All this said, if I accept investment it will prefer to not use the "bankroll investment" system. I tried that on MoneyPot, however am not particularly satisfied with how it shares the various risks and rewards among investors and operator. One model that I'd prefer to try, is selling equity shares where money from the equity sale goes to fund the bankroll. If the company believes the shares are undervalued, it could use money from the bankroll to buy-back some of the shares and likewise if it needs to increase it's bankroll it could sell off more of itself.  And when money is to be taken out of the bankroll, it'll be through dividends on the share.

Probably the biggest thing stopping me, at the moment is just the lack of a robust secondary market for such shares. I haven't heard great things about havelock and crypostocks, and would like to avoid using some altcoin based approach. I read the OpenAssets specification, and it looks quite appealing but without a great secondary exchange it's a bit too difficult.

If anyone has any suggestions or experience, I'd be interested in hearing it.

Just to clarify the investment situation. Currently I do privately bankroll it, and would be interested in taking investors in the future of some shape to increase the max-win and max-bet. I have unfortunately had to (multiple-times) turn back larger players interested in raising the limits, simply because the limits are too low. The reason I have in the past (and continue to) take money out of the bankroll on occasion is just to reduce the amount of money I have at risk and spend on expenses and what not.

As arallmuus said, the moneypot investment feature isn't very suitable for bustabit as bustabit is very non-atomic, and moneypot allows investors to invest and divest at all times (including other constraints) which make it difficult. It's worth noting at the moment, bustabit is 100% separate from MoneyPot with the only common thing a shared owner =)

If you click "Payout Editor" on their site, it shows the estimated max bets (~1.4 BTC on green, 0.13 on yellow, 0.003 on red). Actually though, these are very conservative estimates (they seem to be based on the simplified kelly, not the generalized kelly that MP uses) especially in the red case but bets of those size should go through no problems.


I'll try put together a blog post with some code on the full generalized kelly (multi-outcome, like plinko) as it's pretty hard to come across any good information about it and for things like the max-payout it gets pretty interesting

The commission charged isn't on profits, but rather on the amount wagered. i.e. Even if a gambler breaks even, investors will be a loss. Investors will only make a net profit when gamblers lose > 0.25% of what they wager (which over the long run they should, as the house edge is 4x that)

Actually, not quite. The best way to think of house edge is the average amount of MONEY (not bets) you'll lose, over a very big number of bets.

For better illustration, take a fair coin-toss. If you win, you'll double your money. If you lose, you'll lose it all. This has a house edge of 0%, yet you'll still lose 50% of the time. Or lets imagine a fair (normal) 6-sided dice. If you win, you get 6x your money, yet you'll still lose 83.33% of the time.

Or lets lets take a state lottery with a 20% house edge. You're not going to win the million-dollar jackpot 80% of the time.

I hope this makes it clear that there's no implied relationship between house edge and win probability. Also, I verified some of the bet payouts and they do indeed have the advertised house edge.

Pretty exciting day today at bustabit. Over 1000 BTC wagered, with both some really big winners and losers, and some good bonus hunting opportunities.

No, they're all completely different. The chance of losing 30 in a row on bustabit would 1 in 1,683,362,473. While on a 1% house edge dice site might be 1,451,590,215. Although one thing to consider with bustabit, is even if you lose you are still eligible for bonuses. So the actual house edge on bustabit is about 0.5% when playing at 2x.  While on MP games, the house edge is configurable by the app developer.

As announced on slack, we're temporarily running on 0 commissions for app developers. This means devs should be now making x2 as much as normal for the next couple of days.


Also in the coming hour, we'll support setting our own app thumbnail and description. So app devs please checkout the admin panel, and make sure thats set.

Ctrl+f for 'api', it's in the footer:

https://primedice.com/api

Unless it's stopped, I don't see any problem on this end. We're reporting 95% of bets responses processed under 12ms, so it's hopefully not here.

Wikipedia has a page for everything:

https://en.wikipedia.org/wiki/Risk_of_ruin

and the original meaning of:

https://en.wikipedia.org/wiki/Gambler%27s_ruin

A great end of day result for almighty:



clawed back all the way from a ~6 BTC loss, ending with the famous bustabit instant withdrawal =)

Finding investors isn't hard, it's convincing them you can deliver returns that is.

I would clearly present your idea, plan, roadmap  (don't worry, no one steals ideas. If you have a good idea, you have to shove it down peoples throats) and sell people on it. Research the hell out of it, get some estimates on how much development and design is going to cost. Put yourself in the position where everything is ready to go, you just need money.

If you are going to the marketer, here is your first challenge. Market yourself and your plan to investors. Check out some successful kickstart projects to get an idea of what it takes to get people to part from their cash.

As my statement was not misleading, I am going to say it again an application has complete control of all money deposited to it until that money is withdrawn from the app, which is why the deposit UI shows that warning in every place a user can fund an app:



MoneyPot does allow you to deposit and withdraw from an app instantly, but when funds are on an app the app can do what ever it wants with it. That's how the app makes bets, and can tip accounts. The fact that the app doesn't have physical bitcoins on it is pretty irrelevant as it can just withdraw the money  (as the attacker did today).

So please do not try subvert the warnings MoneyPot gives, or misrepresent the relationship.

There's a good chance the person who exploited the site is reading this, so if you let us know how you did it before we figure it out I'll give you a nice bounty =)

Just include your address. Feel free to post the details publicly here, or email it to me: ryan@moneypot.com

Yes, hence the warnings =)

An app owner has full control of the money that is deposited on their app, until it's been pulled back into the moneypot account. We authenticate the app using an app_secret, so if this ever gets leaked the person can make any API request on behalf of the app

I'm not entirely sure they used the app_secret was leaked, unfortunately i don't log the authentication mechanism in the API. Either that, or all the betterbet's users access tokens.

I'm waiting on the betterbets developer to get online so we can discuss and finish piecing together exactly what happened

Hey sammyfoir,

I've been spending the last few hours trying to piece what happened. It appears to be that someone got betterbet's app_secret and used it to tip themselves and withdraw the money.

The list of affected users are:

http://privatepaste.com/download/0936bdb007

Edit: All affected users have been credited in full.

When ever you deposit money in an app, you give it control as you see in the standard warnings we issue:



that said, I am told betterbets have already sent the full amount to your main moneypot wallet in compensation.

The current max bet is 1 BTC, and max profit (per game) is almost 9 BTC.  (The max profit per game is dynamic, but the max bet is fixed)