Question for the industry guys (RTP)

andym

Experienced Member
Joined
Jul 6, 2013
Location
germany
Hi guys,
i have one question for you and would be happy if someone can answer that.
Lets assume there is only one online casino out there. This online casino licensed only one slot. This one slot has a theoretical RTP of 96%.
In that online casino only two players are playing the whole time.

Player1 made a really good win on the slot and get an RTP of 104%. Then Player1 stopped immeadiately playing.
Now whats the most likely real-RTP-outcome for Player2 who still continues playing with a very lot or unlimited spins:

1. Player2 will end up at 96% RTP
2. Player2 will end up at 88% RTP

Thanks
Andy
 
Last edited:
Solution
You are talking about the concept of compensation if you're thinking Scenario 2 is the answer, online slots are random and are not compensated. therefore Player 1's results and Player 2's results are entirely independent of each other.

Eventually Player 1 and Player 2 will, given a sufficient number of spins, reach the T-RTP of the slot, but their results along the way do not influence the other at all.
Wondering what the heck RTP is? Find out here at Casinomeister.
I guess there must be a technical/purity difference, maths is not a strength so it would probably go over my head!
If you go to random .org you can get more info if you are interested.

But in short:

"Perhaps you have wondered how predictable machines like computers can generate randomness. In reality, most random numbers used in computer programs are pseudo-random, which means they are generated in a predictable fashion using a mathematical formula. This is fine for many purposes, but it may not be random in the way you expect if you're used to dice rolls and lottery drawings.

RANDOM.ORG offers true random numbers to anyone on the Internet. The randomness comes from atmospheric noise, which for many purposes is better than the pseudo-random number algorithms typically used in computer programs."

So apparently we can get true random, but not by using a mathematical formula.
 
In simple terms you need a truly random external 'input' to the computer (known as entropy) for it to generate a truly random number.

Pseudorandom will generally use a seed and an algorithm, if you know the seed and the algorithm, and have some of the sequence of results, you can theoretically work out what's coming next. (Back in the 80s some guys famously worked out the P-RNG on US casino video poker machines and were able to work out where in the sequence the machines were and therefore when winning hands would be dealt, they made a lot of money.)

Pseudorandom is actually fine for a lot of applications (do you really care if the random play on your CD player is 'true' random or pseudorandom?), but for true randomness you need an external input of a random value.
 
In simple terms you need a truly random external 'input' to the computer (known as entropy) for it to generate a truly random number.

Pseudorandom will generally use a seed and an algorithm, if you know the seed and the algorithm, and have some of the sequence of results, you can theoretically work out what's coming next. (Back in the 80s some guys famously worked out the P-RNG on US casino video poker machines and were able to work out where in the sequence the machines were and therefore when winning hands would be dealt, they made a lot of money.)

Pseudorandom is actually fine for a lot of applications (do you really care if the random play on your CD player is 'true' random or pseudorandom?), but for true randomness you need an external input of a random value.

Glad you used simple terms lol.
 
who decides the 'solution' and is this a new forum feature as I haven't see it before?

edit: not disagreeing with Chop's post, other than I believe the rng is 'pseudorandom'

I don't know who did that, I didn't ask for it or influence it! :)

Still, make hay whilst the sun shines and that.

CHOPLEYIOM SOLUTIONS PROVIDER AT YOUR SERVICE - Solutions cost £250 each, payable to my Neteller account.
 
The thing is that everything is so hidden in so many conditions/preconditions/terms/hidden formulars,processes and "unlimited" proof of spins that you will be never able to judge if a casino cheated on you or if everything is just the "randomness"...
I mean how you want to proof that?
Example:
i make a big win on one day on one slot in an online casino. After that day i played 7 days in a row every day on different slots in the same casino. All slots i played have a T-RTP of 94-96%.
After these 7 days i ended up with an overall RTP of 62% for the complete week.
While i played a huuuuge amount of spins with an huuuge amount of euros.

Did the casino now cheated on me to get the money back? Impossible as every slot for itself is just random and not influenced by other slots/players or the casino... Or is it just random and bad luck?
And what, if i tell you now that exactly that pattern happened after my last 2 wins in that casino also.
So everything just random? how many spins/euros do i need to proof that its not random?...
Impossible guys. You have to live with that or stop gambling!
 
I have a video on this subject that people might find helpful. Will link it at the bottom.

However as was already stated, one players play has no impact on another. The VIP manager you spoke to was just explaining the concept very poorly. Conceptually he's correct, the RTP of a game over all its spins over all players will converge on the TRTP because some players will win big and others will not, but that is based on individual luck for those players, not because the game is 'keeping track'.

The more important thing these days is 'volatility'. Games these days are highly volatile, meaning there will be big winners and (more) big losers and means that a convergence on the TRTP takes millions and millions of spins, so any one individual would never be able to prove it based on their own play.

P.S, I 'know' this rather than 'think' this, because I used to make these games for a living :) And we used to simulate around billion spins to prove RTP's I believe (this is about 10 years ago, so games were less complicated back then).

If you have any specific industry questions, feel free to ask.

 
I wonder what the chances were of that guy getting those identical bonuses on Diamond Mine, and suspiciously identical spins previously, didn't they payout the same also? On a slot that randomly produces mega ways :)

For games that are supposedly extremely random and unique for each and every single player, what are the chances eh :p

It's no wonder people wonder, that's just one guy that noticed it and records his sessions.

Edit: This one.. This'll Make the Sh*t Hit The Fan - Random Megaways? - Casinomeister Forum
 
Last edited:
Where are you guys getting the info on how many test spins were done at NetEnt etc?

As you may know, RTP is calculated and games tested over millions, billions and trillions of spins. This is recorded on game sheets and inevitably, some information gets out.

Some Providers, like Peter & Sons and Blueprint, have started listing full stats in the game menu and hopefully it won't be long before others follow such ethically pioneering ways :)
 
I have a video on this subject that people might find helpful. Will link it at the bottom.

However as was already stated, one players play has no impact on another. The VIP manager you spoke to was just explaining the concept very poorly. Conceptually he's correct, the RTP of a game over all its spins over all players will converge on the TRTP because some players will win big and others will not, but that is based on individual luck for those players, not because the game is 'keeping track'.
Sorry but I don’t get your statement. If the rtp is calculated over all spins and all players, then your clear answer in my case is player1-rtp affects player2-rtp.
please note that in my example you only have two players playing the slot.
 
Sorry but I don’t get your statement. If the rtp is calculated over all spins and all players, then your clear answer in my case is player1-rtp affects player2-rtp.
please note that in my example you only have two players playing the slot.
You are missunderstanding RTP. It isn't a target, and it isn't something the game actively tries to hit. It's just the natural percentage the mathamatical model outputs. A single player, with enough spins, will converge on the TRTP. All players play together (with enough spins) will converge on the TRTP. It's just mathamatical statistics.

The play from one player does not effect another. If player 1 does two spins and wins big, they will have an RTP well over 100%. That win doesn't change the chance of anyone else winning or losing.

Month on month, the total RTP of the game is likely to vary above or below the TRTP, but it's just a number caluclated based on all the money staked vs all the money returned.
 
If you flip coin, you should have pretty equal chance for both sides but you can get quite many heads/tails in row but even you would have 5 heads in row, on next flip it's still again 50/50 chance for both without any weighting to hit tails after 5 heads in row.

It's expected that what bigger sample of these coinflips you take, you end up nearer of 50/50 but there can be still quite significant discrepancies even after quite many throws but what bigger is your sample, nearer of "right" result you end up and in very big grand total it's getting really near of actual assumed result (like @The Reel Story clearly showed in his video).
 
The vip manager has no idea what is going, if you ask me.
It's a typical manager explanation.
"If you don't win, the other one win. Next time (maybe tomorrow) you might be the winner.
So don't be sad. It is just luck! "

The realistic point is:
There are slots giving very low rtp for a whole month maybe more...
So t-RTP might be 94,5% or whatever and everyone who is playing
this slot on this casino in this month just get 91% in average back.

You often can check this on videoslots homepage

btw.: The really high rtp's for a month are always no-name slots.
 
So, in the testing stage, who decides how many spins should be done? Surely to get a true RTP there should be a set number otherwise it can be manipulated?
For example, Netent say it took 11 trillion spins.
What if they started thinking, ok, we will do 5 trillion, at the end the game shows 70% RTP, they think, shit, do another trillion see if we can get it up a bit, then another, then another, then another until they hit the 96% they need.

They could even have a stop limit, something like if number spins>1 Trillion and RTP = 96.2% then stop. If number of spins >1 Trillion and RTP <96.2% continue. Then eventually they will hit the RTP they require and stop, even though another trillion could see it drop to say 80%

Surely all slots should be tested over the same amount of test spins, otherwise the RTP means nothing?
No, all slots don't need to be tested over the same number of spins. It depends on many many factors, but the main one is the standard deviation. You need to run enough spins in order to be confident in the RTP, and this will vary down to many factors.

And yes, the VIP manager is NOT a games designer or provider, so their knowledge of how games work may well be very limited, and judging by what they have said, their knowledge IS very limited. And to be honest, is exactly what I would expect someone to say who doesn't understand how slot games work.
 
yes, but thats why the model based only on two players in that topic.
The only question i wanted to point out is: if a rtp of one player effects the rtp for another player in the same casino on the same slot.
In my opinion it shouldnt, but referring to a comment of a vip manager it does (see my posts above).

And thats just an additional point why i in the meantime have huge doubts about that whole industry...
Just out of interest, why do you believe your VIP manager but not me. Or @The Reel Story or @Halvfor example.

if you have huge doubts about the industry, honestly you should stop playing.

But categorically I can tell you that what happens with one player has no impact whatsoever on another player.
 
No, all slots don't need to be tested over the same number of spins. It depends on many many factors, but the main one is the standard deviation. You need to run enough spins in order to be confident in the RTP, and this will vary down to many factors.

And yes, the VIP manager is NOT a games designer or provider, so their knowledge of how games work may well be very limited, and judging by what they have said, their knowledge IS very limited. And to be honest, is exactly what I would expect someone to say who doesn't understand how slot games work.
You said it was ok to run different math models for different betsizes, and that gives quite a bit of control over exposure etc if you want to use it like that (which i assume casinos would want to do)
Is it also allowed to have more than one math model for the same betsize, or can i be sure that i always play using the same math model as long as i stay on the same betsize?
 
You said it was ok to run different math models for different betsizes, and that gives quite a bit of control over exposure etc if you want to use it like that (which i assume casinos would want to do)
Is it also allowed to have more than one math model for the same betsize, or can i be sure that i always play using the same math model as long as i stay on the same betsize?
I said its possible. I didn't say its probable. But having different maths at the same bet size would not be allowed.
 
I said its possible. I didn't say its probable. But having different maths at the same bet size would not be allowed.
For me, the fact that its possible, and that its allowed no less, makes me believe its something that happens.
What would the downside be? (for casinos i mean, the downside for players is obvious)
Limiting big hits on big stakes would reduce the risk for casinos, and since its all above board for some weird reason, why wouldnt they?
 
For me, the fact that its possible, and that its allowed no less, makes me believe its something that happens.
What would the downside be? (for casinos i mean, the downside for players is obvious)
Limiting big hits on big stakes would reduce the risk for casinos, and since its all above board for some weird reason, why wouldnt they?
There would be more testing required, more cost (each bet would have to be certified separately, and this is not cheap) and there is little point. We tell the casinos what the max liability is as part of the par sheets (maths documentation), and most casinos have a maximum payout limit anyway which would negate the need to reduce max liability on higher bets
 
There would be more testing required, more cost (each bet would have to be certified separately, and this is not cheap) and there is little point. We tell the casinos what the max liability is as part of the par sheets (maths documentation), and most casinos have a maximum payout limit anyway which would negate the need to reduce max liability on higher bets
Do you mean that some sites limit how quick they have to pay out a big win, or are there sites that limit the size of wins aswell?

Im not saying there is a need to reduce max liability, i think there is a greed to reduce max liability.
;)
 
Do you mean that some sites limit how quick they have to pay out a big win, or are there sites that limit the size of wins aswell?

Im not saying there is a need to reduce max liability, i think there is a greed to reduce max liability.
;)
I think all sites limit the maximum win, as some (many) games have infinite potential (although its highly improbable).

The platform I ran could configure min stake, max stake, default stake, stake selections and maximum win (as a monetary value) all easily in the back end.

Games these days seem to limit on an X value (5000x for example) which is probably a more sensible method than we used.

But Yeh, all casinos will put a cap on their liability (and some will have terms and conditions about how fast they pay out big wins, like 5k per month or something), so always read the terms.
 
I think all sites limit the maximum win, as some (many) games have infinite potential (although its highly improbable).

The platform I ran could configure min stake, max stake, default stake, stake selections and maximum win (as a monetary value) all easily in the back end.

Games these days seem to limit on an X value (5000x for example) which is probably a more sensible method than we used.

But Yeh, all casinos will put a cap on their liability (and some will have terms and conditions about how fast they pay out big wins, like 5k per month or something), so always read the terms.
I still dont think that 'negates' the upside of reducing liability with gimped math models on the biggest betsizes on popular slots.
Sure, you may have no problem paying out that 500k (or whatever the limit is) but surely it would be better if that 500k hit never happened in the first place?

Im not likely to hit any payout limits anytime soon.
Even a max win on doa2 would probably be paid the same day with the stakes i play at.
=)
 
Wondering what the heck RTP is? Find out here at Casinomeister.

Users who are viewing this thread

Meister Ratings

Back
Top