If you make this bet OTR only for value you will get exploited with alot of fold -> not GTO
If you fold your bluff/value hand for just 1 chip you will likely be correct, but you will likely get exploited later on when the opponent knows that he gets 5 001 : 10 000 fold equity (33.4% for easier calculation).
If you get "exploited" more than 1 in 15 001 and you fold you lose a lot. So 1/15 001 * 33.4% or
in about 1/45 000 you should fold?
In % its something like 99.99998% call?
Originally posted by nsavov
If you make this bet OTR only for value you will get exploited with alot of fold -> not GTO
If you fold your bluff/value hand for just 1 chip you will likely be correct, but you will likely get exploited later on when the opponent knows that he gets 5 001 : 10 000 fold equity (33.4% for easier calculation).
If you get "exploited" more than 1 in 15 001 and you fold you lose a lot. So 1/15 001 * 33.4% orin about 1/45 000 you should fold?
In % its something like 99.99998% call?
First, you don't get to control the initial river bet. Just assume it is GTO. It's part of the premises
Next, if you always fold to that $1 raise the first few times, yes you are likely to be correct (add: in the result-oriented way).
And yes, the GTO expert is allowed to change his strategy (by the premises) to exploit any overfolding or underfolding.
Then you go mathematically wrong.
HINT: The correct answer is not an absurd number like being extremely low or high.
So what is the correct percentage, and how do you justify?
(sorry about adverse comments in previous posts :-)
PS, Give me X x Quadrillion examples where a pot size river bluff is GTO?
Originally posted by YohanN7
In practice, some have actually used a wristwatch (where is the second hand when they look) in live play to randomize. Online, you can probably legally have a program on you own running ready to produce a "call" or "fold" based on a percentage. Most use their holecards, calling with the best bluff catchers and folding the worst.
right, but firstly, folding the worst bluffcatcher would mean folding too much here, and secondly, i don't want to use external software, i want something i can do myself.
i guess i could use a watch somehow, wonder if that's the best method, and what watch-based methods there are
I don't know really, and I fully agree with your dislike of having a program running.
This is probably a topic of its own. My guess is that the holecard strategy isn't that bad in practice.
Firstly, it is possible to "break down" a hand like AK. (Say there is a runner-runner flush on the board and you face a big bet holding the nut straight.) Quite likely, it is incorrect to either always call or always fold AK because you (for some reason, let it just be true) have few flushes in your range to bluff catch with. Then you could subdivide the 15 AK combos (one AK would be the nuts) according to suits. So, some granularity is possible.
Secondly, it has the advantage that sometimes your best bluff catcher, by freak accident, beats your opponents value bet (not all play well), whereas your worst don't.
The disadvantage I see is that it takes a hell of a lot of work. Memorizing "good" and "bad" suit combos (of AK in example) is pretty quick. But there are a quite a few situations in NL Holde'm?( It's not always easy to see how "high up" in your range your actual holding is. The wrist watch could be a reasonable low-tech alternative, at least in very "polarized" situations.
In the scenario ITT, or playing Fixed Limit versus Cepheus there is no escape from an RNG.
Originally posted by GoOnCal1
...
(sorry about adverse comments in previous posts :-)
...
Well, too bad for the thread, it could have been good without your contributions. Now new readers will have to parse through your bullshit before coming to real replies. They might not bother.
But don't worry, I am used to attracting crack-pots and cranks. Such people hate logic and clarity, and attack it when they see it. Apology accepted. I know it wasn't personal.
imo my grasp of logic is near infallible
the turbidity of your opening premise a little confusing perhaps
Crackpot, well yes .. what can u do though...
Note also: The only proof you offer that your incomplete answer is correct is only your opinion and say so, not even an argument purporting to follow the rules of logic. What is the percentage (actual number and how is it derived to produce a poisitive EV?
:heart::heart::heart:
Originally posted by GoOnCal1
...
imo my grasp of logic is near infallible
...
Such beliefs are part of the defining conditions for crack-pottery and crankery.
I have not presented an answer or a proof. I'll do that later.
Please don't post more.
one quack always deserves another
Just stating that statistical relevance is not required in your open post, demonstrates you are clearly unread on even the definition of statistical terms. In any case as the situation is always going to be identical very few repetitions would be required for confidence level for 99.9%. Also having an RNG and a % constitutes TWO parameters, not one as you ask for.
Quack Quack Quack
Statistical relevance is needed, otherwise almost pure chance will determine whether you win your $100 000 000. You provide one parameter, the percentage, not the RNG. The RNG is part of the computer program. It is, for the present purposes, assumed perfect.
The number or "trials" or "hands played" is, as explained in OP, irrelevant. It could be the quadrillion, more than that, or less than that. Just so that it is enough to establish statistical relevance.
I can put it differently just for you. How often should you call to be unexploitable in the in OP described situation? We don't need computer programs or a $100 000 000 prize for that question. (But it would shun away most potential posters, hence the OP)
@GoOnCal
You post
Just stating that statistical relevance is not required in your open post, demonstrates you are clearly unread on even the definition of statistical terms.
The OP states
Note: The quadrillion hands may be replaced with a possibly larger number for it to be statistically significant, but this is of of no particular importance for the problem at hand.
This is about the tenth time you have posted flat out lies.
I have to ask you:
Are you drunk?
Are you trolling?
I call for a moderator. This individual is with or without purpose messing up a forum thread. This thread has thereby cost me a lot of time.
Call the mods, and stop replying.
How can a troll not have the last quack.
There is a method to calculate the number of trial needed for statistical relevance for a required confidence level. It involves running a few trials and measuring the variance of the result, as trials continue the variance about the result reduces. Not as you believe this trial could be set up to run as there is not enough information about action prior to the scenario for a competent programmer to set it up.
10th time I have told lies...in bold like all your mistakes in fact logic and clarity, name one or apologise or be quacked for ever more
Now way is statistical relevance "irrelevant" in any trial situation.
Hi all,
It appears that this thread has outlived its usefulness.
Please let's keep the coversation civil
I suggest reading our forum principles. People are going to be mistaken sometimes, and it is always OK to point out the mistakes, but it is important to do so politely.
If someone points out a mistake of ours (or what they think is a mistake) it is important to keep our rebuttals respectful.
If the tone of this thread doesn't improve, it will be closed.
Best regards,
VS
There is a method to calculate the number of trial needed for statistical relevance for a required confidence level. It involves running a few trials and measuring the variance of the result, as trials continue the variance about the result reduces
Exactly. (But that is not necessary, because you can calculate the required number of hands without "running a few trials". The probability distributions are known.) But this is not part of the problem for you to solve. As the OP says, enough hands will be played for the outcome to be for all practical purposes be free of pure luck/statistically relevant or whatever you want. You should realize though, that the critical situation might not occur that often. A "few trials" will do you no good.
Why do you misunderstand EVERYTING on purpose. Even the "amusement thing" about getting raised $1???
Another lie:
Not as you believe this trial could be set up to run as there is not enough information about action prior to the scenario for a competent programmer to set it up.
This is so vague and whimsical that it is impossible to reply to. "...action prior to..." - LOL
Now, let us forget this for a moment.
The question for you to answer is: How much should you call in the described situation to be unexploitable?
Can you do that? Can you for a start avoid misunderstanding the QUESTION OF THE THREAD?
If not, please keep out of this thread.
Yes Vorpal, please close this mess. I'll start a new thread with the same OP. Hopefully, it will be clean. And mind you, he wasn't "pointing out mistakes". It was downright trolling.
