[Phil153]

[ QUOTE ]
50K +

[/ QUOTE ]
This is silly. The meaningfulness of results isn't given by a cutoff or even a ballpark figure. It's simple math. The 95% Confidence Interval around your results is given by

winrate ± 1.96*(standard deviation per 100 hands)/ sqrt(hands/100)

For example, if you managed 10BB/100 over 10K hands with a standard deviation of 50BB/100, your confidence interval is:

10 ± 1.96*50/sqrt(10K/100)

= 10 ± 9.8

So your true winrate is anywhere between 0.2 and 19.8BB/100. You can tell from this sample that you're a winning player.

For 50K hands, your winrate is given by:

10 ± 1.96*50/sqrt(50K/100) = 10 ± 4.4 BB/100.

You can see that all you have is ranges of confidence in your actual winrate. There's no such thing as a cutoff or a starting point.

As for OP, looking at his 17K hands for PL $100 with a 5.95 BB/100 winrate and assuming a 50BB/100 standard deviation, he has a 95% CI given by:

5.95 ± 1.96*50/sqrt(17K/100) = 5.95 ± 7.5 BB/100

so there's a 95% chance his true winrate falls somewhere between -1.5 and +13.5 BB/100, with the values toward the middle having a higher probability than those toward the extremes.

Of course, there are other ways to determine true winrate. One is by observing the skills and mistakes of opponents relative to your strategy. Someone experienced can estimate their winrate fairly quickly, but it's not rigorous, and given the intelligence of most professional poker players, probably not recommended.