Question Are slot RTP values constant across a particular slot, or can casinos adjust them?

[mathsboy1975]

I am curious about the RTP values for slots. I mainly play microgaming slots and I was wondering if the RTP values are hard coded by the company that create them or whether they are adjustable by the casinos that use them. If the latter is true, are the casinos obliged to document the RTP and if so where can you find the information on it. For example, if I play Thunderstruck at 3 different casinos, will they ALL have the same RTP?

As a related question, I have seen a number of posts where members refer to the high variance of a particular slot. Is this based on user experience or is this also documented anywhere - if so where could I find it??

Thanks in advance for any assistance.

 

[ChopleyIOM]

MG are on record as stating that their slots pay out identically everywhere, my own personal experience over the years is that this is true.

Variance is very rarely documented by casinos/software providers, it's basically down to the players to work it out for themselves - which is a bit of a sorry state of affairs IMO.

 

[philderby]

Thats an interesting question. Ive always presumed that all slots payed out at the same percentage. In fact why should there be varying degrees of pay out percentage? Hope others post on this thread because Id like to learn more about this subject.

 

[mattsgame]

Thats an interesting question. Ive always presumed that all slots payed out at the same percentage. In fact why should there be varying degrees of pay out percentage? Hope others post on this thread because Id like to learn more about this subject.

Many factors can come into a slots RTP. You will find slots that are linked to jackpots to have a slightly lower RTP to there non jackpot counter parts as a % is distributed across the Jackpot pool as an example.

 

[mathsboy1975]

slot layout

On a related matter, I was wondering if the suppliers of these slots release the reel layouts or if they are kept secret. With this knowledge, one could compute theoretical probabilities of each win outcome. To do so would require the following assumptions:

1) The reels are in constant and representative of a mechanical reel, therefore do not consist of randomly selected symbols on each win line.
2) There is no weighting of particular symbols on the reels.

Are either of these assumptions valid and is it possible to get hold of the reel layouts anywhere. In theory if you played enough and if assumption 1) above holds, then you could work out the physical layout of the reels although you would be unable to detect if any weightings were present.

If nobody knows this I may attempt to contact the suppliers directly, as it does not seem unreasonable for players to actually know the probabilities of each reel win event. For instance, the national lottery in the UK , we know roughly that there is a 14,000,000 to 1 chance of getting all 6 numbers matched, and yet these poor payout odds do not stop people from buying the tickets. I would imagine most slots players would play regardless so I cannot see why slot providers would WANT to keep this knowledge to themselves - but I just do not know. Does anyone else??

 

[ChopleyIOM]

On a related matter, I was wondering if the suppliers of these slots release the reel layouts or if they are kept secret. With this knowledge, one could compute theoretical probabilities of each win outcome. To do so would require the following assumptions:

1) The reels are in constant and representative of a mechanical reel.
2) There is no weighting of particular symbols on the reels.

1) They're not.
2) There is.

That's where slots even use what we'd understand to be 'reels' at all, and my personal opinion is that the 'reels' on modern video slots are simply an illusion (some providers, such as WMS, don't even pretend otherwise).

I don't see it as a problem though, as long as the RNG behind the scenes is truly random and fair.

(My favourite online slots are WMS, which have 'reels' that are clearly nothing other than fancy graphics and are in no way intended to present any sort of representation of the result selection process.)

 

[mathsboy1975]

1) They're not.
2) There is.

That's where slots even use what we'd understand to be 'reels' at all, and my personal opinion is that the 'reels' on modern video slots are simply an illusion (some providers, such as WMS, don't even pretend otherwise).

I don't see it as a problem though, as long as the RNG behind the scenes is truly random and fair.

(My favourite online slots are WMS, which have 'reels' that are clearly nothing other than fancy graphics and are in no way intended to present any sort of representation of the result selection process.)

That is very helpful to know. So basically then there is no way that we as players can ever attempt to calculate particular win event probabilities. However, there are particular patterns that do seem to repeat a great deal in particular games - like the same combinations of symbol across a reel on win line 1,2,3 that I have noticed a lot that give the illusion that there is a constant reel structure, but why would they do that if it is not the case.

I agree about the RNG and I am not suggesting it is a problem, I am just interested in calculating probabilities just to compare the value aspect of particular games. For example if to £3 pound stake slot A pays £100 for a 5 of a kind on symbol B but the actual probability of hitting that 5 of a kind is only 1 in 200 then you are getting poor odds. I know the RTP figure is the key thing for value of a slot, and probably the equivalent of a bookies overround statistic. I am just interested in finding a way to calculate these probabilities, but seems that I am probably out of luck.

 

[KasinoKing]

On a related matter, I was wondering if the suppliers of these slots release the reel layouts or if they are kept secret. With this knowledge, one could compute theoretical probabilities of each win outcome. To do so would require the following assumptions:

1) The reels are in constant and representative of a mechanical reel, therefore do not consist of randomly selected symbols on each win line.
2) There is no weighting of particular symbols on the reels.

Are either of these assumptions valid and is it possible to get hold of the reel layouts anywhere. In theory if you played enough and if assumption 1) above holds, then you could work out the physical layout of the reels although you would be unable to detect if any weightings were present.

I partly disagree with Chopley; IMHO SOME slots DO have fixed reel strips which can be analysed to work out the chances of hitting any particular win and the RTP of the slot.

A lot of work was done on this by CM forum members several years back, and I created a website to pass this information on to players.
However, there was not a huge amount of interest and therefore I could not justify the time & commitment to carry on producing these reel-strip layouts. (Each one is at least 2 days work)
You can view the reel-strips for 26 Microgaming slots (including Thunderstruck) via links from the table you can access by clicking the "Reel Layouts and Jackpot Odds" link in my signature below.
I do have the data for a few more, but never got around to doing the graphics & probabilities.

I think you have absolutely ZERO chance of getting the reel-strip layouts from any software supplier.
(But if you do - I'd like to see them too please!)

KK

 

[ChopleyIOM]

I partly disagree with Chopley; IMHO SOME slots DO have fixed reel strips which can be analysed to work out the chances of hitting any particular win and the RTP of the slot.

Some older slots yes, I agree with you.

Anything more current I don't think that's the case, the most recent MG one I can think of that was basically proven is Battlestar Galactica which was obviously weighting some symbols in a bonus round.

 

[ChopleyIOM]

That is very helpful to know. So basically then there is no way that we as players can ever attempt to calculate particular win event probabilities. However, there are particular patterns that do seem to repeat a great deal in particular games - like the same combinations of symbol across a reel on win line 1,2,3 that I have noticed a lot that give the illusion that there is a constant reel structure, but why would they do that if it is not the case.

I agree about the RNG and I am not suggesting it is a problem, I am just interested in calculating probabilities just to compare the value aspect of particular games. For example if to £3 pound stake slot A pays £100 for a 5 of a kind on symbol B but the actual probability of hitting that 5 of a kind is only 1 in 200 then you are getting poor odds. I know the RTP figure is the key thing for value of a slot, and probably the equivalent of a bookies overround statistic. I am just interested in finding a way to calculate these probabilities, but seems that I am probably out of luck.

You might want to take a look at this page, as it's unique in the entire world of online casinos AFAIK. They quantify RTP, variance, standard deviations, chance of each spin to be a winner, chance of hitting a bonus round, and so on.

Link Removed ( Old/Invalid)

They even give you the exact odds of hitting the progressive jackpots on their progressive slots.

You can find Galewind's games at the Pinnacle Casino (and a full demo casino at their own website). Personally I find the slots a bit 'dry' for my tastes but in terms of RTP (97.8% RTP with 0.3% cashback on the slots) and transparency they are head and shoulders above every other casino software provider in the world IMO. (Only 3Dice come close in the transparency stakes.)

Galewind's CEO is also active on these forums, goes by the name of binary128, a thoroughly decent chap.

 

[mathsboy1975]

I partly disagree with Chopley; IMHO SOME slots DO have fixed reel strips which can be analysed to work out the chances of hitting any particular win and the RTP of the slot.

A lot of work was done on this by CM forum members several years back, and I created a website to pass this information on to players.
However, there was not a huge amount of interest and therefore I could not justify the time & commitment to carry on producing these reel-strip layouts. (Each one is at least 2 days work)
You can view the reel-strips for 26 Microgaming slots (including Thunderstruck) via links from the table you can access by clicking the "Reel Layouts and Jackpot Odds" link in my signature below.
I do have the data for a few more, but never got around to doing the graphics & probabilities.

I think you have absolutely ZERO chance of getting the reel-strip layouts from any software supplier.
(But if you do - I'd like to see them too please!)

KK

Thanks very much for that - this is exactly the kind of thing I would like to do for myself. I am very interested in how you did this - did you just play a lot and piece things together gradually? I am amazed that there was not a lot of interest in this as it gives players some insight into the kind of value for money they are getting. To be honest in that link you provided the only slot I recognized was Thunderstruck. I tend to stick to the same 3 slots at the moment - Battlestar Galactica, Immortal Romance and The Dark Knight. I would be interested in getting the reel layout for these as they APPEAR to be reel based. Perhaps this is why I tend to think that the slots are reel based and perhaps I should branch out a bit to get a wider feel of slots.

Would you suggest the best way to do this is to just play the games and try to piece together the reels the hard way?

Thanks very much for your reply anyway.

 

[KasinoKing]

Thanks very much for that - this is exactly the kind of thing I would like to do for myself. I am very interested in how you did this - did you just play a lot and piece things together gradually? I am amazed that there was not a lot of interest in this as it gives players some insight into the kind of value for money they are getting. To be honest in that link you provided the only slot I recognized was Thunderstruck. I tend to stick to the same 3 slots at the moment - Battlestar Galactica, Immortal Romance and The Dark Knight. I would be interested in getting the reel layout for these as they APPEAR to be reel based. Perhaps this is why I tend to think that the slots are reel based and perhaps I should branch out a bit to get a wider feel of slots.

Would you suggest the best way to do this is to just play the games and try to piece together the reels the hard way?

Thanks very much for your reply anyway.

Yes, I did Tomb Raider, Thunderstruck and Cashapillar by playing loads of spins, writing down the result of each spin, then piecing them together like a jigsaw puzzle!
But this was INCREDIBLY time consuming: Cashapillar took over a week!

Then some bright spark came up with a way of extracting the data from the feed used by the casino when loading up flash games in your browser, so all the rest of the reel layouts on SlotBeaters.com were based on those findings.
This is the thread where it all took off: https://www.casinomeister.com/forums/threads/cracking-the-slot-reels-here-you-go.19302/

Regarding those 3 newer slots, personally I believe the base games are also done with "true reel strips" (meaning they have a fixed sequence of symbols and the chance of each symbol landing on the centre win-line is equal). So if you plotted them out you could work out the chances of hitting the bigger combinations and 3 or more scatters in normal spins. But, and this is a big BUT - you couldn't work out the T-RTP of the slots due to the unknown formula's used in the bonus rounds.
e.g. On Battlestar the "timer feature" in the middle phase is obviously not created by a fixed symbol on the strip just happening to fall in view, but I think it is triggered randomly and then the reels are generated by a separate "sub-program" if you like to product the pre-determined result.
And on IR & TS2 the "spreading wild" feature gives an unknown number of wilds, and is obviously weighted towards certain reels over others.

So in essence, you can only work out the T-RTP of slots which have a fixed known bonus round, such as 15 spins @ x3. And then, only if you know the reel set used in the free-spins (which on some slots is different from the base game (e.g. Moonshine)).

The reason that this attracts so little interest is "so what"? Even I have the same attitude myself these days. I don't need to know the T-RPT or chances of hitting a feature before I start playing a new slot - I just play it and if I enjoy it and/or get some nice wins on it, chances are I'll play it again. If I think it's shite, I wont.

KK

 

[mathsboy1975]

Yes, I did Tomb Raider, Thunderstruck and Cashapillar by playing loads of spins, writing down the result of each spin, then piecing them together like a jigsaw puzzle!
But this was INCREDIBLY time consuming: Cashapillar took over a week!

Then some bright spark came up with a way of extracting the data from the feed used by the casino when loading up flash games in your browser, so all the rest of the reel layouts on SlotBeaters.com were based on those findings.
This is the thread where it all took off: https://www.casinomeister.com/forums/threads/cracking-the-slot-reels-here-you-go.19302/

Regarding those 3 newer slots, personally I believe the base games are also done with "true reel strips" (meaning they have a fixed sequence of symbols and the chance of each symbol landing on the centre win-line is equal). So if you plotted them out you could work out the chances of hitting the bigger combinations and 3 or more scatters in normal spins. But, and this is a big BUT - you couldn't work out the T-RTP of the slots due to the unknown formula's used in the bonus rounds.
e.g. On Battlestar the "timer feature" in the middle phase is obviously not created by a fixed symbol on the strip just happening to fall in view, but I think it is triggered randomly and then the reels are generated by a separate "sub-program" if you like to product the pre-determined result.
And on IR & TS2 the "spreading wild" feature gives an unknown number of wilds, and is obviously weighted towards certain reels over others.

So in essence, you can only work out the T-RTP of slots which have a fixed known bonus round, such as 15 spins @ x3. And then, only if you know the reel set used in the free-spins (which on some slots is different from the base game (e.g. Moonshine)).

The reason that this attracts so little interest is "so what"? Even I have the same attitude myself these days. I don't need to know the T-RPT or chances of hitting a feature before I start playing a new slot - I just play it and if I enjoy it and/or get some nice wins on it, chances are I'll play it again. If I think it's shite, I wont.

KK

Perfect - that is exactly the kind of info I was looking for. I spent about an hour yesterday cracking the first reel on IR then went to the second one. Its designed quite sneakily with many 3 symbol combos that comprise a visible part of the reel having 2 common symbols so that there are a number of different permutations. In fact its not even possible to discount the possibility of multiple occurrences of the same permutation across the same sequence.

If there is no interest then there is no interest - I can kind of understand that. I will pursue this via software and write a TCP traffic sniffer and see what I can come up with. I would have thought that the http would be encrypted however, but I will have a got at it and see what I can find.

Thanks for the pointer to that other thread anyway it will probably be very helpful.

Cheers

 

[KasinoKing]

Perfect - that is exactly the kind of info I was looking for. I spent about an hour yesterday cracking the first reel on IR then went to the second one. Its designed quite sneakily with many 3 symbol combos that comprise a visible part of the reel having 2 common symbols so that there are a number of different permutations. In fact its not even possible to discount the possibility of multiple occurrences of the same permutation across the same sequence.

Yeah we had that problem with Thunderstruck I think it was (or TR1) - it has the same sequence of 3 symbols on one of the reels - took a while to suss it. Couldn't even work it out by watching the reels spin like on the newer slots, cos on those old ones all you see is a blur!

KK

 

[mathsboy1975]

Yeah we had that problem with Thunderstruck I think it was (or TR1) - it has the same sequence of 3 symbols on one of the reels - took a while to suss it. Couldn't even work it out by watching the reels spin like on the newer slots, cos on those old ones all you see is a blur!

KK

Actually I have decided to give up on this approach due to this problem but also due to the fact that I am sure that the weighting on the second reel is not constant. I must have sat there recording about 500 spins and the distribution of particular reel patterns was disproportionately large for a group of them. Having said that 500 spins is not much of a sample size but I would be willing to bet that the distrubution was not uniform. Reading around on other threads it appears that these slots do not even work in the way I had envisaged. I thought that the reel position would be generated randomly with a RNG, and that the distribution and amount of symbol types would be chosen in such a way that the probability and payouts for particular combinations resulted in the machines RTP. From opinions in other threads it appears that this is not actually the way in which it works.

Still this has been an interesting topic for me. I will still have a go at the HTTP packet sniffing approach though and see what happens.

In your opinion KK do you think that these reels are designed intentionally to contain symbol sequences that make piecing the reel together difficult or that it is just a coincidence? It did occur to me that you could easily design a reel that was unable to be cracked just by using repeating sub patterns within the main reel. That way you could not even be sure of the frequency of the symbols as you would never know how many repeated sub patterns there were.

From the other threads, it seems that people have opinions as to how slots work internally but there are no factual documents or links. This makes it even more interesting to me now and makes me want to get a job with one of these companies even more just to see how they actually do it.

 

[KasinoKing]

In your opinion KK do you think that these reels are designed intentionally to contain symbol sequences that make piecing the reel together difficult or that it is just a coincidence? It did occur to me that you could easily design a reel that was unable to be cracked just by using repeating sub patterns within the main reel. That way you could not even be sure of the frequency of the symbols as you would never know how many repeated sub patterns there were.

In my opinion, it is just coincidence.
If software suppliers REALLY wanted to hide how their slots work it would be pretty easy for them.
The fact that they show the "true reel-strips" during the spin and allow some operators to publish the RTPs indicates to me that they have nothing to hide.
I just wish they would really open up & publish this information publicly - I mean, what have they got to fear?

KK

 

[mathsboy1975]

I just wish they would really open up & publish this information publicly - I mean, what have they got to fear?

KK

I totally agree. I suppose it could be argued that if people knew the probabilities of each combination they could end up feeling cheated in some way. For example if the major symbol 5ok pays out a paltry 50-1 but only has a 0.001 probability of coming home then you would feel you are getting poor value. Of course it has to be offset by the implied wins that would occur over other symbols based on permutations of the reel output plus all the jackpot/feature stuff but it is hardly a good advertisement if taken in isolation. Of course I am just clutching at straws here but in that contex I can see why they would favour a single RTP figure that covers all aspects of the game. Still, releasing the reel layouts or even just the relative symbol frequencies would be harmless enough you would think...

 

[SoftSwiss]

Reputable software developers typically certify their software with independent test labs. RNG and RTP output audits are among the evaluations test labs perform in certification process.

 

[IanO]

I just wish they would really open up & publish this information publicly - I mean, what have they got to fear?
KK

FWIW, I said exactly this to them when down in their offices recently... to as many people as would listen to me. No promises were made, but most people did at least seem to see the merit in what I was suggesting. Who knows if my words will have the desired effect, or how quickly, but hey... I tried

 

[dunover]

I partly disagree with Chopley; IMHO SOME slots DO have fixed reel strips which can be analysed to work out the chances of hitting any particular win and the RTP of the slot.
A lot of work was done on this by CM forum members several years back, and I created a website to pass this information on to players.
However, there was not a huge amount of interest and therefore I could not justify the time & commitment to carry on producing these reel-strip layouts. (Each one is at least 2 days work)
You can view the reel-strips for 26 Microgaming slots (including Thunderstruck) via links from the table you can access by clicking the "Reel Layouts and Jackpot Odds" link in my signature below.
I do have the data for a few more, but never got around to doing the graphics & probabilities.

I think you have absolutely ZERO chance of getting the reel-strip layouts from any software supplier.
(But if you do - I'd like to see them too please!)

KK

Yes and no. It's my belief that reels strips (even on WMS) are ALL fixed and can all be analysed to work out the RTP and win probability on online slots. Before your good self and Chopley jump on me for a massive U-turn, I say this in conjunction with my view that reels are irrelevant; they are simply video graphics, but the video graphics ARE constructed to match up with the random win pool and RTP after the fact. So, although the reels are not relevant, they are accurate.
PS Having said that, I did point out once that on Hitman (MG) you can see the Hitman white-wild spinning round on reels 1 and 5, where you can't actually get it! So maybe this demonstrates that on some MG slots they haven't even bothered to make them accurate. As for Chopleys observation on Bruce Lee, maybe the reels are reflective of RTP and win-probability if you view the strips, but simply the video graphics have been done as he says so sloppily that certain wins or features have poor graphics, i.e. the freespins behaviour when you can tell they're coming.

 

[kktmd]

Immortal Romance 1.052.399 normal spins (freeplay, 0.30 bet). Frequency is the number of times the weel landed on this position. The distribution is close to perfect and NOT weighted IMHO. (If they wanted to rig the slots, it would be almost impossible to detect anyway).

 

[mathsboy1975]

Immortal Romance 1.052.399 normal spins (freeplay, 0.30 bet). Frequency is the number of times the weel landed on this position. The distribution is close to perfect and NOT weighted IMHO. (If they wanted to rig the slots, it would be almost impossible to detect anyway).

That is very informative not to mention impressive. I assume you must have used a traffic analyser to get that volume of data? If so, would you mind sharing what language you used to do it? I am a C++ man but I can see that perhaps a scripting language might be better to get hold of and parse http packets.
By the way I wasn't trying to imply that the slots are rigged at all, merely that if they really didn't want people to know the reel layout then they would easily be able to design an undecipherable layout (based on 3 symbol views that we get to see in the game).

Thanks for that post anyway, if for no other reason that it confirms my own layout for reel 1.

 

[dunover]

I look forward to the BDBA results with great anticipation....