Meta game works - how to train it?

Do you mean in terms of establishing a loose image so they misadjust when you start playing your normal game? I like the idea of that.

I don't like the idea. In short term it might sound good, but in long term people will know what are you doing, and some might not even adjust to what you do, i.e. if they play X way when you are loose, they will keep X style of play when you switch gears, therefore your effort was worthless.

Just exploit their weaknesses in the beginning and you will make much more money.

Rihard

Unless you're playing midstakes+ the playerpool is prob too big to worry about this sort of thing without a sample.
Just play your default gameplan until you develop reads imo.

Thanks for the response guys!

I think, you are right about sticking to game plan and
adjusting to players. I often times will 3bet some players w/
Ax to understand how they play, if I don't have stats yet and
that sometimes helps me to get an action few orbits later,
when I have a premium hand. So, it looks like I'm not a total nit.

However, if hands are just not coming for first 40 hands f.e. and my VPIP/PFR
is like 13/11, then I just better quit that table, because my image screams =
ROCK!!!
I'll start a new one and my first BU might be a 3bet vs CO if I get lucky to
have Ax, Kxs, Qxs ... some blockers.

There is a difference, if you look like a ROCK or TAG.

I saw a video from coatch insider 2 years ago where
He starts playing maniac style to create the image
later slowsdown with similar arguments as you
said. I use same approach and mostly i profit in the begininning and at some
point you notice how people dont give you credit anymore then you can start being value heavy and
after some time standard.

I think begin playing tight and then use that image to bluf a lot could work as well, couldnt it?

Originally posted by william8787
I think begin playing tight and then use that image to bluf a lot could work as well, couldnt it?

Yes it could work. I think it is better to play agro until unknown because people tend
To play tight vs unknown. Also depends on the limit imo, if you play nl200 you will hardly benefit much with agression because people will play gto, there your approach is imo much better.
Gto guys will burn money bluff catching you in beginning, and later you will bluff them when they think you are tight.

Originally posted by GingerKid

Originally posted by william8787
I think begin playing tight and then use that image to bluf a lot could work as well, couldnt it?

Yes it could work. I think it is better to play agro until unknown because people tend
To play tight vs unknown. Also depends on the limit imo, if you play nl200 you will hardly benefit much with agression because people will play gto, there your approach is imo much better.
Gto guys will burn money bluff catching you in beginning, and later you will bluff them when they think you are tight.

That would be very hard to do against someone really playing GTO (not just thinking they play GTO) because GTO can't be exploited by definition.

EDIT: Specifically, a GTO player does not make or use assumptions about the opponents bluffing tendencies. I'm not arguing against your reasoning, but your use of terminology (GTO) is plain wrong:)

Originally posted by YohanN7

Originally posted by GingerKid

Originally posted by william8787
I think begin playing tight and then use that image to bluf a lot could work as well, couldnt it?

Yes it could work. I think it is better to play agro until unknown because people tend
To play tight vs unknown. Also depends on the limit imo, if you play nl200 you will hardly benefit much with agression because people will play gto, there your approach is imo much better.
Gto guys will burn money bluff catching you in beginning, and later you will bluff them when they think you are tight.

That would be very hard to do against someone really playing GTO (not just thinking they play GTO) because GTO can't be exploited by definition.

EDIT: Specifically, a GTO player does not make or use assumptions about the opponents bluffing tendencies. I'm not arguing against your reasoning, but your use of terminology (GTO) is plain wrong:)

Hi,
Thats why i wrote that in limits where most of unknowns are regs and play gto, i would not bluff until i am unknown, because anyway ev of bluff is zero, and they cant exploit my value heavy ranges because i am unknown. When i collect stats vs gto player, he would see i am very tight and will exploit it by folding, where i can start playing more agro and exploit his wrong assumption.

Originally posted by GingerKid

Originally posted by YohanN7

Originally posted by GingerKid

Originally posted by william8787
I think begin playing tight and then use that image to bluf a lot could work as well, couldnt it?

Yes it could work. I think it is better to play agro until unknown because people tend
To play tight vs unknown. Also depends on the limit imo, if you play nl200 you will hardly benefit much with agression because people will play gto, there your approach is imo much better.
Gto guys will burn money bluff catching you in beginning, and later you will bluff them when they think you are tight.

That would be very hard to do against someone really playing GTO (not just thinking they play GTO) because GTO can't be exploited by definition.

EDIT: Specifically, a GTO player does not make or use assumptions about the opponents bluffing tendencies. I'm not arguing against your reasoning, but your use of terminology (GTO) is plain wrong:)

Hi,
Thats why i wrote that in limits where most of unknowns are regs and play gto, i would not bluff until i am unknown, because anyway ev of bluff is zero, and they cant exploit my value heavy ranges because i am unknown. When i collect stats vs gto player, he would see i am very tight and will exploit it by folding, where i can start playing more agro and exploit his wrong assumption.

Hi!

Again, I'm not arguing against your logic. I am arguing against your terminology. If you put out a GTO player there, then there is no way at all of exploiting him. No way at all! This is the definition of GTO, and by Nash's theorem, GTO exists. There is no way around this. You need to check out your terminology

Hi!

Again, I'm not arguing against your logic. I am arguing against your terminology. If you put out a GTO player there, then there is no way at all of exploiting him. No way at all! This is the definition of GTO, and by Nash's theorem, GTO exists. There is no way around this. You need to check out your terminology smile

Yes, gto can't be exploited by bluffs. When I said player who is playing gto, I expect that guy to also fold vs guys who are value heavy and not to burn money. Or to call down wider vs bluff heavy guys. So my assumption is that when player who plays gto sees that villian is having very passive stats, that he will not defend by nash's theorem but will exploitatively fold more often. We are talking here about real poker application, how to exploit people, and not about abstract theory about gto.

Originally posted by GingerKid

Hi!

Again, I'm not arguing against your logic. I am arguing against your terminology. If you put out a GTO player there, then there is no way at all of exploiting him. No way at all! This is the definition of GTO, and by Nash's theorem, GTO exists. There is no way around this. You need to check out your terminology smile

Yes, gto can't be exploited by bluffs. When I said player who is playing gto, I expect that guy to also fold vs guys who are value heavy and not to burn money. Or to call down wider vs bluff heavy guys. So my assumption is that when player who plays gto sees that villian is having very passive stats, that he will not defend by nash's theorem but will exploitatively fold more often. We are talking here about real poker application, how to exploit people, and not about abstract theory about gto.

You can argue about that all week. Being a black member, I'm pretty sure you are an excellent poker player. But the fact remains, you don't seem to know what GTO is, due to the way you formulate things. Here is a starting point for GTO: Nash equilibrium

I'm not doing this to convince you or "get right" at any price. I just don't want other readers of this thread to get the wrong impression about GTO. If you mean real poker played by real people, just don't use the term GTO and say that you can exploit GTO. You can't. You can just defend against it and play GTO yourself (you both eventually going broke due to the rake). The other point to understand is that the GTO player doesn't assume any particular ranges on your part. He's not changing style - ever.

When are you changing from exploitable play to solid play against an opponent? Costs of getting caught playing exploitably and being counter exploited can easily outway the profits from earlier exploits. I want to design a better strategy than 'at some point I notice and change my exploit'.

I think there are two schools of thought on exploiting.

1. Make very small adjustments and pick up a little EV, around say 10bb/100 per exploit. Here our exploits are very hard to discover and even if they are, they're very close to GTO anyway, so it's no big deal.

2. Make max or near maxEV exploits and pick up masses of EV, around say 200-300bb/100 per exploit. Here our exploits are blatantly obvious and when picked up on can be counter exploited and cost us even more than our profits over a lightening quick period, before we can notice.

I think ideally, against most players and situations (yes everything is different, but common situations can be modelled and extrapolated for similar situations) somewhere in the middle is going to be the most +EV.

I haven't seen anyone give a concrete reason for picking how large they're willing to exploit? Or reasons for when to change the exploit?

Things to consider for choosing how large to exploit and when to change our exploit.

Is he capable of noticing?

- Has Villain seen a showdown of our exploitative play verse him? How good is he? Does it matter? Does he need more than 1? Is it a showdown of an obvious or non obvious exploit?

- Have our exploits verse him started to skew our stats? i.e. Our fold to river bet verse him is 100% after X occurrences, how big is X before we have to give up on our exploit?

- The larger we're exploiting, the faster he's going to realize it and adapt.

Anything else to consider?

Originally posted by Anger86
Anyway, someone feels the same way as me?
I mean, that it is a great way to start playing like monster mode
and then switch gears after ~5-10 min?

I recall trying to do this versus certain villains and when I first started getting into HU. Unfortunately I think the pitfalls are too great and at least for me, to take the extreme example: the negatives of 'creating an image by 3-betting/4-betting etc w/ 72o' well outway the positives.

I prefer a style of smaller adjustments and exploitations. A levelling style game in general is not for me, but I do of course love to make moves in those lovely situations where it just becomes obvious what the opponent is doing/thinking.

Originally posted by YohanN7

Originally posted by GingerKid

Hi!

Again, I'm not arguing against your logic. I am arguing against your terminology. If you put out a GTO player there, then there is no way at all of exploiting him. No way at all! This is the definition of GTO, and by Nash's theorem, GTO exists. There is no way around this. You need to check out your terminology smile

Yes, gto can't be exploited by bluffs. When I said player who is playing gto, I expect that guy to also fold vs guys who are value heavy and not to burn money. Or to call down wider vs bluff heavy guys. So my assumption is that when player who plays gto sees that villian is having very passive stats, that he will not defend by nash's theorem but will exploitatively fold more often. We are talking here about real poker application, how to exploit people, and not about abstract theory about gto.

You can argue about that all week. Being a black member, I'm pretty sure you are an excellent poker player. But the fact remains, you don't seem to know what GTO is, due to the way you formulate things. Here is a starting point for GTO: Nash equilibrium

I'm not doing this to convince you or "get right" at any price. I just don't want other readers of this thread to get the wrong impression about GTO. If you mean real poker played by real people, just don't use the term GTO and say that you can exploit GTO. You can't. You can just defend against it and play GTO yourself (you both eventually going broke due to the rake). The other point to understand is that the GTO player doesn't assume any particular ranges on your part. He's not changing style - ever.

I dont understand gto fully, but the things you wrote i understand. Probably i was using wrong terminology as you said. When i said gto player, i didnt mean player who plays perfect gto, but player who tries to play as close as possible GTO vs unknown frequencies or vs other gto players, and will play exploitative vs people who have leaks. So basically realistic player from high limits.

P.S.
Nice that you are thinking about orher people

Now we are certainly on the same wave-length. I thought we were already at the outset.

Fortunately, for all of us, nobody knows a full GTO solution for NL Hold'em. There are pretty good robots for FLHU, but these are still beatable (by a so-called Nemesis, particularly designed to beat it (and nobody esle) - but hardly by any human player). For NL Hold'em - just forget it. Computers aren't nearly fast enough to compute what GTO exactly is (given a million of years). All we know is that GTO exists.

I know also about that. We know that there is GTO for poker because there is a theorem from Game Theory saying that each game has GTO, which is proven. Just the fact is that for many games the decision tree is so complicated that it is in reality not possible with current knowledge to find gto, and poker, chess belong to this group of games.

Robots are not perfect YET for 6max but are surely today better than humans. And there is plenty of them especially in ipoker. I can tell you that in low limits speed poker you can hardly find humans (nl5, nl10). There are fish and bots, some rare regs as well who go away very fast when they see that nobody makes mistake there. This bots there are not playing perfect, they will not find many good spots to exploit except easy spots, but you will so hardly find leaks vs them which you can exploit.

Originally posted by GingerKid
Robots are not perfect YET for 6max but are surely today better than humans.

Well, I agree bots are better at GTO poker than humans. They are not better at poker though. The winrate of a MES (maximum exploitive strategy) is much higher than that of a GTO strategy.
After all, GTO is not designed to maximize our winrate but to minimize the winrate of our opponents.