whats a solid bb/100 winrate for NL?

[fimbulwinter]

i think wayfare understands variance.

several 2K+hands/day (i take this to mean 5+) losing days in a row at NL screams holes in your game. unless you're misposting here and play a very tough, very high-stakes game that belongs in the mid-high forum, these games don't offer much variance to a solid player. that's often the reason people play them, namely that the winrate/bankroll in NL is many times better than in limit, where variance is much, much higher.

fim

[fimbulwinter]

[ QUOTE ]
NL has many different formats which could lead to grossly exaggerated or underestimated winrates.

That why you need a big sample.

[/ QUOTE ]

um, think about that one for a moment...

you play $1000NL with $1/2 blinds and i play 50NL with the same structure. we're both the best player at the table and equal in skill. no amount of samples will make my winrate equal to yours.

do you understand?

fim

[jfresh]

damn, and i was just gonna chalk this one up as a zero-response, stupid-question post...

so now i'm confused again -- lemme see if i get this straight: pokertracker lists winrate as BigBet/100 correct? So for 100NL party with $2 big blinds, the Big Bet would be $4? so if i win X BB/100 I win $4 times X per 100 hands?

So a good/great player wins >10 Big Bets/100?

[MrFroggyX]

Yes.. off course.. You are supposed to have a higher win rate at the 100xBigBlinds games than the 50xBig Blinds game.
But that have nothing to do with your sample size.
You still need a big sample size before you are sure about your winrate.

[fimbulwinter]

last post for the night.

beating a NL game for 10big blinds/100 is an admirable goal. This is the standard rule of thumb. There are certainly (reputable) posters here who have claimed upwards of 50big blinds/100; i am not one of them.

the real measure of NL success is if you are ahead and if so by how much when the money goes in. on a shorter stacked game like party, a smaller winrate is expected; in a deeper game like prima/intercasino, your winrate will be bigger in terms of big blinds simply because of the structure.

in reality, anything above 0 is great; with the rake it means that you're doing better than the majority of your opposition.

oh yeah, and i'm a liar, froggy is dumb and wayfare punches kittens. purplemonkeydishwasher.

fim

[cornell2005]

about this variance stuff. the reason i say 30k hands is good enough is because that will get you to within 1 of your "real" win rate. some people often say that they want to know to the tenth what their wr is, and consequently they need a much larger sample. well, getting that accurate is pointless. when you get down to the tenth, there are other factors that distort statistical win rate that far outweigh sample size problems. A copule of examples are improvement in skill, increase in average opponent quality, improvement in notes on other people, and other peoples improvements in notes on you. further, saying you need one huge database is also pointless because the win rate is most often used to predict future successes. if you have 100,000 hands where you were a 2BB/100 player and 100,000 hands where you were a 8BB/100 player, it does not mean that you are currently a 5BB/100 player.

[meow_meow]

[ QUOTE ]
about this variance stuff. the reason i say 30k hands is good enough is because that will get you to within 1 of your "real" win rate.

[/ QUOTE ]

This isn't true.
Of course, it's entirely dependant on the SD/100hands. For me, at the 100NL, my SD/100 is 37.5BB (I will use the big bet convention throughout). I'm pretty aggressive, so let me know if this number is way too high.

Lets assume your play and the playing conditions do not change throughout the sample.

After 30k hands, the 95% confidence interval is +/- 4.2BB/100 from your true winrate( e.g. if your true winrate is 5BB/100, after 30k hands there is a 95% chance that your actual winrate will be between 0.8 and 9.2BB/100).

After 100k hands, the 95% c.i. is true winrate +/- 2.3BB/100, so you are still not nailing it down to within one BB.

The long run is very long.

[snowbank]

Could you please tell me one person who claims to make over 50 bb's/100 hands. Also, how deep are the stacks at Prima?

[cornell2005]

my variance is like 18 so i guess that is where our discprepency comes from

[fimbulwinter]

use the search function, im not going to put words in anyone's mouth.

the stacks at prima are 100Xbb

andrea beats the $5NL game there for (much) >100BB/100.

fim