| This is a discussion on Sample Size within the online poker forums, in the Learning Poker section; What do you consider an adequate sample size, in # of hands, to determine if you are a winning player or not?... |
| | ||||||
![]() |
| |
|
#1 | ||||
| ||||
| Sample Size What do you consider an adequate sample size, in # of hands, to determine if you are a winning player or not? Last edited by chattin35 : 17th November 2009 at 4:49 AM. Reason: Typo in title |
| Play Texas Hold'em Online Poker | Sample Size | |
|
|
|
#4 | ||||
| ||||
| re: Sample Size poker It depends how big your win-rate is. If you're winning at 40bb/100 over 5000 hands, you're likely a winner. If you're winning at 1bb/100 over 15000 hands, its hard to tell if you're a winning player or not. It also depends on the game. Games with lower variance (i.e. less short term luck) require fewer hands to determine if you're a winner. So Razz for example, would require fewer hands than limit hold'em to determine if you were a winner. Just post your win-rate, # of hands, and your preflop VPIP, PFR, and aggression percentage stats. From that data alone we can usually tell if you're a solid player or not. |
|
#5 | ||||
| ||||
| The question I'm asking really isn't whether or not I'm a solid player. Although, I appreciate the offer to help analyse my game. I might take you up on that later. I am asking what a resonable sample size is to account for variance while doing a self analysis. How many hands does it take to be confident in a trend rather than a statistical anomaly? Thanks. |
|
#6 | ||||
| ||||
| it really depends on a lot of variables. in your particular case, at 2NL, you could have a million leaks and still be crushing the game. i know that from personal experience. that said, normally 7k hands is not enough to give a *definite* answer to your question for any limit. some people play that many hands in just a few days. you still could have some analysis made from that sample though, so it is not completely useless. (and if it will make you feel better about yourself without getting overconfident, then you surely can consider yourself a winning player at 2NL.) good luck! |
|
#9 | ||||
| ||||
| You just have to kibitz one hand at 2nl to see that you will be a winning player at that level. On a more serious note. The problem is that its not a matter of simple single factor variance. There are a lot of factors, and if you only have the # of hands and the BB/100hands as input you would need a huge sample size. |
Number of Posts: 9
Number of Authors: 7