ALERT: Poker is a LONG TERM game!

[jrz1972]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
18/root(X) = 0.5

[/ QUOTE ]
This is on the right track, but not quite right. You've actually found X (or n) needed to get your standard error within 0.5. What you want is for your true rate to be within 0.5.

[/ QUOTE ]

I said I wanted to just pin my winrate down to plus or minus one BB. That means I need a SE of ~0.5. (Plus or minus two SEs from my sample mean).

[/ QUOTE ]
I guess we're using different terminology. In probability, the sampling error for a mean is Z*(standard deviation)/n.

[/ QUOTE ]

No, you and I are talking about two different things.

[numeri]

[ QUOTE ]
[ QUOTE ]
Not quite right [img]/images/graemlins/tongue.gif[/img]
Z = 1.96 for 97.5% confidence interval

[/ QUOTE ]
Nope. We want P(Z>|Zo|) = 0.05 = 2*0.025 = 2*P(Z>Zo), so Zo = 1.96. It's a +/- thing.

[/ QUOTE ]
Fixed my own equation, cause I was typing too quickly between classes.

[numeri]

[ QUOTE ]
No, you and I are talking about two different things.

[/ QUOTE ]
OK. I'm not trying to be a jerk about it. I'm just missing something.

EDIT: OK, I finally figured it out. You posted above that you wanted your standard error to be ~0.5. That's the same as saying I want 1.96*SE = 1. So I'm just being a nit about it, where you're doing a rough calculation.

Mine should change to:

1.96*SD/sqrt(n) = 1

So n ~ 1,245 or about 124,500 hands.

Sorry for being a nit. [img]/images/graemlins/laugh.gif[/img]

[CaptainCrunch]

[ QUOTE ]
It had to be repeated to me a few times when I first started reading the forums. And as tempting as it is to say 'quit whining' I think we need to remember where we came from.

[/ QUOTE ]

Well, as one of the guys this thread is undoubtably aimed at, thanks for this thread.

I know its damned annoying to see "newbs" whining non stop. I try to not whine so much as I plead for info... anyways!

Regarding this info needing to be repeated, I think it absolutely needs to be repeated down here in the micros! Not everyone is comfortable with math, or able to understand past a superficial level what SD will mean to their play. I myself am an artist... I ain't never been good at math (borderline learning disability) but damn it all if I'm going to let that stop me from comprehending this stuff! The average new player to the micro forum is not going to catch every cricical thread even in his first month of reading. (After all, you have to know what to look for before you can look for it) And once they've caught up with the critical reading, how long is it going to take before experience gives them a framework for comprehension, before they think ".....oh! Now I get it!" I think maybe I'm somewhere around that point.

It also helps that this thread was couched in more comprehensible language than they tend to be. Thank you.

Oh, and I did have a question re: Standard Deviation. Is that a statistic that is at all controllable? I don't mean "I am going to shave 5% off my SD" More like, how much of the SD rate emerges from our choices at the table, and how much emerges from the randomness of the cards dealt all around, and how much is the choices the other players make?

[afk]

Another thing worth repeating is the same goes for individual hand statistics.

"OMG I have 20k hands and I'm LOSING with KJs!"... Remember kids you've only been dealt KJs like 100 times (I'm making that number up) and it doesn't mean much.

[numeri]

[ QUOTE ]
Oh, and I did have a question re: Standard Deviation. Is that a statistic that is at all controllable? I don't mean "I am going to shave 5% off my SD" More like, how much of the SD rate emerges from our choices at the table, and how much emerges from the randomness of the cards dealt all around, and how much is the choices the other players make?

[/ QUOTE ]
I'd say it's most certainly both. More aggressive and higher VP$IP players will have bigger swings, hence a bigger standard deviation. On the other hand, no matter how TAG you try to play, there will always be a significant component of your standard deviation simply due to the cards and your oponents.

Think of it this way: If you played AA against 5 other random hands, you'll win about 50% of the time. So that's half of the time you win, and half you lose! In that case, even though you're playing the best hand and playing it correctly, you'll still lose - just because of the cards and your opponents.

Hope that helps! [img]/images/graemlins/laugh.gif[/img]

[KingOtter]

Well, I hope the math in this thread doesn't scare off anyone thinking 'I have to know THAT to play?' ... it is really just further proof of how long term poker is, and meant as bolstering evidence rather than required knowledge to be good at poker.

I only half understood what they are saying and discussing, but in all cases they came out with Really Big Numbers (tm), which is proof enough for me.

And I graduated with an engineering degree. [img]/images/graemlins/smile.gif[/img]

KO

[KingOtter]

[ QUOTE ]
Another thing worth repeating is the same goes for individual hand statistics.

"OMG I have 20k hands and I'm LOSING with KJs!"... Remember kids you've only been dealt KJs like 100 times (I'm making that number up) and it doesn't mean much.

[/ QUOTE ]

You mean the fact that my 22 is outperforming AJs is not statistically significant?? Bollocks!

KO

[afk]

[ QUOTE ]
[ QUOTE ]
Another thing worth repeating is the same goes for individual hand statistics.

"OMG I have 20k hands and I'm LOSING with KJs!"... Remember kids you've only been dealt KJs like 100 times (I'm making that number up) and it doesn't mean much.

[/ QUOTE ]

You mean the fact that my 22 is outperforming AJs is not statistically significant?? Bollocks!

KO

[/ QUOTE ]

Hahahah I think I just said that because those types of threads bother me. I don't mind them if you're just dicking around seeing what wacky results you have - but there have definitely been threads where people say "oh hay guys A8s has lost me a lot of money and A9s has made me a lot. fix my leaks."

[KingOtter]

[ QUOTE ]
Hahahah I think I just said that because those types of threads bother me. I don't mind them if you're just dicking around seeing what wacky results you have - but there have definitely been threads where people say "oh hay guys A8s has lost me a lot of money and A9s has made me a lot. fix my leaks."

[/ QUOTE ]

Oh, I agree... I was making a bit of fun of them, too.. [img]/images/graemlins/smile.gif[/img]

I think MY most embarassing post (now) was looking at the Misc. Stats screen and wondering why High Card and Top Pair were losing me so much money! Or was it when I was convinced that I was losing wayyy too much money in the blinds.... hrmm...

KO