Some thoughts on continuation bet sizing (long)

Panthro · #1 05-19-2007, 01:37 AM

Most 2+2’ers are familiar with the 4xBB + BB/limper preflop betsizing methodology, but what about post flop? Do we always bet the pot? Min bet? Shove? On one hand, if you always bet too much, you’re going to lose the most when your opponent calls with a hand that’s better than yours. On the other hand, when you always bet too little, you’re offering your opponent correct pot odds to continue with drawing hands and suck out on you. Obviously, we want our bet sizes to encourage our opponents to make mistakes against us; Big mistakes; Mistakes that maximize our expectation. So how much do we bet? What factors should we consider before sizing our bets?
In this article I plan on discussing some of the more common methodologies regarding sizing your continuation bets and some of the advantages and disadvantages that accompany the various tactics. Note that this article will not cover the “ifs” and “when’s” to continuation bet, or what conditions make a bet +EV or –EV, types of villains or flops to cbet, etc; I’ll leave that for another time. Rather, I’d like to focus explicitly on bet sizing strategies and the rationales behind why we may find it optimal to utilize a specific approach over another. It’s also worth noting that I don’t consider any one strategy to be any more or less favorable than any other; This write-up is simply an attempt to get people thinking as to why they’re betting a precise amount and the implications related with the size of their bets in different situations.

Methodology #1: Sizing your bets based on the strength of your hand
Seems straightforward enough; When you have a strong hand, you want to maximize your winnings and play a big pot, and when you have a weak hand, you want to minimize your losses and play a small pot. So you bet as the smallest amount your opponents will let you get away with when you’re bluffing, and the largest amount you think your opponents will call when you have a strong hand. On the one hand, risking a large amount just seems silly when you can bet smaller and achieve the same result; On the flip side, you’re losing a ton of value when you make a small bet when your opponent would have called a much larger bet. Theoretically, this approach would be almost flawless and poker would be much easier if our bets somehow didn’t represent the actual strength of our hand.
The obvious drawback to using this tactic is that you become very exploitable to observant opponents, as they’ll soon realize that your bet sizing represents the exact strength of your hand. A solid villain will correctly fold his marginal holding facing your pot sized bet, and will interpret your smaller bets as weakness and raise you off your hand, or float with the intention of taking the pot away it away on a later street. By telegraphing the strength of your hand with the size of your bet, you’re allowing your opponents to play perfectly against you, and you’ll wonder why you always get raised when you have nothing and always get folds when you flop the nuts.

Works best against: Weaker villains, villains who call preflop raises to play fit or fold, villains who don’t show aggression against weak bets without a hand.

Disadvantages: Against good/observant villains you lose action on your big hands, and have pots stolen from you when you show weakness with small(er) bets. Lack of balance.

Methodology #2: Vary your bet sizing based on the texture of the board
The general idea being to vary your bet sizing based on the texture of the board(duh) – bigger bets on more coordinated boards and smaller bets on less coordinated or dry boards. Since the texture of the flop impacts the shape of the hand distributions, that is, hands that the board ‘hits’ are much greater on coordinated boards than on dry boards, we make our bets larger on drawy boards and smaller on dry boards Since dry boards miss most hands, we can bet a smaller amount (~½ to 2/3rds of the pot) that will often achieve the same result as a larger bet with less risk. On draw heavy boards that hit a wide range of hands, we can make larger bets (~2/3rds to the full size of the pot) to deny drawing hands proper odds to continue that a smaller bet may not have. This strategy is very advantageous against the type of villain who will raise smaller continuation bets with drawing hands sensing weakness, but is more liable to flat call a larger bet sensing strength. After all, the semi-bluff can difficult to defend against with marginal one pair type hands, and we’d much prefer our opponents to play passively against us. Also, when we make large bets on drawy boards with strong hands, we’re anticipating getting a lot of money in the middle before the river when many drawing hands in our opponents range become worthless. It is also worth mentioning that by using a bet size methodology that advocates using consistent bet sizes relative to the texture of the board and not your hole cards, you effectively disguise your hand to your opponents.

Lets look at 3 examples: a dry board, a somewhat coordinated board, and a very coordinated board, and some bet sizes we may decide to fire into each pot.

Example 1: a dry board
Hero (BTN): $100
SB: $100

Preflop: Hero is dealt X, Y (6 Players)
4 folds, Hero raises to $4, SB calls $3.5, BB folds

Flop: ($9) A[img]/images/graemlins/club.gif[/img] 2[img]/images/graemlins/diamond.gif[/img] 8[img]/images/graemlins/spade.gif[/img] (2 Players)
SB checks, Hero bets $5.50

Example 2: a middling drawy board
Hero (BTN): $100
SB: $100

Preflop: Hero is dealt X, Y (6 Players)
4 folds, Hero raises to $4, SB calls $3.5, BB folds

Flop: ($9) 8[img]/images/graemlins/club.gif[/img] 4[img]/images/graemlins/diamond.gif[/img] 9[img]/images/graemlins/diamond.gif[/img] (2 Players)
SB checks, Hero bets $7

Example 3: a very drawy board
Hero (BTN): $100
SB: $100

Preflop: Hero is dealt X, Y (6 Players)
4 folds, Hero raises to $4, SB calls $3.5, BB folds

Flop: ($9) 9[img]/images/graemlins/club.gif[/img] Q[img]/images/graemlins/heart.gif[/img] J[img]/images/graemlins/club.gif[/img] (2 Players)
SB checks, Hero bets $8

Works best against: all villains.

Disadvantages: Small increase in losses with weaker hands/bluffs from balancing bet sizing with stronger hands and vice-versa.

Methodology #3: Adjusting your bet sizing based on the depth of the stacks relative to the size of the pot
Without divulging into too much theory, stack sizes can help us extrapolate an optimal percentage of the pot to bet. The whole idea of geometric growth (credit “The Mathematics of Poker”) is essentially betting a consistent percentage of the pot on each street to get our effective stack ‘all in’ comfortably by the river, a concept critical to success in NLHE. It should seem obvious that when we flop a monster, we want to get as much money as possible in the middle to maximize our winnings.
One of the most important aspects of stack sizes in NLHE is how they affect implied odds. When we adjust our bet sizing based on the effective stack sizes, we can reduce our opponents implied odds to call bets on early streets with inferior hands in hopes of drawing to a better hand. Also, if we assume 100BB effective stacks, it does make a lot of sense why we would prefer to bet a smaller percentage of the pot when the pot is large, and a larger percentage of the pot when the pot is small, both from a practical and mathematical standpoint.

A really really basic example:
Hero (BTN): $121
SB: $200

Preflop: Hero is dealt X, Y (6 Players)
4 folds, Hero raises to $4, SB calls $3.5, BB folds

In this example, the pot is $9 and the effective stack sizes are $117 (13 times the size of the pot), so betting 100% of the pot on all three streets should get our stack all in by the river.

Flop: ($9) 8[img]/images/graemlins/club.gif[/img] 5[img]/images/graemlins/diamond.gif[/img] J[img]/images/graemlins/heart.gif[/img] (2 Players)
SB checks, Hero bets $9, SB calls.

Turn: ($27) 3[img]/images/graemlins/heart.gif[/img] (2 Players)
SB checks, Hero bets $27, SB calls.

River: ($81) Q[img]/images/graemlins/club.gif[/img] (2 Players)
SB checks, Hero bets $81 and is all-in

Lastly, I should point out that ponicaraux made a cool write-up entitled Get it in where he mentions similar concepts.

Works best against: all villains.

Disadvantages: Small increase in losses with weaker hands/bluffs from balancing bet sizing with stronger hands. Not always optimal/feasible with deeper stacks.

Methodology #4: Always bet the pot
Many UB/FT’ers religiously use the “bet pot” button to size their bets. By making the same sized bets relative to the size of the pot, you again completely disguise your hand to your opponents. Against weaker/passive players, always betting the pot seems optimal because you build bigger pots when you’re the aggressor, and you can set yourself up for larger bets on later streets with strong hands. Weak-tight players will give up easier against you knowing they’ll be facing large pot sized bets on every street.
The drawback of this approach is that consistently building large pots results in very high variance, especially for a LAG who is playing a high number of marginal holdings. Since the pots are bigger, you will often find yourself facing difficult situations with medium strength hands, as building big pots against aggressive opponents who will check-raise with air and draws and try to push you off your hand can get expensive. Also, playing big pots regardless of hand strength reduces your ability to utilize pot control, as every pot you play in as the aggressor will be of the same size (number of opponents is relevant here). Lastly, by always betting pot you’re risking a lot to win a little when you’re bluffing.

Works best against: all villains

Disadvantages: Higher variance; Small increase in losses with weaker hands/bluffs from balancing bet sizing with stronger hands.

Methodology #5: Always bet x% of the pot, where x is some fixed predefined number
I’ll refer to this as the multitabelers theorem because many 2+2’ers use a bet pot script to handle their bet sizing. They usually have it set up to bet some arbitrary percentage based on the current size of the pot (often in the neighborhood of 60% to 100% depending on preference). As with the previous methodology, it can be very difficult/impossible for opponents to deduce the strength of your hand if you’re always making the same size bet whether you’ve hit the flop or not. With a smaller continuation bet size than the ‘bet pot’ advocacy, you can take a stab at more pots while risking a smaller amount of chips. Most of the same disadvantages associated with the aforementioned bet pot methodology are apparent in this theorem.

Works best against: all villains.

Disadvantages: Small increase in losses with weaker hands/bluffs from balancing bet sizing with stronger hands and vice-versa.

Methodology #6: Overbetting the pot
Making a large overbet doesn’t have to work all that often for it to be a profitable play. Some opponents will interpret your overbet as weakness or a bluff and call down with a marginal hand. Using this strategy extracts maximum value from ‘calling stations’ who won’t fold with any piece of the board, and the so called ‘chasers’ who won’t fold any kind of draw on any street. You can extract a large amount of chips before the river when missed draws become worthless and induce river bluffs from villains who have missed their draws and find themselves pot committed. I’ve had some success using this strategy after losing a large pot when my opponents perceive me to be on tilt, or in blind battles where opponents always think an aggressive player is FOS and trying to buy the pot.

Works best against: Weak players, calling stations, chasers, villains who don’t like to fold, villains who like to make hero calls and pick off bluffs.

Disadvantages: Bigger bets may lose action when a smaller bet would not have. Balancing overbetting strong made hands with weaker hands/bluffs can be difficult and/or suboptimal.

Methodology #7: Adjust your bet sizing on the objective you’re trying to achieve
-- credit Ray Zee, soah and a few other HSNL/MSNL 2+2’ers

The idea being to adjust your bets to manipulate your opponents into playing not only the pot size of your choosing, but the actions you may want your opponents to take against you. Sometimes you want to bet smaller when OOP to price yourself into seeing cheap cards, or to pick up the pot with minimal risk, or to induce a raise; Or bet larger to force your opponents to fold, or at the very least force them into a difficult decision as to whether or not they should continue with the hand. As an example, we might choose to make smaller ½ pot bets when you want action or want to induce a raise, and bet the pot when you don’t want action or want to discourage opponents from making plays or calling down lightly.
The disadvantage of any complex bet sizing methodology is that our opponents may or may not construe the information we had hoped to associate with our bet sizes as we have intended. Say for example, we make a bet of size A because we want our opponent to take action X; In response our opponent may instead decide to take action Y because they interpret our bet of size A much differently than a bet of size B, despite the fact that we anticipated this opponent to be much less likely to take action Y over action X if we made a bet of value A instead of value B. Still with me? OK, enough theory mumbo jumbo, let’s look at a really simple example where we might be able to manipulate our opponents into taking specific actions against us by varying our bet sizes.

Hand 1:
Hero (UTG): $100
BTN: $100

Preflop: Hero is dealt T[img]/images/graemlins/club.gif[/img], T[img]/images/graemlins/spade.gif[/img] (6 Players)
Hero raises to $4, 2 folds, BTN calls $4, 2 folds

Flop: ($9.50) 8[img]/images/graemlins/club.gif[/img] 3[img]/images/graemlins/diamond.gif[/img] A[img]/images/graemlins/heart.gif[/img] (2 Players)
Hero bets $5.50…

We’ll play the role of hero, a 18/14 thinking TAG who uses a highly varied bet sizing methodology. We open UTG open get called by the 20/10 button. In our short history, we view the button as an aggressive villain who is capable of making a move. With a dry Ace high flop, we decide to make slightly over half pot size continuation bet expecting the button to fold all worst hands and call or raise all better hands.

Flop: ($9.50) 8[img]/images/graemlins/club.gif[/img] 3[img]/images/graemlins/diamond.gif[/img] A[img]/images/graemlins/heart.gif[/img] (2 Players)
Hero bets $5.50, BTN raises $20

Easy fold right? Well, maybe. Could the button have interpreted our ~½ pot sized bet as weakness? Would a 2/3rds or near pot sized bet be more sufficient in representing a strong hand? Would the button be less likely to make a play facing a larger bet?

In the actual hand hero folded, the on collected the pot and exposed his hole cards, J[img]/images/graemlins/heart.gif[/img] Q[img]/images/graemlins/club.gif[/img].

So what have learned from this hand? The actions taken by this particular villain may or may not have been prejudiced by the size of our bet. Maybe the button was planning on making a play regardless of our bet size, and maybe he wasn’t. Nevertheless, it is something to keep in mind. On to the next meeting…

Hand 2: (same villain)
Hero (UTG): $100
BTN: $100

Preflop: Hero is dealt A[img]/images/graemlins/heart.gif[/img], Q[img]/images/graemlins/club.gif[/img] (6 Players)
Hero raises to $4, 2 folds, BTN calls $4, 2 folds

Flop: ($9.50) 5[img]/images/graemlins/heart.gif[/img] K[img]/images/graemlins/diamond.gif[/img] T[img]/images/graemlins/heart.gif[/img] (2 Players)
Hero bets $8…

In this hand, we have a read that this villain may have interpreted our smaller continuation bet as weakness, so we adjust by making a larger bet with a hand that we don’t necessarily want action with.

Flop: ($9.50) 5[img]/images/graemlins/heart.gif[/img] K[img]/images/graemlins/diamond.gif[/img] T[img]/images/graemlins/heart.gif[/img] (2 Players)
Hero bets $8, BTN folds.

The button folds and hero collects the pot. We now have a stronger suspicion that this villain may perceive our smaller continuation bets as weakness and larger continuation bets as strength, and we intend to exploit our read by manipulating our opponent into making a mistake. Now again, this may or may not be the case; Maybe the button decided to fold in this hand given he recently made a play against us, or maybe he had a timing tell and felt we were stronger in this hand than in the previous one. However, in a game with no absolutes, we still plan on experimenting with our newly gained information in an effort to size our bets to incite specific responses against this opponent in the future.

Hand 3: (same villain)
Hero (UTG): $100
BTN: $100

Preflop: Hero is dealt 7[img]/images/graemlins/club.gif[/img], 7[img]/images/graemlins/diamond.gif[/img] (6 Players)
Hero raises to $4, 2 folds, BTN calls $4, 2 folds

Flop: ($9.50) 7[img]/images/graemlins/heart.gif[/img] J[img]/images/graemlins/spade.gif[/img] 2[img]/images/graemlins/club.gif[/img] (2 Players)
Hero bets $5.50…

Using previous history and our read on villain, we can now make the same small continuation bet we made back in hand 1 in an attempt to induce either a call from a weak hand or a bluff raise.

Flop: ($9.50) 7[img]/images/graemlins/heart.gif[/img] J[img]/images/graemlins/spade.gif[/img] 2[img]/images/graemlins/club.gif[/img] (2 Players)
Hero bets $5.50, BTN raises $20, Hero calls $14.50.

Turn: ($49.50) K[img]/images/graemlins/spade.gif[/img] (2 Players)
Hero checks, BTN bets $38, Hero raises all-in, BTN calls all-in.

Ship it. Button mucks K[img]/images/graemlins/heart.gif[/img] T[img]/images/graemlins/heart.gif[/img] and hero collects the pot.

Now obviously, our static strategy won’t work forever on this particular villain, so we’ll need to make continuous adjustments. The next time we flop a strong hand we might make a larger bet, or make a smaller bet when we’re bluffing. The idea is to adjust better than your opponent does, and yet still be able to manipulate them into doing what you want them to do.
Do note, that with this simple illustration we’ve only begun to scratch the surface in realizing how powerful a highly varied bet sizing methodology can be. We’re not necessarily always trying to deceive our opponents with our bet sizes, but rather, betting an amount that maximizes our expectation which is the size of our opponent’s possible mistake times the chance he will make that mistake (Sklansky, Miller; NLHETAP; p57).

Works bets against: villains who we have a very good read on.

Disadvantages: By varying our bet sizes based on a specific set of objectives, we may face difficult decisions on later streets as opponents reactions may or may not be influenced by the size of our bets. Requires constant adaptation.

There are some things I haven’t covered that are worth noting, such as adjusting bet sizing based on your opponents estimated hand ranges, adjusting your bet sizing based on position, number of opponents, your image, history, etc;

Feel free to add any thoughts/criticisms.

kekedarius · #2 05-19-2007, 01:50 AM

jesus christ thats long. reading now

The_Bankroll · #3 05-19-2007, 02:05 AM

intersting, even thought I usually take the opposite of methodology #1. I c-bet a little smaller with my strong hands, and more with my weak ones. my reasoning being that smaller bet size= higher calling %. when I say small, I usually make a 1/2 pot-2/3 pot bet, leaning more towards 2/3. a big bet is usually the pot.

there are a few exceptions to this though. if the board is really coordinated, I tend to bet higher w/ good holdings. if someone wants to call w/ a draw or float me, they are going to have to pay for it. on the flip side, if it is a situation where I feel a c-bet is less likely to work, but it still works enough that I should go for it, I'll bet w/ weak hands a bit smaller. for example, I tend to bet less with 3 players than with two.

josh_x · #4 05-19-2007, 03:36 AM

great post, but sooo long [img]/images/graemlins/smile.gif[/img]

D.L.M. · #5 05-20-2007, 11:02 PM

liked your last example.

cbloom · #6 05-20-2007, 11:30 PM

Good post.

It made me think about something, which is related to Method #2. I know a lot of 2+2ers advocate this method. Anyway, the idea is to bet more on drawy boards regardless of your hand.

The thing is, most villains don't fold a draw even if you bet pot. So what exactly is the point? So that you can bet bigger when you do have a hand to protect it I guess. But even then, your equity sucks. Let me see if I can explain :

Hand 1 :

Hero opens 7[img]/images/graemlins/heart.gif[/img]7[img]/images/graemlins/club.gif[/img]
Button calls

Flop 7 Q 3 rainbow

Hero cbets ...

Hand 2 :

Hero opens 7[img]/images/graemlins/heart.gif[/img]7[img]/images/graemlins/club.gif[/img]
Button calls

Flop 7[img]/images/graemlins/diamond.gif[/img] Q[img]/images/graemlins/spade.gif[/img] 3[img]/images/graemlins/spade.gif[/img]

Hero cbets ...

So this method would say to bet more in hand 2 because of the draw.

The thing is, in hand 1 our hand is actually a lot stronger in hand 1. Say his calling range is a Q or a draw - in hand 1 our equity is way better, so don't we want to put more money in when our equity is better?

carnivalhobo · #7 05-20-2007, 11:35 PM

i used to change mine based on texture, now i just use a random system that makes sense to me that ends up usually being close to 2/3 pot

reup · #8 05-21-2007, 12:11 AM

grunch: yeah #7 all the time, as much as you can get right. i suppose if you don't have a read you have to resort to some bet, but how would you know when you have no read? when this happens to me i look at the general percentages that people actually hit flops. It's not much, so i'm c-betting quite a bit. if i don't have the slightest clue how he plays, to leave me in dark as to whether i could push him out with a certain sized bet, or set up a turn bet where he folds a the proper amount of time, i just size my bets big whua, get em out, if you want to play pay, or raise if you gotta big hand.

makes my decision easier, and what else can i assume if he's unknown, that he's super tricky. no stay simple until you know better, bet big he calls so what c/f.

but then if i pick up a hand the next time i'm betting big into the pot again. if he calls i consider check/raising a lot of turns, but usually just bet out a lot. then it's right back into 'adjusting your bet sizing on the objective you're trying to acheive.' i believe that is the style that accounts for the most dynamics. it includes the others but can go as deep as the hand plays out.

DjSkyy · #9 05-21-2007, 01:24 AM

great post, do we have any idea of the relative strength/ev between the different methodologies - like which one is the best? is it even possible to determine something like this?

rakes.a.beach · #10 05-21-2007, 02:23 AM

nice post sir.