http://forumserver.twoplustwo.com/showthread.php?p=2678035
http://pokerazor.com/

This could be huuuuuuuge.

looks pretty impressive, luckily knuckle heads like me will never be able to figure it out.

If this software can be linked to the PT databases, then there's just a small step left to build NL deepstack robots, right? I mean, if from PT you can extract villain's strategies for a given board and action (with enough sample hands), then you could use this software to find the strategy that maximizes the EV. Would this be possible?

I think this is an interesting point. I mean, doesn't it get to a point where you're taking the thinking out of the game by making software like this? Pokertracker allows us to track certain stats that help us remember what kind of "by the numbers" player someone is, but we still have to make our own decisions on what plays to make versus them. Pokerstove allows us to run our hand vs our opponents range of hands, which helps us know our odds and equity in certain situations that we're likely to encounter.

I dunno, it looks cool and everything, but when they make programs that are telling people WHAT PLAY is most optimal, it takes the skill out of the game IMO.

To be blunt, this type of thing will slowly kill higher and then even mid-stakes poker both in terms of the number of players playing online and the skill of those players. This is what happened in chess, there was a boom around the time of Kasparov vs. Deep Blue but now the best machines (Fritz 11.0 and Rybka) can easily best the world champions. The lustre is gone and chess is all but solved.

Don't get me wrong, I think it will take a long time. Many players won't know this technology is around and others will and still won't use it/care. But when machines that can almost perfectly and instanteously calculate the best move (they'll call them 'hints' or 'tips' or something) saturate the high-limits it can only damage the appeal of poker to the lesser players.

This is one step away from computers auto-evaluating players skill-level, auto-evaluating their ranges, insta-calculating the odds and expected EV and then clearly presenting the results in a manner untraceable by the websites. Anyone who foresees that has got to be worried.

well, there is definately going to be the point where all the online sites just forbid this.

i think you are jumping the gun here a bit, but "Find odds of your opponent getting strong hands relative to the board." is pretty much the hardest part for a computer imho.

also you ranges aren't going to be very accurate unless you have a TON of hands on each player.

Is this all the rage in teh England?

Voila.

LOLOL.

http://pokerazor.com/blog/2008/02/04/under...n-reraised-pot/
http://pokerazor.com/blog/2008/01/26/simpl...m-a-tournament/

This is why this program is so useful imo.

Hmmm...

I think this is a bit fantasy. Maybe I am to much believeing in the skills of professional poker players here, but I think there is a slight but subtle difference between

a.) chess, where a machine being able to calculate the next 835.000.000 possible combination of moves (lol at own estimate for chess combinations) will certainly have an advantage over the human player and

b.) poker, whre a pro needs to think through the betting preflop and on 3 streets.

After all, the machine can only "learn" what the pro does and call if it has cards "good enough" by the odds given by the pot.
On the other hand the pro should be able to figure out what the relatively "dumb" machine here is doing and "learning" from him rather quickly and should be able to read it good enough to provide "bad odds" for the machine often enough with entirely empty hands.

I agree that it will kill fun on lower and medium stakes levels though.

Hi,

I am one of the Pokerazor authors and I think that's a good moment to chime in and calm down some concerns a bit.
First, this software will not make any decisions for a player nor say what the optimal play in given moment is.
I agree that this is very powerful tool and it will be even better soon as we have some more features in the plans but it won't solve anything for anybody. You need to choose settings yourself. You need to have good read what possible range of your opponent is and more importantly HOW HE BEHAVES with diffrent part of this range.
Sure once you know it Pokerazor will give the answers in 1 seconds and spare you 30mins of calculating it by hand which many players (me including) did for many months. There is long way from this to actually making correct decision though.
You need to analyze many many situations to get the edge. This is huge effort and only the most dedicated player will get through it. To this day getting knowledge about how to solve holdem problems required either years of experience, brilliant intuition or tremendous effort. True, now it will be simpler to get through math part of the game but there is still huge area left : psychological war. There is no way programs like ours will help with it in near future.
I am sure level of play of average "regular" will improve greatly there is other side of this though. People who make money playing poker will have much less stress. They will have solid mathematical fundamentals and will understand from when they edge come from which will allow them to get through downswings with undamaged confidence.
I also agree that sooner or later purely "feel" players will all go broke but it's nothing new. It was happening for some time now and our software will just speed up the process :-)
Thanks all for sharing your thoughts (I found out about this thread today and I am happy our software is getting that much interest). I will answer any question you may have as long as it's within the rules of this forum.