welcome to the rock.

[Aaron W.]

[ QUOTE ]
Aaron, you're completely right that if Villain has been running very cold for 150 hands that AKo/s is a possibility. Regardless, I still don't see how you're raising for value. If we assume Villain's range will include any AK combo, suited or unsuited, without being discounted, our equity is around 30%:

Text results appended to pokerstove.txt

19,870,449,984 games 150.337 secs 132,172,718 games/sec

Board:
Dead:

equity (%) win (%) / tie (%)

Hand 1: 30.2506 % [ 00.30 00.00 ] { JJ }
Hand 2: 53.9030 % [ 00.54 00.00 ] { AA-JJ, AKs, AKo }
Hand 3: 15.8463 % [ 00.16 00.00 ] { random }

We go to around 23% if he'll only raise AKs from this position. Of course, the limper will likely have a better than random hand as well.

...

Unless he's raising more than this hand range, I can't see how we get value out of a 3-bet.

[/ QUOTE ]

And this is where we get back to the whole question of how reliable your stats are after 150 hands. It's very possible that villain is actually raising somewhere between 5-7% of his hands, but we've only seen him raise once or twice because he's not getting those hands. If you had a read to suggest that he's very rocky (and I do mean a read), then you can infer from that information that he's not as likely to raise those hands. The value of knowing he open-limped QQ in one particular hand is far more greater than knowing that he's raised twice in 150 hands.

Edit: Similarly, if you've seen him raise 99 UTG one hand, then you can infer that your 3-bet probably has good value.

[VBCurtis]

If we assume his raising range is independent of table conditions (i.e. number of limpers doesn't matter), then estimating his "real" PFR% is easy, statistically speaking. Raise/Not Raise is a binomial event, and the variance of such an event is given by np(1-p), where n is the number of trials (hands observed), p is the "real" PFR%. Std Dev is the square root of variance.

For this player, let's pretend "real" PFR is 2%. Then variance is (150)(.02)(.98), which is about 3. The square root of 3 is 1.7, roughly. This means after 150 hands, we would expect this 2% PFR rock to raise 3 times, with std deviation of 1.7. 90% confidence is about 2 Std Dev's, which means 0 raises and 6 raises (which would be 4% on a HUD) are within 90% interval for a "real" 2% PFR person.

Repeat for 3% "real" PFR: (150)(.03)(.97)=4.5 variance, sq root is about 2.2. Thus "avg" raise 4.5 times in sample, with 90% interval no raises and 9 raises. If you are an eternal optimist, 3% is the highest PFR you could assume, which is roughly AK, QQ-AA. Pokerstove that, and you're in trouble.

I don't think we need to be 90% certain of actual PFR to judge that we're WAY behind here, and should probably fold. Our pot equity vis-a-vis pokerstove might nearly justify cc'ing, but nearly all of us are either flopping a set or folding, which is clearly less equity than pokerstove implies. If you're playing JJ here because it's pretty and you hope to flop a set, you should also play 88-TT for same reasons; in fact, 88 might play better against his range, since more flops give you straight draws while totally missing him.

I'm interested in how anyone thinks this changes with QQ. I see no change, since his PFR range remains KK, AA, maybe AKs or AK. We're either coin-toss or dead meat. How powerful is the voice in your head: "QQ is pretty, I must play it"? If you think TT is an obvious fold, ask what QQ is OK against that TT isn't of his possible holdings.

This is my first post in micro- flame gently.
-Curtis

EDIT: Aaron's "maybe he's a 5% running bad": (150)(.05)(.95)=7.2 Sq root is 2.7, putting 90% confidence interval as 2.1 raises minimum (150*.05 minus 2 std dev's). This means there is a 5% chance he is a 5% PFR. (90% interval means 5% are below the interval and 5% are above the interval)