View Full Version : Analysis Project

07-22-2005, 03:23 AM
As soon as I find the time I would like to analyze my 22 playing style. I thought of computing the following table:

Level - my VPIP - Avg No of Players at the end of level - my stack size end of level if still alive - my bust out ratio - No of me being all-in - % of all-ins won - final rank if alive at end of level. That would give me nice raw data to analyze if I push enough depending on the status of the tourney.

I would also do a special for the bubble situation and for HU.

1) Would that be worth while?

2) Any further suggestions what I should look into?

3) Any suggestions/ tricks how I could do this with the least effort? Restriction is here that I don't know access, sql and such. I thought of cutting and pasting that analysis page from the SnGPT of all tourneys that I played into an excel sheet and using excels database functions for the compiling of data.

07-22-2005, 05:15 AM
Shameless bump.

Anything unclear?

07-22-2005, 05:23 AM
ok, just to give you a response, sounds like a waste of time. You would be better off spending the time learning nl ring, watching hand hist from better players, reading the archives, comparing hand histories/ideas with someone else at your level, etc.

Not trying to be rude, it's just time is limited for all of us, I dont want you to spend alot of yours on something that when you get done you will say, hmmm, and still not know much more other than your all in win % in level 3 etc.


07-22-2005, 05:50 AM
Ok, fair statement, no offense.

07-22-2005, 11:34 AM
I think the question is a deeper one for those of us who play at low levels. That is, "How can I best analyze my play?"

Proper self-analysis is critical for any serious game. It's the key to higher levels. I had this problem in my chess game for a long time. I reached a certain level of competence, but at some point, to improve, you have to be able to sit down and analyze your games and those of mster players. But, understanding the criteria by which to do this requires some serious effort, thinking, reading, and playing against better players. Only after blood, sweat, and tears, do you have the skill to properly analyze your game and hope to understand the games of those playing way above your head.

The same is true with poker, I think. I'm going through that now...how do I analyze my play, how do I sift out the erroneous thinking when I can't recognize the errors to begin with? It's especially difficult in poker due to the short-term results factor. I had the same thought as the OP once...whip up a perl script to correlate my stack sizes at each level with finishes and all sorts of stuff. I'm inclined to agree with Lacky though.

07-22-2005, 05:00 PM
Nicely said -- had a similar take coming from a chess background as well. The nice thing about poker -- especially using something like PokerTracker is that you can go over the hands you got busted on and apply Skalansky's Fundamental Poker Theorem on it -- knowing the cards, would you have played your hand differently. Harder to find are the hands you won by luck over +EV (you sucked someone out) and hands that you weren't in but should have been ( pot odds, position, reads, etc ).

07-22-2005, 05:40 PM
I think there's a lot to be gained from an analysis of this type. I don't like the suggestion to switch to a ring game, that doesn't do you much good if your weakness is in short tables or on the bubble. Tourneys are very different from ring games.
I don't have any good suggestions for how to separate your games by level other than querying, but I would imagine that most players have weaknesses at some level in a SnG tourney. Clearly, the way you play has to change based on the size of the blinds, the number of players and the size of your stack relative to the blinds. To the extent that those average out at a given level, an analysis of your play on average could detect where you need to look to find a major weakness in your game. It won't tell you the weakness, but it will give you an idea of when it occurs. Once you've found a level where you seem to struggle relative to your skill level, then I would suggest using poker tracker to analyze hands at those levels.

I'm not great with Pokertracker, but it seems to me that if you wanted to analyze your play once the tables got smaller, you could filter only hands where you played and there were x or fewer players, then you could sort by various items of interest, perhaps your stack size, maybe sort all hands where you played by $$ won or lost and analyze your big winners or losers.

Alternatively, you could do the same type of filters and analyze your wins/losses by hand type and see if there is any type of hand that you are playing at a loss on average at different levels, or very profitable hands that you might not be playing often enough.

Macro analysis isn't going to fix any weaknesses ,but I'm a big fan of using it as a starting point.

07-22-2005, 06:13 PM
Tourneys are very different from ring games.

[/ QUOTE ]

really? After playing several hundred thousand hands of each I should have realized that by now. I feel dumb.....


07-22-2005, 09:12 PM
I've though about something similar. A session analyser that takes a SnG summary from Party and runs a purly mathmatical analysis of each action and rates each player based on their plays.

While I have the skills to do it ... well y'all know.

07-23-2005, 06:15 AM
best thing you can do is have a good player review like 10 HHs. no amount of self analyzation can do justice to a solid player offering input on even TWO hhs. no matter how hard you try, its a waste of time to analyze yer play if you dont know what yer looking for. if yer having trouble with yer game, sure, u can look and analyze and make observations and maybe come to some useful conclusions.. but until you know what yer looking for, yer wasting yer time. holla