|
YuujiA
|
02-27-2008, 03:44 AM
Post subject: Pokertracker and Limit Hold em
|
#1 (permalink)
|
|
Join Date: Nov 2006
Posts: 25
|
|
I wasn't sure if there is a separate forum for this topic, but I figured I should post it here since most of my data is on playing limit anyway.
So we always talk about plugging leaks and how PT can assist us in doing so. I was wondering what exactly are you looking for when searching for a leak? I understand the basic stuff like VPIP and AF but they can only tell you so much right? Do you mainly use PT to dissect specific hands to see if you could have saved/gained bets?
|
|
|
Play for FREE and practice your game at...
Join the FTR Poker Forum to disable these banners and start posting!
|
|
VinceSincere
|
02-28-2008, 02:53 PM
Post subject: Re: Pokertracker and Limit Hold em
|
#2 (permalink)
|
|
Join Date: May 2007
Posts: 24
|
|
Quote:
|
Originally Posted by YuujiA
I wasn't sure if there is a separate forum for this topic, but I figured I should post it here since most of my data is on playing limit anyway.
So we always talk about plugging leaks and how PT can assist us in doing so. I was wondering what exactly are you looking for when searching for a leak? I understand the basic stuff like VPIP and AF but they can only tell you so much right? Do you mainly use PT to dissect specific hands to see if you could have saved/gained bets?
|
With PT, you can show your +/- totals for any given starting hand. I remember when I first started using PT, I looked at this information and was surprised to see that the hand that I had lost the most money with was AQo, followed closely by AQ. I began to play these hands much more carefully, and my results have shown improvement as a result.
You can also see your +/- totals by position. In my case, I'm net positive in every position except for the blinds. However, that wasn't always the case. After noting that I was net negative in UTG+2 and UTG+3, I started playing tighter in those positions and saw my results improve.
Keep in mind that any results you do see when doing this type of analysis will be limited in value until you have a good sample size in your database. I would take any information that involves less than 10K hands with a grain of salt.
|
|
|
|
Hermann the Lombard
|
|
Flush
Join Date: Aug 2007
Posts: 270
|
|
From Pat at pokertracker.com:
Posted: Sat Oct 06, 2007 8:15 am Post subject:
--------------------------------------------------------------------------------
Free video
http://www.stoxpoker.com/ptredirect.php
Free articles
http://www.bet-the-pot.com/poker-tra...ne-page31.html
Pay for guide
http://www.pokertrackerguide.com
_________________
|
|
Oh, no! Not another learning experience!
|
|
DrivingDog
|
02-28-2008, 04:14 PM
Post subject: Re: Pokertracker and Limit Hold em
|
#4 (permalink)
|
|
Full House
Join Date: Jul 2007
Location: UK
Posts: 923
|
|
Quote:
|
Originally Posted by VinceSincere
I would take any information that involves less than 10K hands with a grain of salt.
|
I haven't actually done the math on this, but my guess is you need more like 50-100k hands to even get a reasonable idea of how you do in any given position, and something like 300k hands to see how individual cards perform (depending on the cards themselves). Some hands will come up more often than others, e.g., AQ0 > AA > AQs, and others will not often be played in certain positions, so the latter need larger overall samples than the former. Also, some hands are inherently variable in their performance, e.g., suited connectors, so these need larger samples also.
To get a better idea of when your numbers become reliable, you should compare similar hands and see if they give comparable results.
One shortcut to looking at individual hands' performance is to pool the data from several similar hands, like 98s, 97s, 87s. This will give you a better overall idea of how those hands perform than looking at any one of them individually. In principle, these hands are more or less of equal value and should perform more or less the same. However, overall this month for me i've played about 60k hands. 98s is -0.1 BB/hand loser over 199 hands, 97s is -.31 over 175, and 87s is + 0.24 over 185. I don't think this really means 87s is a better hand than the other two - i'm much more inclined to think the average of -0.05 is closer to the truth for all these hands (it's been a bad month). But this just illustrates why you need such large samples to examine certain types of hands like suited connectors.
So overall my advice is if you are going to look at individual hands, it's best to pool similar hands together and see how they perform. This solves some of the sample size issues as these hands should play about the same anyways.
As for me, I mostly use PT to gain information on my opponents. I did use it to analyse individual hands once but it was a lot of work and didn't really tell me anything i didn't already know.
|
|
"You can fool some of the people all of the time, and those are the ones you want to concentrate on." (George Bush).
|
|
YuujiA
|
|
Join Date: Nov 2006
Posts: 25
|
|
well i guess the question that has really been on my mind (in light of this terrible losing streak I'm having) what would be the most sound method by which to analyze your PT stats to find out if you are playing like a losing player or just suffering from variance?
|
|
|
|
DrivingDog
|
|
Full House
Join Date: Jul 2007
Location: UK
Posts: 923
|
|
try this thread:
http://www.flopturnriver.com/phpBB2/...te-t67768.html
if your 95% CI does not include a positive number then it's time for concern.
this one also might be enlightening:
http://www.flopturnriver.com/phpBB2/...ds-t67765.html
|
|
"You can fool some of the people all of the time, and those are the ones you want to concentrate on." (George Bush).
|
|
MTTGuy
|
|
Join Date: May 2008
Location: new york
Posts: 11
|
|
|
|
|