Continuation Bet Article

[DWarrior]

[ QUOTE ]
It's already been published

[/ QUOTE ]

where?

[Skuzzy]

[img]/images/graemlins/grin.gif[/img]

I think your other posts will get flamed more often if this is how you respond to it.

wp - excellent

I think a lot of us would like to see the full version - where is it published or will u send it on request?

[binions]

[ QUOTE ]

1. At small stakes No-limit Holdem the most optimal play is to vary your continuation bet sizes between ¾ of the pot and the full size of the pot.


2. Keep in mind also that 70% of flops are missed by our opponents.

[/ QUOTE ]

1. Some players prefer a standard sized bet on the flop no matter if the flop hit them or not. Same amount, every time. Very hard to read. Others vary their bet amount depending on the texture of the board. Either works. As does betting less than 3/4ths of the pot. I know from experience.

2. And an unpaired hand will flop at least one pair 32% of the time. If the hand is suited and or connected, it will hit the flop with a draw another 5-15% of the time. So, your 70% figure is a little high.

[FreakDaddy]

[ QUOTE ]

1. Some players prefer a standard sized bet on the flop no matter if the flop hit them or not. Same amount, every time. Very hard to read. Others vary their bet amount depending on the texture of the board. Either works. As does betting less than 3/4ths of the pot. I know from experience.


[/ QUOTE ]

That's partly the point of the article, as I used to be one of these players as well. What I'm stating, if you read the whole thing, is that there is a more optimal way to make continution bets that maximizes your wins when you're ahead, minimizes your losses when behind, and is still impossible to read.
[ QUOTE ]

2. And an unpaired hand will flop at least one pair 32% of the time. If the hand is suited and or connected, it will hit the flop with a draw another 5-15% of the time. So, your 70% figure is a little high.

[/ QUOTE ]

Yes, I think in the final I changed that to 65%, which apparenly I cut and pasted a previous version. Thanks though.

[binions]

[ QUOTE ]


That's partly the point of the article, as I used to be one of these players as well. What I'm stating, if you read the whole thing, is that there is a more optimal way to make continution bets that maximizes your wins when you're ahead, minimizes your losses when behind, and is still impossible to read.


[/ QUOTE ]

I read the article. I don't think there is any way you can draw a conclusion as to the optimal amount to c-bet. Sorry. Way too many variables involved other than game theory. For example:

How big are the stacks
Position
Heads up or multiway
Whether particular foe respects your play or is a calling station
Whether particular foe will fold to a smaller, half-pot/65% pot bet
Texture of the board
How big is the pot (ie how big was the preflop raise)
What gear are you in (TAG or LAG)
The strength of your cards
Reads

Anyone can say - "bet 75% of the pot 50% of the time when you miss, and 100% of the pot 30% of the time when you miss, 75% of the pot 15% of the time when you "hit" and 100% of the pot 85% of the time when you "hit," (whatever hit means).

It's a whole other thing to prove that's "optimal" over the course of one session, much less 100s of sessions.

I mean, since you "hit" your hand a lot less often than you miss, I would simply call or raise every 75% pot bet you made if I knew your criteria, as the majority of your 75% bets will be air.

I appreciate your effort, though.

[DWarrior]

Well, I think for starters it might help c-betting closer to what you would have done with a real hand.

I usually bet 2/3-pot when I hit a hand, depending on the texture (the standard is 2/3 for draw-free, 3/4-pot for flush), so then duplicating that strategy for c-bets might bring it closer to optimal. Usually I'll also round down for the c-bets, so as to increase my pot odds without affecting the fold equity (I think). Also, I sometimes won't c-bet at all, when I see the villain likes to call, or if my table image is aggressive.

I still have to run numbers on my plays, hopefully when I get ~50K hands, but I would imagine my numbers are sorta around what FreakDaddy suggests.

I don't think he meant strictly "20/30/50" breakdown with say AKo, I read that as your actions depending on the situation should resemble something like that.

I remember Sklansky for his pre-flop strategy to pick two arbitrary cards (I think he said black Kings) and occasionally limp for deception, but I don't think this is what OP meant here.

Same with betting a legitimate hand.

[FreakDaddy]

[ QUOTE ]


I read the article. I don't think there is any way you can draw a conclusion as to the optimal amount to c-bet. Sorry. Way too many variables involved other than game theory. For example:

How big are the stacks
Position
Heads up or multiway
Whether particular foe respects your play or is a calling station
Whether particular foe will fold to a smaller, half-pot/65% pot bet
Texture of the board
How big is the pot (ie how big was the preflop raise)
What gear are you in (TAG or LAG)
The strength of your cards
Reads

Anyone can say - "bet 75% of the pot 50% of the time when you miss, and 100% of the pot 30% of the time when you miss, 75% of the pot 15% of the time when you "hit" and 100% of the pot 85% of the time when you "hit," (whatever hit means).

It's a whole other thing to prove that's "optimal" over the course of one session, much less 100s of sessions.

I mean, since you "hit" your hand a lot less often than you miss, I would simply call or raise every 75% pot bet you made if I knew your criteria, as the majority of your 75% bets will be air.

I appreciate your effort, though.

[/ QUOTE ]

Ummm, No you didn't read it. And in the whole article I cover all the above that you mentioned and explain why the particular bet sizes are optimal in those situations including several examples. I mean every single thing you mentioned above is discussed in the article.

I included a disjointed section that discuss players to bet/not bet against, as well as adjusting based on your playing style. Perhaps you didn't read that post as well.

[binions]

[ QUOTE ]

Ummm, No you didn't read it. And in the whole article I cover all the above that you mentioned and explain why the particular bet sizes are optimal in those situations including several examples. I mean every single thing you mentioned above is discussed in the article.

I included a disjointed section that discuss players to bet/not bet against, as well as adjusting based on your playing style. Perhaps you didn't read that post as well.

[/ QUOTE ]

Dude - Even your math is off.

First, you correctly point out the percentage of the time your bet needs to work given how large your bet is. 50% is breakeven for a pot sized bet.

Then you assume your foe calls 50% of the time.

Then you say you are +60BB over 100 hands if you alternate between 75% pot bets and full pot bets.

I get 75BB, not 60.

To take your assumptions:

100 flops. 50 flops you bet 9BB (75% of 12BB pot) with air and get folds 50% of the time or 25 flops. 25*12 - 25*9 = +75BB

30 flops you bet 12BB (full pot) with air and get folds 50% or 15 times. 15*12 - 15*12 = 0 EV

20 flops you don't bet.

So, over 100 flops, your EV using your mix is 0.75BB.

Using the same 50% call-fold assumption and 80% bet assumption, now let's look at a mix of 100 flops where you always bet 75% of the pot.

40*12BB - 40*9BB = 120BB or 1.2BB per

What about betting 67% of the pot, using the same 50-50 call-fold assumption vs. 80% bet assumption?

40*12BB - 40*8BB = 160BB or 1.6BB per.

Hmmm. Looks to me like the lower your c-bet, the more you make on it given a 50-50 call/fold ratio.

Of course, what you give up is value when you bet that smaller amount with your good hands.

But it is impossible to determine how much value you give up. First, you have to figure out what a "good hand" means. Top pair? 2nd pair? What about draws? Then you have to figure out how often you get good hands vs. air. Then you have to figure out how often you get outdrawn. Then you have to figure out whether the smaller value C-bet can make up the difference in value over the last 2 betting rounds vis a vis the larger value c-bet.

There is no way to PROVE an optimal c-bet size.

All we know for sure is:

1. The smaller the c-bet, the less often it has to work to break even. Likewise, the smaller it is, the more profitable with any given call-fold %.

2. Betting slighly more than half the pot gives the other guy less than 3:1 odds, which is a mistake to call most drawing hands.

3. We want to induce mistakes when we have good hands

4. We want to disguise when we have good hands vs. air.

These are some of the reasons why Harrington and Gordon talk about c-bets in terms of half the size of the pot. See HOHI at p. 279 and LGB at 61:

[FreakDaddy]

The table is formated incorrectly, but thank you for double check regardless. The original article has it's complete format and is correct in it's math, I assure you.

After it is released, I'll will post the entire thing here for review. Thanks.

[Inf1n1tY]

You need to be successful 44% of the time with a 3/4 sized bet and 41% of the time with a 2/3 sized bet to have +EV.

greets