional in purpose and verifiable in form, which is an unusual combination in this trade and worth using while it lasts.
Games Global said that its progressive network paid a jackpot every 12 seconds on average across 2024, and that WowPot! alone (every level of it, and not merely the top one) paid one every 15 seconds in 2023. In May 2026 the same company reported more than 77,000 jackpot wins inside a single month, for a total above £11 million, and both of those figures are its own.
Playtech published the steadiest series of all, averaging 16,037 jackpots a month over 2018 and 2019 (€187,629,964 across the twelve months), which for a network whose fastest level turns over in under three hours is exactly what you would expect.
Relax Gaming is the most precise of the four, because it counted by level: 457 "Major" wins and 25 "Mega" wins between May 2022 and December 2025, with 28 Megas by March 2026 (the counts are the supplier's own). Divide the first interval and a Dream Drop Mega arrives about every 52 days, and a Major about every 2.9 days (both of them averaged across three years and seven months).
Those numbers describe a "network" and not a player, and the distinction is the whole point of printing them. A pot which pays every 12 seconds somewhere in the world may still be a pot that nobody at your own casino will ever see fall!
Two numbers that ought to be thrown away
The first is the claim that Jackpot King pays its top level about once in every 50 million spins, which circulates as an "industry estimate" (the only attribution anybody offers for it). Blueprint publishes seeds and ceilings for its three pots, and nothing whatever about how often the largest of them lands. It is quoted widely, it is attributed to nobody, and it appears in no document of Blueprint's or of any operator carrying the network; we have therefore declined to print it as a figure, and we mention it only so that a reader who meets it knows what it is.
The second is more interesting, because it comes with decimals and an air of precision. A comparison site publishes probabilities for the three levels of Mega Fortune (they are called "Rapid", "Major" and "Mega"), giving the Mega as 0.00576 per cent, the Major as 0.0832 and the Rapid as 8.32, all of them "at maximum stake", and attributing them to no source whatever.
Test the first of those against the measurement and it falls apart. A chance of 0.00576 per cent is about one spin in 17,361 (the division is ours, and the only one the test needs). On a network carried by dozens of operators the Mega would then have to fall several times a day; the observed interval between its wins is 14 weeks 5 days, which leaves a discrepancy of some three orders of magnitude.
That is the kind of number which survives for years because nobody checks it, and checking it costs a division. We print it here with the arithmetic attached, and we print it nowhere else on the site.
A jackpot does not become "due". Each trigger is independent of every trigger before it, so a level which has gone four times its average interval without falling is no likelier to fall on the next spin than it was on the first; what a long drought really measures is how few people are still playing, which is a point we make with figures on the first generation of these games.
What a long drought does mean
If a drought carries no information about the next spin, it carries a great deal about the state of the game, and reading it that way turns the table above into something more useful than a curiosity.
Every NetEnt top level is currently sitting at an interval since its last win between five and twelve times its own historical average (Hall of Gods has gone 3 years 26 weeks against an average of 39 weeks 2 days). At that scale the explanation is not bad luck but "liquidity" which has drained out of the network. The same reading applies to Joker Millions, whose distribution has always been confined to two operator groups (Cherry Casino and Betsson), and whose own pool has been still for years.
The practical use of a "drought", then, is diagnostic rather than predictive, and a reader who takes the interval as a diagnosis instead of a forecast will read the table correctly.
A level running near its average interval is a level with players behind it; a level at eight times its average is one which has been quietly abandoned (King Cashalot stands at nearly four times its own, and is the healthier of the two readings). The size of the counter is no comfort to you at all, and the reason it is none is worked through in detail on when a pot is worth chasing.
We would rather hand you a measurement with a known meaning than a percentage with an unknown provenance, and that preference is the whole of the editorial argument of this page!
Three casinos on the networks measured here
All SlotsAlderney AGCC Category 1, Baytree (Alderney) Ltd · Kahnawake 00892Visit
Euro PalaceAlderney AGCC Category 1, Baytree (Alderney) Ltd · Kahnawake, Baytree Interactive LtdVisit
Amazon SlotsGambling Commission · Alderney AGCC · AGCO and iGaming Ontario, through Six Gaming LtdVisit
18+ · Sponsored link · Terms on the operator's site
Questions
What are the odds of winning a progressive jackpot?
Nobody publishes them. The official pages of NetEnt, Games Global, Playtech and Relax Gaming give the return to player and the hit frequency of the base game, which is a different quantity altogether. The figures circulating on comparison sites are almost all unattributed, and one of the best known of them is arithmetically incompatible with how often the level has really fallen.
How often does the Mega Moolah top level pay?
The trackers put the average gap between wins on the Mega at 9 weeks 4 days, and on the Major one tier below it at 1 day 11 hours. The top level of WowPot! is a great deal slower, averaging 34 weeks 5 days, so a reader weighing the two is weighing five payments in a year against one or two.
Is Jackpot King really one win in 50 million spins?
That figure is an unattributed industry estimate and we do not publish it as a number. Blueprint has never issued a probability for the network, and no operator carrying it has either. The measurable alternative is the interval between wins on the trackers, which is what our table gives.
Does a jackpot become due when it has gone a long time without falling?
No. Each trigger is independent of the last, so a long drought changes nothing about the next spin. What a long drought does change is the state of the game: King Cashalot has gone 9 years 8 weeks, which is a measure of how few people still play it rather than of pressure building up.
