Questions about GTO.

    • Itsnevereasy
      Itsnevereasy
      Bronze
      Joined: 08.08.2010 Posts: 381
      1) Although, we cannot hope to play a full GTO strategy. If we do play GTO on some rivers, then the maximum we hope to win on those rivers is exactly 0. True or False.

      2) Say our opponent bets all in pot on the river. We decide to play GTO, calling with the best 50% of hands. Say our opponent also plays GTO bluffing 50% and value betting 50% and we only have bluff catchers in our range - i.e. we cannot beat his value range. Let the board be Axxxx. Our 50% calling range will have a lot of aces in it thus making him having an ace less likely. So we just made his range inbalanced. Are we both playing GTO and exploting our opponent at the same time? Does this make sense?

      3) What is the aim of GTO? Is it purely defensive? It makes sense to try play GTO against opponents that are better than you but does it also make sense to play GTO vs weaker opponents?

      4) What's the point if GTO in HU? Is it a defensive strategy until we find opponents leaks?
  • 21 replies
    • getdotacom
      getdotacom
      Black
      Joined: 06.04.2008 Posts: 607
      1) True.

      2) Our opponent OTR can't have 50/50 value/bluff, usually it's more like 30/70, depends on sizing. Blockers definitely change ranges a little bit, but I guess it works both ways - bettor can block some bluffcatchers as well. Also I'm sure if we model these ranges, we can take blockers into an account.

      3) GTO is good "default" approach until we know how to adjust. Definitely shouldn't be playing like this if u know how to exploit opponent. This sounds a little bit weird, but actually GTO doesn't work very well in some cases. I've done the maths and it sometimes makes sense to 3barrel our entire range and our opponent still can't call enough, so poker is far away from solving imo.

      4) No idea about HU, maybe to battle for lobby ? :)
    • EmanuelC16
      EmanuelC16
      Bronze
      Joined: 02.01.2010 Posts: 13,897
      GTO = reaching a Nash Equilibrium and does not mean +EV in every spot.

      1) False.
    • dooleslovs
      dooleslovs
      Platinum
      Joined: 17.02.2011 Posts: 481
      1) This is true if oppoment also plays GTO?

      2) Lets say I always bet pot OTR, what vould be my GTO value/bluff ratio? Not 50/50?
    • JCSeerup
      JCSeerup
      Bronze
      Joined: 14.12.2010 Posts: 1,039
      The point of playing GTO is if we don't have a way to exploit our opponent, and we want to find a leak on our opponent that we can exploit. We generally don't want to play against players that are better than us since they probably also play better GTO, so this is primarily something we want to do when we are playing vs unknown players that we don't have reads on yet.

      Playing GTO on the river we play a range that can't be exploited, so no matter what his calling range is the best he can do is breakeven, but when his range is either too strong or too weak we profit.

      At micro limits it's generally not worth the effort to learn GTO since our opponents are so weak that we should be able to exploit them more profitable than what we can achieve by playing GTO. That said I don't think it's a bad idea to get a basic understanding of the concept since we can learn a lot about building default ranges with the concept.

      Don't know if you have tried it, but in poker snowie you can challange the softwear and try playing vs a table of GTO bots and get all your plays analyzed. It's actually lots of fun.
    • EmanuelC16
      EmanuelC16
      Bronze
      Joined: 02.01.2010 Posts: 13,897
      Originally posted by JCSeerup
      Playing GTO on the river we play a range that can't be exploited, so no matter what his calling range is the best he can do is breakeven, but when his range is either too strong or too weak we profit.
      I didn't use Snowie and the rest is not that much off but this is wrong. If you play GTO it doesn't mean your opponent cannot be +EV.
    • JCSeerup
      JCSeerup
      Bronze
      Joined: 14.12.2010 Posts: 1,039
      Originally posted by EmanuelC16
      Originally posted by JCSeerup
      Playing GTO on the river we play a range that can't be exploited, so no matter what his calling range is the best he can do is breakeven, but when his range is either too strong or too weak we profit.
      I didn't use Snowie and the rest is not that much off but this is wrong. If you play GTO it doesn't mean your opponent cannot be +EV.
      Would you care to elaborate, I don't really see how.
    • EmanuelC16
      EmanuelC16
      Bronze
      Joined: 02.01.2010 Posts: 13,897
      Originally posted by JCSeerup
      Originally posted by EmanuelC16
      Originally posted by JCSeerup
      Playing GTO on the river we play a range that can't be exploited, so no matter what his calling range is the best he can do is breakeven, but when his range is either too strong or too weak we profit.
      I didn't use Snowie and the rest is not that much off but this is wrong. If you play GTO it doesn't mean your opponent cannot be +EV.
      Would you care to elaborate, I don't really see how.
      Just a few examples:

      1. You reach that stage with the exact same range and both play it optimally => you split the blinds.

      2. The non-optimal player arrives there with a very strong range so will always be +EV there but he will fold too much in a previous situation since that is why his river range is too strong.

      LE: Probably easier to say it this way: I only bet river with the nuts. Make my bet -EV. :) There will be all sorts of situations where people have suboptimal ranges but because they are too strong/tight. Those are still +EV in that situation but are not part of a EV maximizing strategy.
    • Itsnevereasy
      Itsnevereasy
      Bronze
      Joined: 08.08.2010 Posts: 381
      2) Our opponent OTR can't have 50/50 value/bluff, usually it's more like 30/70, depends on sizing. Blockers definitely change ranges a little bit, but I guess it works both ways - bettor can block some bluffcatchers as well. Also I'm sure if we model these ranges, we can take blockers into an account.


      our opponent bets all in pot


      pot sized bet so GTO is 50/50 value/bluff...

      I've done the maths and it sometimes makes sense to 3barrel our entire range and our opponent still can't call enough, so poker is far away from solving imo.


      Could you give an example? Is it that he CAN'T call enough or WON'T call enough. Why can't he call enough? He cant beat our bluff range???
    • Itsnevereasy
      Itsnevereasy
      Bronze
      Joined: 08.08.2010 Posts: 381
      1. You reach that stage with the exact same range and both play it optimally => you split the blinds.


      On the river the EV of both GTO players is 0!

      1) False.

      2. The non-optimal player arrives there with a very strong range so will always be +EV there but he will fold too much in a previous situation since that is why his river range is too strong.

      LE: Probably easier to say it this way: I only bet river with the nuts. Make my bet -EV. smile There will be all sorts of situations where people have suboptimal ranges but because they are too strong/tight. Those are still +EV in that situation but are not part of a EV maximizing strategy.


      I still don't see how a GTO strategy can be +EV (on the street where we play GTO). :s_confused:
    • metza
      metza
      Bronze
      Joined: 28.01.2012 Posts: 2,220
      Originally posted by JCSeerup
      Don't know if you have tried it, but in poker snowie you can challange the softwear and try playing vs a table of GTO bots and get all your plays analyzed. It's actually lots of fun.
      The bots are not GTO. Yes its fun and maybe you could even improve by playing vs it, but the creators claiming its GTO for marketing purposes when it isn't is really inappropriate.
    • getdotacom
      getdotacom
      Black
      Joined: 06.04.2008 Posts: 607
      Originally posted by Itsnevereasy
      2) Our opponent OTR can't have 50/50 value/bluff, usually it's more like 30/70, depends on sizing. Blockers definitely change ranges a little bit, but I guess it works both ways - bettor can block some bluffcatchers as well. Also I'm sure if we model these ranges, we can take blockers into an account.


      our opponent bets all in pot


      pot sized bet so GTO is 50/50 value/bluff...

      I've done the maths and it sometimes makes sense to 3barrel our entire range and our opponent still can't call enough, so poker is far away from solving imo.


      Could you give an example? Is it that he CAN'T call enough or WON'T call enough. Why can't he call enough? He cant beat our bluff range???
      Pot AI OTR means we give our opponent 1:2 to call. He needs 33% equity for that call, so our value/bluff ratio should be 66:33 or 2:1 to make his call break even. He has to make a call 50% of the time to make our bet b/e. Btw this works like this only OTR.

      Some example : We open BU and SB 3bets us. We defend ~15% by calling (lets assume we even slowplay AA this way). The board runs out 223K9. Then if we want to defend "enough" according to GTO, we will have to call down with something like AT+ on this board, but we can't do it if our opponent 3barrels his entire range. AT is just behind almost any range on runnout like this. ofc we can argue maybe that our preflop 3bet calling range is too wide, but if it's balanced, then our open range is too wide etc. Then we will come with something like 30% BU open range where we can defend against 3barrels on this runnout, but what's the point if our opponent will always fold then ?:)
    • metza
      metza
      Bronze
      Joined: 28.01.2012 Posts: 2,220
      Originally posted by getdotacom


      Some example : We open BU and SB 3bets us. We defend ~15% by calling (lets assume we even slowplay AA this way). The board runs out 223K9. Then if we want to defend "enough" according to GTO, we will have to call down with something like AT+ on this board, but we can't do it if our opponent 3barrels his entire range. AT is just behind almost any range on runnout like this. ofc we can argue maybe that our preflop 3bet calling range is too wide, but if it's balanced, then our open range is too wide etc. Then we will come with something like 30% BU open range where we can defend against 3barrels on this runnout, but what's the point if our opponent will always fold then ?:)
      What exactly do you mean by just behind almost any range? Cos for 1/2 pot or 2/3 pot we can still call rivers when we are behind vs a range.
    • Itsnevereasy
      Itsnevereasy
      Bronze
      Joined: 08.08.2010 Posts: 381
      Glad this is generating some discussion. I hope we can all learn from each other :D

      Some example : We open BU and SB 3bets us. We defend ~15% by calling (lets assume we even slowplay AA this way). The board runs out 223K9. Then if we want to defend "enough" according to GTO, we will have to call down with something like AT+ on this board, but we can't do it if our opponent 3barrels his entire range. AT is just behind almost any range on runnout like this. ofc we can argue maybe that our preflop 3bet calling range is too wide, but if it's balanced, then our open range is too wide etc. Then we will come with something like 30% BU open range where we can defend against 3barrels on this runnout, but what's the point if our opponent will always fold then ?smile


      Interesting example. I think you are just basically illustrating a situation when the bettors range is much stronger than the caller. Let me give another example. By the turn the board is JJ84 (3 spades) and our opponent check raises. We put him on fullhouses and nut spade draws to balance his range. On the river the 4th spade comes and he shoves. Well he has no bluffing range the way the board ran out and according to GTO (im pretty sure because I think I heard Internet say this) we should bluff catch 0% of the time.

      So I think your example is the same. If we cannot beat our opponents bluffs then its the same as he is not bluffing often enough and we can fold AT.

      What do you think?
    • metza
      metza
      Bronze
      Joined: 28.01.2012 Posts: 2,220
      ^What you're talking about is the essence of playing exploitatively (basing decisions off opponents range) though imo, which is the opposite of GTO.
    • Itsnevereasy
      Itsnevereasy
      Bronze
      Joined: 08.08.2010 Posts: 381
      ^ I dont think so. Playing exploitatively would be if we believe he is bluffing too much/not enough. In my example he doesn't have a range to bluff with. Basing decisions off opponents range is the essense of poker (GTO included). In every GTO example, there are assumptions about players' ranges.
    • metza
      metza
      Bronze
      Joined: 28.01.2012 Posts: 2,220
      Originally posted by Itsnevereasy
      ^ I dont think so. Playing exploitatively would be if we believe he is bluffing too much/not enough. In my example he doesn't have a range to bluff with. Basing decisions off opponents range is the essense of poker (GTO included). In every GTO example, there are assumptions about players' ranges.
      Yeh but in the GTO player you'd assume there would always be a balanced mix of value and bluffs.

      Although, actually that makes me wonder because imo true GTO would not have the terms "bluff" or "value" in how it viewed hands, since these terms don't actually exist objectively, its just something that us humans use to simplify things. There is no such thing as a real bluff or real value bet (with the exception nuts and nut low on river.
    • LemOn36
      LemOn36
      Coach
      Coach
      Joined: 07.02.2009 Posts: 1,354
      If you do play GTO your opponent's strategy/range is irrelevant
    • muumionu
      muumionu
      Bronze
      Joined: 25.07.2012 Posts: 816
      Originally posted by JCSeerup
      Don't know if you have tried it, but in poker snowie you can challange the softwear and try playing vs a table of GTO bots and get all your plays analyzed. It's actually lots of fun.
      I just played the bot heads up and it sucked pretty bad.. some hands he played were so stupid imho, maybe he just ran very bad too then but yh idk it seemed terrible to me.

      if thats gto bot id play them all day lol

      edit: for a bot though I guess it was pretty ok and would maybe even beat some nl50-100 regs
    • muumionu
      muumionu
      Bronze
      Joined: 25.07.2012 Posts: 816
      i might've been very wrong :D
    • 1
    • 2