[DICE]Bikinidice, Multicurrency,auto bet, range selector.AUCTION, PLEASE ENTER !!

[zoukenn]

At the end of week I'll invest 1 bitcoin in this website, time to get my salary and bankroll will go up to 5 bitcoin 

[BikiniDice]

At the end of week I'll invest 1 bitcoin in this website, time to get my salary and bankroll will go up to 5 bitcoin 

Thanks zoukenn!

[Dabs]

@dooglus
We will change maximum profit per bet (from 2% to 1%) when we reach 5 BTC of bankroll.
So, what are you waiting for? Help us to achieve the first target (5 BTC) 

I'm not waiting for anything. I have no intention of investing with you.

Why all of this antipathy in confront of Bikinidice? What's wrong with you? I don't understand 

There's nothing wrong with him (dooglus). He's simply stating that he has no intention of investing in this site. You go ahead and take the risk if you want to, and if you think it is worth it.

[BikiniDice]

Google 2-Step verification added:
https://www.bikinidice.com/settings

You can add this on: login, withdraws, divest.

I hope you appreciate it 

[eden1]

nice
when is the sig campaign?

[dooglus]

I'm not waiting for anything. I have no intention of investing with you.

Why all of this antipathy in confront of Bikinidice? What's wrong with you? I don't understand 

I have no reason to trust that they will return any investment I make.

You're asking me what's wrong with me because I won't trust an anonymous stranger with my coins? Does he even WoT?

[dooglus]

So the variance dosen't increase linearly, right, but it should be my only concern, or not? In the long run, if bankroll is replenished and don't default, I would expect a return equal to 1% of wagered amount.

Oh, OK. I see the point you're making.

If you intend to keep depositing more if you lose part of your investment, then the 2% risk is fine. Risking 2% of your whole bankroll on a 1% house edge game is a very bad idea. But you're not - you're depositing a small part of your bankroll at a time and keeping it topped up.

Is a kelly number of 2 equals to a max win of 2% of the bankroll? I think your graph is about fractions of f*, not of the total bankroll.  Seems to me that they are pretty different concepts.

The 2 means you're risking twice as much as Kelly recommends. Kelly recommends that if the house edge is 1% then you risk 1% of your bankroll per roll.

You need to think about the bet that the house is making, not the player. The "max profit" is the stake from the house's point of view, and the player's stake is the potential profit for the house.

I have all the variables to calculate kelly for a single bet, but how I can calculate f* for many consecutive irregular bets? ?

Kelly will tell you how much you should risk per bet. As the house, you don't get to decide how much the players are going to bet each roll, but you can set the maximum you are prepared to risk. If you can figure out the optimal amount to risk per bet, then risk that amount on each and every bet to maximise your return over multiple bets.

If you're asking whether it might be better to risk too much on the highest bets, too little on the smallest bets, and hope they average out, that's probably true, but you don't know what the distribution of bet sizes is going to be in advance, and so I prefer to set the maximum profit to the optimal stake size and leave it at that. Basically, none of this is going to matter too much until the day some crazy player comes and spams you with max bets. At that point you're going to want to make sure you aren't risking too much per roll.

[waterpile]

Black Coin with an image of a Black Gal... I find it Racist 

[dooglus]

Black Coin with an image of a Black Gal... I find it Racist 

Racism: the belief that all members of each race possess characteristics or abilities specific to that race, especially so as to distinguish it as inferior or superior to another race or races.

How is showing an image of a black woman next to "blackcoin" racist? Or are you using a different definition of racism?

[erre]

Oh, OK. I see the point you're making.

If you intend to keep depositing more if you lose part of your investment, then the 2% risk is fine. Risking 2% of your whole bankroll on a 1% house edge game is a very bad idea. But you're not - you're depositing a small part of your bankroll at a time and keeping it topped up.

The 2 means you're risking twice as much as Kelly recommends. Kelly recommends that if the house edge is 1% then you risk 1% of your bankroll per roll.

You need to think about the bet that the house is making, not the player. The "max profit" is the stake from the house's point of view, and the player's stake is the potential profit for the house.

Kelly will tell you how much you should risk per bet. As the house, you don't get to decide how much the players are going to bet each roll, but you can set the maximum you are prepared to risk. If you can figure out the optimal amount to risk per bet, then risk that amount on each and every bet to maximise your return over multiple bets.

If you're asking whether it might be better to risk too much on the highest bets, too little on the smallest bets, and hope they average out, that's probably true, but you don't know what the distribution of bet sizes is going to be in advance, and so I prefer to set the maximum profit to the optimal stake size and leave it at that. Basically, none of this is going to matter too much until the day some crazy player comes and spams you with max bets. At that point you're going to want to make sure you aren't risking too much per roll.

Ok, i got it.

But I still can't understand how you calculated " optimal kelly"  to be exactly 1% of the bankroll. I also tried some online calculators, but I guess I have to use a different formula than the previous one who I linked from wikipedia, who works only for single bets. How did you adjust the formula for multiple bets?

I also noticed that for 1/2 kelly you have little volatility, bit almost halve the earinings! Is it the best solution, in your opinion?

A last precisation: when I spoke about replenishing bankroll,  I meant that variance would not be a problem:even if I lost 99% in the short term, I would expect to gain 1% of the wagered amount in the long term. The only risk is the risk of default. So...can we calculate it for kelly and fractional/double kelly? And how?

[dooglus]

But I still can't understand how you calculated " optimal kelly"  to be exactly 1% of the bankroll.

It's only 1% if the bankroll if the house edge is 1%.

Start with the f* = (bp-q)/b formula that you started with. Substitute in the value for b, and simplify it. You end up with f* = e.

I guess I have to use a different formula than the previous one who I linked from wikipedia, who works only for single bets. How did you adjust the formula for multiple bets?

You don't need to. All bets are the same - risk the optimal amount on all of them and you can expect to perform optimally.

I also noticed that for 1/2 kelly you have little volatility, bit almost halve the earinings! Is it the best solution, in your opinion?

Isn't it more like 75% of the earnings? What is "best" depends on your tastes. How much do you long to win? How much do you fear to lose?

A last precisation: when I spoke about replenishing bankroll,  I meant that variance would not be a problem:even if I lost 99% in the short term, I would expect to gain 1% of the wagered amount in the long term. The only risk is the risk of default. So...can we calculate it for kelly and fractional/double kelly? And how?

There's a problem with this. If you lose 99%, some other guy will come along after your loss, invest the same amount you initially invested, and get around 99% of all future profits. You'll never recover because you were too easy to "dilute" into insignificance. You never go bust using Kelly betting, since you only ever bet a fraction of your remaining bankroll. But you can get arbitrarily close to busting.

Here's the math I did a couple of years ago when first thinking about max profits:

Code:

house edge: e (a fraction)
house win probability: p
player win probability: q = 1-p
player payout multiplier: (1-e)/q
when player bets 1, he stands to win (1-e)/q - 1
house risks (1-e)/q - 1 to win 1, so b = 1/((1-e)/q - 1)

f = (bp-q)/b
  = bp/b - q/b
  = p - q/b
  = p - (1-p)/b
  = p - (1-p)((1-e)/q - 1)
  = p + (1-p) - (1-p)(1-e)/q
  = 1 - (1-p)(1-e)/q
  = (q - q(1-e))/q
  = (q - q + qe)/q
  = qe/q
  = e

[waterpile]

But I still can't understand how you calculated " optimal kelly"  to be exactly 1% of the bankroll.

It's only 1% if the bankroll if the house edge is 1%.

Start with the f* = (bp-q)/b formula that you started with. Substitute in the value for b, and simplify it. You end up with f* = e.

I guess I have to use a different formula than the previous one who I linked from wikipedia, who works only for single bets. How did you adjust the formula for multiple bets?

You don't need to. All bets are the same - risk the optimal amount on all of them and you can expect to perform optimally.

I also noticed that for 1/2 kelly you have little volatility, bit almost halve the earinings! Is it the best solution, in your opinion?

Isn't it more like 75% of the earnings? What is "best" depends on your tastes. How much do you long to win? How much do you fear to lose?

A last precisation: when I spoke about replenishing bankroll,  I meant that variance would not be a problem:even if I lost 99% in the short term, I would expect to gain 1% of the wagered amount in the long term. The only risk is the risk of default. So...can we calculate it for kelly and fractional/double kelly? And how?

There's a problem with this. If you lose 99%, some other guy will come along after your loss, invest the same amount you initially invested, and get around 99% of all future profits. You'll never recover because you were too easy to "dilute" into insignificance. You never go bust using Kelly betting, since you only ever bet a fraction of your remaining bankroll. But you can get arbitrarily close to busting.

Here's the math I did a couple of years ago when first thinking about max profits:

Code:

house edge: e (a fraction)
house win probability: p
player win probability: q = 1-p
player payout multiplier: (1-e)/q
when player bets 1, he stands to win (1-e)/q - 1
house risks (1-e) / 1q - 1 to win 1, so b = 1/((1-e)/q - 1)

f = (bp-q)/b
  = bp/b - q/b
  = p - q/b
  = p - (1-p)/b
  = p - (1-p)((1-e)/q - 1)
  = p + (1-p) - (1-p)(1-e)/q
  = 1 - (1-p)(1-e)/q
  = (q - q(1-e))/q
  = (q - q + qe)/q
  = qe/q
  = e

here is a better idea re-open your site and apply those things that you posted, so that everyone will be happy 

[Dabs]

here is a better idea re-open your site and apply those things that you posted, so that everyone will be happy 

That's not likely to happen. But when he does, I'll have to close mine down because I'm going to run out of players. hehehe. (or investors)

[EdenDice]

Good to see you are listening to the community's suggestions. They definitely like and will respect that. Have you managed to change it already or are you still in the coding phases?

Is there a was to see the amount invested or will that come in the changes?  

We are currently working on. A little more time and we release new Invest system.

Great, I can now see the total invested and the stats. It looks like a lot has changed in your investment strategy, also the addition of investment of multiple coins. I would think this would be a lot of admin to handle. How is it going so far?

[erre]

Here's the math I did a couple of years ago when first thinking about max profits:

Code:

house edge: e (a fraction)
house win probability: p
player win probability: q = 1-p
player payout multiplier: (1-e)/q
when player bets 1, he stands to win (1-e)/q - 1
house risks (1-e) / 1q - 1 to win 1, so b = 1/((1-e)/q - 1)

f = (bp-q)/b
  = bp/b - q/b
  = p - q/b
  = p - (1-p)/b
  = p - (1-p)((1-e)/q - 1)
  = p + (1-p) - (1-p)(1-e)/q
  = 1 - (1-p)(1-e)/q
  = (q - q(1-e))/q
  = (q - q + qe)/q
  = qe/q
  = e

Ok, now I REALLY got it. I also got why, because of "variance riders" (people who divest when the bank is winning too much) or " later investors", a too high volatility is not a good thing.

[allcoinminer]

Bikini Dice = XXX + Gambling.
A really nice but dangerous combination for any hardcore gambler.

[erre]

12 loss martingale 


free image upload

[waterpile]

12 loss martingale 


free image upload

common story, i got a X15 losing streak in bitdice and also in PD

[Josepht]

I rolled a 69.69
Seems like I have won the jackpot

[Jerome?]

I rolled a 69.69
Seems like I have won the jackpot

Lol.. Whats your strategy when you play dice? Are you going for 50%? or what?