Bizarrely condescending and authoritative posts for someone who said this in another thread 2 days ago:

"The stake I'm playing is NL10 rush and cash. I'm completely new to the game and would really like your input here, I feel like I can improve my game a lot here if just one of the stats is completely off. Thanks".

I have no idea why anything I've said is perceived as condescending. Clearly people here are used to apologetic tones and anything else is construed as arrogant.

I think maybe you are just not adept at presenting a hypothesis.

Thanks this is good to know.

Absolutely


I did not know this. But my post wasn't so much about MDF as it was about any approach of "calling enough hands here". So with MDF or without, your calling range there would probably look pretty similar to what I chose in the equilab screenshots, and it would be affected by the opponent's removal just as significantly. Thus you'd think you're calling down optimal, but you'd be not calling enough. Like I said in another post, this example wasn't so good because you don't have anything to call him with anyway. But if he had AJ there, you'd be getting exploited.


I don't know if we're on the same page here, but knowing what your opponent does isn't exactly theory based approach. If I knew for sure my opponent does X I wouldn't even consider my range in the first place, I would simply weigh my current holding against X. So I do agree with your point obviously that we wouldn't do it if we knew, but my post details a situation where hero plays a strategy where he doesn't care what opponent does. A la Doug Polk style.

Didn't know this. So in my example you'd be calling tremendously less than me, because many of the hands don't even count towards MDF as they're not bluffcatching material. Further you'd call even less than that because as you said villain's range still has equity as there are cards to come. Out of curiosity, what would you be calling from the range I posted?

If that was the case you'd be criticising the hypothesis and not my tone

I think you'll find it's the opposite.

Good clean example, thank you. But what if OOP doesn't know which of his hands are bluffcatchers and which aren't? What if IP also has 33, 44 to bet as bluffs? Then OOP would be overfolding, right? If we are discussing a theory based approach, shouldn't there be an optimal solution where you can do the correct thing without making assumptions about your opponent's range? Because otherwise aren't we just playing another kind of exploitative poker?

Cool then; going through my post history seemed like a different thing

This is specifically a toy game and isn't meant to represent reality. It just represents where MDF and EV comes from.

You're asking a question that is not possible to answer as it is something along the lines of, "If I know nothing and have no idea what is going on then how do I know what to do?" Well, the answer is you don't in that case.

In the situation where OOP doesn't know what hands his bluff catchers are then there is likely some other fundamental flaw with the thought process.

In the situation where IP has additional bluff combos to choose from it doesn't matter as he only balances out his range with 9 of them in this example.

You're essentially referencing "game theory optimal" with your statement of,

Which would encompass the entire game tree and is not relevant to a toy game scenario.

There are lots of situations in which a player can be balanced, not be playing GTO, and also not be playing exploitative.

Not to barrage you with questions, but if it doesn't represent reality, how can this example be implemented? What is the moral of it? Can you describe a situation in your own poker career where you used this kind of thinking in a river spot?

I think I just thought that MDF refers to a part of a GTO approach and thus thought that you should be able to implement it without making assumptions about your opponent's range. But now I understand that MDF is just an overall concept of balance


Good to know, I used to think that GTO can be used for isolated individual scenarios and not just for a full solution to the game.

I never considered this. I always thought that if you're balanced, then your unaxploitable, if you are unexploitable - you are GTO. But yeah balance does make sense even if it's outside GTO

Right, I realize the spirit of your post was focused on a different topic, so I apologize for taking it off course.

I think the aspect that is causing difficulty here is the focus on specific hands rather than a range of hands. You're essentially highlighting the blocking effects that occur in poker - and I suppose your observation is correct to some extent once you zoom on a single holding of your opponent. However, as you mentioned in a later post - if you knew your opponent held a single hand, you wouldn't consider MDF at all. Once you allow for your opponent to hold a range of hands, now the blocking effect you've highlighted gets minimized as they don't always block your calling range.

Perhaps a good thought experiment is: say you're right and these blocking effects cause you to overfold to the bet and you wanted to correct for this. What would you do?

All good man .

Not sure which post you're referring to, but if it's the one where I quote answered you last time, by X I didn't mean specific hand but his entire strategy. So the problem I am highlighting is so big when we don't know that our opponent does this with only AJ. Obv if we know that he does it with only QJ or AJ, it becomes a very easy problem to solve.

You got me good here man, I honestly do not know what to do here. The point I was driving with the original post - as I already mentioned here somewhere - is that our range is always a function of our opponent, the cards we hold in our hand never are. So step one to implementing this piece of theory is placing a little bit more value on the hand we currently hold than to the hands we should have in our range. So for example if we multiply the value of the hand we are currently holding by 1.25, or 1.4, this would be a sustainable solution? Obviously these are purely arbitrary numbers, but feel-wise it is a pretty implementable piece of theory I think.

A toy game is just a model to illustrate elementary aspects of game theory. Just because it doesn't take 100% of thinks into account, does not mean it isn't useful in improving our understanding. Many models for many different systems do this.... That is literally the definition/use case of a model.

In the example I gave you could definitely have similar scenarios where one player is mostly polarized on the river and then constructing a reasonable number of value:bluff would be important. A scenario very similar to this could be reached on the same run out in a 3 bet pot at ~100bb eff. For example, UTG opens and CO 3bets and UTG flats. If the board runs out 22233 then CO is going to be pretty polarized between essentially JJ+ and air.

Okay, thank you for all the advice. Have a good one

Nash ftw, basically. MDF doesn’t give a complete picture, just explains an aspect of the problem. But to know the precise GTO ranges (which can include mixing/playing hands with certain non-0/non-1 probabilities), you have to work out the equilibrium solutions, and this isn’t as simple as top X% of range though that is a good start in practice.