Topic: Just-Dice.com : Invest in 1% House Edge Dice Game - page 49. (Read 435353 times)

I expect that doog will keep private any details of the mechanism for emergency return of funds which might undermine the security of that mechanism.  (Ideally it should be wrench-proof, but it's more important that it should deal thoroughly with practical possible adversities than that it should be theoretically perfect in contrived scenarios.)  However, having more confidence in that mechanism would allow some participants to increase their participation substantially.  

@Doog: Another point that I think needs discussing is the layers of security surrounding the cold wallet.
In particular, would it be possible to know how you plan to activate the dump to our emergency withdrawal addresses in the case something "extreme" were to happen to you?

Would it be possible to setup some sort of multi-signature transaction whereby the top 5-10 investors can combine, say, a hash of their 2FA passwords to activate the dump transaction? This way, if the site were suddenly shut down or all communication with Doog or Deb vanish, the top 5-10 investors can attempt to contact each other privately and construct the passphrase to activate the dump transaction remotely.

Obviously the dump transaction would consist in a redistribution of the cold wallet funds to all investors proportionately to their stake in the bankroll.

Like I said, this wouldn't be a measure to protect investors from the stealing of funds by Doog but simply to give investors a Plan B should greater forces interfere with the site.

Overall, like I've mentioned in previous posts, the site functioning should be structured around principles that incentivize long-term, patient, investing so that investors might relax and think of things to do with their time.  

One thing i can see is that, when the house took a big hit of taking away 70% then it cant go back. Thats the difference to throwing a dice because all results have the same chance there. If the max profit is only 30% of the previous then the chance to win 70% back is slim because it would need more losing bets because of the lower max profit. So it might be that stochastics dont work here perfectly because the chances arent even. But when throwing a coin it would have to work. I mean when the first results are 100 times head and then it moves on with head, number, head, number and so on then the theoretical result of head and number thrown the same time in average will become more and more near. If youre at 100000 throws then the 100 times from the first throws doesnt matter much because constantly the results go nearer to the theoretical value.

It might be that it doesnt work here because the chances arent fully even.


Right. But so it is with the JD house too. It is set together of many small investments that werent there from the start. So if you invest 1BTC in your account and it gains 20% profit, then you add 99BTC, then the profit of your 100BTC Investment is 0.2% only anymore. Simply because its 2 different investments and the biggest investment didnt have the time yet to earn so many profit. Might be that nakowa has a share in the 0.25% but the rising house has to have a share in it too.


Was the profit jump higher after october 7th nakowa too? If so then nakowa only should have taken the house from 6000BTC to 4500BTC only. Thats nothing were i would await that it leads instantly to the theoretical profit being 7.5 times higher than the real profit on october 21. I mean when i see the theoretical profit how could it develop that far away from the real profit? That doesnt fit to me yet. When it grows so high while in fact he did win way less at the end then the variance should be high since the profit only shows the 1%. Thats some extraordinary luck then.
But when i see the y-axis... i realize those are only the stats starting from the house value. It doesnt say how big the house was. That would be interesting to see to judge his luck.

Then... when i see the graph http://i.imgur.com/0Ikj2PC.png it seems that all bigger wins (cant judge what part of the house was taken) are doing an irreversible damage to the profit since the chance to win it back is lowered. Winning back the loss of half the house is only possible with much more luck than the winner had because you need more wins at max profit. That "luck" might already been eaten by wins with lower max profit. Max profit can eat 100BTC when the player wins at a house of 10000BTC. But it would need more than one bet at max profit to win these 100BTC back then. So the house is vulnerable to big losses because it cant grow back easily?

I wonder why the irreversible damage doesnt seem to apply to the new graph. Even when there werent so big losses it should have to have at least some effect isnt it?

The graph shows that nakowa played so much that in theory JD Incestors would have won 30,000BTC. And thats only 1% house edge. Wouldnt that mean he played bets worth 3,000,000BTC when they would have been even bets? And the variance of so many bets win-lose wasnt enough to stop this winning streak? Its hard to imagine.

---

And when nakowa wins again then the investors doesnt only have a big loss from nakowa but JD would think they made a fortune. 30.000BTC additional profit so the fee for all investors would be 3000BTC on top. It would be a really big hit to the investors on top.

Im uncertain if the profit can go as high again at all. Something along the line that you can win high with a high investment, but when your investment is shrinked very much you will struggle to win it back because you have less investment now. Bitcoins make bitcoins, without them you need way more luck. Though i cant explain it properly.


Im not sure what to think. When the small bets nowadays dont tend to go to the real theoretical profit why should another huge better tend to that line? His bets are exactly even again. He could be the next nakowa winning as he could be nakowa losing.
When he really won 10K, how big was the house at that time? And would it be possible at all to make the same in the opposite direction with the current size of the house and half kelly?

---

If you think that way you need something like kelly 0.01% instead kelly 0.5% as it is set now globally. Now you risk potentially a half percent of your investment with each bet. This is what whales are for. I didnt want to miss likashing because i won with him. And guys like him move the whole site. I doubt he would be happy playing millions of bets because the risked amount of investors became so small that the max profit is very small.
Though i dont know what to think about nakowa yet. I dont know how much of the house he really took. It shouldnt be too much when max profit is 1% of the house. And max profit gets lower everytime he wins. Maybe theres a graph for the house as y-axis too.

With regards to the fees, I'd rather have Doog take on the variance (current system) instead of passing it on to investors. However, I can understand if he would like to change the fee structure so his compensation is more stable.

Hello JD community.  I've been an investor and player at JD nearly since inception.  I've following the latest lively discussion regarding how to remunerate Doog for all his hard work.  First, I appreciate that Doog tries to discuss his thinking with the JD community.  The current system seems to incentivize Doog's sensitivity to investor issues.  The proposed system rightly rewards Doog with a more steady return for his NOW proven genius insight of allowing investors in the first place. For me, which-ever way Doog thinks would be best is aok.

Please keep up the great work Doog & Deb.  I very much enjoy Doge-Dice, JD poker and contests.  I'm not an artist, but everyone should take a look at the great submissions. Such fun!

Best regards: Player/Investor Gerty

We're back up.  Sorry for the worry!

Check here http://just-dice.blogspot.ca/2014/01/j-d-down.html for more updates on site issues.

Deb

Hey guys why don't we try to make a brainstorming of possible ways to reduce variance for investors but still attract new whales? I can think of the following:

- put an higher max bet but with higher edge (example: a whale can bet 2% of the bankroll but with 90% of expected return). Variance should be reduced by the higher edge. (bets smaller than 0.5% bankroll have still the 1% edge)
- every time you wage more than 100000BTC you get back a "special" bonus of 100BTC. Variance should be reduced by making someone willing to wage more.

ideas/comments?

Just-Dice is experiencing some technical difficulties. Doog is on it! Come chat on Doge-Dice for updates.

Deb

No, where would that higher return come from? You're not suggesting that your proposal would increase the expected return are you? That makes no sense.

It almost sounds like you are a victim of the gambler's fallacy when you say that the next lucky whale will likely lose. The likelihood of that happening is exactly the same as it was the first time.

What I am saying is that Nakowa had no way of losing 10k because he never had that kind of money walking in. He could only lose 10k after winning. The same is true with almost all players. On the other hand, any player with 100 BTC or so, can accumulate 10k if playing aggressively and with a bit of luck. The chance of that happening is small but not insignificant. The law of large numbers more or less guarantees that we never make much more than 1%, but a lucky player who manages to stop at the right time can bring us down significantly. This means that the actual expected return is less than 1%.

@Tucenaber

As one of the major investors, I completely agree with what you said. Not all volume is created equal. As an investor I would much prefer to see 1,000 btc wagered over 1 million bets rather than 5-10 bets, as the variance is sure to be much less. The proposed system misaligns management incentives with investor interests. Specifically, Dooglus will have a disproportionate incentive to expend resources attracting whales rather than on advertising likely to bring the smaller, more desirable traffic. Furthermore, the smaller traffic is more demanding for the server and therefore more expensive for the site owner, further putting management at odds with investors. I can't say whether the change would cause me to pull my support for the site, but I would have to carefully reconsider its merits as an investment.

@zipmaster: @"He will statistically lose next time."  fallacy much?

@Tucenaber: I do not agree.

If anything, switching to the new system would STATISTICALLY signify a higher return in the long term. We are currently at 0.25% profit. If another Nakowa walks onto the site, he will statistically lose next time. Eventually in the long term the true 1% house edge should be reached.

If I am right, investors would make a lot more money under the new system as opposed to the old since Doog just takes a cut of wagered and not of profits.

@Sebastian: Of the 5 million BTC wagered on this site, 1 million of them were wagered in one day by Nakowa. That's 20% of all wagered BTC on JD. You understand that the 1% house edge protects investors in the long run via the law of large numbers. Unfortunately for us, Nakowa was an isolated circumstance thus deviating reality from statistical expectations. If more Nakowa's were to play on the site, we could expect to reach the true 1% line in the forseeable future since Just like one managed to win 10k, another stands to lose 11k etc...

As one of the top ten investors I would be ok with, but not enthusiastic about the new proposed reward system. Provided that the fee already paid is credited first. It has the advantage of being simpler and also making it easier to plan for advertising, referral programs and the like. I also think that Dooglus can do whatever he thinks is right. If we as investors don't like it we can just divest.

Having said that, I have no problem understanding why several investors have been voting for the old system. The new system is worse for investors since we take on more risk with no increase in expected return. If you really want investors to prefer the new method, you have to offer an incentive.

The additional risk is also skewed downwards. There is almost zero chance of being well above the 1% line ever, since very few people even have 10k to lose, but there is a small but significant risk of a player being lucky and winning 10k again (maybe 5k is more reasonable now since the decrease in max winnings), that he potentially can walk away with like Nakowa did.

Since a whale can take his winnings and walk, the expected value for investors is in fact not 1% of wagered amount but smaller than that. So perhaps it is reasonable to offer to take a smaller cut with the new system?

What happened in the past does not affect the outcome of new bets. Nakowas wins caused the site to be a large amount below the expected profit and the expectation is that the site will stay that amount below the expected profit. A numbers example, suppose 1M coins wagered, 1% edge, but no profit was realized (so 0 profit). This is 10K coins below the expectation. Lets say that the next 1M wagered has the expected 10K profit. Total profit is now 10K, with 2M wagered. So the profit percentage has grown from 0% to 0.5%, coming closer to 1%, but the site is still 10K below its expected profit.

Now 8M more is wagered and again, the profit matches expectation: 80K. Total site profit is now 90K, which is 0.9%, already quite close to 1%. But the site is still 10K below the expected profit.

Profit-percentage is calculated by dividing profit by the amount wagered. There's is an important distinction between this notion of profit ("1% of the wagered amount was gained by investors!") and the notion of a profit on your investment ("My investment in JD has grown by 20%!"). Both statements between brackets are compatible with eachother, since they concern different notions of profit.

Where do you get that the house was down 70%? If I recall correctly, it was never that bad.

Winning 70 times in a row is very unlikely, yes. But Nakowa didn't do that. He lost many bets, but played so much that he edged out ahead. The nice chart by Peter R. shows the progression of Nakowas bets and the likelihood of this occurring. The conclusion is, and has always been, that Nakowa was quite lucky, but not extremely so. Rolling a pair of sixes on two six-sided dice has comparable odds, to give a point of reference.

Down for me too, after an apparently badly timed deposit!

The chart looks really bad. Though i dont understand why the chart at bitcoinproject is working so fine. I mean the real profit really is going tight to the expected profit. Its strange that in fact the real profit should be 4 times higher. There wasnt a glitch with nakowa? I mean even with the high bets of likashing its not moving to the real theoretical profit line and remains at the theoretical profit line that doesnt take nakowa into account. Till now i only knew the other graph and when i watch the theoretical profit and the practical profit it convinced me really that mathematics work fine there. But that it doesnt move to the real theoretical profit even after so many bets were made in the meanwhile is making me insecure... at least there should be a pressure to the top... but that the bets from nakowa are "forgotten" by luck... i dont know what to think.

Isnt the site profit calculated solely from profit divided through house? If so is it possible at all that it would be at 1%? I mean when the site profit would be calculated for the last month only then it would be very near to the 1%. But for example when JD runs with 1000 house and makes 10 profit it would mean 1% profit. But if you add 30000 to the house at that time you would have 0.00032% profit only. The fact that its already at 0.25% shows that the house grew slower. But it could only at 1% with much luck. If im wrong where is my error in thinking? I cant see the JD-Stats at the moment because cloud-flare is blocking me. I only can get a cached website without numbers. DDOS or is cloudflare blocking HMA-VPN standardized? Im using it because i use a relatively open WLAN here and i want to protect against being hacked from other wlan-users.


That would mean that your 3000 more fee would have been earned and taken from the investors that were invested while nakowa played. The fee would have hit the investors very hard additionally i guess. 70% house crash? Wouldnt the theoretical profit that high that any investment would be wiped out with the fee then? Probably even going into negative then...

The house was crashed 70%? And the graph Peter R posted supports that it was quite normal? At 1% kelly, which means nakowa only could win max 1% of the house at one bet means that he had to win more than 70 times at full profit. And that should be a quite normal happening? Of course im not a genius in math so i might think wrong here somehow...


Really? I thought implementing risk settings would be impossible because the server would have to calculate too much. I thought till now you calculate the profits only when divesting or weekly and that the profit shown at the website is some javascript running on the users computer that is guessing the profit from the bets.
Then the server load wouldnt be an argument against risk settings anymore too...


Right... i meant it more in terms of server load. The amount of calculations that have to be done will decrease to the point where no calculation will happen anymore. Until one of those investors come back.

I hope i can use JD with HMA again and its only temporary. But the nakowa thing, the luck he had to have and that the profit isnt moving back makes me think. I dont want to bring back a "maybe" old discussion but i dont see how thats possible. Practically winning 70 times in a row?

Afaik he's waiting since weeks to get the historical data from you to create complete charts....

It's still down from here? Anyone know anything going on?

Why is the site down?

Just as I made a deposit 

I also prefer the current method.


You can also offer that kind of deal now. But instead of giving a fixed 0.05%, you would just give 50% of whatever you win, once per month. In the long run it's the same.