Is my approach to GTO useless?

A

Aleh00

Rising Star
Bronze Level
Joined
Aug 17, 2022
Total posts
4
FI
Chips
0
So, I'v been doing some solving and studying GTO. My choice of solver is GTO+. I have solved situations for 3bet pots, single raised pots and part of 4bet pots. All of these have been solved for each position. I know the limitations and the fact that playing GTO does not equal to most EV gained. Thats not what I'm asking here. Previously mentioned situations have been solved for 3 bet sizings (33%, 67%, 140%), no donking (Im aware that there is boards where you can have donking range) and 1 check-raising size (75%). I still use 3 sizings on turn and river.

After solving, the solver obviously for many situations showed varying bet sizing and frequencies, which in many boards are pretty much impossible to remember or execute. I started playing around and editing the game tree on the flop by removing one or two bet sizings on different flops. I noticed that in some cases reducing the bet size to only one sizing made negligible loss to EV and in some cases reducing even gained some EV. I know that there is many boards you should have more than one sizing. I have not just blindly stared at the EV. I have tried to understand why the solver plays as it plays with the given ranges involved in each situation. For some flops I have left 2 sizings, since I realize that you preferably need 2 sizings for protection, getting more value etc. This has been my approach on simplifying game trees to be more pragmatic while playing. All im trying to do is understand the reasons behind solvers outputs and to create "baseline" for myself to fall back on while on playing, which still would be GTO approved.

After all of this long introduction. My questions is that am'I heading to wrong direction by reducing bet sizings on different flops by reasoning that it doesnt lower the EV that solver spews out? What kind of a techniques you guys have used to simplify GTO to be more usable while playing? I would love if you could point me to right direction if my approach here is wrong. And ofcourse please rip my approach to shreds if needs be :D

Oh and btw, solved for 100bb effective stack and 25 flop subsets.
 
Luvepoker

Luvepoker

Lost in the twilight zone
Community Guide
Joined
Feb 21, 2018
Total posts
5,296
Awards
23
US
Chips
436
I dont think you are going in the wrong direction. While the difference at time may be small its not just about EV. Its more about how to best play a hand in a situation. Another thing is you also need to adjust to the player your against. The issue with playing GTO is it assumes your playing against a player who is playing perfect GTO.
 
A

Arnakk2424

Rock Star
Bronze Level
Joined
Dec 25, 2021
Total posts
252
Awards
1
Chips
42
I wish I could link few videos I watched lately
You are in the right direction
I don't know what level your but you can have 2 bet sizings, 1 for heavy draw boards and 1 low draw boards. Some pro does
Like you said you won't get the most value vs X types of opponents but at some point you will be able to define certain players type by taking notes and/or with hud stats.
You know you have to bet small on As 8d 2h board. Do it if you have no info on your opponents
You know the guy is a calling station bet larger or even overbet vs him.
You know a guy is playing match or fold, bet small
Always think in mind the GTO approch, but adjust.
 
A

Aleh00

Rising Star
Bronze Level
Joined
Aug 17, 2022
Total posts
4
FI
Chips
0
I dont think you are going in the wrong direction. While the difference at time may be small its not just about EV. Its more about how to best play a hand in a situation. Another thing is you also need to adjust to the player your against. The issue with playing GTO is it assumes your playing against a player who is playing perfect GTO.
Absolutely, this is something im trying to keep in mind when solving. Since nobody plays perfect GTO the most EV gained will be made exploiting known weakness on a given player. Also I guess you have to have understanding about GTO to effetively exploit leaks. Could'nt agree more with you here, thank you.
 
A

Aleh00

Rising Star
Bronze Level
Joined
Aug 17, 2022
Total posts
4
FI
Chips
0
I wish I could link few videos I watched lately
You are in the right direction
I don't know what level your but you can have 2 bet sizings, 1 for heavy draw boards and 1 low draw boards. Some pro does
Like you said you won't get the most value vs X types of opponents but at some point you will be able to define certain players type by taking notes and/or with hud stats.
You know you have to bet small on As 8d 2h board. Do it if you have no info on your opponents
You know the guy is a calling station bet larger or even overbet vs him.
You know a guy is playing match or fold, bet small
Always think in mind the GTO approch, but adjust.
Im trying to see the underlying "reasons" why solver uses the sizes it does. I know this just my own simplification/abstraction since solvers dont have any "reason" behind the outputs. Altho im finding it hard on some flop textures to see why the solvers output is giving such sizings. When in doubt I have just left all the 3 sizings as they are and im planning to get back at them in very near future. On many cases when simplifying to only 1 sizing solvers output EV is only -0,1% to -0,5% less than with the full 3 sizings solve. In these kind of a spots I always try to look for reasons to use 2 sizings since not every hand in your range bets for the same sizing preferably. Also even tho EV decline seems very small (-0.1% to -0.5%) I guess it makes a big difference how often the given flop texture and situation comes up in practice. Small EV loss but when it happens often, makes a big EV loss overall. Thank you for the response, got few things to look into.
 
John A

John A

Poker Zion Coach
Silver Level
Joined
Apr 12, 2012
Total posts
6,492
Awards
3
Chips
37
Not the wrong approach, but def pick two sizings and try and simplify as much as possible.

I understand GTO very well, but I don't really use almost any of it when I'm playing 100nl or below, and I crush. It's pretty pointless.

I think the most important thing you can do w/ your poker learning is defining your goals. Do you want to be the best player of all time? Just looking to win some vacation money? Want to be a solid pro? What's the goal. Once you do that, then figure out a roadmap that will get you there the best.

It sounds like you're trying to get to the understanding behind GTO, which is great... it's what you should be doing. I'd recommend this video series we did which delves into that why a lot more in an entertaining way -

 
A

Aleh00

Rising Star
Bronze Level
Joined
Aug 17, 2022
Total posts
4
FI
Chips
0
Not the wrong approach, but def pick two sizings and try and simplify as much as possible.

I understand GTO very well, but I don't really use almost any of it when I'm playing 100nl or below, and I crush. It's pretty pointless.

I think the most important thing you can do w/ your poker learning is defining your goals. Do you want to be the best player of all time? Just looking to win some vacation money? Want to be a solid pro? What's the goal. Once you do that, then figure out a roadmap that will get you there the best.

It sounds like you're trying to get to the understanding behind GTO, which is great... it's what you should be doing. I'd recommend this video series we did which delves into that why a lot more in an entertaining way -

Will definetly watch the series over the upcoming weekend. I had these 3 sizings as "baseline" where to start from. After doing some solving and diving deeper into the outputs of solver on many flops I have boiled down to 2 or 1 sizing on which to use. I do understand that there is no "right" sizing on any given flop. What im having trouble with is that on some flops having 2 sizings is very close to having 1 sizing EV-wise. Still not have managed to totally reconsile on picking one over the other. Altho I have been trying to look at it from the perspective of reality in the tables. Lets say having a 1 sizing on certain flop loses EV 0.15% compared to having 2 sizings on the same flop. Lets also say that the sizings would be 120% pot vs 33% pot + 120% pot. I can see the benefits of both of having 1 sizing (120%) and 2 sizings (33% + 120%), one of these will make the game tree much more simpler and the other would probably be harder to people to play against but would make game tree more complex. Obviously the 1 sizing can work but I feel like its pretty easy to play against some one who only bets 1 overbet sizing on certain flop texture. That said, one should know the defend ranges very well against this overbet sizing to play correctly against it and I feel like atleast at my stakes most people wouldnt know how to defend correctly. Any advice on how to resolve this dilemma?

Have done loads of thinking on what it is that I want to achieve in poker. If I had to choose, solid pro would be the best. I mean who wouldnt want to be the best in the world, but I think the sad reality is that not everybody can reach it. So on that front the odds are stacked against me.

Absolutely, im playing in such a small stakes that playing GTO would be like burning money since there is so many leaks to exploit. Im just trying to arm myself with the knowledge and understanding on GTO. Also I do want to play higher stakes and if Im not mistaken the edges just get smaller the higher you get. I guess it also means that you have to know and understand GTO to have something to fall back on, especially against good regs.

Anyhow, thank you for pointing me to the series and taking time to comment.
 
John A

John A

Poker Zion Coach
Silver Level
Joined
Apr 12, 2012
Total posts
6,492
Awards
3
Chips
37
Sure... pick the sizing that has lower variance, because at your stakes, you won't need it. At higher stakes, it's sometimes needed.

Simple example from the other day when I was reviewing a poker friends play. This guy is one of the biggest online MTT winners of all time, and we had an argument about a turn raise spot. I was advocating for a call because of his field advantage in the game, and he thought it was higher EV to just raise. Turns out, we were both right according to the solver, the EV was almost identical, but obviously calling neutralizes the times he is against the top of someone's range (it was a broadway paired board and he had PP). Keeping a lower variance line made more sense in this case.

That said, generally most people use 1/3 and 3/4 c-bet sizings. It's more the turn where you'll have over bets.. usually 50% and 120%.

But again, honestly, I don't even think most players understand min defense frequencies enough to justify sticking to these sizings. Especially at micro stakes where player turnover is high and you won't have had many hands on the regs as you do when you move up stakes.
 
Top