fuzzy thinking?

[tewall]

I agree with your comments. One value of the discussion is it can help isolate what the really important factors are. Obviously in real time one (not including Sklansky) couldn't even begin to do an analysis like this.

[tewall]

Simplifying this some, the approximate cost of the bet is P * (prob crazy ahead of limon) * (prob crazy calls). The approximate profit is P * (prob limon beats old man) * (prob crazy folds). Let p1, p2, p3 and p4 be the respective probabilities. Noting that p4 = 1-p2 gives:

P * p1 * p2 = cost
P * p3 * (1-p2) = profit

Equating the two gives: P * p1 * p2 = P * p3 * (1-p2).

Using your assumptions of p1, and p3 being 1/2, this simplifies to p2 = 1-p2.

So the bet is profitable if crazy is more likely to fold, and not so if he's more likely to call.

This is a nice elegant result, and one the jibes with one's intution. The bet is good or not depending on how likely crazy is to fold.

[Ulysses]

I don't care about the model or what the right numbers are, so I'm ignoring anything you wrote about that. Two comments re: your assumptions, though.

[ QUOTE ]
For example, I give it a 50% chance that he's ahead of old man (which I think is very off).

[/ QUOTE ]

I agree that 50% is way high for that. I came up w/ a lot more possibilities than you did in my initial response to you, but 50% seems very high.

[ QUOTE ]
The 80% only indicates that someone is playing the cards and not some random number generator. If you say he's only 50% to call if ahead and 50% to fold if behind, then you might as well say that crazy is playing blindly with no bias as to what he's holding.

[/ QUOTE ]

Here I disagree. limon thought that the guy had a small pair and would likely fold that small pair to a $500 turn bet. Given the fact that limon made this bet, I think the chances that crazy has a better hand that he will call the bet with are far less than 80%.

The other night, there was $130 in the pot. Flop was Queen-high. Someone bet $60. I was sure he had a weak Queen. I made it $200 straight w/ pocket Tens on the button. If someone tried to analyze that like you're analyzing this hand, they would probably come up with something like an 80% chance that he'd call the raise with a better hand than mine. But the very fact that I made that raise knowing the guy likely has a Queen makes that probability way less. That's why I think your 80% is way off. Anyway, as expected, he folded. And, as expected, he had QT. And I'm sure you know who it is based on that description.

[jen]

Almost - you're missing one factor:

P * p3 * (1-p2) * (p1) = profit

Which changes your equation to this:

P * p1 * p2 = P * p3 * (1-p2) * (p1)

Using the assumption: p1 = p3 = 0.5

p2 = 0.5 (1 - p2)
p2 = 1/3

This means that if crazy calls greater than 1/3 of the time when crazy has the best hand, then the bet becomes unprofitable.

[tewall]

The first time I did this I came up with 1/3. I didn't publish that result, but I have it in my notes in front of me.

[jen]

Yes, a nice, elegant oversimplified result...

I don't think I'm going to play with you anymore, tewall. Are we done? [img]/images/graemlins/smile.gif[/img]

[tewall]

There's still Ulysses to deal with.

I'm done. My head hurts.

[limon]

i agree and 9 out of 10 i do this exactly but i think i just got greedy. i had visions of tripling through and it was possible given the way the table was playing. crazy guy was jamming so hard that even if a flush arrives if he has a decent hand say a pair with a QD hell go all in and someone with a smaller flush or a set would never be able to get away. when i just called on the flop i asked myself the question: is letting people in more likely to brak me or get me paid off huge...i opted for the latter. i also KNEW that crazy guy would not lay down any type of hand to an over the top raise so why do it? there are also very aggressive players in that game who were still tring to bully crazy guy and if one of them moves in there as likely on a big draw as a big hand so im prepared for that. on the turn when he checks, which is generally a straight forward play by him in my short observation i have better reason to beleive i can get him to fold something. ill admit the game was such i had truoble making comfortable plays. how often do you see raised pots 8 handed and a total x factor in the game causing everbody to shift? its hard to be totally prepared so i was just not totally confident i maximized here.

[Matt Flynn]

definitely like playing for the whole enchilada every once in a while. seems that it's a question of percentage, and 10-15% feels comfortable in nut draw situations for no math reason at all, just comfortable. more if the stacks are deeper and people are nutty. your hand's great for it naturally: you have multiple potential underdraws and weak come-alongs and are willing to go all-in. just didn't work that time. on the turn you weren't happy with your bet and me neither.

when the game gets revved up with huge multiway action it's easy for me to lose table sense. too many people. my best game is 5-6 handed plhe with 3-4 players seeing the flop. when I can find it that is, which is just about every Saturday night late nowadays. hence my refound enthusiasm for poker. we have several crazy and semicrazy players making a medium x. hard for me to adjust and i sometimes find myself palying back hard at the crazy callers. better to lie back and pick them off with the big hand / big draw and press when there's weak blood in the water. still, nice to play just almost well and still make a killing.

btw very much enjoy your sense of humor and general grump. especially liked your "maybe with my bankroll" response to the guy who said you weighed 315 lbs.

matt