winrate Q (i've read the FAQ)

[Cosimo]

[ QUOTE ]
if your standard deviation is the same as mine, then after 12K hands at 0bb/100, statistically there is a 95% chance your winrate is somewhere between 6.92 and -6.92 BB/100. You have absolutely no idea if you're a winner in this game or not

[/ QUOTE ]

I have to disagree.

I think it's also important to pick the nit that, if some say 8 is sustainable, then the <2.5% chance that I'm playing that well says that I'm not that good. The evidence suggests that I suck at $50.

My winrate at $25 is between -3 and 11 BB/100. I'd have to say that there is evidence that suggests that I fare better against the $25 players than against the $50 players. It is more likely that I do worse at $50 than any other combination here. Just because I could be a "real" 6BB/100 player doesn't mean that it's likely. It's actually very unlikely. There's a greater chance that I'm in the (-1,0) range than in the (-2,-1) range.

As for "gospel" ... this is still money we're talking about. There's no guarantee that I'm really only a breakeven player at $50, but there's evidence that suggests that I'm not good. It is more likely that I'll make money at $25 than at $50. Furthermore, posts everywhere on this forum (in SSNL, general, the zoo, limit micros, etc) suggest that putting in more hands at $25 will let me improve my game and help me build a bankroll to attack $50 again later.

You seem to be suggesting that, because I might actually be a winner at $50, that I should be playing there. Or that, because I might actually be doing the same at both limits, that I should be at $50. I think this is a confusion between precision and accuracy... or that you think that I'm trying to read precision where there is none. I acknowledge the imprecision of these stats, but I don't doubt their accuracy.

What I see says that the most likely explanation is that I have figured out how to beat the $25 players (but not by much), and the $50 players play slightly differently and I haven't figured out how to adjust to them.

[beeyjay]

im certain 8/100 is possible at 100 6 max. those games are so beatable in non peak hours and during peak hours they are unreal. the variance, especially during peak hours tho makes it hard to believe that anybody would have a big enough sample to prove this.