Just-Dice.com : Invest in 1% House Edge Dice Game

[mechs]

Has a stop-loss feature been discussed before? Surely I'm not the only investor who has doubts about being exposed to this kind of variance, especially just after big losses. Why not have a user-settable investment threshold, and if a user's investment falls below that level they'd be auto-divested?

mechs placed a 7 BTC bounty on anyone who makes a script that does this.

Link? Sounds like easy money to me.

https://bitcointalk.org/index.php?topic=263522.0

"easy money"

I actually have one written with python.  Wish I had been using it.

[1715623337]

1715623337

[utens]

At the end , if he did 100k wagered,   he should have won 99k back ,  but he won like 107k   ( in this last week)

Edit :  i think he did more about 200k wag   , so instead of 198k he got 206k

[dooglus]

At the end , if he did 100k wagered,   he should have won 99k back ,  but he won like 107k   ( in this last week)

Edit :  i think he did more about 200k wag   , so instead of 198k he got 206k

Also, he gets to choose when to stop.  He could have chosen to stop when he was down, but instead waited until he was up.  The site doesn't get to make that choice.

"Ok, you've lost a couple thousand; no more bets for you.  thanks for the coins".

It wouldn't fly...

[utens]

Of course..

Was saying that he run like 8%  above ev ( looking at the wag and expected of today).. So yes he got lucky today, but nothing supernatural

[mechs]

Of course..

Was saying that he run like 8%  above ev ( looking at the wag and expected of today).. So yes he got lucky today, but nothing supernatural

It the fact he gets lucky everytime he plays which is painful (for investors and Doog)

[Lohoris]

Nakowa ticking the site again.  I was going to make a rape joke about it, but then I saw this:



Kyle, you should really study this flow chart!

I like this one better:

[RationalSpeculator]

Applying the Kelly Criterion gives you the best risk/reward ratio.

No, it doesn't.  The Kelly Criterion maximizes the reward, regardless of risk.  Read the Wikipedia article.

The "best" risk/reward ratio is a complicated subject.  There are a number of approaches, such as Sharpe ratio, Treynor ratio, etc.  Your mileage may vary.  But the Kelly Criterion doesn't address this at all.

Isn't a strategy that has maximized the reward, per definition also a strategy that has minimised the loss?


In the wikipedia article is mentioned: "One of the most important ideas in Kelly is that betting more than the Kelly amount decreases the probability of very good results, while still increasing the probability of very bad results."

So it is clear that betting more than the Kelly amount is unwise but is it wise to bet less?

Here what I found elsewhere: "Betting half the Kelly amount, for example, reduces bankroll volatility by 50%, but growth by only 25%."

The big question is, is volatility the same as risk? If yes, then you are right, reducing max bet reduces risk more than the loss in growth. But if volatility does not equal risk, then you just lost growth in return for nothing.

I would say, volatility does not equal risk. If I have a statistical certain win, then volatility is not risk, since loss is always temporary with statistical certainty.

[tucenaber]

Also, he gets to choose when to stop.  He could have chosen to stop when he was down, but instead waited until he was up.  The site doesn't get to make that choice.

"Ok, you've lost a couple thousand; no more bets for you.  thanks for the coins".

It wouldn't fly...

Perhaps surprisingly, a stopping rule doesn't change the expected return, nor does it change the variance. The variance is affected, and if I'm not mistaken the variance is E(X)^2 V(T) + E(T)V(X) >= E(T)V(X), where T is the number of bets according to the stopping rule and X is the net winning of each bet.

EDIT: fixed my stupid mistakes

[Oleander]

Here what I found elsewhere: "Betting half the Kelly amount, for example, reduces bankroll volatility by 50%, but growth by only 25%."

The big question is, is volatility the same as risk? If yes, then you are right, reducing max bet reduces risk more than the loss in growth. But if volatility does not equal risk, then you just lost growth in return for nothing.

I would say, volatility does not equal risk. If I have a statistical certain win, then volatility is not risk, since loss is always temporary with statistical certainty.

Wrong.  By your logic, the risk is always zero, regardless of the max bet, therefore there is no reduction in risk when the max bet is reduced.

"Risk" and "volatility" usually refer to the standard deviation of the return, and I think in your quote "bankroll volatility" is the standard deviation of the bankroll, which would be the same.  In this thread we also use the term "variance", which is technically the square of the standard deviation.

Not to get all pedantic on everyone, but we use "risk", "variance" and "volatility" imprecisely, to capture the idea that the investor's balance can be up one day and down the next.  And that's OK.  Now, for something like "reduces bankroll volatility by 50%", then we need to get specific and ask whether we're talking about a standard deviation or the square of a standard deviation, because that 50% number can't be correct for both of those at the same time.

Even if you persist in your colloquial use of "risk", we can talk about a volatility/growth ratio, and see that an investor might reasonably prefer a situation with half the volatility and three-fourths the growth.

[ColdHardMetal]

Well seems like not many people have experience with grinding poker otherwise you would know this is not even close to extreme in variance world 

Quoted for truth.

[chriswen]

Is that enough?  Do you need counts for the sub 1 BTC bets too?

http://just-dice.com/grouped.txt.bz2 has the same information for all bets, down to 0 BTC staked.

I removed all the formatting, so it's just 3 numbers per line.

It goes down to 502 satoshis staked.  Which should be fine.  I'm gonna try working on this for fun.

[chriswen]

Allright, I believe that the correct equation for the JD-standard deviation is:



For an individual bet, sqrt(p * (1 - p)) is the expected standard deviation of the chance to win. The expected standard deviation of the return is obtained by simply multiplying by the expected return, which is 0.99 * b.

Summation of standard deviations for independent random variables is done through taking the root of the sum of the squares. Each term is multiplied by how often it occurs (c).

That brings me to the numbers obtained from the dataset posted by dooglus:
# of bets = 33,257,221 (currently posted on site: 33,524,569)
Total BTC bet = 556,351 (currently posted on site: 557,909)
Expected BTC return for gamblers = 550,787
Expected house profit = 5564 (if only ...)
Expected house edge = 1.0000881%
Expected standard deviation of amount returned to gamblers = 2754

Current house profit is at -228 BTC, which is 2.1 times the standard deviation away from the expectation value. If the profit follows a normal distribution, the probability of this deviation occurring through sheer chance is approximately 2.5%.

Conclusion: The site has been very unlucky, but the deviation is not large enough to be implausible.

Well, someone should check my work.

That brings me to the numbers obtained from the dataset posted by dooglus:
# of bets = 65,711,328 (currently posted on site: 136,541,000) (only counted bets 502 or more satoshis)
Total BTC wagered = 1,296,828 (currently posted on site: 1,299,808)
Expected BTC return for gamblers = 1,283,860
Expected house profit = 12,968 (if only ...)

Expected house edge = 1.0000881%? dunno what this means.

Expected standard deviation of amount returned to gamblers = 4223 (This was the calculation)

Current house profit is at -168 BTC, which is 3.1 times the standard deviation away from the expectation value. If the profit follows a normal distribution, the probability of this deviation occurring through sheer chance is approximately 0.2% (I think, using wolfram alpha, not exactly sure how to do this).

Conclusion: The site has been very unlucky, and I don't have the background to make any further comment.

Here's the excel: https://db.tt/7x9532Hs

[organofcorti]

Current house profit is at -168 BTC, which is 3.1 times the standard deviation away from the expectation value. If the profit follows a normal distribution, the probability of this deviation occurring through sheer chance is approximately 0.2% (I think, using wolfram alpha, not exactly sure how to do this).

The profit doesn't appear to follow a normal distribution, it's much more like white noise with drift. If I can get a hold of a dataset of the house profit with time, I can probably estimate forecast confidence intervals for the investor profit ( regressed to total btc bet).

[molecular]

Has a stop-loss feature been discussed before? Surely I'm not the only investor who has doubts about being exposed to this kind of variance, especially just after big losses. Why not have a user-settable investment threshold, and if a user's investment falls below that level they'd be auto-divested?

mechs placed a 7 BTC bounty on anyone who makes a script that does this.

Link? Sounds like easy money to me.

https://bitcointalk.org/index.php?topic=263522.0

"easy money"

not quite... it's only 4 BTC and stuff like...

6.  Must provide support for at least 1 year (should changes me made to website which "break" the bot).

kills it for me ;|

[naphto]

All profits are gone? Wtf?
- 1.2 k btc

[maxmint]

Nakowa is doing an AMA at reddit:
http://www.reddit.com/r/AMA/comments/1my8fc/im_the_whale_on_justdice_id_love_to_answer/

[romerun]

he does not answer shit, just to promote his site apparently

[Rannasha]

Well, someone should check my work.

Couldn't find the parsing-program I used last time, but your spreadsheet seems to be okay at first glance.

Current house profit is at -168 BTC, which is 3.1 times the standard deviation away from the expectation value. If the profit follows a normal distribution, the probability of this deviation occurring through sheer chance is approximately 0.2% (I think, using wolfram alpha, not exactly sure how to do this).

Yes, 3.1 sigma is roughly 0.2% (on both ends added up).

[integrity42]

At this point, I would seriously be concerned that someone at amazon has done a memory dump of the server stolen the server seed.

Dooglus should probably move the dedicated server to his house and run a dedicated line to it.

nakowa did an AMA on reddit and said his last 2 wins were

deposit 500, win 2500
deposit 200, win 1000+

i guess thats the same as people who deposit 5 and win 25, or 2 and win 10, but the difference is those people dont do it with max bets of 2-3btc at 50% they hit some low odds wins along the way

[chobopeon]

At this point, I would seriously be concerned that someone at amazon has done a memory dump of the server stolen the server seed.

I'd be amazed if that was the case if only because he's so brazen about the whole thing. If it was a theft from inside a company like Amazon, you'd think it'd be a little more subtle.

But then maybe I'm way off.