How many (allin-ev adjusted) hands do you need to be 80% sure you are a winning player? What about 95%?

It depends on your expected winrate ("edge") vs the player pool, and the standard deviation in your results. i.e. Having an expectation of 20bb/100 would mean you are likely to be winning even over a small sample, but an EV of 1bb/100 could mean you have to play hundreds of thousands of hands before you're certain of profiting. In addition, playing 200bb PLO, your results would be "swingier" than playing 10bb push/fold NLH.

Put some numbers into the variance simulator at http://pokerdope.com/poker-variance-calculator/ and read the text underneath.

How many mistakes do you see at the poker table, and how many mistakes could you possibly be making without knowing that they are mistakes? You should have a rough guess after, perhaps, 10000 hands, qualitatively speaking. You could lose after 10000 hands and still know you're a winning player. The problem that arises with qualitative analysis is that so many people in poker are freakin' delusional.

The allinev has been adjusted at around HM3 (also in HM2 since), so it means about nothing, or it shows accurately if you are above ev, I think, but the below ev is no longer below ev but the same as your winrate.

In practice, one finds out pretty fast if the game is good enough or not, it only being difficult when the opponents are extra aggro or sit on their hands more than optimal. It taking time to find where their leaks are, although you should already be confident that if you play optimally and adjust a bit here and there, that you will have an edge.

For variance, they have some standard deviation formulas somewhere, or might even be in that software like these days (in HM3 perhaps). People have done such calculations and simulations and analyses and some stuffs are online, and at least online tourneys are still tracked by one or more tracking sites.

"You could lose after 10000 hands and still know you're a winning player. The problem ... is that so many people in poker are freakin' delusional."

Never a truer word spoken

Look at your account balance and if it's greater than what you have put in then you're a winning player. If it's less then you're a losing player.

What your talking about is a confidence interval. You can
Calculate this I believe.
Statman should be able to chime in on how to do it otherwise Google is your friend.

I'm chiming.

Assume you have sample data that indicates you have a positive win rate of W/100 with a standard deviation of S/100. To determine what sample size you need to assure with a specified confidence that you are a winning player, you can use the following formula:

N =100*( Zc*S / W)^2

where Zc is the normal distribution factor corresponding to a one-sided C% confidence interval: Z80 = 0.84, Z90 = 1.28, Z95 = 1.645. Note that N increases with C and S, and decreases with W.

Example: You have an estimated win rate of 4 per 100 hands with a corresponding standard deviation of 75/100. To be 80% confident you are a winning player,

N = 100*(0.84*75 / 4)^2 = 24,800.

For C=90%, N = 57,600 (approx.) and N = 95,100 for 95% confidence.

The above assumes your win rate is distributed normally (valid by the Central Limit Theorem) and that your win rate and standard deviation are relatively constant over the sample period. If your win rate is not constant but is increasing and the standard deviation is non-increasing, then the sample size result can be considered to be conservative.- i.e., it is more than you need.

Thx statman