Poker and Stockpicking

[ira1313]

I am doing a project, that will attempt to find what can make a better risk adjusted return on investment, online poker, or stock picking. Obviously this experiment will not be perfect, because of time constraints and other variables that cannot be controlled.
What I need help with, is which ratio I should use to calculate this risk adjusted return, between poker and stocks(I've been looking into the Sharpe Ratio), and whether a fair assessment could be made with any ratio. Also how should I be tracking my results, and what statistics do I need to compile though out my experiment?
Any help would be greatly appreciated.

[emon87]

I would say it will be incredibly difficult to compare them. They are so very different.

[ira1313]

I agree, however from a mathematical standpoint, a ratio should still give me some sort of ROI or something that is comparable correct?

[Shoe]

For small amounts of money = poker (can easily turn $100 into $200 at low stakes in the easy games).

For large amounts of money = stock market. Can make more in the stock market than you can even put into play on a poker table. Not to mention how tough the competition gets at the highest stakes.

[kimchi]

You're trying to compare poker to stockpicking. I would specify stock trading over stock investing when trying to draw parallels with poker.

I assume you'll be using some sort of broad stock index as a comparison for risk-adjusted ruturn, but you'll be better to use individual stock traders' accounts versus individual poker players' accounts.

This info will be very difficult to obtain, and then how will you select whose accounts (and a standard account size) to use?

[gull]

Poker is -EV, on average.
The stock market is +EV, on average.

Poker stakes are capped in the thousands.
Stock market investments can be billions.

[blackasthma]

Unless you're being paid a large sum for this study, don't waste your time. Your typical "Joe the Degenerate Gambler" won't break from his busy schedule long enough to breeze through or comprehend this study, if it's even valid at all.

How do you quantify online poker threats such as cheats and fraudulent sites? US Government sure as hell won't protect online poker for its citizens. Online speculators/gamblers are their own worst enemies, anyway, as they're usually only a click away from financial ruin.

I think a more appropriate study in this case would compare setting one's money ablaze to flushing it down the toilet...

[ira1313]

[ QUOTE ]
You're trying to compare poker to stockpicking. I would specify stock trading over stock investing when trying to draw parallels with poker.

I assume you'll be using some sort of broad stock index as a comparison for risk-adjusted ruturn, but you'll be better to use individual stock traders' accounts versus individual poker players' accounts.

This info will be very difficult to obtain, and then how will you select whose accounts (and a standard account size) to use?

[/ QUOTE ]
I'm doing this on an individual basis. I realize that if I wanted the best experiment, I would have numerous traders and poker players accounts data, but it is not possible for my situation. I am in fact using my own trading and poker account not a index.

The problem I find is trying to compare poker and stock returns in a statistical sense. Say I double my money in poker, but it takes considerable more risk, as opposed to gaining 10% in stocks, but taking far less risk. Which one is better, and how do I determine that? Obviously, some people can withstand more risk, but if I were to be able to take the risk to reward on stocks compared to poker in my experiment, and do it a million times, which would produce the best returns? Often times if you can double your money easily, it also means you can go broke easily, but if I double 60% of the time and go broke the other 40%, it's a pretty good return over the long run.

Essentially what I am trying to determine is, Taking both risk and potential returns into consideration from a statistical stand point which will make me more money, if the test was run millions of times.

[T50_Omaha8]

A number of variables and functions come into play:

You have cash P at time t=0.

You choose to allocate n hours per time period to making money (work and poker).

Your wage rate per hour is w if you get a &#*%$ job.

Let G(x) equal your expected hourly rate with a poker bankroll of size x (ie it will quite possibly be negative when you get above some certain point). Let's assume you are properly rolled.

Lets assume you spend portion q of your alotted work time time playing poker and (1-q) of your alotted work time working.

Let S*(P-x) equal your projected return on the remaining P-x dollars in your portfolio. (That is, your expected rate of return in the stock market is S.)

Your income from poker is G(x)*q*n
Your income from work is w*(1-q)*n
Your income from investing is S*(P-x)

Your trouble will be 'risk-adjusting' the rate of return you get in poker to make it similar to that of the stock market. I don't feel like going into that right now. This also requires maximization with two variables--x and q--which isn't helpful. In reality, q probably isn't that practical to consider as a continuous variable anyways. Just approximate it for your preferences/situation and assume its constant.

But I think these variables, as well as decent approximations of your winrate with respect to stakes/bankroll, in addition to SD/hour playing poker, should be enough to fingure it out.

If you get it to that point, the probability forum would probably be helpful.

[emon87]

[ QUOTE ]
Essentially what I am trying to determine is, Taking both risk and potential returns into consideration from a statistical stand point which will make me more money, if the test was run millions of times.

[/ QUOTE ]


The problem, mathematically, is that skill is involved. So whatever you are better at would yield you more, unless you are limited to not using your bankroll optimally in poker.