Implementing Game Theory

[Marlow]

I'm rereading parts of the Reuben/Ciaffone book, and it got me thinking about working some game theory into the mix. But I have a problem with this. How do I use Game Theory when most of my decisions are read- and player-dependent?

Marlow

[FatalError]

a good game theory example is bluffing frequency where you are overbet moving in on a flop with a nut hand vs complete air/semibluff in such a frequency that your opponent cannot call with any hand range or frequency and make a +ev decision

can you elaborate on this? examples etc?

[Marlow]

Yeah, sorry about the content-light OP.

Let's say that I raise with AK OOP and get a caller on the button. The flop misses me. No pair, no draw. I will make a continuation bet depending on a number of factors including:
-my opponent's tendencies (LAG, etc)
-board texture
-my table image, etc.

So how does Game theory fit into this picture? Do I only use it if my opponent has no leaks that I feel I can exploit, or is it something that should be adjusted depending on the situation (i.e. bluff 10% against LAG, and 25% against solid TAG, etc.)? Also, this is a very simple example. Once we get to the river, I feel that I should be making all of my decisions based on the information I've got. Or is there some way that GT can be helpful?

Marlow

[ninjia3x]

[ QUOTE ]
a good game theory example is bluffing frequency where you are overbet moving in on a flop with a nut hand vs complete air/semibluff in such a frequency that your opponent cannot call with any hand range or frequency and make a +ev decision

[/ QUOTE ]

shouldn't it be: a bluff frequency that will not effect the ev no matter what your opponent does (call/fold/raise) is the optimal one.

[fimbulwinter]

This was one of the take home problems in the second part of my how-to EV posts in SSNL.

fim

[async]

[ QUOTE ]
[ QUOTE ]
a good game theory example is bluffing frequency where you are overbet moving in on a flop with a nut hand vs complete air/semibluff in such a frequency that your opponent cannot call with any hand range or frequency and make a +ev decision

[/ QUOTE ]

shouldn't it be: a bluff frequency that will not effect the ev no matter what your opponent does (call/fold/raise) is the optimal one.

[/ QUOTE ]

There are situations where bluffing is +EV, regardless of what your opponent does. Sklansky touches on it in TOP.

[Marlow]

Ok, does anyone here even actually USE game theory?

Not being a jerk, I'm really curious. /images/graemlins/smirk.gif

Marlow

[thabadguy]

I dont mean to sound rude,but i think you're bumping your own thread again (as you have done in the past) which is not accepted at MHNL.

[fimbulwinter]

[ QUOTE ]
I dont mean to sound rude,but i think you're bumping your own thread again (as you have done in the past) which is not accepted at MHNL.

[/ QUOTE ]

what? since when? I see posters doing this kind of thing all the time.

crap like this is as annoying as people posting short 2/4, not fatal but annoying.

fim

[9cao]

I personally use the principles when putting together a general gameplan but in the heat of the moment no.

[Marlow]

[ QUOTE ]
I dont mean to sound rude,but i think you're bumping your own thread again (as you have done in the past) which is not accepted at MHNL.

[/ QUOTE ]

No, I don't bump my own threads. I'm trying to get a question answered. I thought maybe the question was a bad one, and I was trying to rephrase the OP or redirect the thread altogether.

I'm certainly content to see it die now...

Marlow

[thabadguy]

When you post " No takers??" when no1 replies to ur post, or advertise your thread in another thread, asking that person to look at ur post, thats bumping in my book.

[Marlow]

[ QUOTE ]
When you post " No takers??" when no1 replies to ur post, or advertise your thread in another thread, asking that person to look at ur post, thats bumping in my book.

[/ QUOTE ]

Thanks for the opinion. All of what you say is wrong or out of context. I sent you a PM. I don't care to carry on publicly.

Marlow

[Leptyne]

Game Theory is a useful tool in optimizing strategy when faced with a decision. I think it's application to poker is limited to HU play, and you will quickly learn the optimum strategy is to vary your tactics so that you are unpredictable. The IBM game theory team that decided to write a program for a super-computer that could play chess thought it would be challenging but not overwhelming. Since the number of moves is finite it should be easy for a super somputer to examine all moves possible and select the best move each time. Simple decision making, classic game theory. In the first match against Big Blue in 1996 Kasparov won 3-1-2. IBM was able to study these games and write a "deeper" program which was named "Deep Blue" and the rematch was on for 1997. It was a tie after five games, 2.5-2.5, and Deep Blue won the final match.

My point is that delving deeper (pun intended) into game theory will not help your poker game.

[Leptyne]

[ QUOTE ]
I dont mean to sound rude,but i think you're bumping your own thread again (as you have done in the past) which is not accepted at MHNL.

[/ QUOTE ]



It sounds to me like you do mean to be rude. Even if you are correct I thought that criticisms like this were to be sent to AZK for evaluation. I'm not sure about that, but it would seem appropriate for the moderator to address your concern.

[Subfallen]

[ QUOTE ]
Yeah, sorry about the content-light OP.

Let's say that I raise with AK OOP and get a caller on the button. The flop misses me. No pair, no draw. I will make a continuation bet depending on a number of factors including:
-my opponent's tendencies (LAG, etc)
-board texture
-my table image, etc.

So how does Game theory fit into this picture? Do I only use it if my opponent has no leaks that I feel I can exploit, or is it something that should be adjusted depending on the situation (i.e. bluff 10% against LAG, and 25% against solid TAG, etc.)? Also, this is a very simple example. Once we get to the river, I feel that I should be making all of my decisions based on the information I've got. Or is there some way that GT can be helpful?

Marlow

[/ QUOTE ]

I think you're missing the point. Yes, you *have* AK, but from a game theory perspective, your holding is the range of hands you raise for this amount in this position. Do you see why?