Open Side Menu Go to the Top
Register
Game Theory Optimal style at /5, does it work? Game Theory Optimal style at /5, does it work?

03-05-2016 , 02:23 PM
I've been practicing this style for quite a while now but have applied it to the 2/5 level for only like 20-30 hours now (too small a sample size to draw conclusion).

For those not familiar, playing GTO means you try to call often enough (mostly just in heads up or 3 way pots) to prevent villains from profitably bluffing with ATC. That means calling 3-bets preflop at least around 30% of the time and calling down against each postflop bet at least round 60-70% of the time. Also calling raises postflop the huge majority of the time. This means the "Baluga theorum" is ignored, since following Baluga means folding to raises the huge majority of the time.

GTO also means bluffing at the right frequencies. The "right frequencies" are roughly the majority of the time preflop and on the flop, about half the time on the turn, and approximately 33% of the time on the river if betting pot.

I've found at the $1/2 level, the loose calling part of GTO loses since players don't bluff often. At the $2/5 level so far, I've found a much greater proportion of the player pool bluffs more frequency.

Would you guys say GTO should be used at $2/5, or is this level still too low to apply it?
03-05-2016 , 02:29 PM
Use GTO frequencies against Asians and 20 something whites wearing RUN IT UP hoodies and use Baluga against old white men.
03-05-2016 , 02:32 PM
Quote:
Originally Posted by A_C_Slater
Use GTO frequencies against Asians and 20 something whites wearing RUN IT UP hoodies and use Baluga against old white men.
Is that the calling/defending part of GTO only?

What about the bluffing part? Should I aim for GTO bluffing frequencies against say an unknown, random 2/5 player?
03-05-2016 , 02:37 PM
It takes two to play GTO. Pretty sure that if you play this way, your villains will be happy to exploit you collectively, even if they don't realize it.
03-05-2016 , 02:41 PM
I don't even know where to begin.

I'd suggest reading CMVs work on GTO in th COTM section.
03-05-2016 , 02:44 PM
I think you are probably losing a metric-ton of value by not playing exploitable poker against people who will never exploit you, if that makes sense. At 2/5, that would be a large percentage of the player pool.
03-05-2016 , 06:09 PM
Quote:
Originally Posted by BenT07891

For those not familiar, playing GTO means you try to call often enough (mostly just in heads up or 3 way pots) to prevent villains from profitably bluffing with ATC. That means calling 3-bets preflop at least around 30% of the time and calling down against each postflop bet at least round 60-70% of the time. Also calling raises postflop the huge majority of the time. This means the "Baluga theorum" is ignored, since following Baluga means folding to raises the huge majority of the time.
This paragraph is exactly how fish play in my games. I am dead serious. They dont use fancy words but they call 3 bets way too much. They call post flop way too much. And they call turn raises way too much
03-05-2016 , 06:29 PM
^ lol'd
03-05-2016 , 07:44 PM
Quote:
Originally Posted by squid face
This paragraph is exactly how fish play in my games. I am dead serious. They dont use fancy words but they call 3 bets way too much. They call post flop way too much. And they call turn raises way too much


But are they doing it in HU and 3 way pots or more multiway? This is a huge difference, almost all 2/5 games the avg player to the flop is 3 or greater. Online fullring the avg player to flop is 3 or less. On the rare occasion you find yourself in a tight live 2/5 game you certainly should be playing very sticky postflop, except for the maybe calling the turn raises part, that's almost always super strong value oriented when playing live.

Last edited by A_C_Slater; 03-05-2016 at 07:51 PM.
03-05-2016 , 07:48 PM
Gto is a defensive strategy that will not yield the best EV at $2/$5

Most players make so many mistakes so you should play exploitative to make the most EV

Also most Gto modelling is done on heads up pots so is largely irrelevant for the multiway pots at $2/$5
03-05-2016 , 07:50 PM
I will say though knowing what Gto strategy is should help you exploit people because you know where they are deviating from optimal play.
03-05-2016 , 08:03 PM
A few points:

1) You're not using the term "GTO" correctly. All you're talking about is thresholds, which is only a very small part of GTO and is sometimes even at odds with equilibrium strategies. For example, against someone who plays poker perfectly, you should fold to their flop cbets far more often than thresholds suggest (trying to continue often enough that it's not profitable for the opponent to bet with ATC will often make you lose more money with your range as a whole).

Not sure how deep you wanna dive into this theory discussion, but just thought you should know that continuing with 60% of your range to a 2/3 PSB flop cbet on a board that sucks for you is generally a disaster.

2) GTO, by definition, is a style that would never lose you money regardless of what strategies your opponents employ. This is an extreme, pie-in-the-sky ideal that not even nosebleed crushers should expect to have achieved (and is certainly WAY more complicated than just continuing/bluffing at threshold). But if it were achieved, then of course it wouldn't lose money against bad players. It would rake money in at an extremely impressive rate.

3) That being said, your opponents' in these games play extremely exploitable styles, so that GTO is rarely the most profitable approach (though, again, it would still make you a crusher). Baluga Theorem is a great example of an exploitative heuristic that applies to a lot (not all) of players in LLSNL games, and there are many others.

Whether it's better to start with GTO play as your base strategy and adapt to reads from there or if it's better to start with a strategy that's geared toward beating players that fall under a set of safe assumptions is up for debate, but in the end, I don't think it matters as much as everyone pretends it does. Most of the GTO vs ABC debate is between people who have very very little idea what GTO even is much less what its strategies look like but just assume it's a bunch of FPS rubbish, and on the other side of the debate is people who know one or two very basic GTO tenants and think that this is the magic solution to beating everyone who's ever played the game.

IMO, the more you study any angle of the game, the better you will be long-term.

4) This is the most important part: the nature of GTO changes drastically when you talk about multiway pots. One practical reason is because all solvers at the moment only deal with heads-up pots.

The big theoretical reason, though, is that in MW pots, your own strategy is not the sole determinant for what plays are profitable for your opponents. In particular, when there are fish in the pot whose folding frequencies are so screwed up that it can never be profitable for anyone to bluff with ATC, then no one player in the pot needs to employ a defensive strategy. This is what we call a protected pot.

As lame and straight-forward as it sounds, you can just have a pure strategy of always value betting hands that are good enough to value bet, never bluff hands that aren't, and can only profitably continue to bets with hands that have a very good chance of winning at showdown against a strong range. Your profit, quite simply, just comes from the fishs' (or fishes') spewy calls and raises when you have a strong hand, and there are no plays you can make with the other half of your range to improve the overall EV of your range.

In LLSNL, we spend a great deal of our hands in these pots. Ironically, the further you depart from the LLSNL ABC style (overlimping in LP, making priced-in calls in the blinds, opening for a smaller size that isn't deadset on getting the pot headsup, etc), the more these situations arise.
03-05-2016 , 08:28 PM
You post this on LLSNL thread, so I dont think GTO work that well
illinois mainly using flat drop I believe....
if $5-8 (include tip) per hand win, then GTO is getting rake too much
03-05-2016 , 08:29 PM
You are massively not understanding what GTO means. You read in a book that GTO means calling X% blah blah blah, but all those books are written with assumptions on how the V will play because GTO shifts with how your villain plays.

For example: if your V cbets 100% for Xpsb, GTO says that you call A% of the time, if he cbets 10% for Xpsb, you would call B% of the time. But in any book you read, they outline that (usually) the V will be playing GTO back at you to cause an equilibrium so the equation can be solved.

All you're doing is being a fish and calling too light
03-05-2016 , 09:42 PM
I think good points were made above. Particularly by Slater, RA and Squid.

I read Janda's book and Miller's 1%. I did some of the ranging work using the frequencies they suggest. I know Miller isn't the authority, in fact he is careful not to even use the term GTO instead referring to it as a "frequency based approach". But he has some interesting things to say.

First Miller suggests that as soon as your opponents "break the rules" you too than can (should?) break the rules. Meaning essentially that at LLSNL villains are so insanely unbalanced that you can and should exploit.

Miller actually did a video on Red Chip Poker about GTO strategies in Live poker. The conclusion is mostly to use it as a tool to set up some base ranges and to use it as a plumbline from which to evaluate the ways your opponents are out of balance and as a result use it to develop exploitive strat against them.

I do think in Live Poker the suggested defending frequencies regarding multi bet preflop pots and on the turn sort of turn you into a station simply because villains are generally very unbalanced and biased toward passive actions. So an equilibrium solution is sort of improperly "calibrated" as your range becomes weaker vs their aggressive range. To this point it is quite possible that those frequencies become more correct "in general" at your 2/5 game than the 1/2 game. You will likely be encountering a little more floating, bluffing, preflop aggression and barrelling.

Miller accounts for what he calls "events". Where villains for example show unusual aggression. By the time you subtract the multiway pots, the "events" and the "they broke the rules so I can too" exceptions there isn't a ton of opportunity left where this is a superior to a good exploitive approach.

Again I come primarily from a live back ground and I've only been looking at this stuff during the last year so take it for what it's worth.
03-05-2016 , 10:09 PM
Quote:
Originally Posted by johnny_on_the_spot
You read in a book that GTO means calling X% blah blah blah, but all those books are written with assumptions on how the V will play because GTO shifts with how your villain plays
This post is entirely wrong. GTO operates on no assumptions whatsoever about villain. The whole point is to derive strategies that can't be beat by any opponent.
03-05-2016 , 10:22 PM
Quote:
Originally Posted by RaiseAnnounced
This post is entirely wrong. GTO operates on no assumptions whatsoever about villain. The whole point is to derive strategies that can't be beat by any opponent.
There is an inherent assumption though. If a strat is based on being indifferent to an opponent bluffing us with any two cards but we are up against an opponent that NEVER bluffs for example it is essentially making an assumption that the opponent will bluff some % of the time when in fact he wont.
03-05-2016 , 10:29 PM
Quote:
Originally Posted by BenT07891
For those not familiar, playing GTO means you try to call often enough (mostly just in heads up or 3 way pots) to prevent villains from profitably bluffing with ATC.

GTO also means bluffing at the right frequencies. The "right frequencies" are roughly the majority of the time preflop and on the flop, about half the time on the turn, and approximately 33% of the time on the river if betting pot.
No. This is part of GTO, but not even close to all of it.

Quote:
Originally Posted by RaiseAnnounced
1) You're not using the term "GTO" correctly. All you're talking about is thresholds, which is only a very small part of GTO and is sometimes even at odds with equilibrium strategies.

2) GTO, by definition, is a style that would never lose you money regardless of what strategies your opponents employ. This is an extreme, pie-in-the-sky ideal that not even nosebleed crushers should expect to have achieved (and is certainly WAY more complicated than just continuing/bluffing at threshold). But if it were achieved, then of course it wouldn't lose money against bad players. It would rake money in at an extremely impressive rate.
1) Yes.
2) No. Just because it is unexploitable does not mean that it would win you a lot of money. It may not win you any money, as if you are indifferent to your opponent's action as they all carry the same EV, that EV can easily be nagative in a raked game.

Quote:
Originally Posted by RaiseAnnounced
This post is entirely wrong. GTO operates on no assumptions whatsoever about villain. The whole point is to derive strategies that can't be beat by any opponent.
Yes

Quote:
Originally Posted by spikeraw22
I don't even know where to begin.

I'd suggest reading CMVs work on GTO in th COTM section.
This. /thread.
Closed Thread Subscribe
...

      
m