10K Post: Blind Steals on 4th Street

[Dendrite]

So, checking the turn makes me vulnerable to value bets, but, betting the turn, putting the same amount of money in as calling the river, and potentially more on the river, doesn't?

[MrWookie]

Yes, because on the turn, there's still value in your bet. And if you're behind, there's still another street to improve.

[Caddy_4_Life]

Wookie, looks like an excellent post. I'll review it more in detail tomorrow when I can see better. I appreciate the time you must have spent on it.

On a side note, what did you spill on your shirt and why did you do it?

-Caddy

[josh_x]

Wookie,

What happens if we check behind to induce a bluff, but then on the river only call 50% of the time or some other amount (less than 'always', anyway). I'm not good enough with the math to figure this out, but is it possible that inducing a bluff with the intention of not always calling is more optimal than calling every time (with regard to your point on allowing them to value bet too strongly). I geuss this is highly dependant on your hand, i.e inducing a bluff when you hold 6 high with the intention of calling it is obv bad. I admit that i skimmed over some bits of your post, so apologies if you adressed this.

EDIT: Great post btw,

[Nfinity]

I get to play Devil's Advocate on a lot of Wookie's awesome posts. You really don't get to see the results of my fine work though, because he usually deletes the 10,000 occurences of the word "though" in his run on sentences when I tell him to.(I kid the Wookie)

I thought I would chime in with the answer to Josh_x's question, since I had already thought this matter through when proof-reading.

First of all, checking behind to induce a bluff that you only intend to call 50% of the time is a relatively arduous task. The reason being is, how would you go about basing your calling frequency? I would suggest lunar cycles, because that's bound to get you as close as anything else. The problem is, to get optimal results you would want to call more bluffs than value bets. I consider this a bit of a pipe dream. It takes a fairly decent read to evaluate what specific cards will actually improve my opponents hand. To be able to discern what cards will not hit my opponents, yet he will decide to bluff on, well that read requires a relationship with my opponent so intimate in nature that I feel flowers and candy should be involved.

The best you can hope for is that now we only call 50% of his value bets, yet only catch 50% of his bluffs. You would think this wouldn't change much, but it actually results in a fairly substantial loss in EV. if I'm not mistaken( your right, I didn't do the math)

This is because the 50% we fold to his value bet, we end up saving 1 bet, but the other 50% we end up folding to his bluff, we lose the entire pot.

I'm pretty sure that's right. If it's not I blame wook for writing an epic post and not taking the time to be thorough.
[img]/images/graemlins/smile.gif[/img]

[MrWookie]

Nfinity is right. If our opponent is bluffing more than the game theory answer to bluffing (>25% in this case), then we are giving up substantial EV by ever folding, precisely because folding to a bluff loses us 3 BB, but calling a value bet only costs us 1. If our opponent is bluffing precisely 25%, though, then it doesn't really matter if we always call or always fold OR call with some random frequency. Our EV is going to be the same, and it's going to be at a minimum with respect to any other bluffing frequency from our opponent.

You do make a good point about reading the board, and I didn't address it. If we take the checking line, we should be much more inclined to call a bet on a card that pairs the board or a 3 or a 4 than we should if a J or a Q drop. Based on our opponents range, the hands we were beating are frequently improved by a J or a Q, but any pair to the board can't help them, and there aren't a whole lot of 3's or 4's in his range, either. Working this out precisely would make for a much more complicated calculation, but it could be done. I think it could increase the EV of checking against the 100% bluffer.

Also, I do address the point about inducing a bluff with 6 high. I haven't worked it out, but it might actually be better than inducing a bluff with A high, in that you have no intention of calling the value bets. Instead, we'd call it "taking a free card" rather than inducing bluffs. This is particularly true if you are looking at a board that would imply you have very little fold equity to go with your draw. The example I gave was a gutshot royal flush draw on a board of AKxy. If your opponent called the flop with an A and a K showing, he's almost certainly got something he's taking to showdown or a legitimate draw he's not folding, and there certainly aren't a lot of 6 outers against our JT still lingering. Thus, we have a similar number of outs to the example I worked out here (zero for our pair outs), but we have zero chance of picking up the pot on the turn. Also, we have no intention of calling any bets on the river. In that case, we can take our free card, see if we hit, and if not, we can comfortably fold to any bets. The value bet is dead to us because there's no value in it.

And lastly, Caddy,

If you have not seen Airplane!, this is something you need to immediately rectify.

[F Paulsson]

"I have a drinking problem."

I learned something today. Good post.