Can you try sending a tiny bit less, like 1 satoshi less than your balance? That's a bug i need to fix ASAP

Just a reminder to everyone, please make sure to use unique, strong passwords. I've been mitigating both some combo-list logins, and brute-force dictionary attacks on select user accounts. While for now all offending ips have been banned I likely will up needing to add a captcha on login.

Also, it's a good idea to set your email so that I can get contact with you. There's been been some accounts which have been specifically targeted, but I've been unable to give a heads up to the owners due to not having a contact email.

where computeKelly the kelly criterion. In the case of a simple binary bet, it returns quite literally the house edge of that bet.

We actually bisect this. We have a function which applies the generalized kelly, which says if it's acceptable to investors or not . So we start 1.25x,  and are told it's acceptable. So we decrease this to say  1.1x, and are told it's unacceptable. So we increase it, to say 1.17x and told it's unacceptable so we increase it again. We try 1.21x and told it's acceptable, so we decrease it etc.

And eventually we converge on what is the minimal good deal, and use that.

There's no average, as even in the simple case of a binary bet (win or lose), it depends on the current bankroll and the house edge on the bet. I could build a debug calculator, where you enter the details and it tells you exactly based on the bankroll at the time.


The criteria for that is well defined, we use a generalized kelly to determine how much of our bankroll we should risk, for a given bet. If investors are risking less than that, we consider it acceptable. If investors are risking more than that, we consider it unacceptable.

The investors will never be given a bet that is -EV, as it's a requirement that it is "acceptable" as per the kelly criterion. The reason we don't charge a normally commission, is because imagine there are two MoneyPot casinos, one who offers a game of 1% house edge dice and another offers keno, with a 20% house edge.

 From the investors perspective it doesn't matter what game people are playing, but from the casinos perspective it's huge. The keno game should expect its players to run out of money 20x faster, and have much more trouble getting customers. If we gave that huge edge to the investors, the business would be unviable for the keno casino and they wouldn't be able to use MoneyPot. So the way it works, is use the extra money to pay the casino rather than the investors. But the important thing is investors still get a good deal, they just get the worst good deal.

We hope this translates into more bankroll turnover, which should makeup for it.

It does But PM me your username, and I'll send you some play money to give it a try. We'll add a faucet early next week

If you get the chance, could rename 'MoneyPot' to 'bustabit'. The only thing we've changed is the domain (to bustabit.com) so all the previous stats and history should be fine!

We've also created a new site: dustdice.com if you want to take a look. Tomorrow we'll add some scrapable stats for it as welll

I don't see the problem, but if one still exists could you please contact support and we'll sort it out.

And congratulations to SnailMan, who just won over a bitcoin!

There's a good reason it shows like that, but I might change the behavior to make it less confusing. It's just a
view-layer rounding of a single satoshis, and investing and divesting all works fine:





Ah thanks, nice catch. We show  < 0.01% if your stake is less than 0.005 but in reality it should be if your stake is less than 0.00005. I just fixed that, and it should display properly now

That's correct, MP protects investors from malicious apps. We check every bet, and require them to play against our provably fair secret, so we know they're not cheating. By "strange bets" I meant bets, I was referring to it from a users perspective. If you authorize an app, and it starts gambling away your money, those are "strange bets". MP itself has no way of knowing if you intended those bets to be placed or not (after all, you authorized the app), so it allows them through.

Originally we imagined it as you describe, and required you to put limits in place such as "It can lose up to X, wager up to Y, play with a house edge of less than Z" but it was simply too confusing to users, and was very hard to explain. Now we settled on a much simpler "balance" system, where you transfer money to the app and play with it. The app can only spend from its balance, but nothing more. Since you can do instant transfers, it's extremely easy to top up this balance on demand. People have had no problems understanding it (unlike a permissions based scheme) as that's how they're already used to using a casino. But we're not opposed to the idea of re-adding some more fine-tuned permissions for advanced users.

That's right! Even malicious apps have no access to your invested money, or balance. They can only spend what you give them.



We guarantee every bet is fair (as in, the outcome is not tampered with), but have no way of knowing if it's the bet you wanted to make. You can protect yourself by limiting the amount of money you put in each app, and making sure there's no strange bets


There's no technical measures that stop me from running away with your investments/balances, but I can (and do) prove that I'm solvent, and I'm happy escrowing and verifying personal details so that no one thinks it's a simple exit-scam. But if I was trying to scam you, these are the exact things I'd be saying (including this very disclaimer) so I highly recommend you keep the bulk of your money yourself, in something like a hardware wallet. We'll both sleep better that way. =)

We hope to provide a lot more utility to a casino than just a bankroll: a unified login system, 0-conf deposits, chat servers, security and what not. If a 50-50 profit share ever seems burdensome, we'd be happy to negotiate that down =)


Thanks, I really hope so.


It's not as bad as it seems. All coders I've dealt with, want a fixed amount. The investors don't ask for much, and provide you an extremely useful service (they eat the variance in exchange for a fee). So really, it's mainly just a 50-50 profit share between Vault and yourself, which isn't terrible considering just the security stuff, maths and bitcoin stuff easily took us and established team 3 months to properly build.

It really needs to come out of your pocket, although we're pretty reasonable and depending how attractive it looked would be willing to split the costs of promotion 50-50 with you, or at the very least apply a credit to our commissions so that you get 100% of your earnings until you're at least profitable

The API is already out, we're just in the process of documenting it. If you check out toydice and view source, you'll find a very minimal and simple walkthrough of how to create a casino with just a handful of jquery

Well, you could fork dust dice and rebrand it as your own app. We could make this process much smoother if there was some demand. So far DustDice has made 0.11 BTC in risk-free commissions over the last couple of days, so promoted right there could be a business opportunity. =)

It's not something we have much desire to support, however we've found shapeshift makes it pretty effortless to convert back and forth from bitcoin.

Just tried on android and it lags a little but otherwise is pretty usable. Silky smooth on an iphone. Some of the margin and hidden buttons don't look as great as they should, but that's something we'll keep improving

That would be me

That's right, I reworked the example where the numbers this time should add up. The issue was only affecting the example, not the code (which does these calculations properly). Thanks for pointing out the mistake, btw )