I'm not quite sure this is true, from app developers perspective the new scheme is both simpler and more profitable.


ELI5: You get 50% of the house edge of all bets  (the only exception to this is when this would be too risky for investors otherwise).


ELI5: You get paid more than last time

Set the max-profit to 0.5% of the bankroll, and you'll be guaranteed half the house edge (more than before).

Well, I'm a person and sole owner of MoneyPot S.L.R. which is the sole owner and bankroller of bustabit.

Honestly, one reason I prefer not to take investments for BaB is it keeps the stress and pressure down. I have reoccurring nightmares about some sort of vulnerability/hack/breach and the less money in the bankroll, the less liability I have.

Actually, that's why I'm conceptually a lot more inclined to want to do something like equity shares. That makes everything a lot more obvious/fair in the disaster cases. I spent a while looking at colored coins, but the secondary markets for them aren't great.

Thanks NLNico for taking the time to explain it. I'll try put it in different words for JackpotRacer. But let me prefix it with by saying for his case the new scheme is strictly better. There are no circumstances when he'll get paid less than before.

Bets can be put in three categories:

* Less than 0.5x kelly bets

In the binary-bet case (e.g. dice), this is when someone makes a bet trying to win less than half of the house edge of the bankroll. So with a 3% house edge, if the player is trying to win less than 1.5% of the bankroll.

* 0.5x to 1x kelly bets

In the binary-bet case (e.g. dice), this is when someone makes a bet trying to win between 0.5 and 1  of the house edge of the bankroll. So with a 3% house edge, if the player is trying to win between 1.5% and 3% of the bankroll.

* Greater than 1x kelly bets

In the binary-bet case (e.g. dice), this is when someone makes a bet trying to win between >1 of the bankroll. So with a 3% house edge, if the player is trying to win between >3% of the bankroll.


The commissions for placing each type of these bets vary:
* Less than 0.5x kelly bets

Apps get 50% of the house edge on every bet (regardless of outcome). This is paid by investors.

* 0.5 to 1x kelly bets

Apps get between 50% of the house edge, and 0% of the house edge. This is paid by investors in such a way that keeps them kelly-complaint. By default you can make this type of bet, but it has been requested to have an option in the admin panel to block them. As long as you don't offer a rakeback on these type of bets, accepting them is perfectly safe, though.

* >1 x kelly bets

Apps PAY investors to accept the bet.  (Don't worry, you can't accidentally do this type of bet, it can only be done via a confidential app using an explicitly subsidy argument).

--

We think in general apps should restrict themselves to <= 0.5x kelly bets, where their earnings are very predictable: 50% of the house edge. For promotional purposes they might want to use the other sort of bets, or to satisfy whales, or sell magical bet tokens etc.


While at a glance the investor:app profit split might look worse for app developers, to be clear MoneyPot was taking a 50% profit split on app developers earnings, now it takes 0%! All commissions will be taken from investor profits instead.

This makes it strictly better for app developers, although the story for investors is a bit mixed. Most bets investors will now make more, but there are circumstances when they may make less.

Well, it's really just different. It'll be better or worse depending on how player bet (and win). Sometimes now investors will be getting more than they used to (small bets) and sometimes investors will be getting the same as they used to (big bets, but having to pay 10% on profits).

But all round I think it's quite positive and investors are happy with this change though. Their volatility should be considerably less now, as instead of giving a constant commission to apps and MP they only give a constant commission to apps, and MP only takes a cut on the investor profit.

BIG UPDATE TO APP AND INVESTOR PROFIT SHARING

Apps will now get a constant 50% of the house edge, with investors continuing to take the risk. This means investors give the extra 50% of the house edge to apps. MoneyPot will no longer take a commission from apps, but now a 10% commission on all investor profit (but not on app-profit).

If apps place a bet that would put investors (after giving the commissions) at greater than 1x kelly risk the investors will give apps a reduced commission to limit themselves to a kelly. For this reason, we recommend apps limit bets to a half kelly for predictable earnings. If there is demand, we can create a setting in the admin panel for apps to reject any bet that exceeds this risk.

The 10% investor profit commission will kick in once the investor profits reach their previous high levels (~50 BTC). From then on, investors will pay a 10% fee on further profits, this will be deducted upon divestment, or on a monthly basis.

I think the lower the max win actually decreases the amount the site makes, but it's really quite essential to avoid bankruptcy.

Sorry, I just paid the people who "bustabit" picked, but reading the OP:

"Bustabit is hosting a contest to see who can make the coolest illustration in their profit chart! Use your wins, losses, and differences in amounts and cashout points to create cool drawings and pictures!"

and the intention of the competition was to get people to actually play the game, and play around with controlling their loses and wins, so it seems fair it went to BertBot  (although btw it was a great find)

I guess your sig pays per post? 

Not content with a mere 25 BTC in profit, phoking has come back and taken his spot on the all-time leaderboard!



That's 41 BTC of net profit, over only 159 BTC wagered.

 

Yeah, I just moved as part of my on-going campaign to reconcile my finances.  If anyone does however know the private key, (or any of the future hashes) I'll more than happily give you a lot more than that in a bounty (i offered Dexon 50 btc if his brute force sha1 program actually does find it) as if someone did find it, it'd easily cost me more than that (as they would know the outcome of all future games)

Quite a nice win we had today:



Turned a 24.7 BTC profit on his 7 btc deposit, and ended the day by withdrawing the whole thing cleanly (only confirming the deposit/withdrawal details, as phoking already shared this on the chat).

Well played.

Sorry about that, wasn't checking on the thread. (Just send me a PM next time)

I credited the two accounts 50k each for the delayed payment

The chat is now hugely improved. Multi-lingual people can subscribe to multiple channels at once, and see the unread messages of each.  And now all users are assigned a random color!

I don't have any experience with that sort of thing, but I imagine one the biggest difference is their costs are astronomically higher. Everything about running a physical casino seems expensive.


It's really a 1 person job, the most time-consuming part of the whole thing is just stopping the chat degenerating too far, and helping the newbies out. Every 2 to 3 months, there tends to be a sever/datacenter hiccup that needs emergency looking into it. And other than that, there's just some part-time improvements we do.

Thanks, I think we've fixed the issue. Can you give it a try and confirm?

Well done! It's great to seeing new interesting stuff like this!

The expected value is the same, but the risk profiles are very very different.

One way to think about it, is if I backed the first bet, I'd have a 50.5% chance of losing all 100 BTC. But if I backed the second bet (sequence), I'd have a 1 in 3463220205675693173883308162838 of losing my full 100 BTC. With the second option, the majority of the time the outcomes are going to live a lot closer to the mean than the single-bet. There's a lot of resources available online to analyze this further, I've found some really good youtube videos that explain the study of variance quite well.


If my example wasn't extreme enough, lets push it a bit further:

1 bet of 10000 BTC @ 2x with 49.5% win chance
vs
1000000 bets of 0.01 BTC @ 2x with 49.5% win chance


Assuming I had 10,000 BTC (which I don't), I'd rather start sacrificing body parts before I backed that bet.  The second option, however, has me salivating just thinking about it.

Ahh, I think that's what he's asking. Investors will definitely make a lot more on a small large bet, then lots of little bets. The reason for this is simple though, take an extreme example:

1 bet of 100 BTC @ 2x with 49.5% win chance
vs
100 bets of 1 BTC @ 2x with 49.5% win chance

they both have the same EV for investors (+1 btc) but in terms of risk profile, they're in different leagues. Ask me to back the first bet, and I'd wince and refuse. Ask me to back the latter sequence, and I'd have trouble concealing my boner.


So I think it's makes a lot of sense the way it is now. The more you risk, the more you need to be compensated to take it.

Hmm, I don't understand your example? The risk to the investors on the first one is 100x as much. For instance, imagine you had a 1% house edge, and a 100 BTC bankroll. Taking the first bet would be absolutely insane (hugely negative expected bankroll growth), while the second bet is a great deal!

Or put another way, if you had a 100 BTC bankroll you'd probably want a 10% house edge to even consider the first bet. But you'd happily take the second bet even if it had a 0.1% house edge. This is exactly how MoneyPot works, investors are rewarded proportional to their risk in such a way that ensures every bet they accept is "kelly compliant"

Fair point. One thing worth noting, is that there's no rule in place like: "App Developers make x3 as much as investors", but rather investors share is proportional to the risk taken. For instance if someone is going for max-profit, then investors make 100% (and apps 0%), which app developers routinely complain about =) But if someone is going for a very small win compared to the max-win, the bets constitute very little risk to the investors so I think it makes sense to give that extra money to app-developers.

What I'll do is do is create a dedicated thread for the investing side of MoneyPot, so we can keep the conversation going there as I'm totally open to changing the system, but just getting tired of hearing complaints from both sides in something that is inherently zero-sum =)