mikeavida

I would love to know more about why you have that feeling you have. You have given me a lot to think about and allowed me to clarify some ideas I already had and explore some new territory as well. I think we probably could come to more of an agreement with more discussion, but as much as I love theoretical discussions, I value practicality and pragmatism more. The point for me, is to play better poker and win at a higher rate and move up in stakes. So, if we are at the point where this is akin to religious differences, it is of course better to move on to things that actually improve our game.

I do wonder where we started talking past each other though. In my mind it is self evident that the range matters because solvers require two ranges to work.

Thank you Shaun. I have some research to do. First, I want to check out how much pool's 3 bet percentage differs from 14%. Then dig in at least a little to see more in detail where the difference comes from. Also, I will check out the Equity Realization video. I had heard of Hand2Note's Range Research and was wondering if it could be used for this. I don't have a huge sample of hands, but I might still check out to see what I can do with it.

GTO Warrior, I think there is something really interesting to explore about what you said. You said that the opponent's range doesn't matter when playing GTO, but I think that needs to be modified to something more like the opponent's frequencies with a given range don't matter when playing GTO. Range, of course, does matter, which is why PIO needs the opponent's range as well as your range. PIO outputs the equilibrium strategy given two ranges. The defining feature of the equilibrium strategy being that neither player can reduce the other player's EV by unilaterally changing the frequency with which they take certain actions within that range.

The interesting question to me, and something I have been wondering about, is how much does the GTO strategy suffer if the starting range differs from the assumed starting range. If the answer is not much, than you really can just try to make your frequencies match what GTOwizard outputs. But if the answer is quite a bit, then it is imperative that you have a good sense of what the opponents range is.

Thinking about this a bit further, the distinction I'm making of frequencies with a range vs a different range altogether really only matters if you haven't solved preflop. But, I think we are still at the point where preflop is not truly solved because preflop is multiway and we don't really have solvers that work multiway yet. Is that correct? If you do have a perfectly solved game tree from preflop on, then any deviation in range from your opponent is really just a frequency deviation.

Just wanted to expand on this a little bit. Obviously, villian is playing a really terrible strategy in this example and is massively overbluffing. Our continues are going to make a ton of extra EV because of this. If we somehow didn't know that villian was getting to the turn with every hand and was pot sized bluffing all of the extra combos, we would fold our normal folding range, which seems unfortunate from the perspective of knowing his crazy strategy, but I don't think in this situation it "ruins" our strategy, because we make so much extra EV from all of our good hands that continue.

I've been wondering about this question though. When hero is playing either a somewhat or massively different range than we think he is, does the GTO strategy based off a different villian range ever become quite bad?

rosaa, at first, I thought you did say something about overfolding to an opponent who overbluffs, but I re-read your post very carefully and I think I understand better the situation you are describing. Let me try to clarify to see if this is what you mean:

Hero has a GTO derived strategy that says to fold vs, say, a pot sized bet OTT. Villian makes a pot sized bet OTT, but he is doing it with a hand that isn't even in the range that we used to derive the GTO strategy for hero. Your question, basically, is is this good for hero and bad for villian or the other way around, correct?

Interesting question. Obviously we know, that hero can never be exploited by villian if he changes his frequencies from the GTO strat, but what if villian has a completely different range? I think the answer is that with a sufficiently different range, hero sticking to the "GTO strat" could be quite bad.

I like to go to extreme positions to work things like this out in my head. Imagine villian somehow got to the turn with every single hand and is now betting pot with all of those extra combos. Does folding still seems like a good strat just because that is the GTO strat vs the range villian is "supposed to have?" It doesn't seem like it to me.