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

[dooglus]

Thanks for the question RationalSpeculator.  To calculate properly, we need an updated list of all Nakowa's bets (a new nakowa.txt file from dooglus).

It's here:

https://just-dice.com/nakowa3.txt.bz2

[dooglus]

I just don't see the point of implementing the simple fractional kelly system when you can get the same thing from just investing less.

You don't get the same thing by investing less.

There was an example earlier: A: invest 500 at 1% or B: 125 at 0.25%.

You'd think they were the same, but they're not.  After a few thousand losses, A has lost nearly 500 coins and B has lost nearly 125 coins.  That's not the same at all...

[dooglus]

Simple answer just leave it at 0.50% where it should be 1% was to much, the only thing about the change that was done wrong (went to far) and was when without an announcement to players and your "investors"

You're right, and I apologise for that.  I don't know if you saw the run-up to increasing the risk from 0.25% to 0.5%, but I announced it in this thread giving 24 hours' notice, and had the site chat announcing the change automatically every 10 minutes in the run up to the change.

I'll do the same for any future changes, too; give notice both here and in the on-site chat.

[integrity42]

There's absolutely NOTHING that indicates he is cheating. The wild swings he is experiencing looks 100% legit. Look at how he plays, its just a completely random walk.

Here's a chart of the site's profits for the day, with the expected profit (minus 20k, to bring them closer together!) overlayed.

The random walk goes horizontally, whereas the expectation goes up at 30 degrees to the horizontal.  How?  And why?

Well doog, at least the chart, since 18:00 , is making higher lows

[mechs]

I just don't see the point of implementing the simple fractional kelly system when you can get the same thing from just investing less.

You don't get the same thing by investing less.

There was an example earlier: A: invest 500 at 1% or B: 125 at 0.25%.

You'd think they were the same, but they're not.  After a few thousand losses, A has lost nearly 500 coins and B has lost nearly 125 coins.  That's not the same at all...

I don't see anyone screaming for 1% kelly anymore.  Just leave it at 0.5%, the variance is plenty.  I just don't get how after bidding for 12 hrs straight 1.6M bitcoins between 3 whales (mostly coming from 1), we end yet another session down significantly.   It seems the edge is too small to overcome the variance at such high bids.  I cannot explain it, but the results continue to confound logic and be 3+ SDs off from the expected results

[willphase]

I just don't see the point of implementing the simple fractional kelly system when you can get the same thing from just investing less.

You don't get the same thing by investing less.

There was an example earlier: A: invest 500 at 1% or B: 125 at 0.25%.

You'd think they were the same, but they're not.  After a few thousand losses, A has lost nearly 500 coins and B has lost nearly 125 coins.  That's not the same at all...

I presume you mean the other way round - A: 125 at 1%  and B: 500 at 0.25%?

Will

[dooglus]

Doog:

Is it possible to get a lifetime graph of AIEV vs Actual rather than just the one day?  That would be interesting to look at.

Thanks!

I'm not sure what 'AIEV' means, but if you mean 1% of the total wagered, then this is it.  It really shows how huge the volume was today:

[dooglus]

I presume you mean the other way round - A: 125 at 1%  and B: 500 at 0.25%?

Yes, I meant whatever makes sense...  it's been a long day and I'm tired.

I'd like to answer the posts about how best to make the variable risk stuff work, but I'm too tired to get it straight in my head, and this error just proves it.  I'll take a look tomorrow.

[theskillzdatklls]

Doog:

Is it possible to get a lifetime graph of AIEV vs Actual rather than just the one day?  That would be interesting to look at.

Thanks!

I'm not sure what 'AIEV' means, but if you mean 1% of the total wagered, then this is it.  It really shows how huge the volume was today:

Perfect, thanks so much.  AIEV (all-in expected value) is a poker term, probably shouldn't have used it here because it doesn't even make sense.  'Expected value' by itself is more appropriate.  But ty again for the chart

[usagi]

There's absolutely NOTHING that indicates he is cheating. The wild swings he is experiencing looks 100% legit. Look at how he plays, its just a completely random walk.

Here's a chart of the site's profits for the day, with the expected profit (minus 20k, to bring them closer together!) overlayed.

The random walk goes horizontally, whereas the expectation goes up at 30 degrees to the horizontal.  How?  And why?

img width=900]https://i.imgur.com/X9fL34G.png[/img]

Need more info. Where was that list of all his bets and how much each one won/lost? https://just-dice.com/whale.txt has old info.

EDIT: Ahh I see you just posted that while I was typing. Thanks.

[maqifrnswa]

I just don't see the point of implementing the simple fractional kelly system when you can get the same thing from just investing less.

You don't get the same thing by investing less.

There was an example earlier: A: invest 500 at 1% or B: 125 at 0.25%.

You'd think they were the same, but they're not.  After a few thousand losses, A has lost nearly 500 coins and B has lost nearly 125 coins.  That's not the same at all...

I don't see anyone screaming for 1% kelly anymore.  Just leave it at 0.5%, the variance is plenty.  I just don't get how after bidding for 12 hrs straight 1.6M bitcoins between 3 whales (mostly coming from 1), we end yet another session down significantly.   It seems the edge is too small to overcome the variance at such high bids.  I cannot explain it, but the results continue to confound logic and be 3+ SDs off from the expected results

no one is screaming because we thought doog was implementing variable max bet. No need to argue over what is "correct" - let people choose on their own.

[David Chen]

Yep, Nakowa just lost and reloaded 1,790BTC. A degenerate gambler is a degenerate gambler. That's why casinos exist

Now,do you think he is  still a degenerated gambler, you guys always judge a person by results,huh?

[Peter R]

Thanks for the question RationalSpeculator.  To calculate properly, we need an updated list of all Nakowa's bets (a new nakowa.txt file from dooglus).

It's here:

https://just-dice.com/nakowa3.txt.bz2

Thanks doog.  Here's the new plot:

[chriswen]

I just don't see the point of implementing the simple fractional kelly system when you can get the same thing from just investing less.

You don't get the same thing by investing less.

There was an example earlier: A: invest 500 at 1% or B: 125 at 0.25%.

You'd think they were the same, but they're not.  After a few thousand losses, A has lost nearly 500 coins and B has lost nearly 125 coins.  That's not the same at all...

I don't see anyone screaming for 1% kelly anymore.  Just leave it at 0.5%, the variance is plenty.  I just don't get how after bidding for 12 hrs straight 1.6M bitcoins between 3 whales (mostly coming from 1), we end yet another session down significantly.   It seems the edge is too small to overcome the variance at such high bids.  I cannot explain it, but the results continue to confound logic and be 3+ SDs off from the expected results

no one is screaming because we thought doog was implementing variable max bet. No need to argue over what is "correct" - let people choose on their own.

No because if dooglus goes forward with the current implementation, the 1% investors will increase the max bet and increase variance for all users.

[chriswen]

Thanks doog.  Here's the new plot:

https://i.imgur.com/Rk031ix.gif

Nice!  So it seems nakowa is staying in the same sigma band.  So he's not getting even more lucky.  He's just a bit lucky.

[dooglus]

Thanks doog.  Here's the new plot:

https://i.imgur.com/Rk031ix.gif

Nice chart.

Here's a plain "lifetime of nakowa" chart from the same data:

[drawingthesun]

Here's a plain "lifetime of nakowa" chart from the same data:

At one point he lost 10,000 bitcoin.

Crazy.

[Rampion]

Well, before Nakowa started playing my b/e point was roughly -2k in house profit.

I never divested and now I'm significantly down (more than 10%), and my break even point is significantly higher (+1,600 in house profit).

I can only assume the reason for this is the "invest/divest" day traders, including Dooglus (who divested when the house was up 3k), on average got lucky and screwed the ones like me who didn't divest. I'd say the "invest/divest daytraders" were gambling against those that did not divest, and the former won big time with a simple strategy: divest when you are up, invest again when the house is down.

What do you think, doog? So much for the "long term, stable investment" some day dreaming investors are looking for...

[Rampion]

One more comment: after looking at the insane betting volume we had yesterday, I'm starting to think a tiny 1% edge is not enough to make decent money for investors unless we have whales pushing max profit non stop, every single day. That might be the reason all other casinos have much higher edge.

Nevertheless I think that for Dooglus the site is good as it is, as he is taking a healthy cut of all the gains on the site, and with such a low edge he can count on a high volume and thus a nice, riskless and clean profit based on commisions. On the contrary, investors (who are actually the guys providing the bankroll of the casino) have a very small edge working on their favour, which means that they can be down for a long time, which is made worse by having their shares diluted when they are at a loss.

If you ask me, I'd just wait and see, but I'd think about taking house edge to 1.5%, or to think of another system of rewarding investors, helping out the tiny edge they have.

[dooglus]

I can only assume the reason for this is the "invest/divest" day traders, including Dooglus (who divested when the house was up 3k), on average got lucky and screwed the ones like me who didn't divest. I'd say the "invest/divest daytraders" were gambling against those that did not divest, and the former won big time with a simple strategy: divest when you are up, invest again when the house is down.

I guess that's the case.

On first view it seems that 'day trading' nakowa has a zero expectation.  But if he keeps going up and down like today then "divest at +1000, invest at -1000" is a profitable strategy.

You lose when you divest at +1000 and the profits go to +5000 and never come back.  Or invest at -1000 and profits drop to -5000 and don't come back.  But if they keep going past both your trigger points (-1000 and +1000, say) then you win, and the committed investors lose.

My "day trading" today was accidental.  I pulled out as soon as I saw nakowa was playing again.  Luckily for me he was down at the time.  I've not reinvested.

He asked to withdraw his winnings, but I don't have the cold wallet with me.  He's agreed to wait around 24h until I'm home again to withdraw.  I don't know if that means he won't play until after that or not.  There's no point asking him, because there's no correlation between what he says and what he does.