Please help me analyze my SB and BB win rate

[RiverJohn]

I ran a limited number of hands through PokerStat and come up with win (LOSS) rates for my small and large blinds.

Of course, theoretically, these should be the only positions where I am losing money. (and they are, thank God)

Granted, I have a small sample size, but here are my results:

SB...Average per hand: -0.157 big bets (2844 hands)
BB...Average per hand: -0.171 big bets (2833 hands)

I narrowed my analysis only to the games which I usually play (1/2 and 2/4). Are these decent win (LOSS) rates? Considering that if I folded every hand from the SB, I would lose 0.25 big bets and from the BB, 0.50 big bets, the money I make back by playing the way I do is as follows:

SB: +0.093 big bets
BB: +0.329 big bets

As a side note that may possibly factor into my following questions, I calculated that my big blind was raised 51.5% of the time. This surprised the Hell out of me. How relatively aggressive would that make this game?

My first question is: Does this indicate that I am playing too tightly in the SB? From the SB I recoup a mere 28% of the amount I recoup in the BB. I realize that there is twice as much to recoup from the BB. And I also realize that I cannot possibly recoup a full half the amount from the SB, as I am in bad position relative to the BB. (Are there other similar reasons that prevent me from doing this?)

My second question is: Am I able to estimate what my MINIMUM win rate from UTG may be based upon these figures? Or can I tell the amount of hands which I should be playing in this position? These may sound like ludicrous questions, but think about them for a while.

Another similar question, maybe more on track, if I had an infinite number of BB hands to analyze, could I infer that I should be playing every hand from early position that breaks even (or better) from the BB?

[Bozeman]

The BB rate looks very good to me, the small blind rate looks pretty good. Are these short-handed games?

I don't think that you can estimate UTG EV from these numbers.

A minimal model for EV in each position would include: baseline win rate, position dependent correction (one for linear, more for non-linear models), live big blind correction, small blind correction (possibly a discount of the BB correction). From your two data points, you can't extract 4+ unknowns.

Willing to give out your #'s for each position?

Craig

These are full games (8-10 players).

I would be more than happy to give out my win rates, etc., from each position, except that I just installed WIN2000 and my PokerStat is not accepting the old registration code. I've sent an email to Tony to correct this problem.

I'll post them when I'm able.

Now I realize why I should have paid more attention in my college higher math courses.

ThanX for your help!

[RiverJohn]

OOPS. OK...I'm logged in now! Perhaps another reason to do away with anonymous posting?