Mass Database Analysis with holdem manager.
Posted by WM2K
Posted by
WM2K
posted in
Gen. Poker
Mass Database Analysis with holdem manager.
Hi everyone. Been getting back on the poker wagon after a hiatus and this mass database stuff coupled with piosolver sims makes a ton of sense (high fives to nick howard!).
I use holdem manager and am wondering if anyone has any insight on where to start with this idea. The problem I keep running into is that in the reports every hand and stat involves me. How would I even begin to make use of a larger database to see even simple things like average 3 bet positional tendencies from the player pool? Or reactions to c bets?
Any advice and pointing in the right direction is very much appreciated!
Cheers!
Loading 10 Comments...
With notecaddy you can create an alias for player profiles like shown here:
http://forums.holdemmanager.com/notecaddy/532096-notecaddy-2-6-beta-what-he-knows-throws-blows.html
Ya I guess this is the only or best way atm other then something like hand2note which I can t justify 45$/month for and doesnt work easily with the HHs I m using. Thanks!
Anyone have any exp doing this with stuff with pokertracker4? This is of course another option I could be open to. Always used HEM2 so theres I d have to learn to drive another program.
H2N is the best tool for MDA work. HEM and PT4 can't really compete in that arena. If you don't want to part with HEM for play, use H2N as a study tool, you won't regret it.
And welcome back! :-)
Its more that I m only playing micros at the moment till I get my teeth back and stop spewing $$ everywhere and 45$ a month is steep. On top of this Im playing microgaming so HH conversion is an additional cost. Im working a job these days on top of all this so finding the 1-2hrs a day to do hardcore studies to justify those costs is also tricky. Splurged on pio basic already cause that program is too cool not to have. Just seems worthwhile to do at least some basic database analysis to get a general framework of what to assume a readless opponent is up to.
I'm interested in this as well. Is there any material out there on what exactly H2N does so well as a study tool?
Main selling points are speed and configurability. Importing and calculating stats for big databases is blazing fast.
You can also build custom stats very fast, using a technique of first defining base stats (building blocks) that you later clone and tweak to more specific stats. Once you get the hang of that you can build extensive reports (using the Range Research feature) where you dissect the tendencies of your opponent pool.
You might start with a generic situation (say, how often a pre flop caller in position takes various lines post flop) and branch out from there to see how that varies with board texture, different bet sizings, and so on.
It feels a little like programming. Whatever stats you need you can build fast. Then you can build more stats from those, branching out from general cases to more and more specific cases.
Documentation is sparse at the moment, that's the only minus. Hopefully that will come soon.
Hmm.. no takers.
So clearly Hand2note is better. I ve tried the notecaddy hack with HEM2 and ya its slow. Better then nothing for sure as you can get the basics done and thats where I m at when it comes to this kind of work.
Anyone done research from several micro and low stakes pools? Any noticeable difference between say the 888 pool and the pokerstars pool? Just pondering the relevance of doing research using a database from one site and then playing on another. Im strictly taking about pools with large numbers of players (200nl and lower). I could see the leaks you find over 5 million+ hands are so general that it would be relevant no matter where we play. Maybe even gaining insight into the thought processes of live players?!
HEM2 is really slow, the good news is that HEM3 is in the works.
Be the first to add a comment