Topic: Just-Dice.com : Invest in 1% House Edge Dice Game - page 130. (Read 435357 times)

Understanding variance is not in realm of possible for a vast majority of investors. All they want is easy and fast money but it doesn't really exist. They would have been better forgetting about JD for 2 years and then come back.

Oda, you are investing in a Bitcoin casino, you are not depositing your money in a bank with a fixed yearly interest. 1% max profit can obviously lead to high volatility, but its the best way to get mid and long term profits. If you fear your investors do not understand they are investing in a casino, or what 1% max profit means, you just add a big disclaimer explaining them what variance is, and where they are investing. You just cannot expect 100% yearly returns with no variance and no risk. Do we agree on that?

As nicolaennio said, we are roughly in the middle of that graph, and instead of letting math do its thing so it can go up again, we scared away our best customer, who is the only one that has the volume that allows to climb up again quickly, effectively condemning investors to stick with a huge loss for months.

Will investors be OK with MUCH lower yearly returns but less variance? Don't think so, because the counter-party risk (eg: trusting your BTC to a third party which happens to be an unregulated gambling site) is too high for such a low return. As soon as a trustworthy competitor copies J-D's "original" model, investors will flee (mark my words).

Finally, the very bad thing that happened here is that the change was made unilaterally while our best customer was playing, pissing him off and condemning investors to stick with a loss in the mid term. Changing the default 1% max profit might be a good thing, having a "market for risk" so everybody can decide their risk exposure is certainly an excellent idea, but the way in which yesterday this decision was taken is very counter-productive.

For fuck's sake, finally. Thanks for putting this sentiment into precise words. Fuck Kelly criterion. Fuck "trust the math". Investors are humans. Pretending they are not, pretending there's a zero chance that the site will receive irreparable damage in trust if the bankroll starts approaching new lows is delusional.

Maybe the sudden change from 1% to 0.25% was not the most wise action. But people need to pull their head out of their ass and stop pretending leaving it at 1% was THE ONLY MATHEMATICALLY CORRECT ACTION DO YOU EVEN LIFT BRO.

If you open a casino in your platonic realm of mathematical objects maybe that's justified. In the real world, you better take into account your fellow human's irrationality, because otherwise you're being irrational yourself.

"invested" isn't stored in the database.  It stores your percentage of the bankroll, and the bankroll.  So when someone loses a satoshi, you get your percentage share of that satoshi.

The bankroll is stored to 8 decimal places, and wins are rounded down to the satoshi.

As an example of why, if you bet 1 satoshi at 60% chance to win (and win), your profit should be a little less than 1 satoshi.  Rounding to the nearest satoshi would give 1 satoshi, giving the player a 60% chance of doubling up.  So we round down.

Here's raw data of 'bets minus nakowa', alongside the regular 'wagered.txt' list:


I selected "uid not in (2548,9075,31791,46591,113828,118977,119016,136175,143341,145625,150486,153338)".  That's a pretty complete list of his uids I think, though I may have missed some, or included some false positives.

The problem with this is that those investors aren't risking anything at all.  When the house needs coins to pay out a winner, these investors aren't going to give up their coins.  They may as well not even bother depositing their coins, because we're just going to put them into cold storage and then give them back, with interest.

My point is, what do such investors bring to the table?  What use are they?  We pay them (60% of 1% of some percentage of all bets) but in return they give us nothing.

A similar idea was suggested in the JD chat some months ago, I think by uvw, but maybe by somebody else.  He proposed having 3 tiers of investors, the regular ones we have now, the "coward" tier (like the ones you propose) and the "hero" tier, who take the 40% 'risk cut' that the cowards don't get, but also pay the cowards when the house loses.  This idea fails for the same reason - the 'cowards' are effectively parasitical at the expense of the 'heroes'.  They get a guaranteed return on their investment, but we're not free to use their investment for anything because we've promised never to lose it.

Some say that he'd have lost more.

So...gambler wins lots of money on a house edged game and you tell him to stop betting big after he's won lots of your money?

You do realize your running a casino; not a retirement fund?

Cheating has been taken into consideration, and Nakowa's betting "system" has been thoroughly analyzed. There's absolute NO evidence on him cheating, and on the contrary so far EVERYTHING indicates that he just got lucky.

I've just checked JD chat logs, and I see that Dooglus acknowledged math was on our side, but "we couldn't count on Nakowa coming back after winning".

When somebody pointed out that gamblers always come back, Dooglus said that "Nakowa already walked away with 4k in August".

Wow. Just wow.

First, is obvious that Nakowa didn't "walk": he just stopped for a few weeks and CAME BACK. For Christ sake, he was playing yesterday, and you say" he walked in August"? Doog, as a casino operator you should know how gamblers minds work. They might stop for some weeks after a big win, but the more they win the more they will think their strategy is unbeatable. Nakowa will come back to "prove his points", until variance hits him hard, and at that point he will lose everything very fast because he thinks he can "spot patterns", and he won't be able to accept he might lose.

As a poker player I've seen that story over and over and over.

IMO a very poor management decision was made based on irrational fears and highly unlikely scenarios, I'd dare to say totally negligible scenarios from a statistical point of view. And we all know our business is based on probability, right?

The hard cold facts:

We have pissed off and treated badly our best customer ever

Investors who did not divest are left holding huge losses that won't be covered for months

If dumb investors do not understand that a casino is NOT a "stable" investment, you just add a big disclaimer explaining what the word "variance" means. You do not screw up with your best customer who, additionally, is giving a HUGE promo to your site, condemning most of your investor to remain holding a bag indefinitely.

Well, just no. If you played at 49,5%, statistically and probably your next roll will be a loss 50,5% of the Time. Like every roll of you was before. There is no "memory". Whats so hard to understand about that?

Statistik and probability like this only work when you look in the future.

The probability for loosing 6 Times in a row is ~1,65%, but if you already have lost 6 Times in a row it is still ~1,65% for loosing even another 6 Times in a row.

If we were in the world of mathematics I would have no doubt that keeping the max bet to 0.01% is the only choice. But in the real world people have, blame them, feelings. Now I run some simulations with a player who is playing at max bet, lets say that three over four of them are satisfactory, in those cases we are making money with no problems, but one over four is more or less like that (initial capital is one bitcoin):



Now think we find ourselves in the middle, the heavy gambler has taken half of our pot. I want to know how many people will believe in mathematics (waiting for the inevitable return to gaining) and how many people (normal people having emotions and knowing that the world is not mathematics) will think that actually the website is scam.

Our best customer has 1/4 probability (lets say, I would like to have a more precise computation) to make us think the website is scam. So mathematically, taking into account that humans are complex, that is actually not our best customer.

[/quote]

 No matter what, if you play long enough you will end up losing simply because this game is -EV for you and +EV for the house.

Don't you guys grasp these simple concepts?
[/quote]

I am aware of this obvious maths/fact. and many more are aware of this fact. but there is still an option that a casino with house edge can lose. it is called "Cheat" and it looks like that You guys just push this option to far away aside. it doesnt hurt to take this also in consideration.

Wow. You simply do not understand gamblers fallacy and you created a "gambling security"? Please tell me that nobody ever invested in it.

Secondly, you cannot explain the patterns you spot because there are no patterns. No matter what, if you play long enough you will end up losing simply because this game is -EV for you and +EV for the house.

Don't you guys grasp these simple concepts?

But ... any other player or investor playing the game would be at a different point in the sequence than I would. I'm essentially selling my seeds: the combination of the current server seed, the client seed, and the next nonce is going to win.

I just lost 6 times in a row, and statistically, and probably, the next roll will not be a loss but a win.


What I'm saying about my own account is roll number 1334414 is going to win.

Anyways, I've stopped begging a long time ago. The ones who "invested" in my "gambling security" did it willingly and contacted me directly. No one ever contributed to the "Group Bet" I started even earlier.


As for the security of SHA2 (SHA256 to SHA512) and HMAC. They are useful for the purpose they were created and designed for. As a message digest that can reliably be used to confirm that the original input is exactly the same.

I have a slight suggestion for dooglus in the way he implements the lucky number generation. Use the nonce twice. Once as concatenated input to the "secret", and second as concatenated input to the "message".

If he does this, I've also lost my winning roll #1334414, which previously had a max bet of 3,910.93 at the chance I play (87.7779%). Now I'm limited to a max bet of 642.33.

This can be proven, if indeed dooglus changes the implementation of lucky.txt, by getting my prior server seed, the client seed, and the specific nonce 1334414, and if I would have won that roll or not using the original implementation. However, I'll not get to claim the win cuz it won't happen. (oh sure, I can restart my run for another 1.3 million rolls, maybe ...)

Site owner has much more to lose than average investor, so his decision is understandable. Besides, there is always a non-zero probability of a bug or a server hack, and whale's win streak odds were quickly approaching that limit.

And that's how Nakowa/allover/celeste graph would have looked like. But, you know, we decided to send him off for good when he was 13k BTC ahead because some investors do not understand what math and variance are, and what does it means to invest in a Casino.


And exactly this this is what I've been screaming in the J-D chat since the moment Doog changed the max profit - and now we hear it from the whale's own mouth.

I guess the fact that Dooglus was fully divested means that for some reason I personally cannot understand he "lost faith" in math, and he gave up on the idea of recouping losses from Nakowa. Maybe that was one of the reasons he decided to mess up with our best customer ever, while crippling the expected profits by taking max profit to 1/4 while investors that didn't divest were on HEAVY losses.

Yes, given your way of defining D (I was defining D as what you actually sign over to JD, which is why my logic table looks a bit different).  But maybe your first line is wrong:

"If I increase D, holding R and M fixed, then P stays constant, V stays constant (if bet sizes grow relative to it, which won't be the case, so it probably decreases) and C goes up."

Wouldn't increasing D lead to an increased share of the pie and thus larger values of P, V and C?

When you talk about increasing M, I noticed that you are applying the Kelly Criterion based on the sum of the reserves held at JD and those held offsite (your definition of "D").  So, I think proves that it is OK--and in fact necessary for profit maximization--to have a max bet % greater than 1% of the reserves held at JD (i.e., the number dispayed on the website).  It's OK because there are other reserves that will flow into JD should the ones currently controlled by JD become depleted.  

And participants who "fakes reserves" by more than a factor of 2, like you point out, would be expected to burn through the smaller amount of funds that actually possess and bust.  Darwinism.  

Thanks for the further info on the Kelly criterion, BTW!

Depends on exactly what you define D, R, P and V to be.

IF:

D= total deposit, including coins held offsite
R= deposit reserve ratio where R is % of coins held AT justice
P= expected investment growth rate
V= investment variance

THEN:

If I increase D, holding R and M fixed, then P stays constant, V stays constant (if bet sizes grow relative to it, which won't be the case, so it probably decreases) and C goes up.

If I decrease R, holding D and M fixed, then P and V stay constant (given I don't get auto-divested), but C decreases (fewer coins help at JD).

If I increase M, holding D and R fixed, then (due to Kelly Criterion):
       a) if M increases and is <1%, then P and V go up, and C stays fixed.
       b) if M > 1% and increases, then P goes down, V goes up and C stays fixed.
       c) if M < 1% and increases to >1%, then P can increase or decrease depending on the exact numbers, V goes up, and C stays fixed.

Does that make sense? If you change the definitions of DRP&V I'll redo it.

Sometimes I come to this thread just to hear the angry cries of the villagers that thought it was/is riskless to invest in a casino.

/duck