Poker Tracker - Starting Hands Evaluation

I decided to give Poker Tracker another chance. After doing much reading over the summer, going back to playing (low) Limit ring games, and really trying to play disciplined poker, I thought that it only made sense to add Poker Tracker back into the mix as a tool that I can use to help me grind it out for long term profit or long term loss.

Because I had such a hard time gaining any insight to my game last time I used this program, I figured I had better take a different approach for this go around. One of the biggest things I had a problem with was being overwhelmed by the vast amount of information that was available. I really did not understand how to use the tool to find leaks in my game. I was more interested in getting information on my opponents and seeing how they were playing. I spent very little time on my own game. As Ying would say "To the Google Mobile!".

As an IT professional, I have had the (mis)fortune of being exposed to a myriad of acronyms that are certainly confusing to anyone not in the know. Being in a management position puts me in a sometimes awkward position of having to know "geek-speak" and be able to interpret it for execs and other non-techies so that it is useful information. Why am I telling you this? Because there are some acronyms in the world of poker and particularly when discussion statistical data within Poker Tracker that would make a government IT professional gawk. A couple of examples; Vp$IP (Voluntarily put Money in the Pot), W$WSF (Won Money When Seeing the Flop), PFR % (Pre-flop Raise Percentage). I will do my best to explain acronyms as they show up, but there is a good "Poker Glossary" that you can refer to if you need it.

So I found a great site, "bet-the-pot.com" that details how to pick apart your staring hands and figure out where you might have leaks in your starting hand play. It explains in great detail how to use PT to setup filters so you can periodically review your play and check for leaks often. I won't go into the actual steps, think that the bet-the-pot site does and excellent job on that front.

The only suggestion I have is that instead of setting up the filters based on hands that are +EV (positive expected value) for you specifically, it may be best for a beginner to setup filters based on known "core hands" and known "marginal hands" that are more generally accepted as +EV. Just because 97s is generating +EV, doesn't mean it is a hand you should be playing it.

When I say "known hands" I am referring to hands that are generally accepted as optimal starting hands. You can find these hands in almost every instructional poker book. So as an alternative to the "finding your boundary method" as instructed offered by bet-the-pot.com (which works just as well if you know what you are doing), I offer the following as a guide: based on my readings and about 1 year of playing low-limit games with moderate success. Note that these hands exclude anything that you would play from the blinds.

Core hands (Hands that I would play from the from the first 3 seats to the left of the blind):
AA-77
AK-AJ
ATsKQ
KJs-KTs
QJs-QTs
JTs

Marginal Hands (Hands that I would play from middle to late position)
Any Pocket Pair
Any Suited Ace
AK-AT
KQ-KT
K9s
QJ-QT
Q9sJT
T9s-54s

Junk Hands would be everything that is not listed above. These are the hands I know I shouldn't play but do anyway.

Using the tool you can obtain very general information such as; I found out that AKo is a big time loser for me, more than any other hand. I am probably playing this too aggressively, treating it the same as AKs when it is clearly not as strong. This was a surprising find.
You can also get very specific information. For instance I found that QQ in early position is more profitable than AA in any position for me. In fact, QQ is almost twice as profitable.

I should mention that my database is not where it needs to be for accurate statistical analysis. I only have 3,355 hands in it so far and everything I have read says that you can not really begin to get accurate analysis of your play until you get to 10,000 hands. Just by filtering out the hands that I played as blinds from this database I am down to 2,478 hands so you can see why a large sample is so important to get accurate results. I just wanted you to be aware of this before we dive too deeply into my analysis. I am still analyzing my starting hand play, but after several hours of research these are my conclusions:

I am seeing the flop 27% of the time, a little high, but not too bad. This number falls to 17.6% when you don't count the blinds.

KK is my most profitable starting hand with a 92.31% win rate compared to only 57% with AA.
AKs and AQs are treating me well, don't change a thing with those.

A8s is losing lots of money, why the hell am I playing this hand so much?

QTs is surprisingly profitable winning 67% of showdowns with it.

AKo and 88 in early position are a big leaks - those are my two biggest losers over all.

TT in middle position is a big leak, I need to stop being too aggressive with this after the flop.

ATo, JTo, 55, and 44 are big leaks in early position, I need to tighten up a bit in that position.

JTo is a big loser - stop playing this hand unless the situation is perfect, like being first in the pot from late position.

KJo from middle may be a winning hand for me statistically but I do not like my chances of this holding up in the long run. I noticed that I limp in with this hand 40% of the time, Maybe I should raise with it a bit more often when the situation presents itself. I am also riding this out a bit to far with 41% of the time going to the river.

98s from middle is a big leak - one of the biggest leaks over all in my game. Suited connectors haven't been nice to me according to these numbers, but this was the worst of them.

Small pairs are a problem for me but 44 is bad in particular - it is one of my biggest leaks. I need to learn to let go of small pairs when they miss on the flop.

So check this out. I used my "junk hands" filter to check my play of the last week where I have lost the most money since July and found that during this time I saw the flop with exactly 5 hands that were junk. That is EXCELLENT. It doesn't put money in my bankroll, but it makes me feel better that I am not just going on random tilt.

The fact that I am not losing money on junk hands outside of the blinds made me think about the blinds themselves. It has been a fear of mine that I am playing the blinds too loosely and that this could be a serious leak. I pulled up Junk hands in either blind and confirmed my fears. Holy crap this is a big problem. Without looking too deeply at the numbers (it was a gruesome site) I could tell that I was catching small pieces of the flop and riding it out to the river. K3o is draining my account. I need to figure out how to fix that.

Sorry this was so long, technical and, well, boring probably. I needed to go through this process and I thought you might be interested in how it was done. I also thought you might like to get a peek into how I have been playing from a more analytic view rather than just "I'm playing good/bad."

Next, I am going to look at some tilts! That should be revealing.