So before we all get to mess up the turn, we somehow need to decide how big we want our cbet to be

Since there's no "correct" way to do it, I've been dabbling with different philosophies.

One size:

I've looked at my cbet ranges (semi bluffs, value hands) and derived some odd 41% of the pot from that using flopzilla, a couple dozend of sample flops that I would cbet and the break even point.
I can use this size for every single situation I want to cbet, it's hard to exploit, easy to implement and since it's so small, villains are likely to overfold to it anyways.

Problem is, there's no flexibility and since I'm basically flipping a coin with an edge in this single spot, I cannot abuse other players and it messes up potsize on the other streets. If I cbet a value hand with 40%, the pot is too small on the turn and as soon as I try to bloat the bot there, it gives my hand away.

I play like this on a table full of nits to play just for the redline. besides that, I don't like that approach. It's really difficult to keep frequencies under controll and rather easy to slip into 100% cbet mode.



Two sizes based on hole cards:

Big bets (2/3 pot) for unstable equity: semi bluffs, combo draws and TPTK .
Small bets (1/3 pot) for stable equity: nutted hands, top/mid sets, trips, 2pair.

Also not too difficult to implement and a little more flexible, but rather easy to get a read on. What's really problematic tho, is the fact that some of the small bet hands should be in my checking range, like sets or flopped str8s.

The big advantage is that frequency is on autopilot and controlled by the hand you have, similar to preflop play.


Multiple sizes based on board structure:

Bet small and often on dry and static boards, bet big and seldom on wet and dynamic boards.

For example, I would cbet 1/3 pot with any2 on 2K7r, however I would bet big with TPTK, combo draws and semi bluffs on 6h8hQd and check 6's, 8's and air.

I feel this is the most flexible line for exploiting villains, on the other hand it's the most difficult way to play since it allows for so much discretion. I'm trying to work towards that, but there's a lot of stuff to work on...


Situation based:

As seen on TV. Just bet what you think is optimal in this current situation.
Definitely not for me and I think as a beginning player I should have kind of a framework that I can come back to or to derive from when I have a read.




So what do you guys think? What's your approach to optimal cbet sizing and frequencies?

I click 50% pot button then scroll up or down as I feel necessary. So I think that's the third one. If I'm feeling a bit confident I do have a 35% and 75% button that I can click and adjust from too.

I should add that the board is only part of deciding your betting range and sizing. Yours and villains range on that board is the important thing. UTG vs MP on K72r is different to BTN vs BB which is different to CO vs SB in a 3bet pot etc.

How did you get there? I mean I played in a similar fashion when I started out but quickly realised that I lack the understanding to make those decisions on the fly.

Kind of a "I didn't know what I didn't know" situation which has changed into "I know what I don't know". So now I'm trying to build a couple of drills to get better at cbetting, since it's impossible for me to make an educated decision within like 10seconds online and shooting from the hips just results in massive spazz outs from time to time.

It depends on your range

33% usually is a range bet
50% is a safe bet most of the time
68% usually when you're going for value on the turn
78-88% on the river with value/bluffs

you can also go smaller with thin value in certain spots

Dunno you get better at it with time. I'm not a good poker player really and I guess a lot. When I used to play a bit more seriously I was quite good at poker theory so i get the jist of what's going on but I've always been far too lazy with putting work in and learning the spots.

I suppose the answer is I play low enough that it doesn't matter as I'm playing better than enough of the player pool. There are spots that I'm lol bad at like flop raising ranges, turn sizing so if I played like 100nlz regs would laugh at how bad I was. You probably play as high/higher than I do.

Lol, honestly I used to play 100NL 15y ago. Now I'm having a hard time beating the russian grinder mob at 25NL and sometimes even 10NL. Everyone but the worst fish knows a little about poker. Everyone cbets now...and everyone floats cbets. Some 5NL guys even know why to check middle pair or TP low kicker.

So that's why I decided not to play based on feel anymore. I had a good run up from 2NL to 10NL in a very short time, but mainly due to good table selection and BRM. But boy do I get hurt when I run into a table of 24/22 regs...


So right now, I'm trying to get a foot in the door. I know the basics and why certain things have to be that way. I don't have enough data to construct my own ranges, so I'm using Pokersnowies' with a couple of twists.


The hartest thing IMO is finding a proper way to practise. I always wondered, why 50-60 year old farts play golf with the same handycap as the younger guys who are on the green literally every day. So I asked a trainer and he said, the old guys don't have as much time, but they train different.


The young guns play every day, but they don't do isolated training drills and they don't structure their training. The older guys have a putting green in their office and practise putting in their free time. When they don't have time to play even 9 holes, they work on chipping for two hours or go to the driving range.


And what do we do for poker? We grind endless hours, watch youtube videos and discuss single hands. This is like playing 18 holes every day...you just learn nothing from it.
And honestly, those hand analysis threads are completely worthless to me. Get Flopzilla, disect your hand and forget about opinion. But to me it's just not effective, because it's one single hand that never ever happens the same way again.

The game has over 2.5 million different combos and even if we condense them to 500k, we would still need 10 years to analyse every one of them within 10 minutes using a solver or equity calculator.

So I try to work on groupings or decision trees, something you could actually use when you're at the table. You know, getting behind the mechanics of it and in the end put a short action note on a post it.

Like "bet dry flops small" or "don't be too afraid when the flush completes, 60% of the time he has only a pair", "don't cbet air on dynamic flops", when you have a str8 draw and a flush draw and a pair (eg Kc2c on a 3c4cKd board), you have 60% equity vs TPTK so the right play is to shove after he bets"

I mean, poker is more complex than chess, yet you only have 30 seconds max to make your decision.

For flops I typically only use one size based on board texture ... 30% or 75% and 30% isn't always a range bet.

Turns in SRP its whatever, 30-200%. Turn sizing is a function of flop strategy also, FYI.

When cbetting we are rarely playing the cards we actually have, we are plying the cards we don't think the vil has and the cards we want them to think we have.

As such, the first consideration is always "do they think?" Do they know if I bet 25% of the pot I am trying to tell them something different from a 80% of the pot bet. If not, then I am just playing bet sies that best fit my desires (small if I have nothing, 75%+ if I think I am good). I will drop small to as small as it can be and still work, if one single BBB will do it, I will bet a single BB.

My default cbet is half pot but I will adjust it based on the board and what I am trying to tell the vil I have. If I want to pretend I have a protecting hand, I will bet 75%, if I want to act like I flopped the nuts, I will drop my bet sie to under half (I might try one or two bigs if its a AA9/ KK5 sort of flop and i have nada)

If I flop strong (2 pair, strong top pair) I generally play strong (75% or so cbet) and if I flop super strong (set/ nuts) I will usually put in smaller bets, more in the half pot plus a few bigs range.

If there is any reason a player may be rattled (recent suck out or dealt KK vrs AA etc) I will play against them in the same way I'd play a non thinking player, assuming there is some chance they will make overcalling mistakes (so, value good/bluff bad).

I will also make adjustments base on hands I showdown. If I am looking bluffy, I will obviously bet stronger on good hits. If I've shown down some big hands, I will cbet bluff cheaper.

No, the key to good poker for the most part is playing a solid strategy and understanding the game, not trying to mindgame your opponents.

I mostly use two sizes (around 1/3 or 2/3 pot) and just try and remember what I was taught 5 years ago, and then do the opposite.

To whit, if the board is very connected and I expect villain to continue very often, I usually go small with my made hands and draws. If the board is drier and/or I expect villain to fold a lot, I go big.
My in-game bet-sizing decisions probably depend more upon villain's continuance range/frequency than my own range. I'm rarely gonna go big if villain will call with a lot of high equity hands. I go big when I want/expect him to fold.

I also use the old "big hand => big pot, small hand => small pot" concept. I was intrigued to see that solvers/bots often seem to follow that philosophy. They won't make massive c-bets with AA on 754tt (BTN v BB) if the only hands that continue are those that will often beat AA. In single-raised pots, it's more typical to bet small if your hand/range isn't going to win very often if all the money goes in, or when there will be a lot of ugly runouts. You don't want to play a big pot with one pair if the board is likely to get super-ugly. Betting small with one pair means you get called down by draws and other hands you beat, but prevents you from getting stacked by straights/sets/2 pairs, or from being bluffed off the best hand when villain represents those hands.

It's all too complex to summarize though. A solver might have an entirely different bet-size solution on Tc6c2x and 9c6x2c, even though the two flops look fairly similar at first glance.