Archive for Tagpoker Forum A friendly place for winning poker players to share their strategy to making money playing poker.
 


       Tagpoker Forum Forum Index -> Heads Up Poker Strategy
Simba

Looking For Thoughts of Good Players

Hey everyone Smile . Recently I've been exploring game theory to get a grip of the 'best' heads-up strategy, focusing on pre-flop at the moment, planning to move to post-flop once I have pre-flop cracked.

As part of my research I'm curious as to how close good players play to optimal, so if you consider yourself good, I'd be extremely grateful if you'd fill in some or all of the following:

Questionnaire:

(IN A VACUUM/FIRST HAND)

1) What % of hands do you believe it is optimal to open pre-flop assuming you come in for a 3x raise each time you want to play a hand?

2) Which hands do you believe it is optimal to 3-bet (to 10BBs), assuming the button raises the percentage you gave in Q1.

3) Which hands would you flat the raise with assuming Q1?

4) Which hands do you believe it is optimal to 4-bet (to 25BBs), assuming the big blind 3-bets the percentage you gave in Q2.

5) Which hands would you flat the 3-bet with assuming Q2?

6) Which hands do you believe it is optimal to 5-bet all-in (for 75BBs), assuming the button 4-bets the percentage you gave in Q4.

7) Which hands would you call the 5-bet all-in with after you have 4-bet as in Q4?

~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~

Thanks a lot to anyone who participates - I'll share some of my findings in exchange as a thank you (and to encourage people to fill that out) Smile . If you'd rather PM me the information, that's also fine.

Much appreciated!

~~Simba
kolonel

Re: Looking For Thoughts of Good Players

Hope that all makes sense..........

Simba wrote:


Questionnaire:

(IN A VACUUM/FIRST HAND)

1) What % of hands do you believe it is optimal to open pre-flop assuming you come in for a 3x raise each time you want to play a hand?


~ 70%

Simba wrote:
2) Which hands do you believe it is optimal to 3-bet (to 10BBs), assuming the button raises the percentage you gave in Q1.


99+,AJs+,KQs,QJs

Simba wrote:
3) Which hands would you flat the raise with assuming Q1?


88-22,ATs-A2s,KJs-KTs,QTs,J9s,T8s+,98s,AJo

Simba wrote:
4) Which hands do you believe it is optimal to 4-bet (to 25BBs), assuming the big blind 3-bets the percentage you gave in Q2.


QQ+,AKs,AKo

Simba wrote:
5) Which hands would you flat the 3-bet with assuming Q2?


TT-JJ, AQ

Simba wrote:
6) Which hands do you believe it is optimal to 5-bet all-in (for 75BBs), assuming the button 4-bets the percentage you gave in Q4.


KK+

Simba wrote:
7) Which hands would you call the 5-bet all-in with after you have 4-bet as in Q4?


KK+
YATHINNK

It seems like you're asking how to play optimally against someone who is playing just as optimally? Unless I'm totally misunderstanding your point.
Simba

Yes, precisely, that's how optimal is defined, though I'm not asking how to play optimally (I've already calculated that from the mathematics I've done over the past few days), I'm just curious how close to optimal good players play by default against an unknown (hence wanting to compare responses in this thread to the optimal ranges I've calculated). I'd really like to hear responses from top players, people playing $30s+. I fully expect that everyone below the $15 level will be playing massively sub-optimally by default, I'm just curious as to how quickly that changes. Are the $30 players playing near optimally by default? The $100 players? That's what I'm curious about.

Kolonel: Thanks a lot for that, that's great Smile much appreciated.
Brokerstar

I would guess that anyone playing "optimally" will be playing much higher than $30 games.

Most $30 players are fish or players on their way up and just passing through Wink
Simba

Thanks Broker Smile I appreciate your reply. Part of what's motivating me to do all this mathematics is your view on reads/tells and such - I want to see how far it's possible to get using pure mathematics alone, the flip side of the coin if you will. Finally something my maths degree is useful for!

Some of the results and ranges I've calculated are very surprising - I think I'm beginning to see why I'm not doing so well myself in games (as mine look nothing at all like they do)...

I've finished calculating the optimal pre-flop strategy now for 3x open-raises pre. Next task is to do the same for minraises and for open-limping, then to experiment a bit with varying stack sizes. Then I'll move onto analysis of optimal play post-flop. I'm aiming to have 'solved' no-limit hold'em heads up in a sense by the end of October. Then there's going to be a lot of memorisation (yay for having a good memory), and then to the tables to put it all to the ultimate test Smile . I've run simulations of quadrillions of hands on the calculations I've done so far, and all looks correct up to now! My inner maths geek and poker geek are both very excited. Razz
aggsyb

Simba wrote:
Thanks Broker Smile I appreciate your reply. Part of what's motivating me to do all this mathematics is your view on reads/tells and such - I want to see how far it's possible to get using pure mathematics alone, the flip side of the coin if you will. Finally something my maths degree is useful for!

Some of the results and ranges I've calculated are very surprising - I think I'm beginning to see why I'm not doing so well myself in games (as mine look nothing at all like they do)...

I've finished calculating the optimal pre-flop strategy now for 3x open-raises pre. Next task is to do the same for minraises and for open-limping, then to experiment a bit with varying stack sizes. Then I'll move onto analysis of optimal play post-flop. I'm aiming to have 'solved' no-limit hold'em heads up in a sense by the end of October. Then there's going to be a lot of memorisation (yay for having a good memory), and then to the tables to put it all to the ultimate test Smile . I've run simulations of quadrillions of hands on the calculations I've done so far, and all looks correct up to now! My inner maths geek and poker geek are both very excited. Razz


Your preflop play is in direct correlation of villains postflop play there is no "optimal stratergy" there are so many variations and for one u ask what hands at 10bb u are 3betting, 3b/c or 3b jam? there is a huge difference in ranges between 10-13bb and 13-16bb etc , I would just get to grips with a basic style and get some post flop experience behind you, then start working on adjusting preflop sizings to be able to play post flop optimally vs villains tendancies
StasKo

"I want to see how far it's possible to get using pure mathematics alone"
IMO, no further than your opponents tendencies!
if your opponent changes one of his tendencies even by a small fraction it will change the math exponentially.

its far more "optimal" to focus on playing the opponent rather than to solve the game mathematically.

ive been there myself, trying to figure out how much math involves in becoming a winning player and its always the same answer:
the math (most of it) is dependant on opponent's psychology and not the other way around. so "playing optimally" is solving your opponent rather than solving the math behind the game
sandman369

NLHU is not the game you want to play with pure math. Turn your focus to limit cash games (not HU) to utilize optimal math wizardry.
two2go

aggsyb wrote:

Your preflop play is in direct correlation of villains postflop play there is no "optimal stratergy" there are so many variations and for one u ask what hands at 10bb u are 3betting, 3b/c or 3b jam? there is a huge difference in ranges between 10-13bb and 13-16bb etc , I would just get to grips with a basic style and get some post flop experience behind you, then start working on adjusting preflop sizings to be able to play post flop optimally vs villains tendancies


I think Simba is talking about the optimal strategy with no reads. Like where the best an opponent can do against you is break even (so you both lose equal rake). I think in game theory is called GTO or something like that and is a similar idea to a nash chart.

Simba, can you share a general method for how you went about getting your solution.
blsmur

Re: Looking For Thoughts of Good Players

Simba wrote:
Hey everyone Smile . Recently I've been exploring game theory to get a grip of the 'best' heads-up strategy, focusing on pre-flop at the moment, planning to move to post-flop once I have pre-flop cracked.

As part of my research I'm curious as to how close good players play to optimal, so if you consider yourself good, I'd be extremely grateful if you'd fill in some or all of the following:

Questionnaire:

(IN A VACUUM/FIRST HAND)


1) What % of hands do you believe it is optimal to open pre-flop assuming you come in for a 3x raise each time you want to play a hand?
3X RAISE  I AM USUALLY LOOKING SOMETHING 50% PLUS
FOR MIN RAISE 75% FROM BTN

2) Which hands do you believe it is optimal to 3-bet (to 10BBs), assuming the button raises the percentage you gave in Q1.
I WILL 3 BET WITH ANYTHING TO GET FOLD  IF 4 BET THEN I NEED pp OR AK AQ  MAYBE AJs
3) Which hands would you flat the raise with assuming Q1?
I RARELY FLAT  CALL  BUT IF I DO IT WILL BE A GOOD HAND 40% PLUS

4) Which hands do you believe it is optimal to 4-bet (to 25BBs), assuming the big blind 3-bets the percentage you gave in Q2.
4 BETS ARE ALL IN FOR ME IT DEPENDS ON MY STACK

5) Which hands would you flat the 3-bet with assuming Q2?
IF I FLAT A 3 BET IT WILL ONLY BE AA QQ  KK

I DONT SET MINE IN HUSNGO


6) Which hands do you believe it is optimal to 5-bet all-in (for 75BBs), assuming the button 4-bets the percentage you gave in Q4.

I DONT


7) Which hands would you call the 5-bet all-in with after you have 4-bet as in Q4?
AA PERHAPS KK
~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~

Thanks a lot to anyone who participates - I'll share some of my findings in exchange as a thank you (and to encourage people to fill that out) Smile . If you'd rather PM me the information, that's also fine.

Much appreciated!

~~Simba


I  WOULD RATHER BE ALL IN THAN TRY AND ARGUE MY WAY THROUGH A FLOP TURN AND RIVER   AFTER BEING 3 BET  OR 4 BET  OR 5 BET
YOU ALWAYS NOT KNOW WHERE YOU ARE IN THE HAND

SO ALL IN CURES THE PROBLEM
IF i DONT HAVE A TOP HAND I FOLD TO 3 BETS
Simba

sandman369 wrote:
NLHU is not the game you want to play with pure math. Turn your focus to limit cash games (not HU) to utilize optimal math wizardry.


Out of curiosity, why do you suggest that NLHE HU is not the game I want to play with pure maths?

two2go wrote:
aggsyb wrote:

Your preflop play is in direct correlation of villains postflop play there is no "optimal stratergy" there are so many variations and for one u ask what hands at 10bb u are 3betting, 3b/c or 3b jam? there is a huge difference in ranges between 10-13bb and 13-16bb etc , I would just get to grips with a basic style and get some post flop experience behind you, then start working on adjusting preflop sizings to be able to play post flop optimally vs villains tendancies


I think Simba is talking about the optimal strategy with no reads. Like where the best an opponent can do against you is break even (so you both lose equal rake). I think in game theory is called GTO or something like that and is a similar idea to a nash chart.

Simba, can you share a general method for how you went about getting your solution.


Yes indeed, this is what I'm doing. It's kind of like an endgame Nash chart as you mention but a heck of a lot more complex.

To get the solution, essentially you have to simultaneously optimise one player's range against the other's, stage by stage. It's a bit long to explain fully and I'm heading to work in a moment!
Brokerstar

Re: Looking For Thoughts of Good Players

blsmur wrote:


I  WOULD RATHER BE ALL IN THAN TRY AND ARGUE MY WAY THROUGH A FLOP TURN AND RIVER   AFTER BEING 3 BET  OR 4 BET  OR 5 BET
YOU ALWAYS NOT KNOW WHERE YOU ARE IN THE HAND

SO ALL IN CURES THE PROBLEM
IF i DONT HAVE A TOP HAND I FOLD TO 3 BETS


I think this is being a bit too simplistic in your approach to thinking about heads up and thus will make you exploitable to aggressive players.

When you say "If I don't have a top hand I fold to 3 bets" then we all know that you don't get 'top hands that often in a heads up match so an aggressive guy can profitably just 3 bet you a lot if you're opening even a reasonable range of hands.


If you're being 3 bet a lot by an aggressive and observant player who will likely fold when you go all in on him as you're only doing it with as you say 'a top hand'  then how do you adjust to this type of player?

Think about it, what he's doing to you and how you should realistically think about fighting back?

Broker
blsmur

Re: Looking For Thoughts of Good Players

Brokerstar wrote:
blsmur wrote:


I  WOULD RATHER BE ALL IN THAN TRY AND ARGUE MY WAY THROUGH A FLOP TURN AND RIVER   AFTER BEING 3 BET  OR 4 BET  OR 5 BET
YOU ALWAYS NOT KNOW WHERE YOU ARE IN THE HAND

SO ALL IN CURES THE PROBLEM
IF i DONT HAVE A TOP HAND I FOLD TO 3 BETS


I think this is being a bit too simplistic in your approach to thinking about heads up and thus will make you exploitable to aggressive players.

When you say "If I don't have a top hand I fold to 3 bets" then we all know that you don't get 'top hands that often in a heads up match so an aggressive guy can profitably just 3 bet you a lot if you're opening even a reasonable range of hands.


If you're being 3 bet a lot by an aggressive and observant player who will likely fold when you go all in on him as you're only doing it with as you say 'a top hand'  then how do you adjust to this type of player?

Think about it, what he's doing to you and how you should realistically think about fighting back?

Broker


yes your right Dom


i dont fold to all 3 bets  just most of them

of course  the  question was  "blind"

ie we have no reads

so i opt to fold when i have no read
if you  ie the opponent  3 bets me every hand  
i do it right back in position

i dont just fall on my sword and i happy  to fight them to the river in position after i have a read
but not before

does that make sense
kierkegaard1

Re: Looking For Thoughts of Good Players

Simba wrote:
Hey everyone Smile . Recently I've been exploring game theory to get a grip of the 'best' heads-up strategy, focusing on pre-flop at the moment, planning to move to post-flop once I have pre-flop cracked.

As part of my research I'm curious as to how close good players play to optimal, so if you consider yourself good, I'd be extremely grateful if you'd fill in some or all of the following:

Questionnaire:

(IN A VACUUM/FIRST HAND)

1) What % of hands do you believe it is optimal to open pre-flop assuming you come in for a 3x raise each time you want to play a hand?

too many variables. it simply depends. 100% if they fold 50%+ . and reduced the more they 3bet, if they flat a lot, but give up post, 100% again. if they flat a lot but play well post, getting rid of a lot of stuff like K2o etc that flops terribly.

2) Which hands do you believe it is optimal to 3-bet (to 10BBs), assuming the button raises the percentage you gave in Q1.

i dont think i ever 3bet to 10bbs (unless shoving) premiums of AT+, pairs 77+, bluffs with equity. Including a wider value range against an aggressive player.

3) Which hands would you flat the raise with assuming Q1?

suited pictures +, 8T+, small pairs (if early), if they fold to a lot of 3bets, some of my value range will be in here, too.

4) Which hands do you believe it is optimal to 4-bet (to 25BBs), assuming the big blind 3-bets the percentage you gave in Q2.

i dont know. i think this is tough against my ranges. i'd need to work out the maths. hard to work out espec because my value range may be wider than i said, and i may flat some of my value range based on the reasons i said earlier. very difficult question because of this.

5) Which hands would you flat the 3-bet with assuming Q2?
again, difficult to say without knowing the frequencies of the bluffs with equity. and def depending on gameflow of what to flat and why (instead of 4betting)

6) Which hands do you believe it is optimal to 5-bet all-in (for 75BBs), assuming the button 4-bets the percentage you gave in Q4.

all of my value range if there is a 4bet dynamic.

7) Which hands would you call the 5-bet all-in with after you have 4-bet as in Q4?

im not folding any of my value range

~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~

Thanks a lot to anyone who participates - I'll share some of my findings in exchange as a thank you (and to encourage people to fill that out) Smile . If you'd rather PM me the information, that's also fine.

Much appreciated!

~~Simba


you simply can't solve this because frequencies of ranges change too much with thinking players.
blsmur

Simba wrote:
sandman369 wrote:
NLHU is not the game you want to play with pure math. Turn your focus to limit cash games (not HU) to utilize optimal math wizardry.


Out of curiosity, why do you suggest that NLHE HU is not the game I want to play with pure maths?

two2go wrote:
aggsyb wrote:

Your preflop play is in direct correlation of villains postflop play there is no "optimal stratergy" there are so many variations and for one u ask what hands at 10bb u are 3betting, 3b/c or 3b jam? there is a huge difference in ranges between 10-13bb and 13-16bb etc , I would just get to grips with a basic style and get some post flop experience behind you, then start working on adjusting preflop sizings to be able to play post flop optimally vs villains tendancies


I think Simba is talking about the optimal strategy with no reads. Like where the best an opponent can do against you is break even (so you both lose equal rake). I think in game theory is called GTO or something like that and is a similar idea to a nash chart.

Simba, can you share a general method for how you went about getting your solution.


Yes indeed, this is what I'm doing. It's kind of like an endgame Nash chart as you mention but a heck of a lot more complex.

To get the solution, essentially you have to simultaneously optimise one player's range against the other's, stage by stage. It's a bit long to explain fully and I'm heading to work in a moment!


limit is more maths because you cant go all in
therefore you can call with correct odds all the time and you will win in the long run
of course if try and bluff or play against the odds you will be in trouble
maybe not this hand or that hand but overall you got to go with the odds
pot odds mostly as implied odds are not there like in NL
sausage

[/quote]

Yes indeed, this is what I'm doing. It's kind of like an endgame Nash chart as you mention but a heck of a lot more complex.

To get the solution, essentially you have to simultaneously optimise one player's range against the other's, stage by stage. It's a bit long to explain fully and I'm heading to work in a moment![/quote]

I don't know if you have it already, but if not I am pretty sure that the CREV program will do what you are doing with a lot less effort.
Simba

sausage wrote:

I don't know if you have it already, but if not I am pretty sure that the CREV program will do what you are doing with a lot less effort.


I'm using a similar program to that. The program doesn't construct the ranges, but it's good for testing strategies against each other and such. This is what I'm using for my simulations.
       Tagpoker Forum Forum Index -> Heads Up Poker Strategy
Page 1 of 1