GTO vs. Exploitative Poker | Weekly Coaching

[@GTOWizard]

Watch other free coachings: The Mechanics of Preflop: kzbin.info/www/bejne/g4HElmWdptCYiNk The Mechanics of Cbet Sizing: kzbin.info/www/bejne/moe9YnaaeriGitk The Mechanics of Fold Equity: kzbin.info/www/bejne/eJ3PZKmVibCqhNU

[@naeco5778]

Thanks a lot for these tutorials. These concepts are mostly already known by many pros but you have a way to synthesize it and make it very clear and understable

[@nomilknosugar4900]

Thanks for a great video. It all just reinforces the maxim that in poker we make all our winnings from opponent mistakes and that we need to do everything we can to note and respond to their errors (without them catching-on!)

[@tsarrast]

This is a concept with lots of subtlety and I enjoyed your treatment. Thanks for the video.

[@oriolciurana274]

Pure gold. Thank you so much guys. The best free poker content i've ever seen.

[@camilotm]

About what you say on @46:55 .I think the idea of calling with all of our bluffcatchers when we believe the agressor is overbluffing is not that bad in reallity,because if we are right,our opponent will usually overbluff with way more combos that those QJ combos used in the example (that gave that 4.9 and 2.6 ev results) ,giving us a way higher ev when overcalling.And yes,if we are wrong we are going to lose a lot (for example if the agresor counter exploit us) but if we are right I think that in practice we are for sure winning way more than 4.9 or 2.6 so I don't think that the cost of overcalling there (which is being counter exploited) is really that high .

[@thesneakingninja1]

Most helpful lecture yet. Thank you so much

[@looper6394]

I imagine GTO as decoupling our strategy from the opponent. No matter how Villain plays, if he makes a pure mistake, we win.

[@awesomenasavideos1499]

Looking forward to the new coaches, Saulo did a fantastic job 👍

[@andrzejkondracki9796]

It seems to me that this is a very complicated way to say you need to spot your opponent's mistakes and exploit them :) The concept of an expected value and 'minimax' (which here is called an equilibrium) as a GTO basis is in a way similar to bridge gaming strats. The more the deviation, the more you penalize provided your read and interpretation is correct. I like the job you have done.

[@dfdice1]

great video. please keep content like this coming

[@ewallt]

For example 1, since you’re making villain indifferent to calling or folding your bluffs by your strategy, it follows immediately from that that it can’t matter to your results if they call or fold all the time. Another way of looking at it is that your EV when you bet is the pot, regardless of what they do.

[@ncannavino11]

Amazing video yet again 👏

[@Cyberhhhh]

Great lecture but confused by what was said at 29:35, how would OOP gain by overfolding when IP is overrbluffing? Am I missing something or was it just slip of the tongue?

[@GTOWizard]

Good catch! I meant to say, "How much does OOP gain when IP over-bluffs and OOP over-calls", not over-folds.

[@24yroldchessimprover84]

I liked the toy game with the Queens and Aces vs Kings. I will refer back to this video in the future.

[@AmongUsAcademy]

Amazing lecture, thanks

[@Scottjf8]

Is this Aschwartz? Sounds like him. Great content

[@GTOWizard]

This is our content creator, Tombos21. Glad you enjoyed the video!

[@DoubleBassX2]

Highly recommend watching this one twice.

[@postmur625]

very nice channel

[@dschungelheissmann]

Very good video. Besides the approach , idea and options offered, I really like how well done the app is in all versions (PC, phone, browser). You have a great development team there. Can you disclose which framework or environment did you use for development applications that work so well and similar in all platforms?

[@GTOWizard]

Hello, thank you so much! We are happy you enjoy GTO Wizard. We used a lot of javascript to achieve that.

[@beam5655]

How common are pure mistakes? Wouldn't most of a NIT's mistakes be frequency mistakes, and therefore if they played against a GTO bot would almost be 0 EV? Because even if they overfold, as long as the solver says to fold over 0% of the time, then it is a 0 EV play vs a perfect GTO bot. I'd guess bet sizing are were most pure mistakes occur, due to the large amount of possible options. I only play poker casually against friends though, but I'm curious about the math.

[@jonathanwolter1502]

Amazing content!

[@3vilMonk3y]

So if an AI played perfectly unsing the Nash Eq GTO strategy (which is balanced vs all players) and never adjusted, would it have a long term EV of 0 (excluding rake loss)? does GTOW have a feature allowing you to adjust villains ranges like PioSolver to show how to adjust and exploit to become profitable vs unbalanced players?

[@GTOWizard]

We're adding Nodelocking (like pio) in the next few weeks! The EV of playing Nash Equilibrium would depend on who you're playing against. In practice, most people make many "pure mistakes", so a perfect GTO strategy would print.

[@jorisdemunck9406]

GTO will make money from pure mistakes and not from frequency mistakes. What are pure mistakes? Is it wrong betsizings (too small, too big). Playing hands that dont belong in our range (wider vs tighter range)?

[@kyleandrews3911]

In the last example, when this situation comes up in live low stakes, my strategy would never include a flop overbet, and I suspect that leaning towards range betting small instead is an unwitting exploit for the live low stakes player pool. The reason being that these players rarely 3 bet, so there is less value to be made with the top of our range. (They are only calling overbet with better, and overfolding the rest of their range in general, but especially to an overbet)

[@GTOWizard]

As the video says, every adjustment can be met with a counter-adjustment. If you think they are overfolding to a large bet, then why not overbet all your bluffs as an exploit?

[@AspiringPokerPro]

Love this!

[@michelparmentier3063]

awe - some ! 🧡🧡🧡

[@TG-fh7mc]

23:59 : How is AA EV 15$? Some of the time, Villain puts money in, some of the time he doesn't. In both cases we win and if we only consider one case, here we consider only half of the time, we shouldn't arrive at a correct result, should we? Case 1: Villain calls 50% -> I win 50%*30$ = 15$ Case 2: Villain folds 50% -> I win 50%*10$ = 5$ Means to me AA EV should be 20$, shouldn't it? Otherwise it would be like flipping a coin in the following way: If it lands on heads, we win 30$, if it lands on tails, we win 10$ and if this game wasn't strange enough already, we would only count heads as contributor to winnings?

[@GTOWizard]

When villain calls AA wins $20 (the $10 pot + their $10 call). When villain folds AA wins $10 (the $10 pot). EV (AA) = (50% * $20) + (50% * $10) = $15

[@TG-fh7mc]

@@GTOWizard Hey, thank you so much for going through the comment sections and replying to our questions! That showcased some limit for a formula. For everyone who has seen the following formulas and was as confused as me: Total pot := pot + hero_bet + villain_bet, where we define b := hero_bet = villain_bet = risk EV = Total pot * equity - risk = (pot + 2b) * equity - b = pot * equity + b * equity + (equity - 1) * b = (pot + villain_bet) * equity - (1 - equity) * hero_bet = $Win * %Win - %Lose * $Risk where $Win = how much money you can win (=pot+villain_bet) and $Risk = hero_bet. If we insert the according values into the first and last line, we This equality can only hold if %Lose > 0, of course, because if %Lose = 0, there is no risk. This shows that the first equation isn't really well-defined or at least has to come with mentioning that it only applies before River and for when villain isn't dead. Or more generally: For when the outcome can still change/there is risk to hero. It needs to be stated as EV = Total pot * equity - hero_bet

[@tombos211]

@@TG-fh7mc The intuitive approach is to define wins/losses as the change in your stack size from the point of decision. That frame of reference always works regardless of actions, and is easy to interpret. So if you bet AA and they fold your stack increases by $10. If they call your stack increases by $20. No need for extra complexity!

[@michaelsheehy9466]

Way too much for my brain to handle. Understood about 25% of this. Still very entertaining.

[@intheairtonight-btd3651]

So watch it three more times! 👍👍👍

[@PeterSleiman]

Tombo can you please rephrase example 4

[@GTOWizard]

The basic idea is that the "Simple Solutions", which prevent cold-calling, handicap our opponents. We are removing ways they can counter a wide HJ opening range, and therefore, the new solution opens wider.

[@ДмитрийНесяев-у7е]

What is the name of your discord channel?

[@GTOWizard]

discord.gg/jfdm9vk5WN

[@daniellojuas2518]

Balancing river part is mind blowing

[@kaaristotelancien3005]

hi possible to détail the multistreet in an other video ? it was interesting but too short

[@GTOWizard]

Check out the Toy Games lecture from tombos21 next month!

[@jrm8206]

Good video! But had to to put it on x0.65 speed to follow. You talk ultra fast my man.

[@GTOWizard]

Sorry about that, we'll try to slow it down next time!

[@cpasa798]

Let’s agree to desagree, GTO is not unexploitable, it’s the strategy you came out when both players are trying to exploit each other. Is you want to beat a perfect GTO player you just play a range much much stronger (a NIT: 5% of hands for example). With this strategy you are going to beat perfect GTO because GTO is a fix strategy and the NIT has too much range advantage in order to make the bluffs profitable. The big wins in poker come from making the best hand, if you almost always make the best hand you are going to win. Of course if you introduce a nitty range and you told a solver to create the best strategy against that range is going to do it and probably crush the nit, but by doing so the solver has to move from GTO, so the solver is no longer playing GTO (he is ultimate playing the best exploitative strategy there is against that particular opponent)

[@GTOWizard]

If you played tighter than GTO you'd lose money to the blinds. As an extreme example, imagine you only play AA from every position. Obviously you'd crush postflop (even if they didn't adjust), but that wouldn't make up for the money you hemorrhage preflop. If it were possible to simply play tighter and beat GTO, then you wouldn't be playing against GTO (by definition)

[@cpasa798]

@@GTOWizard nobody plays GTO. Nobody plays against a NIT and don’t ajust. Imagine using 3-bet and 4-bet polarized GTO ranges against a NIT who only opens 5% of hands, there is no way that the blinds lost can recovered from that. You use the example of a extreme polarized range where the player only has a bluff catcher, the problem with that is that the NIT almost always has value in his range, too much value

[@cpasa798]

@@GTOWizard GTO come from using previous strategy and ajust, if you don’t ajust you can lose. I don’t saying that a NIT can’t be beating and if you ask the solver to come out with a strategy against a NIT it will give you one, but is not GTO anymore, you said it in your videos, equilibrium is very sensitive

[@cpasa798]

@@GTOWizard if you take out the bluffs of a player range, the GTO strategy of paying with the bluffcatchers is no longer profitable, if GTO doesn’t addapt the NIT is going to crush fix GTO. I really can’t understand that people won’t see that. Even GTO wizzard Doug Polk fold to Hellmuth the 2nd nuts straight because he deeply inside understand what I’m saying even if he doesn’t want to admit it

[@Solving_Live_Poker]

@@cpasa798 You are thinking one dimensionally and one sided. If the player never bluffs, he loses as much or more because he overfolds. It's logically impossible to under bluff without overfolding. So, while GTO would be paying off all his value bets, it would be profiting from all his missed bluffs. That would either lead to breaking even or profiting for GTO. That's where the unexploitable part comes from. You have a very elementary understanding of theory, which is very common. Also, in your Polk/Helmuth hand example, Helmuth bet so much over the pot, that even a solver would only advise to call with QT. Just like a solver would only call a UTG open shove for 300bb with AA. You can bet so much, that even at equilibrium, a solver will only call with the nuts. And the reason you "can't understand why people won't see that" is because you're wrong. Generally speaking, when all the educated theory people disagree with you, that's a clue that you should reexamine your logic. Not double down. There's usually a reason they don't agree or see your side.

[@lechero4295]

i keep losing tournaments with gto wtf man

[@jesuslovesyou-matts]

Ok, now can you explain it again as if I was a 5 year old?

[@randomBBL]

I was going to try GTOWizard for a month but I think it's complete nonsense. In theory this stuff would be great if you had all this information available to you at the table which nobody who isn't a super user has which makes it pointless. You can't know what an opponent's range is that you can accurately know exactly how often to do what. There's no way to know the percentages like you put for your bluff and value equity. You have no way to know! You are calculting stuff that nobody knows what they are. You are just pulling numbers out of thin air and then making your game based around random numbers? Stupid. The whole concept of being indifferent also makes no sense. Why would I want to break even??? I want to crush everyone's souls and breakeven is not my goal. This whole thing is dumb.

[@jeremyculot5937]

Come on bro , you obviously didn't watch this video and you dont fucking understand gto vs expoitative plays.

[@GTOWizard]

Thanks for the response, it demonstrates some common misunderstandings that we should address in future videos. The great part about playing GTO is that you don't need to know your opponent's range to profit from it! Those numbers and calculations are used so that we can understand the fundamental workings of game theory, you would never try to calculate that in game. The reason we study indifference isn't to try and break even, rather it's to find thresholds and attack targets that sculpt the overall strategy.

[@Tom_Bee_]

@@GTOWizard if his criticism was that as more and more players study towards Nash equilibrium the profits of everyone decrease as fewer mistakes are made and there are fewer blunders to exploit, then I'd probably have to agree lol