DiceGames:doYouReally own all your chances? Introspection on common DG mechanism

[Danydee]

What is happening on ?

For my example I will take when's the setting is on 33,33% win chance, or x3 payout
the scenario arises when the payout is strictly > (superior) to 2'


So when you set the payout on 3, then you have to roll over or under chosing from one of the ranges [0--33.] and [66--100] ,
- here you cannot set your choice on the third tier/range ( [33--66] ) , in the case the rolled number will be on this range, 50 for example.


So, do is really a problem ?

Basically as you have to make a choice, you must be able choose from any range on the board,,
This is not like letting the machine take your bet and roll a draw randomly to get a result.

(Personally I think it is, that is really there a difference)




So this is a call to brilliant minds to try to get an answer about this !  
Maybe this will be a fist step for getting a definitive say on the Monty Hall paradox


So thank you in advance!






Note:  This is made a self moderated topic, so please do refrain from making non constructive posts and spamming the topic with posts that will not give a real addition for the discussion. Thank you

[1714944915]

1714944915

[1714944915]

1714944915

[1714944915]

1714944915

[1714944915]

1714944915

[1714944915]

1714944915

[1714944915]

1714944915

[Danydee]

Reserved

[o_e_l_e_o]

I don't see why it would make any difference. Provided the game is provably fair, then any of 1-33, 34-66, or 67-99, have the exact same odds of coming up. But if those are the rules the game sets, then those are the rules you agree to when you start playing. It's not entirely different to roulette only letting you bet on the 1st, 2nd, or 3rd 12, and not choosing your own range of 12, for example.

It also bears no relation to the Monty Hall problem. The Monty Hall problem only matters when I pick one of three options, and the dealer then removes one of the two remaining options and gives me a second chance to pick without changing the result. When playing dice games in a casino or online, you only get one bet on each result before the result is changed.

[Darker45]

You have to take note that not all dice games only offer over or under options. And not all dice games do not allow you to choose your range of choice.

I used to play on Windice.io and I remember that on that platform you are to choose in or out instead of over or under. When you roll, your bet would be whether the winning number is inside or outside your chosen range.

In case of 33% winning chance, for example, you have the complete freedom to choose your own range. You're not just given a total of 3 price ranges out of 0-9,999. So you may choose 0-3,299 or 1-3,300 and so on until 6,700-9,999.

[Danydee]

Provided the game is provably fair, then any of 1-33, 34-66, or 67-99, have the exact same odds of coming up.

This is evident,
but the fact that you'll not able to select one of them maybe could decrease your advantage, or increase in loss probability, as the (randomness value) of your choices when they're restricted increase that !  In other words, the disadvantage on the 'probalistic events' when you're restricted on the choice will acts as a increase factor when combined with the probability of occurring a defined draw !
Maybe that needs a mathematical demonstrasion/investigation to see what is really going on!

It's not entirely different to roulette only letting you bet on the 1st, 2nd, or 3rd 12, and not choosing your own range of 12, for example.

There we have it,
on roulette you can choose from all the (three) ranges, what is covering all the possible results, as also you can bet on every single number, letting your choices can theorically cover the whole of the possibilities .

It also bears no relation to the Monty Hall problem. The Monty Hall problem only matters when I pick one of three options, and the dealer then removes one of the two remaining options and gives me a second chance to pick without changing the result. When playing dice games in a casino or online, you only get one bet on each result before the result is changed.

The approach is same, the player is never supposed get know/or a clue of the result.

[janggernaut]

Well you got a point OP. But almost all site is trying to make their game become easier to people to understand how the games work. It will make them confused (maybe) if there is 3rd choice (mid/range), that's why there are only 2 options on almost every casino, which Low and High or (under and over).

[o_e_l_e_o]

but the fact that you'll not able to select one of them maybe could decrease your advantage, or increase in loss probability, as the (randomness value) of your choices when they're restricted increase that !

I don't think so. Provided the casino is provably fair, you have the exact same chance of winning if you select 1-33 continuously on every roll than you do if you change to a different range after each bet, just as flipping 10 heads in a row is exactly as likely as any other result from flipping 10 coins in a row. To think otherwise is to fall victim to the gambler's fallacy.

The approach is same, the player is never supposed get know/or a clue of the result.

In the Monty Hall problem the player does get a clue of the result. By opening one of the remaining two doors, the dealer/host combines the probability of those two doors in to the one remaining door, meaning one door is 33.3% and the other is 66.7%.

[Saint-loup]

What is happening on ?

For my example I will take when's the setting is on 33,33% win chance, or x3 payout
the scenario arises when the payout is strictly > (superior) to 2'


So when you set the payout on 3, then you have to roll over or under chosing from one of the ranges [0--33.] and [66--100] ,
- here you cannot set your choice on the third tier/range ( [33--66] ) , in the case the rolled number will be on this range, 50 for example.


So, do is really a problem ?

Basically as you have to make a choice, you must be able choose from any range on the board,,
This is not like letting the machine take your bet and roll a draw randomly to get a result.

(Personally I think it is, that is really there a difference)




So this is a call to brilliant minds to try to get an answer about this !  
Maybe this will be a fist step for getting a definitive say on the Monty Hall paradox


So thank you in advance!

I think janggernaut is almost right. You only have 2 choices because it's the rules of the game.
In the real Hi-Lo casino game with cards you can only bet on high or low, and you can't bet on "equal" even if you think the next card will have the same value. It's the house edge of the game.
So in order to respect this "rule" or standard, the online casino has to compute a range in order to offer you 33% winning chances with only 2 options.
33% is 1 - 66%
To get a 33% winning game, you must have 66% chances to lose.
If you bet on [1-33] among a [1-100] range you will have 66%chances of losing and 33% chances of winning.
If you bet on [67-100] among the same [1-100] range you will get 66%chances to lose and 33% chances to win.
So the game remains fair at the end. If you think the next number will be in the [34-66] range you are usually free to leave the game round or to change your seed.

[webtricks]

o_e_l_e_o is right! Restriction in choosing the range doesn't effect the probability. The probability of 3.00x payout will remain same no matter if you keep the range constant or change it constantly, provided casino is using strong RNG for generating server seed. The reason for this is that each bet is exclusive. The computation of each bet uses unique inputs which when used with strong hashing algorithm produces entirely different result from the precedent bet.

Anyhow, I created and ran a script to demonstrate the result using strong RNG for roll generation. I ran 5 sets of trial for each case containing 1,000,000 bets each. Here are the results:

CASE 1: Keeping the range constant - Roll under 33.00 i.e. [0.00-32.99] range

            WINS     LOSES
Set 1 - 329894   670106
Set 2 - 329977   670023
Set 3 - 329924   670076
Set 4 - 330425   669575
Set 5 - 329935   670065
Avg. - 330031  669969

CASE 2: Roll in range [0.00-32.99] for first 333,333 bets, roll in range [33.00-65.98] for next 333,333 bets and roll in range [67.01-99.99] for final 333,334 bets

            WINS     LOSES
Set 1 - 329628   670372
Set 2 - 329592   670408
Set 3 - 330024   669976
Set 4 - 330002   669998
Set 5 - 330385   669615
Avg. - 329926.2   670073.8

No significant variance found!

[Danydee]

The probability of 3.00x payout will remain same

That's when taking a fix range, (like when you're playing with auto-bet),  but what of when the player's decision is coming to interact on the game,
I think of for example when the result will be from the (omitted) range,, "(as like in provably fair systems the result is been occurred but is just hidden)", there, whatever what will the player choice the result is sealed to be "false/to loss".
I think giving the player possibility to choose is itself a factor of hazard that influing 'the game'. So do this (randomity) from player acting do not influence on the randomness on the whole (system): [(randomness on draw)+(randomness on player choice/or interaction)] ??



For that, maybe can @webtricks made a simulation with a REG acting as the player choice, so the result will depend on both the number rolled and the choice the player(the REG here) has made;
1- With possibly of choosing on just "over/under"
2- With possibility of choose on all tge three ranges.     

[webtricks]

The probability of 3.00x payout will remain same

That's when taking a fix range, (like when you're playing with auto-bet),  but what of when the player's decision is coming to interact on the game,
I think of for example when the result will be from the (omitted) range,, "(as like in provably fair systems the result is been occurred but is just hidden)", there, whatever what will the player choice the result is sealed to be "false/to loss".
I think giving the player possibility to choose is itself a factor of hazard that influing 'the game'. So do this (randomity) from player acting do not influence on the randomness on the whole (system): [(randomness on draw)+(randomness on player choice/or interaction)] ??

Okay, now I got your point and it is a good one. This is where using the provably fair system in correct way becomes very important. I have seen some casinos which claim to be Provably Fair but change server seed after the player sets new client seed. This system gives undue advantage to the casino to determine the roll result even before player rolls the dice. In last one year or so, I have actively tried to persuade casinos to shift from this system and on my recommendation 5-6 casinos (such as FortuneJack, OneHash, SimpleDice) shifted to more impartial system.

In the current standard system of provably fairness (which is being used by all notable casinos), the player has complete autonomy to pick the client seed before placing the bet. If the player picks a new client seed before making the bet, it will entirely change the roll result leaving no advantage to casino (because casino has already committed the server seed). For example, let assume the casino is trying to cheat player and sets - 'XXAASSXX' as the server seed which when salted with the current client seed of the player generates the roll result of 52.17. But then before placing the bet, user changes his client seed thus changing the roll result and making the result out of the complete-control of both player and casino. If casino is following this system of provably fairness then selecting custom range or not will have zero impact on the probability.

And this also brings me to another important point which I keep on advocating on forum regularly is that the players should actively change their client seed before every betting session and verify some of their bets frequently using the third-party verifier too.

[fiulpro]

I do think auto-bet is not an option that one should go for since most of the times players forget and are not able to control if they are having a loosing streak. For me I do think that dice games when I play for the first time usually give me amazing results , as in I am always in a profit but as the game progresses I start loosing everything. So I feel like I should only play one game in like 1 hour or something. I do think whatever we do it all depends on out odds but we can use some strategies , like doubling if you are loosing and decreasing by half if you are winning things like that.

I think the subject of choosing on anything on the board should be there since it will increase our probability by 1/3rd ! But I haven't played much dice after loosing all my savings 😅 hopefully I can see this thread for good strategies !

[electronicash]

gambler had a hard time winning good amounts with just having just two option, with 3 I'm not sure how i would go, I'll often experience 25+ losing streak here than with 2 options only. i get it that it will result to have more randomness to the dice game and remain fair if there are 3 ranges. if its going to be good for gamblers then why not. or why not make it 10 options too. 1-10, 11-20, 21-30 and so on.

[Danydee]

Around the subject, to try giving an (image)!
Paradoxally, just tell to a roulette player that he can bet just on two ranges [1-12] and [25-36] and not on the [13-24] one for a 1/3 win chance !



I think about the provably fair systems that the client seed should change on evry new bet, this is can settled to done automatically but nothing could guarantee from communicating it to the casino.

So developing an open source app tool, wich will act as the (player's keys), managing the (clients seeds) and also securing them from eventual leaking. This is would very great and give the player assurance that he is really in control !  It will not demand much work and will run on standard protocols, it could be implemented as an installable in app add-on or just provided from an independent online provider. If a such tool is made, casinos will be obliged to providing it's protocols/the implementation.

I have seen some casinos which claim to be Provably Fair but change server seed after the player sets new client seed.

For me I do think that dice games when I play for the first time usually give me amazing results , as in I am always in a profit but as the game progresses I start loosing everything.

About that, they can also just sniffing the client seed, or when the game is settled on Auto-bet just repeating strategically some losing bets(server seed) that have already occurred to precipiss the player's strategy to a loss.
Or maybe could them worked on mathematical models that's favoritise



Edit: 
Where was I,
The nonce of every new bet is added on the draw computation.
Anyway, generally something like a casino would never leave a option to tu turn more advantage to their side!

[fiulpro]

I have seen some casinos which claim to be Provably Fair but change server seed after the player sets new client seed.

For me I do think that dice games when I play for the first time usually give me amazing results , as in I am always in a profit but as the game progresses I start loosing everything.

About that, they can also just sniffing the client seed, or when the game is settled on Auto-bet just repeating strategically some losing bets(server seed) that have already occurred to precipiss the player's strategy to a loss.

Another thing , if I try to reset the seed and play again I still loose. I remember the game I was playing where my 20$ became 40$ after 1-2 hours and I lose the profit I made. So it was 20$ again, but I forgot to clear my seeds and then my amount was again 40$ so I was happy about it but then they **gave me some badge for getting back my investment** but in the next game I was left with !! Nothing. NOTHING at all. Plus I was playing extremely safe. I felt like the badge was a mockery they gave me before their game bot decided to have some fun. Lol but overall I enjoyed it and the amount was not excessive. Your post reminded me to again ** clear the seeds before making any bet**

Plus am not so sure about the open source software though because it's so easy for the developer to make it anonymously and then do something like phishing the attack might even remain unnoticed for long and we all know that many people will try and have fun with it.

[o_e_l_e_o]

I think about the provably fair systems that the client seed should change on evry new bet, this is can settled to done automatically but nothing could guarantee from communicating it to the casino.

You can request a new seed on every bet if you like, but with the additional of an incremental nonce, setting new seeds at the start of each betting session (request the server seed hash before setting your own client seed) is enough to guarantee random and provably fair results.

Or maybe could them worked on mathematical models that's favoritise

To be provably fair, the casino must make public their method for converting the two seeds and the nonce in to the result of roll prior to you starting betting, and this can never change. If it does change or you can't verify it, then it is no longer provably fair.

Another thing , if I try to reset the seed and play again I still loose.

Did you verify your rolls? Any losing streak is simply bad luck unless you can prove that the rolls were non-random, regardless of how long it has been since you reset your seed(s).