Re: High Limits
 

[ QUOTE ]
[ QUOTE ]
If were playing 50-100 nl I might agree. However, given the fact that 8 tabling you do not have great reads, the play is fairly mechanical. With some work a program could be written to beat the 5-10 10-20 nl game.

[/ QUOTE ]

Idiot.

[/ QUOTE ]

I don't get it ... why is this stupid? I don't think it would be hard to play the 10/20 without reads or statistics and beat the game. Now I'm not that smart and I don't know dick about computers but I imagine that someone who did could easily write a program to play without reads and beat the game ... in fact it would probably do better than 90% of people due to the fact that it would never tilt.

Re: High Limits
 

I really dont think a machine/bot could be made to effectively beat those games. They arent creative and adaptive enough. Clones on the other hand..... hmmm seems like a very profitable possibility

Re: High Limits
 

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
If were playing 50-100 nl I might agree. However, given the fact that 8 tabling you do not have great reads, the play is fairly mechanical. With some work a program could be written to beat the 5-10 10-20 nl game.

[/ QUOTE ]

Idiot.

[/ QUOTE ]

I don't get it ... why is this stupid? I don't think it would be hard to play the 10/20 without reads or statistics and beat the game. Now I'm not that smart and I don't know dick about computers but I imagine that someone who did could easily write a program to play without reads and beat the game ... in fact it would probably do better than 90% of people due to the fact that it would never tilt.

[/ QUOTE ]

depends on the players in the 10/20 game. a bot could consistently beat fish but not good players that could adjust. a bot could probably beat 50/100 if the players in the game were bad.

Re: High Limits
 

[ QUOTE ]
bankroll

[/ QUOTE ]

no definitly not it

Re: High Limits
 

I 12 table and its mainly the same players, that you have played 1000's of hands with at all my tables. In many ways it is easier to get reads because you are playing a tonne of hands with people and i can quickly tell how they are playing right at the moment. Then if some donkey comes along, its easy to separate them from their money for a very simple reason: they suck.

Re: High Limits
 

[ QUOTE ]
If were playing 50-100 nl I might agree. However, given the fact that 8 tabling you do not have great reads, the play is fairly mechanical. With some work a program could be written to beat the 5-10 10-20 nl game.

[/ QUOTE ]

I once thought as you do ... and maybe still do.

Some friends and I at work decided we'd write our own poker bots and then hold tournaments between them. I started thinking about the number of situations that the bot would need to deal with. It turns out that a good rule-based bot would require a lot of coding. A lot.

A bot based on a genetic algorithm where the parameters (aggression, hand strength evaluation, hand reading, etc.) can evolve over generations of the bot as it plays could probably start in the lower stakes and move up. Additionally, if you give it access to large opponent profiling databases I think that bot would own.

mj

Re: High Limits
 

Of course a bot *COULD* be programmed to beat the 5/10 - 10/20 games, but it wouldnt easy and probably couldnt be done by an 'average joe' bedroom coder. I think that someone killing higher stakes games with a degree in something like computational neuroscience could pull it off without much difficulty, using reinforcement learning algorithms etc.