Win rate with optimal strategy against limit raise bot

[Ludanto]

"Putting in an extra $3 gave you and extra $0.02 worth of hot and cold equity from the flop action, but it gave up chances to fold with as low as 10.1% equity getting $12:$2 on the river."

Is anybody able to paraphrase it so it sound easier to understand?

[JacksonTens]

Cap pre vs top 50% of hands.
Call with ALL others.
Get crazy with all hands with >50% equity vs RND on flop.
If hand decreases below 50% equity vs RND on turn or river call down.

EASY

JT

[rakemeplz]

[ QUOTE ]
"Putting in an extra $3 gave you and extra $0.02 worth of hot and cold equity from the flop action, but it gave up chances to fold with as low as 10.1% equity getting $12:$2 on the river."

Is anybody able to paraphrase it so it sound easier to understand?

[/ QUOTE ]

He's saying bloating the pot on the flop makes it harder to make a profitable river laydown cuz the pot would be too large.

I'd imagine there are probably spots where you're sub 50%, where you'd want to bloat the pot on the flop to make calling on the turn easier (say if you had a draw), and/or even calling the river more profitable. Pzhon can chime in if I'm wrong or right.

[pzhon]

[ QUOTE ]
[ QUOTE ]
"Putting in an extra $3 gave you and extra $0.02 worth of hot and cold equity from the flop action, but it gave up chances to fold with as low as 10.1% equity getting $12:$2 on the river."

Is anybody able to paraphrase it so it sound easier to understand?

[/ QUOTE ]

He's saying bloating the pot on the flop makes it harder to make a profitable river laydown cuz the pot would be too large.

[/ QUOTE ]
Yes, although it is interesting to see how a fold can be described as profitable. It's common to say that folding is worth 0, but that's not the right base line here.

[ QUOTE ]

I'd imagine there are probably spots where you're sub 50%, where you'd want to bloat the pot on the flop to make calling on the turn easier (say if you had a draw), and/or even calling the river more profitable. Pzhon can chime in if I'm wrong or right.

[/ QUOTE ]
In NL, when your opponent may have an information advantage and your bets now decrease the amount which can be bet later, you can raise as an underdog. Maybe that could be the case in limit, too. However, I don't see how that could happen in this hypothetical situation.

[Ludanto]

What I don't understand is why we are talking in vague term like "makes it _harder_ to make a profitable laydown." The bot from OP is 100% predictable so we can talk in numbers. I assume we have all the time and computing power in the world for every street decision against this bot.
And I am wondering why we are talking about "fold with as low as 10.1%" after I demonstrated that by 4-betting the flop there is no card on turn or river which would make folding the best option.
If the argument is: "By just calling flop you can make FOLDING the best option on turn or river" then I ask: Isn't it in our best interest to not ever make folding the best option?
I still don't know what "winning hot and cold" means.

[helter skelter]

I think the solution to the problem is going to be a decsison tree. Assume that you will see a flop for every hand (I think the implied odds will show that this is the optimal strategy for every possible starting hand)

For each possible starting hand, you will have at least 2 trees for each flop depending on whether you are in the small blind or BB.

I think every possible board/pot combination has to be calculated to determine which path to take on the tree for each hand/board combo. At each decision point, one path will have a higher EV than another path.

Consider a hand like 72 on a flop 772. The most profitable path on the tree will most likely be raising the max from that point on. On a different kind of flop, the path that yields the most profit (or least loss) may be to fold to the next bet.

I think to get an understanding of it, you would have to work it out for at least a couple starting hands and flops.

There may be a way to develop an algorithm or program to examine all the possible situations. I would only consider doing it if this were a real possibility.

The only way this could be a real possibility is if it was awarded as a prize by a casino, like one of those shopping spree prizes where someone is given a shopping cart and let loose in a store for 20 minutes or whatever.

In this case, the prize could be structured one of 2 ways.

A: you are given the prize of being able to play a certain number of hands against the bot

B: you are given the prize of being able to play a certain length of time against the bot

Of course, you must be allowed use your computer where you have already worked out the optimal strategy for each decision point or branch on the tree. Otherwise you will just have to have some kind of approximation of an optimal strategy.

The optimal strategy for B above may be different than the optimal strategy of A since time is now a factor. I think it would be much more difficult to find the optimal strategy for case B.

[rakemeplz]

Ludanto, Uhh yes of course I think you are right...if folding is always neutral EV, and the only EV you can have on the river is by calling does it matter what odds we are getting on the river if we fold anyways? Folding on the river is 0 ev. Making ez folds gets us nothing but making calls with 20:1 pot odds and 5% chance of winning the pot makes us $$$$.

However putting in flop bets might help us in terms of getting pot odds to call the turn bet (making it more profitable), or turning a river fold into a call, or even by simply making the pot bigger on the river, so we can get more profit when we do snap off his frequent bluffs.

For some reason I didn't consider my contention that making the pot bigger to snap off bluffs, didn't contradict pzhon's contention that keeping the pot small so we don't have to call river bets, but they both can't be true, I don't think.

Anyways all of this could easily be wrong/riddled with errors, I'm just trying to learn/figure this out, I'll wait for pzhon or someone to point out all of my errors.

edit: oops replaced "-ev" with "neutral ev"

[pzhon]

If the first 4 explanations didn't agree with your intuition, do you think that is going to change with a 5th? You either need to be willing to fix your intuition, or you should give up.

If you want to understand, reread and think about what I said. Work through the numerical examples I gave. Think about the actual poker situations I mentioned where the same issue comes up. Correct your intuition.

Or, maybe you just don't have the tools to understand this yet. Maybe you should give up, and worry about simpler, more fundamental ideas.

[mykey1961]

Using the assumption that each round is capped at 4 bets:

I let my computer grind on 32o vs this bot to see if the bottom end of the spectrum should fold preflop.

It indicates that 32o should call one raise by the bot preflop.

Next I'll try Q5o, because that's borderline 50%+ preflop to see if it should cap preflop.

[mykey1961]

Q5o should not cap, but just call one raise by the bot preflop.