A software I would love to have

[PokerBob]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
Does this type of software allready exist?

[/ QUOTE ]

yes, but it is my brain.

[/ QUOTE ]

bob let's say I wrote a quiz that looks something like this:

1. You raise A[img]/images/graemlins/club.gif[/img]J[img]/images/graemlins/heart.gif[/img] in the cutoff, big blind calls. After the flop your opponent check/raises you. He will check/raise with any pair or two pair, any flush/straight draw that includes at least one over. With a set or a five card hand he will just check/call.

Now I flash you two flops at a time. for one of these flops you will always have at least 5% higher equity than the other. what percentage of the time do you think you'll pick the right flop?

do you think it matters? If you did this a whole bunch of times do you think you would improve?

[/ QUOTE ]

what are the flops? who is the guy? what is our recent history? lots of stuff matters here.

[nickabourisk]

[ QUOTE ]
what are the flops? who is the guy? what is our recent history? lots of stuff matters here.

[/ QUOTE ]

I don't think you understand the main interest behind such software. It's not to create a program which can always choose the best decision (highest EV) against a specific dynamic opponent who doesn't always play the same. You could answer the exact same way for all poker books containing hand quizzes in them as there is ALWAYS information which is missing (and obv. important for poker decisions).

However, if you make some assumptions:
- I know my two holecards
- I know my opponent's preflop range
- I know what the flop is (3 specific cards)
- I know my opponent's STATIC strategy (just like jba gave as an example)

THEN you can determine the EV of different betting lines (based on the fall of different turn/river cards and the knowledge of your opponent's later-street strategy). This is not intended to be a piece of software which solves poker or anything but simply a tool which would aid very much in postflop analysis (i.e. even making such unrealistic assumptions, using the program over and over again would aid us in our "feel" of what is generally mathematically correct). Basically, it's so that you wouldn't have to do post-analysis calculations for a specific hand by paper-and-pencil.

[ALL1N]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
Does this type of software allready exist?

[/ QUOTE ]

yes, but it is my brain.

[/ QUOTE ]

bob let's say I wrote a quiz that looks something like this:

1. You raise A[img]/images/graemlins/club.gif[/img]J[img]/images/graemlins/heart.gif[/img] in the cutoff, big blind calls. After the flop your opponent check/raises you. He will check/raise with any pair or two pair, any flush/straight draw that includes at least one over. With a set or a five card hand he will just check/call.

Now I flash you two flops at a time. for one of these flops you will always have at least 5% higher equity than the other. what percentage of the time do you think you'll pick the right flop?

do you think it matters? If you did this a whole bunch of times do you think you would improve?

[/ QUOTE ]

what are the flops? who is the guy? what is our recent history? lots of stuff matters here.

[/ QUOTE ]

??? He asked you a pure maths problem and you ask who he is and for history...

[PokerBob]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
Does this type of software allready exist?

[/ QUOTE ]

yes, but it is my brain.

[/ QUOTE ]

bob let's say I wrote a quiz that looks something like this:

1. You raise A[img]/images/graemlins/club.gif[/img]J[img]/images/graemlins/heart.gif[/img] in the cutoff, big blind calls. After the flop your opponent check/raises you. He will check/raise with any pair or two pair, any flush/straight draw that includes at least one over. With a set or a five card hand he will just check/call.

Now I flash you two flops at a time. for one of these flops you will always have at least 5% higher equity than the other. what percentage of the time do you think you'll pick the right flop?

do you think it matters? If you did this a whole bunch of times do you think you would improve?

[/ QUOTE ]

what are the flops? who is the guy? what is our recent history? lots of stuff matters here.

[/ QUOTE ]

??? He asked you a pure maths problem and you ask who he is and for history...

[/ QUOTE ]

of course. these kind of situations are much more than pure math problems, and treating them as such is a mistake.

[Victor]

with enough hands everything is a math problem.

or, without any hands, we can make reasonable assumptions to reduce it to a math problem.

[PokerBob]

[ QUOTE ]
with enough hands everything is a math problem.

or, without any hands, we can make reasonable assumptions to reduce it to a math problem.

[/ QUOTE ]

we can do that, and we do it based on recent history. i don't think a program can do that.

[jba]

bob I'm not disagreeing with anything you're saying but you're missing the point that I and others are trying to make. the point is if you have the best reads in the world but you are 5% off on your equity estimate 10% of the time or don't realize that the line you are taking is less profitable than another given your read you are making mistakes that are significant and that these are leaks that can be easily fixed given the right study and/or tools. maybe you don't have these problems but I know from years of analyzing hands after the fact that I do. as a most basic example: you are on the river and you are 100% sure that your opponent will call with all worse hands and 3bet for value with all better hands and also bluff optimally. you also exactly know every hand in his range. everything that you are good at knowing. what i'd like to do is given that specific info, take you from knowing that you are a 60-70% favorite to knowing that you are a 64-70% favorite. obviously that makes a pretty important difference. or, you used to know that your are a 60-68% favorite but today you've seen some stuff to make you realize that he has JTs in his range and you beat JTs and JTs in his range brings you up to a 64-72% favorite. I can't tell you how many times I've been playing with stove and realized something like "wow I didn't think AT was that important here".

it is an indisputable, provable fact that IF given a certain sequence of actions, and IF you can reduce an opponents hand range to {A} and flop 1 is BCD and flop 2 is EFG and flop 1 has 5% greater equity than flop 2, you are a MUCH better player if you can quickly and consistently tell which is flop 1 or flop 2. you can't convince me otherwise. likewise if you can build a decision tree of "if I take action H, villain will do I with subrange {A1}, J with subrange {A2}, and K with subrange {A3}; else if I take action L, villain will do M with subrange {A1}, N with subrange {A2}" etc etc, it is very beneficial given range {A} and subranges {A1}, {A2}, {A3} what the different EV is between H/L/etc. hope people can follow that because it's a lot of letters and stuff.

I think solving the problem of determining ranges and subranges, and what actions villains will take with each of those is a VERY VERY difficult problem to solve and is the subject a PhD in AI would try to tackle. I'm just barely smart enough to realize that I can't solve this problem with a computer. this is something that Bob is very good at because he's good at poker and it's something that's a LOT easier for a person to learn that it is for a person to teach a computer to do (in real time especially). however determining equity on certain boards between your specific cards and a specific range or subrange, and calculating EV for a line given the same, and providing good ui for a HUMAN to give inputs for those ranges and subranges is a very solvable problem. someone who is really bad at poker could do this.

[Victor]

[ QUOTE ]
[ QUOTE ]
with enough hands everything is a math problem.

or, without any hands, we can make reasonable assumptions to reduce it to a math problem.

[/ QUOTE ]

we can do that, and we do it based on recent history. i don't think a program can do that.

[/ QUOTE ]

i think a program would be an indispensible tool for direction.

[Sushiglutton]

[ QUOTE ]
of course. these kind of situations are much more than pure math problems, and treating them as such is a mistake.

[/ QUOTE ]

You are correct of course that this software wouldn't teach anyone to play great poker. It won't teach u anything about reading ur opp, recent history, how ur plays affects future action etc. But IF u analyzed all the psychology part corectly the software could help u find the 'best' line. Also it could help u find break-even points like: How aggro does my opp need to be before c/c becomes better than b/f?

Just because u know that the decission won't be easy of course. U still have to 'guess' how ur opp play, meta-game, physical reads etc.
Every decission is a cmbination of math and psychology and the software would only help on the math part. But I think that would be a great aid.

[baronzeus]

i think generally if i wrote this tool it would expose TONS of leaks in my game. i would literally be unstoppable.

also i would never ever sell it. its far far more valuable than you think pokerbob.

if someone wrote one that was good enough (i would have to try it first) i would gladly pay several hundred for it