Just-Dice.com : now with added CLAMs : Play or Invest - page 140.

KGambler

full member

Activity: 236

Merit: 100

Quote from: mechs on October 01, 2013, 08:28:34 PM

Quote from: KGambler on October 01, 2013, 08:14:59 PM

Quote from: dooglus on October 01, 2013, 05:56:53 PM

2) if expectation is to lose 25k, to beat it by 40k you 'only' have to win 15k. That means winning 5018 times (5018*400 - 4982*400 = 14400)

Using your numbers, he would be about 8.8% to win 15K or more. That's based on his winning 5018+ out of 10K rolls, with a 49.5% chance to win each individual roll. That's obviously in the realm of possibility.

My understanding is that he has played more than 10K rolls, but the fact he varies his bet sizes adds more variance. As someone already mentioned, some kind of Monte Carlo simulation would be the best way to figure out how probable his win streak is.

We could run his exact same bets through a simulator 10,000 times and see how often his results are met or exceeded. No mathetmatical proofs required

I have read about this situation, but I have never even been to the just-dice site so I don't know how the games work. Is he always betting the "even money" type propositions? Can you bet stuff like "over 25.5" at close to 3-1 odds? How many different types of bets can he make? If he's betting a wide variety of propositions, setting up an exact simulation of his bets is not going to be easy.

If someone does go through the trouble, they should obviously run it more than 10K times.

mechs

full member

Activity: 210

Merit: 100

Quote from: KGambler on October 01, 2013, 08:14:59 PM

Quote from: dooglus on October 01, 2013, 05:56:53 PM

2) if expectation is to lose 25k, to beat it by 40k you 'only' have to win 15k. That means winning 5018 times (5018*400 - 4982*400 = 14400)

Using your numbers, he would be about 8.8% to win 15K or more. That's based on his winning 5018+ out of 10K rolls, with a 49.5% chance to win each individual roll. That's obviously in the realm of possibility.

My understanding is that he has played more than 10K rolls, but the fact he varies his bet sizes adds more variance. As someone already mentioned, some kind of Monte Carlo simulation would be the best way to figure out how probable his win streak is.

We could run his exact same bets through a simulator 10,000 times and see how often his results are met or exceeded. No mathetmatical proofs required

KGambler

full member

Activity: 236

Merit: 100

Quote from: dooglus on October 01, 2013, 05:56:53 PM

2) if expectation is to lose 25k, to beat it by 40k you 'only' have to win 15k. That means winning 5018 times (5018*400 - 4982*400 = 14400)

Using your numbers, he would be about 8.8% to win 15K or more. That's based on his winning 5018+ out of 10K rolls, with a 49.5% chance to win each individual roll. That's obviously in the realm of possibility.

My understanding is that he has played more than 10K rolls, but the fact he varies his bet sizes adds more variance. As someone already mentioned, some kind of Monte Carlo simulation would be the best way to figure out how probable his win streak is.

mechs

full member

Activity: 210

Merit: 100

Quote from: dooglus on October 01, 2013, 05:56:53 PM

Quote from: grau on October 01, 2013, 05:10:57 PM

How lucky one needs to be to beat expectation with 40k?

Lets bet 400 on >50.5 say 10000 times. We only need to win 5100 times, right?

No, I don't think that is right, for two reasons:

1) if you win 5100 times, you win (5100*400)-(4900*400) = 80k. Not 40k. So you're off by a factor of two there

2) if expectation is to lose 25k, to beat it by 40k you 'only' have to win 15k. That means winning 5018 times (5018*400 - 4982*400 = 14400)

So you 'only' need to win 5018 and lose 4982 to be around 40k better than expectation.

How likely is that to occur? Is that the 2.1% on that histogram we always see posted?

saif313

member

Activity: 84

Merit: 10

I set up a account and sed payment from inputs but now my account no working mean I am giving username and password but not working why

dooglus

legendary

Activity: 2940

Merit: 1333

Quote from: grau on October 01, 2013, 05:10:57 PM

How lucky one needs to be to beat expectation with 40k?

Lets bet 400 on >50.5 say 10000 times. We only need to win 5100 times, right?

No, I don't think that is right, for two reasons:

1) if you win 5100 times, you win (5100*400)-(4900*400) = 80k. Not 40k. So you're off by a factor of two there

2) if expectation is to lose 25k, to beat it by 40k you 'only' have to win 15k. That means winning 5018 times (5018*400 - 4982*400 = 14400)

So you 'only' need to win 5018 and lose 4982 to be around 40k better than expectation.

dooglus

legendary

Activity: 2940

Merit: 1333

Quote from: grau on October 01, 2013, 03:37:32 PM

A wagered amount of 3,773,544 with a house edge of 1% implies expectation of bank profit of 37,735.
JD is 40,000 below expectation. That is 3-4 Nakowas. Rather strange.

So, I think it's only one nakowa.

See, he has a profit of 15k but should have a loss of ~25k. That means he's doing 40k better than he should be. It's only 3-4 nakowas if you expect them to break even. And why would you?

* the 25k is a complete guess; I didn't add up how much he bet on all his accounts, but you get the idea

grau

hero member

Activity: 836

Merit: 1030

bits of proof

Quote from: mechs on October 01, 2013, 03:45:02 PM

Quote from: grau on October 01, 2013, 03:37:32 PM

A wagered amount of 3,773,544 with a house edge of 1% implies expectation of bank profit of 37,735.
JD is 40,000 below expectation. That is 3-4 Nakowas. Rather strange.

Everyone agrees the results are improbable. The question is is it so improbable it more likely their a fault with the RNG or a cheat of some sort.

How lucky one needs to be to beat expectation with 40k?

Lets bet 400 on >50.5 say 10000 times. We only need to win 5100 times, right?

How probable is that?

Mathematica says:
In[82]:= Probability[x > 5100, x \[Distributed] BinomialDistribution[10000, 0.495]]
Out[82]= 0.00130587719594

That is 0.1% probability.

Probability of such win gets much lower than that if you bet smaller amounts or more often than 10000.

Most bets on JD were significantly less than 400 and number of bets is magnitudes higher than 10000.

No way is this within "probable".

solex

legendary

Activity: 1078

Merit: 1006

100 satoshis -> ISO code

Quote from: grau on October 01, 2013, 03:37:32 PM

A wagered amount of 3,773,544 with a house edge of 1% implies expectation of bank profit of 37,735.
JD is 40,000 below expectation. That is 3-4 Nakowas. Rather strange.

Indeed. And then consider the investor bankroll is approximately the same as 37,735, and that JD has only been running a few months. So, is it reasonable for all the investors to expect a 100% profit after a few months?

Or is it more likely that there is not 37,735 of coins in gamblers hands which are prepared to be lost and the 3.7m total turnover is much higher than would be supported by organic growth because for some reason the 1% edge is not having the expected effect of draining the supply of gambler coins.

mechs

full member

Activity: 210

Merit: 100

Quote from: grau on October 01, 2013, 03:37:32 PM

A wagered amount of 3,773,544 with a house edge of 1% implies expectation of bank profit of 37,735.
JD is 40,000 below expectation. That is 3-4 Nakowas. Rather strange.

Everyone agrees the results are improbable. The question is is it so improbable it more likely their a fault with the RNG or a cheat of some sort.

grau

hero member

Activity: 836

Merit: 1030

bits of proof

A wagered amount of 3,773,544 with a house edge of 1% implies expectation of bank profit of 37,735.
JD is 40,000 below expectation. That is 3-4 Nakowas. Rather strange.

mechs

full member

Activity: 210

Merit: 100

Quote from: markjamrobin on October 01, 2013, 02:11:14 PM

Quote from: mechs on October 01, 2013, 01:25:33 PM

Quote from: herzmeister on October 01, 2013, 01:21:49 PM

ok so, I think, in short, if he also invests, i.e. has a strategy for gambling and investing, and also has more coins than the house, then he practically controls it. He almost is the house then. But usually houses don't gamble. No wonder why people feel cheated.

Also because he gambles large amounts himself he has first-hand (or rather zeroth hand) knowledge about when to invest or divest, i.e. can react the fastest. It's like insider trading.

He fully divests when he gambles I think.

The problem is the "low bet" action which tends to me profitable he dilutes other investors out to capture. His "high bet" action when he invariably wins at he is divested for and the passive investors see their exposure increase up to 50% between him divesting and active investors divesting.

If he has 50% bankroll, does that change the house edge for him.

No but it has seemed to allow him to overcome the house edge through a mixture of persistent and variance. You would think you could replicate the results with someone with a 1000 bankroll and making bets of 20, as the same with him with a 10,000 bankroll making bets of 200. However, whenever I try to do so in simulators, I cannot get Nakowa's results even after 100 trials. Of course at 1000 and 10000 trials we can get the same results, so his results not impossible but merely improbable. The problem is it is enough to raise serious doubts for investors. Especially since other sites such as coinroll.it and primedice are profitable with a serious edge. However, they do have a much smaller max bet (3 BTC and 40 BTC respectively).
I honestly do not understand it from a mathematical perspective beyond saying improbable occurances occur often at J-D, especially for Nakowa. The site's user luck of 10.49% after 150million bets is also strange since it is an effective historical house edge of 0.51%. I know Doog made some changes to the RNG to try and increase the randomness though arguable it the changes should make little difference. If there a entropy problem, it is in the server seeds and not the algorithm.

markjamrobin

sr. member

Activity: 462

Merit: 250

Quote from: mechs on October 01, 2013, 01:25:33 PM

Quote from: herzmeister on October 01, 2013, 01:21:49 PM

ok so, I think, in short, if he also invests, i.e. has a strategy for gambling and investing, and also has more coins than the house, then he practically controls it. He almost is the house then. But usually houses don't gamble. No wonder why people feel cheated.

Also because he gambles large amounts himself he has first-hand (or rather zeroth hand) knowledge about when to invest or divest, i.e. can react the fastest. It's like insider trading.

He fully divests when he gambles I think.

The problem is the "low bet" action which tends to me profitable he dilutes other investors out to capture. His "high bet" action when he invariably wins at he is divested for and the passive investors see their exposure increase up to 50% between him divesting and active investors divesting.

If he has 50% bankroll, does that change the house edge for him.

mechs

full member

Activity: 210

Merit: 100

Quote from: herzmeister on October 01, 2013, 01:21:49 PM

ok so, I think, in short, if he also invests, i.e. has a strategy for gambling and investing, and also has more coins than the house, then he practically controls it. He almost is the house then. But usually houses don't gamble. No wonder why people feel cheated.

Also because he gambles large amounts himself he has first-hand (or rather zeroth hand) knowledge about when to invest or divest, i.e. can react the fastest. It's like insider trading.

He fully divests when he gambles I think.

The problem is the "low bet" action which tends to me profitable he dilutes other investors out to capture. His "high bet" action when he invariably wins at he is divested for and the passive investors see their exposure increase up to 50% between him divesting and active investors divesting.

herzmeister

legendary

Activity: 1764

Merit: 1007

ok so, I think, in short, if he also invests, i.e. has a strategy for gambling and investing, and also has more coins than the house, then he practically controls it. He almost is the house then. But usually houses don't gamble. No wonder why people feel cheated.

Also because he gambles large amounts himself he has first-hand (or rather zeroth hand) knowledge about when to invest or divest, i.e. can react the fastest. It's like insider trading.

mechs

full member

Activity: 210

Merit: 100

Quote from: DiamondCardz on October 01, 2013, 01:08:48 PM

Quote from: nimda on September 30, 2013, 08:53:52 PM

Did anyone try to explain to nakowa that the site can't possibly be rigged at all because math?

This guy is rich, so his logic is "Fuck logic".

I wonder if his system would continue to work with a smaller max bet. I mean he ran a bot for 12 hrs straight mostly spamming max bets and wound up 2000 ahead.

DiamondCardz

legendary

Activity: 1134

Merit: 1118

Quote from: nimda on September 30, 2013, 08:53:52 PM

Did anyone try to explain to nakowa that the site can't possibly be rigged at all because math?

This guy is rich, so his logic is "Fuck logic".

mechs

full member

Activity: 210

Merit: 100

Quote from: markjamrobin on October 01, 2013, 08:09:08 AM

Quote from: Mooshire on October 01, 2013, 07:59:00 AM

Quote from: Professor James Moriarty on October 01, 2013, 07:40:25 AM

Everyone calling just-dice a scammer.

I mean I just bet , 1 out of 17 was higher than 50 , 16 of them were lower , I won 1 out of a million once but my luck was still %313 , think of how unlucky I need to be in order for that to happen , but it still happened , do I call them a scammer? Well I kinda do , but in your way , I mean you guys create conspiracy theories and all , I just bluntly say they are screwing us in front of us , but at least they are not hiding it right? Its called house edge, if you win long enough you lose all. Simple as that.

It's provably fair, the only part you can't prove is that someone else doesn't know the secrets. In that aspect it may be better to gamble than invest because when you invest there's no provably fair.

I don't think it is ever a good idea to gamble Cheesy

But I don't think Nakowa knows them. He is just a lucky guy, who for some reason, can't take his money and run.

He has no reason to leave when he 16k up all-time and consistently winning. If he lost a big chunk he would stop, at least for a while. Then again, if he ever loss, he would accuse the site of cheating. He just that type of guy. Good results are skill, bad results in a rigged game.

markjamrobin

sr. member

Activity: 462

Merit: 250

Quote from: Mooshire on October 01, 2013, 07:59:00 AM

Quote from: Professor James Moriarty on October 01, 2013, 07:40:25 AM

Everyone calling just-dice a scammer.

I mean I just bet , 1 out of 17 was higher than 50 , 16 of them were lower , I won 1 out of a million once but my luck was still %313 , think of how unlucky I need to be in order for that to happen , but it still happened , do I call them a scammer? Well I kinda do , but in your way , I mean you guys create conspiracy theories and all , I just bluntly say they are screwing us in front of us , but at least they are not hiding it right? Its called house edge, if you win long enough you lose all. Simple as that.

It's provably fair, the only part you can't prove is that someone else doesn't know the secrets. In that aspect it may be better to gamble than invest because when you invest there's no provably fair.

I don't think it is ever a good idea to gamble Cheesy

But I don't think Nakowa knows them. He is just a lucky guy, who for some reason, can't take his money and run.

Mooshire

sr. member

Activity: 322

Merit: 250

Quote from: Professor James Moriarty on October 01, 2013, 07:40:25 AM

Everyone calling just-dice a scammer.

I mean I just bet , 1 out of 17 was higher than 50 , 16 of them were lower , I won 1 out of a million once but my luck was still %313 , think of how unlucky I need to be in order for that to happen , but it still happened , do I call them a scammer? Well I kinda do , but in your way , I mean you guys create conspiracy theories and all , I just bluntly say they are screwing us in front of us , but at least they are not hiding it right? Its called house edge, if you win long enough you lose all. Simple as that.

It's provably fair, the only part you can't prove is that someone else doesn't know the secrets. In that aspect it may be better to gamble than invest because when you invest there's no provably fair.

Topic: Just-Dice.com : now with added CLAMs : Play or Invest - page 140. (Read 454769 times)