Is this a leak?

[DMBFan23]

Party Poker 1/2, XXXXX hands.

WtSD = 28.78
W$@SD = 60.24

I'm not giving you number of hands because you'll just tell me it's not enough. /images/graemlins/wink.gif

my .5/1 numbers over 12,000 hands were about 31 and 55, respectively.

is this something I should be monitoring?

[meep_42]

What's your "Fold to a River Bet?"

-d

[tiltaholic]

I have absolutely no idea.
Furthermore, I have yet to see a coherent argument as to how these numbers mean anything, (ie, as opposed to VP/PFR) except in principle. Not because it hasn't possibly been recently written, mind you.

My 1/2 numbers are 31.6/50.4
That probably is probably a leak. Or a tumor.

[shadow29]

Those are good numbers, tilt.

[davelin]

FWIW my 1/2 numbers are similar...more like 28/58.

[shadow29]

You're probably just running poorly with good cards. You get a premium hand, go to showdown at get outdrawn. Play more hands.

[DMBFan23]

I'll post it when I get home, I had these numbers fresh in my head and I've been thinking about them for a while now. I think in addition to the FRB stat I might be giving up on some marginal hands on the turn too, when we make the decision to "call it down"

of course, the only way to fix this is to post/read hands and play better, I was just probing for a baseline to see if in general I wasnt showing down enough. fixing it is a different matter entirely /images/graemlins/smile.gif

[jay1313]

Glad someone else is in my shoes, I am 29.8 with a 50.8

[Nick Royale]

What should a good "Folded to river bet" be at 1/2$?

[DMBFan23]

excellent question /images/graemlins/grin.gif

[Nick Royale]

Didn't dare to post mine cause I'd look like a fool /images/graemlins/tongue.gif

[tiltaholic]

Your vote of confidence makes me think I don't suck at poker... for about 15 seconds. /images/graemlins/grin.gif

It seems to me that actively monitoring & trying to increase or decrease these 2 numbers could be vastly more detrimental to your winrate than noting the next 5 hands or situations where you might be folding incorrectly on the turn or river and posting it and stopping the action at the appropriate decision point.

-t

[tiltaholic]

I have no shame -- mine is 58.8% (207 out of 352).
12k hands.

[DMBFan23]

well said. I'm just checking to see whether I should focus my hand posts on that aspect of my play /images/graemlins/wink.gif

[Nick Royale]

Ok, then I dare to out my stats... thought my 51.1% was awful... and it might be

[jay1313]

mine is about 52% after 14K hands. Another question: What is your W$SD where you 'just call' the river bet. Mine is 29.74%. One of the things I struggle with in reading these forums is the number of times people fold on the river where I would call. I imagine it is a leak in my game, either I am not aggressive enough on the river (calling instead of raising) or I am calling where I shouldn't. As I use 'simple math' to look at my play, win money at sd when I call makes more then it loses but am I leaving bets on the table too much?

I should add, my River Aggression factor is 2.86

[GrunchCan]

I think they look slightly leaky, but definitely not bad.

For one thing, W$@SD is slightly high. It probably should be closer to 55-58. Your'e not far off, but you may be too tight at the end. Call more rivers with marginal made hands. But not many more.

WtSD is also a bit too low, possibly. I believe (someone correct me here) that this should be closer to 35%. If this is too low, it could also indicate you are too tight at the end. Call a couple more rivers with marginal made hands.

[chris_a]

I spent a few minutes working this out... so I thought I'd share.

How close is W$@SD to converged?

Well say your true W$@SD for your style is about 55% Most people who win decently are in this ballpark I'm guessing.

The standard deviation of the number of times you won $ at the showdown is:

sqrt( 0.55*(1-0.55)*(# of showdowns) )

So the standard deviation of the W$@SD is this divided by # of showdowns.

Thus, the standard deviation of W$@SD is:
sqrt( 0.55*(1-0.55))/sqrt(# of showdowns)

The # of showdowns is
(total hands)*(saw flop %)*(went to SD when saw flop %)

To give some numbers:
If your Saw flop % is 20% and your WtSD = 28% and you have 120,000 hands.. then the standard deviation W$@SD is...0.6% so it's close to converged (off by 1% most of the time).

But if you have 10,000 hands your standard deviation is 2.1% so your W$@SD is within 4.2% (with 95% confidence).

So I wouldn't draw any conclusions about your 1/2 play by looking at W$@SD until you have 40,000ish hands. Then you'll be within 2%.

You already probably know that it's not enough hands (at least you said so already) but this give an idea of how many is enough.

[Sarge85]

I think im in the minority but I would say -

WtSD = 28.78 - is to low

and

W$@SD = 60.24 - is to high.

Are you only willing to take hands that are near nuts to showdown?

Sarge/images/graemlins/diamond.gif

[Nick Royale]

[ QUOTE ]
mine is about 52% after 14K hands. Another question: What is your W$SD where you 'just call' the river bet. Mine is 29.74%. One of the things I struggle with in reading these forums is the number of times people fold on the river where I would call. I imagine it is a leak in my game, either I am not aggressive enough on the river (calling instead of raising) or I am calling where I shouldn't. As I use 'simple math' to look at my play, win money at sd when I call makes more then it loses but am I leaving bets on the table too much?

I should add, my River Aggression factor is 2.86

[/ QUOTE ]
My "Won $ at SD" is abourt 64% (is that the one that you're talking about?)

My river af is 2.8 and I think I'm quite pleased with it. My flop af is about the same, but I would like that one a little higher. Preflop is 1.25, a bit to low also... What's yours?

[DMBFan23]

I'm with you sarge, I think I'm skewed towards being too tight wrt showing down. I like my .5/1 numbers better. but no, I do not need the nuts at all to show down /images/graemlins/smile.gif

I think the problem is I now find myself heads up in smaller pots more often, and I'm folding to aggression where I would normally call down if the pot were a little bigger. I of course am more inclined to call down my decent hands when the pot is bigger. I'll post some river decision hands.

[chris_a]

I beg to differ with anyone who can look at this and say that it's a leak.

http://forumserver.twoplustwo.com/showthreaded.php?Cat=&Number=1293668&page=&view=&s b=5&o=

[jay1313]

I was looking at my Won $ at Showdown where I just call a river bet instead of overall (my overall is 50.8%). I don't have the product in front of me right now but I have a significantly higher win at showdown where I bet/raise around 64% I think, but I can confirm later.

My preflop aggression is 1.4 and seems to be 'right' for me, I was way low for awhile and worked on it a bit kind of going the too soft, too hard, just right thing /images/graemlins/smile.gif

[tiltaholic]

you are my hero for posting this, by the way...but doesn't it assume that W$@SD follows a normal distribution? Why should it? (not saying you are wrong)

[tiltaholic]

sorry. illuminous is my new hero.
that didn't last long for you, sorry.
/images/graemlins/grin.gif-t

[chris_a]

You're right the confidence interval part assumes a normal distribution. But sums of many indepedent random variables always approach normal distributions and it happens quicker than you think with binomial random variables like this one So it's not too bad once you get to 1000's of hands or so.

To compute the exact confidence interval of a binomial distribution would be a bunch harder and would give you almost exactly the same results.

[chris_a]

Analyzing on a hand-by-hand basis is valuable. If you think you are too tight at the showdown on specific hands then that's cool.

Drawing general conclusions from stats that haven't necessarily converged yet is premature.

[tiltaholic]

gotcha.
thanks.