ABCD Theorem - manipulating C

[Mr. Bucket]

First off I want to say this is not a strategy thread. I've just been thinking about this stuff alot recently, and would appreciate any comments on my thought process. Most of the value people will get from this IMO will come from people who haven't ever read Renton's original thread.

If you aren't familiar with the ABCD theorem go to the following link:
http://www.flopturnriver.com/pokerfo...35.html#688308

What I have been thinking about mostly deals with how we should manipulate our C range, and how that should relate to the flow of the game.

Example 1. 100 bb effective, villain raises CO to 4xBB. We have only played a few orbits with villain, and potential spots for us to 3bet him haven't come up yet, but from data mining we know he is a thinking LAG who is opening 25% of hands from CO. We are on the Button. Assuming our villain's continuing to 3b range is JJ+, AK (3% of hands) he is going to be folding to our 3 bet 88% of the time (1-3/25)*100%. Lets assume that the blinds will fold 95% of the time so our 3b takes down the pot roughly 84% of the time.
\/
If we 3bet him to 12bb makes the expected value of our 3b bluffs ( C range):
.84*5.5bb-(1-.84)*12bb=2.7bb

Ie. 3betting any hand will immediately give us 2.7bb (more if we block some of his continuing range) on average just from FE. The problem, is 3 betting against a thinking opponent will cause adjustments. Every time we 3bet, thinking opponents at our table will increase their continue to 3b% against us. This means our 3bets for value gain equity, and our 3bets as a bluff lose equity.

Because we want to keep our C range profitable, we fold a wide range of hands, even though 3betting them would be profitable.

The thing is, this C range can vary dramatically. As Renton put it this range should be "as wide as we can get away with". In my mind, this doesn't mean there is an exact % of hands we can put into our C range and get away with it against this villain. In fact, I feel our C range should be adjusted based on how much we feel this particular 3bet will make our villain adjust.

So lets assume in this hand we had AA and 3bet for value. The next orbit comes around and he raises his CO again to 4bb. My point here is that our C range from the second hand should be a much tighter range our C range in the first hand.

So our C range gets bigger or smaller based on both how profitable a bluff raise is, and how much we feel 3betting will make our opponents adjust. If we haven't 3bet a reg on our right after 10 orbits, our C range should be quite large, whereas if we 3bet a reg on our right who we know is capable of adjusting the last 2 rounds, our C range should be extremely small, or even 0.

From this reasoning I feel it is highly profitable to be a very positionally oriented 3bettor at the micros. Although perhaps at higher limits opponents will be able to pick up on your 3betting range being very high/low at certain positions, microlimit regs are just looking at a 3b% stat. Clearly 3betting buttons is more profitable than other positions so we should already be doing it more here, but what I am saying is that our 3bets from any position cause virtually the same amount of adjustment. For this reason, our bluff:value ratio in less profitable situations can be very low, or even 0 whereas the same ratio should be much bigger from the button. What I'm getting at here, is if this reg on our right has a tight opening range from UTG, we can actually 3bet bluff him 0% of the time here because he's never going to pick up on it. Our 3bets from more profitable positions will cause him to adjust to our 3 bets in general, so there is no reason to balance our range in these spots.

There is probably more I wanted to say here, but I'm tired and need to sleep soon.

Thanks

[Shotglass]

I just briefly skimmed through this and caught a couple of quick things:

So lets assume in this hand we had AA and 3bet for value. The next orbit comes around and he raises his CO again to 4bb. My point here is that our C range from the second hand should be a much tighter range our C range in the first hand.

So our C range gets bigger or smaller based on both how profitable a bluff raise is, and how much we feel 3betting will make our opponents adjust. If we haven't 3bet a reg on our right after 10 orbits, our C range should be quite large, whereas if we 3bet a reg on our right who we know is capable of adjusting the last 2 rounds, our C range should be extremely small, or even 0.

This should really be dependent upon two things: is our opponent thinking above level 1 and did we recently go to showdown after a 3b for value? (hopefully we did not go to showdown after 3b bluff)

Our 3bet range can be fairly wide if we've recently shown premium holdings after 3bets, and this will be less dependent upon our opponent's level of thought. People tend to remember what they are shown and even a non-thinking fish will remember that a few orbits ago you showed AA after 3betting. If, on the other hand, we have not, recently, gone to showdown after 3bets then our range should be significantly smaller.

Before we begin making adjustments to our play we need to determine if our opponent is playing poker (thinking above level 1) or playing cards (level 0 or 1) While it will be profitable to make adjustments to our 3bet ranges against a thinking opponent, if we start appling this same thought process to an opponent thinking along 0 or 1 then we will begin to outplay ourselves by making adjustments to our ranges that are not needed. We can reasonably exploit these low level thinkers based on their previous actions, which should normally indicate future tendancies as they rarely begin to change their actions based upon our actions.

[rpm]

If we 3bet him to 12bb makes the equity of our 3b bluffs ( C range):
.84*5.5bb-(1-.84)*12bb=2.7bb

you seem to be confusing "equity" with "expected value". you are here talking about the immediate "EV" of 3betting some hand.

"equity" refers to the percentage of a pot your hand is "entitled to" at showdown on certain boards versus a range of hands your opponent can have. or put another way, what percentage of the time your hand will win versus an opponents range if you are guaranteed to make it to showdown. i'm sure you understand these differences, it's just a terminology thing.

cool thread idea. i'll save my responses until tomorrow when my brain isn't completely fried from trying to think about numbers for hours on end.

[daviddem]

from data mining we know

Haven't read the post yet, but if you're a data mining cheat, better not mention it here...

Also:

LAG who is opening 25% of hands from CO

I am surely not LAG and I open more than that in the CO.

Other than that, nice post. You could generalize that to how metagame considerations affect your opponent's range.

[NightGizmo]

From this reasoning I feel it is highly profitable to be a very positionally oriented 3bettor at the micros. Although perhaps at higher limits opponents will be able to pick up on your 3betting range being very high/low at certain positions, microlimit regs are just looking at a 3b% stat.

I'm pretty sure that being a "positionally aware 3bettor" is going to be necessary for all levels, not just the micros. What is the alternative? Are you going to try to 3bet the same percentage from UTG+1 as you do from the BTN? At any stake, you should be 3betting more from successively later positions.

I am surely not LAG and I open more than that in the CO.

Ditto.

[Mr. Bucket]

From my original post:
"Clearly 3betting buttons is more profitable than other positions so we should already be doing it more here, but what I am saying is that our 3bets from any position cause virtually the same amount of adjustment. For this reason, our bluff:value ratio in less profitable situations can be very low, or even 0 whereas the same ratio should be much bigger from the button. What I'm getting at here, is if this reg on our right has a tight opening range from UTG, we can actually 3bet bluff him 0% of the time here because he's never going to pick up on it. Our 3bets from more profitable positions will cause him to adjust to our 3 bets in general, so there is no reason to balance our range in these spots."

I'm pretty sure that being a "positionally aware 3bettor" is going to be necessary for all levels, not just the micros. What is the alternative? Are you going to try to 3bet the same percentage from UTG+1 as you do from the BTN? At any stake, you should be 3betting more from successively later positions.




.

[Mr. Bucket]

data mining allows me to put in an exact CO range and allows there to be no comments about the potential for him not being a reg, thats only reason im posting it

Haven't read the post yet, but if you're a data mining cheat, better not mention it here...

[Mr. Bucket]

I really don't think showing down a strong hand allows us to widen our 3b bluff range very much because against regs they know thats in our range and so it doesnt surprise them. Against fish I dont care if they saw me show down strong, I'm not going to have a C range against them. Maybe it allows us to widen our 3b bluff range a little bit, but not much.

And at level 1 v level 0, I'm just not really having a C range against fish again.

Our 3bet range can be fairly wide if we've recently shown premium holdings after 3bets, and this will be less dependent upon our opponent's level of thought. People tend to remember what they are shown and even a non-thinking fish will remember that a few orbits ago you showed AA after 3betting. If, on the other hand, we have not, recently, gone to showdown after 3bets then our range should be significantly smaller.

Before we begin making adjustments to our play we need to determine if our opponent is playing poker (thinking above level 1)

[Renton]

I know it's been a while since you posted this thread, I didn't get to read it until now. You seem to have the right idea, I'll just say a few things.

- You said that "clearly" 3-betting from the button is better than other positions, this is patently untrue. The BB is by far the most profitable 3-betting position, followed by the SB, then button. This has mostly to do with the fact that you are dealing with much tighter open ranges from the cutoff or earlier positions, but also because you are laying a somewhat better price on the steal. Also you are last to act, and people tend to call instead of 4-bet when they have position, so you get more postflop value.

- I also think that your assigned folding frequencies for the cutoff opener and the blinds left to act are far too high. At mid-stakes my 3-bet success from the button is barely over 50%. I'm sure at the micros it would be significantly higher than that, but definitely nowhere near 84%. I would guess the average is around 60-65%.