PDA

View Full Version : Buyins: Mathematics of Confidence Intervals



LowWater
06-24-2010, 07:18 PM
I'm trying to find the right way to ask a question.

I know a winning player who suffered through a 35 buy-in downswing. As I understand it a losing player cannot have a downswing; for them a downswing is an indirect measure of time. As long as it was a given that 95% of the time his downswing would be under 35 buy-ins, what other conclusions can be drawn?

Suppose his win-rate inclusive of this downswing was 5BB/100 over 20K Hands. With these four parameters of Bayesian statistics, what other conclusions can we draw?

1. Worst downswing: 35 buy-ins
2. win-rate = 5BB/100
3. power (N) = 20,000 hands
4. CI (confidence interval), downswing = 95%

What if we wanted to increase our insurance to say that 99% of the time, my friend's worst downswing would be less than or equal to 35 buy-ins? What further parameters do I need to know?

_Loki_
06-24-2010, 09:42 PM
Hi LowWater

uDevils Poker Results Calculator:
uDevil's Poker Results Calculator (http://www.castrovalva.com/~la/win.htm)

Here's an intelligent poker blog you might enjoy:
The Silent Four (http://silent4.blogspot.com/)

A good read (but not exactly Elmore Leonard or George V. Higgens)
LiquidPoker - The bankroll management and variance guide (http://www.liquidpoker.net/pokerarticle/146441/The_bankroll_management_and_variance_guide)

A good place to post your question:
Poker Theory - General Poker Theory Forum (http://forumserver.twoplustwo.com/15/poker-theory/)