re: Poker & 100nl 6max JJ in 3bet pot
As a point of interest, if he just called the turn would you have bet a brick river?
Now, generally speaking I agree with Switch. If this vs me, there's no hand in my range that you beat. Very few decent or semi-decent players will attempt to bluff a player they have no history on. And his line looks sooo much like a queen (QJ ,KQ and AQ are all in a typical player's hijack opening range) or a set. So we have no particular reason to think that a typical player bluffs. A typical player will have it very often.
But the real trick here is that we don't know that this is a typical player, and since not too long ago, I've started being very careful about folding to aggression against unknown, and it's because of this:
If you allow me to make up some numbers, I'm going to say that about 5% of players at 6-max are wild and will commonly do this with nothing. I don't know if you think that sounds high or not, but I think it sounds a bit likely.
95% of players will have it, when they play this way. Of course, the wild players can have it, too (and will play the same way) but we'll factor that in as well.
So the question is: when he raises the turn and pushes the river, how likely is he to be a wild player?
(Feel free to formulate a guess before reading on)
We need some more numbers to be able to figure this out, so we'll invent some more likely figures:
A typical - non maniac - player will have an opening range of about 15% in the hijack, in my experience. If we further trim it down based on what he might call with (and I'm relaxing it slightly because the average player is looser than they should be) I get maybe 10% of all hands. The monsters - that fit the bill for a typical player - are AQ, KQo, KQs, QJs, 88 and 66. That's 30% of his opening range (thanks, PokerStove!).
A maniac player will open maybe 50% of his range in this cut-off, will call a 3-bet with all of them, and will play the monsters the same way, but also bluff, say, half the time. Remember, we're talking about an elusive 5% of the poker population. So the maniacs will actually make this play with ~25% of their range.
(Yes, I'm going somewhere with this)
So what will happen if we put together 10 000 random players and 10 000 random deals?
Well, 9500 of them will be "typical" in the sense that they only do this with their monsters. Of the 9 500 typical players, only 15% will even open, because only that many of them had strong enough hands. But it gets even more narrow, because only 10% of the original 9 500 felt they had a strong enough hand to call a 3-bet. And of these, we know that 30% actually had a monster. So 30% of 10% of 9 500, is 30% of 950 players, or about 300 players.
Now we look at the maniacs:
50% of them thought they had a fun enough hand to open with. Their range is "narrowed" (I use the word loosely) when they raise, to be about half of their trash and the remaining monsters. Their remaining monsters is about 15% of their hands. The same reasoning gives that out of the original 500 maniacs, 250 of them stuck with us to the turn, and 125 of them bet this way. Of these 125, they will actually have it 15% of the time, making the total number of maniacs that we beat ~105.
So when we see an action like this, what is the likelyhood that our unknown player is a maniac and is bluffing?
Well, out of 10k players, and 10k random deals, we have 300 solid players with real hands, and 105 maniacs with trash.
... Given these numbers, about 30%. No, 25%!
If solid players sometimes bluff - even if it's rarely - it's closing in on a call.
... and this is why I don't like folding to unknowns in these spots. Thanks, Mathematics of Poker.