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

[Vani]

He's been making sudden large bets for a while now.  Here's a story I posted about him a week or so ago:

http://just-dice.blogspot.ca/2013/10/time-to-get-stupid.html

It is amazing. Same goes for some of the other things I have seen on JD.

[1715582264]

1715582264

[1715582264]

1715582264

[GOB]

Thanks for the charts.

I don't think it's surprising that there are a small number of large investors, and a large number of small ones.  That's how most things tend to be.  Look at a plot of bank balances, bitcoin address balances, incomes, or pretty much anything else wealth related and you'll see the same pattern over and over.

This is true. It's called Zipf's Law:

https://en.wikipedia.org/wiki/Zipf's_law

Here's the Log(rank) vs Log(Invested amount) chart using JD invested data. It's not exactly like similar plots of, say, bank balances, income, city populations, etc., because it's not a straight line. Perhaps this has something to do with JD (risk seeking or risk averse behavior at different levels of BTC ownership? Use of multiple accounts?), or perhaps it just reflects the underlying distribution of Bitcoin ownership (which we do not know precisely).

[theskillzdatklls]

Doug, still think your site is awesome.  Only thing I would change at the moment is allowing 2% edge bets for 1% the bankroll and 3% edge bets for 1.5% the bankroll.  It would seem appealing to be able to win >900 btc on a micro spin for a gambler.

Site seems well bankrolled so some simple incentives to promote gambling would be good.

[haightst]

I closed my eyes at the moment that guy took 400 profit - but I'm sure we as investors will get over it   Congrats to the winner!

He's been making sudden large bets for a while now.  Here's a story I posted about him a week or so ago:

http://just-dice.blogspot.ca/2013/10/time-to-get-stupid.html

since you logged his ip address care to tell us what country he's from? or is that hush hushh http://smilies.newcastlebeats.com/smilies/badsonar2.gif thankz

[Vani]

Hehe - think you need to ask that to Fox himself    

My guess is that he is from the US.

[b!z]

Just-Dice investors may be interested in reading Bitcoin Reviewer's review of the site: http://bitcoinreviewer.com/just-dice-review/

Great site IMO.

[organofcorti]

Thanks for the charts.

I don't think it's surprising that there are a small number of large investors, and a large number of small ones.  That's how most things tend to be.  Look at a plot of bank balances, bitcoin address balances, incomes, or pretty much anything else wealth related and you'll see the same pattern over and over.

This is true. It's called Zipf's Law:

https://en.wikipedia.org/wiki/Zipf's_law

Here's the Log(rank) vs Log(Invested amount) chart using JD invested data. It's not exactly like similar plots of, say, bank balances, income, city populations, etc., because it's not a straight line. Perhaps this has something to do with JD (risk seeking or risk averse behavior at different levels of BTC ownership? Use of multiple accounts?), or perhaps it just reflects the underlying distribution of Bitcoin ownership (which we do not know precisely).

<img snip>

It turns out that investor amounts are not Pareto distributed (the continuous analog of Zipf's law), but rather [edit] can be modelled as [/edit] a mixture of two log normally distributed random variables. I've posted the details here so as not to junk up this thread. I'll just leave one chart to give you an idea.

[GOB]

It turns out that investor amounts are not Pareto distributed (the continuous analog of Zipf's law), but rather a mixture of two log normally distributed random variables. I've posted the details here so as not to junk up this thread. I'll just leave one chart to give you an idea.

What a great post! Thanks for writing that up!

[side note: What did you use to make your graphs?]

[haightst]

Hehe - think you need to ask that to Fox himself    

My guess is that he is from the US.

shhhh is he in the Army? that sounds like some code ta|kers name? lol

[dooglus]

It turns out that investor amounts are not Pareto distributed (the continuous analog of Zipf's law), but rather a mixture of two log normally distributed random variables.

That's interesting.

The first thing I thought of when I saw that was that when JD first launched, I gave some 250 or so separate gifts of 0.01 BTC to forum users.  Many of them were probably invested and forgotten.  Could those account for the "small investor" group, or isn't it in the right place for that?

[haightst]

It turns out that investor amounts are not Pareto distributed (the continuous analog of Zipf's law), but rather a mixture of two log normally distributed random variables.

That's interesting.

The first thing I thought of when I saw that was that when JD first launched, I gave some 250 or so separate gifts of 0.01 BTC to forum users.  Many of them were probably invested and forgotten.  Could those account for the "small investor" group, or isn't it in the right place for that?

did you hand out tee-shirts too? lol =)

[Rannasha]

It turns out that investor amounts are not Pareto distributed (the continuous analog of Zipf's law), but rather a mixture of two log normally distributed random variables.

That's interesting.

The first thing I thought of when I saw that was that when JD first launched, I gave some 250 or so separate gifts of 0.01 BTC to forum users.  Many of them were probably invested and forgotten.  Could those account for the "small investor" group, or isn't it in the right place for that?

According to the data, the log mean of the "small investor" group seems to be around -2.3. This corresponds to 0.1 BTC

[organofcorti]

[side note: What did you use to make your graphs?]

I use R and the ggplot2 package for the plot, and mclust for the analysis.

It turns out that investor amounts are not Pareto distributed (the continuous analog of Zipf's law), but rather a mixture of two log normally distributed random variables.

That's interesting.

The first thing I thought of when I saw that was that when JD first launched, I gave some 250 or so separate gifts of 0.01 BTC to forum users.  Many of them were probably invested and forgotten.  Could those account for the "small investor" group, or isn't it in the right place for that?

I set out to avoid small low bandwidth groups from having any significant effects on the results. As you can see in the density plots and histograms below on the right, the are significant spikes at round numbered amounts at approximately 0.01, 0.1, 1 and ten btc; also theres a bugle in the shoulder of the (right side) density plot at 100 btc.

These are the same sorts of spikes that were present in the SatoshiDice bet data and seem to be due to human preferences for round numbers (give or take a few percent for wins or losses at the time the data was recorded.





The spikes were not what I wanted to model, and in fact just add noise. It's true that I could have added them in, but then where do you stop? In the end you could create a mixture of gaussians consisting of one per datapoint.

That's why I used the Baysian information criterion (BIC) to decide on the number of gaussians used to model the data, and means that the "power of ten" spikes don't obfuscate the more general trends.

[wolverine.ks]

so other than the .01 free investments, the remaining investments are more like an average of 10btc each?

[organofcorti]

so other than the .01 free investments, the remaining investments are more like an average of 10btc each?

Since the distribution of values is so skew and has such a long tail, an average - and even the median - doesn't really help you understand the data.

If you looked at the density plots (the curved ones) the x-axis is bitcoin, the density (think "percentage") of investors at that amount. You can follow the line along to see what the trends are.

[dooglus]

Last week was a pretty good one for investors, with the site profiting over 1000 BTC:

Another better than expected week, making a little over 600 BTC this time:



The site only took 34 BTC in commission on the 600 BTC profits, which must mean that a lot of investors are still not at the personal peak profit.

Also, I'm not sure if it has been linked here before or not, but http://bitcoinproject.net/just-dice-casino/just-dice-charts was put together by a Just-Dice visitor and has up-to-date charts using data pulled from Just-Dice.

[nimda]

  I missed something. Can I get a link to the discussion where 30K BTC was won?

[meta.p02]

I missed something. Can I get a link to the discussion where 30K BTC was won?

I think you were confused by the "36500 BTC below 1% edge"... simply it means that we should have XXXXX profit but we have only XXXXX-36500.

Nakowa betted a large amount, and was expected to lose a lot, but instead won. This causes the house profit to be much lower than expected.

[eltopo]

Also, I'm not sure if it has been linked here before or not, but http://bitcoinproject.net/just-dice-casino/just-dice-charts was put together by a Just-Dice visitor and has up-to-date charts using data pulled from Just-Dice.

Nice! Can you send him all the historical data so he can create a complete history for JD since June?

[Rannasha]

Also, I'm not sure if it has been linked here before or not, but http://bitcoinproject.net/just-dice-casino/just-dice-charts was put together by a Just-Dice visitor and has up-to-date charts using data pulled from Just-Dice.

Nice! Can you send him all the historical data so he can create a complete history for JD since June?

I believe they're working on that.