Topic: Just-Dice.com : Invest in 1% House Edge Dice Game - page 149. (Read 435353 times)

So far it doesn't look like nakowa is "robbing" anything, he is just gambling and winning big time, it doesn't look like he is cheating at all. I've seen much crazier variance playing poker, what nakowa is doing is unlikely but certainly possible.

Plus, all the profit is already gone, the site and its investors are actually losing money. Nothing too strange, nakowa has a huge bankroll compared to the average player, and with just a 1% house edge variance can really make the house to lose a lot of money to a nakowa-like whale.

Nevertheless, he just needs to keep playing.

This is not a concern, as soon as you divest your money is no longer being used as bankroll. Just because you cannot withdraw a large amount of coin immediately does NOT mean you can't pull all your investment whenever you want.

Ok, but how do I know how much is the "enough" funds

What if earth is flat ?
Nakowa already won all the profit, since site is in -.


JD investment is liquid. That means you can divest at the very moment you like, and withdraw if hot wallet has enough funds (refilled daily).

What if Nakowa will robb all the profit from JD? Can we withdraw all of our money?

thanks for taking the time to explain. so we agree that to overcome 1% house edge is only possible by luck or by cheating 

As an investor I don't blame you at all, this is just variance. I agree that the 1% house edge is low but its also good to attract more players. I would change nothing, nakowa's winnings are driven by variance and math and nakowa's personality ("I'm not a gambler", I found a flaw") are definitely on our side, and he is doing quite a lot of advertising for JD.

Plus, this was never a risk free investment as some implied just a few weeks ago. We need very big numbers, just that.

for those who like to run simulations, make a play with a player always betting max bet
and starting with 2x the size of the site's bankroll. This is very close to the actual situation.

You'll see that we haven't seen the worst yet, but the player will lose all eventually

 I don't think dooglus owes anyone anything , I mean this is something investors take responsibility , what if someone lost 10k? would dooglus win all 10k? no , investors would.

 So basically , if you invest , don't expect wins all the time.

"The odds were in his favour," was a bad choice of words on my part.  What I meant was that given the strategy I assumed (either he reaches his target profit and stops, or he goes bust), that he was more likely to succeed than to fail during his session as cake [uid = 145625].  The expectation value of his profit is still negative, because when he does lose, although it happens less frequently, the loss is bigger in a way that more than cancels out his expected winnings. 

For instance, if he could bet his assumed 10,000 BTC bankroll on a 1.5X payout, he would earn a profit of 5,000 BTC 66% of the time (most the time), but 34% of the time he will lose all his 10,000 BTC.  In this case, his expected profit would be 0.66 * 5000 - 0.34 * 10000 = -100 BTC, even though he would be more likely to win than to lose.  He is applying a similar strategy, just over many bets instead of a single bet (and thus at worse odds than the single "bold move" I just described).

The fun part is that we don't actually know how much he is willing to lose in order to win! 

sorry but I dont understand how a gambler can have an advantage or turn the odds in his favor over an house edge of 1% by risking 10,000 to win 5000. sure that he can win and get his target before losing his bankroll. but how were the odds in his favor? would You be so kind and explain this to me? there are many games out there in many casinos (not online) with less than 1% edge games. interesting what I missed here.

thanks

A player can randomize their server seed as often as they want. Blocks are many minutes apart most of the time. So the blockchain is not a good source of entropy in this case.

Even if you combine the blockchain with another source of entropy (the servers RNG for example), the quality of your random number generation is still limited by your second source, because if that contains a flaw, players have the option to gamble quickly enough and place many bets before the next block is found.

I honestly doubt that the source of entropy used is insufficient in the case of J-D, but even if it is and it needs to be replaced, it needs to be replaced with something that doesn't have a rate-limit like the blockchain.

That actually seems like a pretty good idea. What you think Doog?  Would that be difficult to implement?

why not pull random numbers for use as (additional random to) server seed from the blockchain (every block or so)?

make absolutely sure the server seed is/was safe. Is it an option to change it?

There are only 3 potential security vulnerabilities I can see.  They should be addressed as best as is practical to improve investor and player confidence in the site.

1. The OP is running a scam to fleece investors.  Since there is no way to blind the OP to the server seed, investors have no choice but to assume the OP is honest or to not invest.

2. The implentation of the RNG is flawed.  It would be useful to review the RNG implementation and see if there are any possible ways to further ensure the results are truly random with no predictable pattern.
Some ideas:
A. Perhaps re-randomizing the server seed every roll would help since nonce would always be 1. If this was too taxing on the server, then only re-randomize the server seed every roll for bets over a certain size (eg. 10 BTC).
B. Pull random numbers from a hardware RNG.  Could use a site such as random.org (my concern though would be trading the risk of the OP for risk of a new entity).  This random # would be used as the server seed (cannot be used as the seed # for hashing since would not be provably fair).

3. Security of the server.  This is the biggest concern since hosted on a cloud VPS.  A VPS employee could be compromising the server directly.  Solution would be to move to a dedicated server for increased secured which would seem prudent (even if this is not the issue with nakowa's improbable wins).  Millions of USD worth of bitcoins are at stake afterall.

Here are the results of a simulation, running off the Nakowa = cake bet sizes and odds from Monday morning (PST) that dooglus posted.  The simulation assumes that Nakowa will stop betting as soon as he either hits his target profit (which I took as what he actually won as cake [4494 BTC]) or he busts at -10,000 BTC. 

So, unless I made a mistake, the odds were in his favour (just considering his session as cake):



In case anyone wants to check my methodology, I just modified the Mathematica code I described earlier in this thread to get the bet size by looping through cake's actual bets.  (Cake always went for 2X, so only the bet size was needed).

In[249]:=

betResult := If[Random[] > .505, 1, -1]
betSize := cakeData[[1 + Mod[Length[cakeData] - 1, ++j], 6]]

bankroll = 10000 * 10^8;
target := 4494 * 10^8;
n = 10000;

profitLogs = Table[{}, {n}];

outcomes =
 
  Table[profit = 0;
   
    While[profit < target && bankroll > -profit,
      profit += betSize * betResult;
      profitLogs[[k]] = Append[profitLogs[[k]], profit]];
   
    If[profit >= target, 1., 0],
      
      {k, 1, n}];

100 * (Plus @@ outcomes)/ n

Out[256]= 51.72

This is not true, any new investors will pay commision on their profits.

Last time this happened many new investors came to the site and dooglus was earning commision even though some investors were still negative.

Remember its commission on profits, so a new investor who has not taken a loss will be generating profits.

He doesn't win every time he plays.  He's had losing sessions in the past.  He's got a big bankroll and isn't afraid to risk it.  This is normal variance for low-edge, even-money bets between 100 and 500 BTC.

Are you prepared to do this for the long haul?  After this big loss, you might not be earning commissions for awhile.