Preflop practice software

[Isura]

I'm working on developing a solid preflop strategy for the 6 max limit hold'em. I have trouble sitting down cold at the table and making boarderline preflop decisions (especially facing a raise). Does there exist any software where I can practice with different starting hands. I mean, where the program generates a random hand (and position) and the user can decide on the move. Also, it wouuld be useful to track VP$P, PFR % etc to further analyse the tightness of the strategy. Any recommendations? Thanks.

[sthief09]

Turbo Texas Hold'em is good for this

[Isura]

Thanks.

[Crispy86]

Any poker playing program will do this. Just playing it you will get unlimited random hands, and the position will vary. You can also set the number of players. Personally, I think you'd best be served by Poker Academy Pro (http://www.poker-academy.com), as it also has the best AI of poker bots.

[fire_fly]

ok, just got poker academy pro, and WTF?! 10 years in the making?! For what? The f***ing graphics?

These bots seem like such retarded donks. Who the hell raises pf then folds to a reraise? This happens ALL THE TIME in this "game" or "simulation" or whatever. Haven't played THAT much, but the ai seems downright RETARDED.

Anyone else have any experience with this program?

Oh, and postflop they seem ridiculously easy to push off of hands. Just horrible...

[Huxley]

Fire-fly,

Right click on each of your opponents, click Edit and then in the pre-flop and post-flop windows you can use the sliders to adjust their styles of play to be more to your liking.

Also, NL is a recently added feature in Poker Academy, so the NL AI is not as refined or powerful. The decade or so of research was concentrated almost exclusively on Limit and more recently on HU NL.

From my experience, Poker Academy AI is the best commercially available software, but it breaks down like this:

HU NL: Excellent AI. Sparbot and Vexbot are the nuts. Once you've played enough hands against Vexbot to allow it to build a good opponent model - look out! - the damn bot lives up to its name.

Limit: Good AI. Many of the bots are highly tunable, allowing you to simulate different types of players and tables. You should be able to find a decent challenge amongst the tougher bots, especially after they start to get a read on you (>1000 hands).

NL: Not very good AI (but still IMHO the best available and getting better rapidly). The Poker Academy forum has a lot of info about the new features and improved AI coming in the next version. We'll have to see if the upgrade lives up to its hype.

Using PA does require an initial investment in time and effort to master the analysis tools and become adept at tuning the bots. My experience has been that this effort is well worth the results obtained.

Do some bot tweaking and then play enough hands under the same player profile for the bots to start to get a good read on you. My bet is you will be in for a pleasant (and therefore aggravating!) surprise.

[RedManPlus]

Huxley is obviously a shill for PA.
Do you happen to live in Alberta, Huxley?

That said...
I doubt that FireFly is using PA correctly.

For example...
The original Limit full table...
Where you are one of 10 players...
Has several Strong Bots...
And several Weak or Loose Bots.

If you see a Bot make a poor play...
No need to come on here and FREAK OUT.

It may be programmed to play loose and passive...
Or maybe you chanced upon a leak.

If you want to set up a full table...
That is MUCH tougher than, say, $2/$4 Party Poker...
You build a table...
With 9 PokiBots that are configured to play tough.

Edit the Bot configurations.

There is a special Bot for head-to-head play.

As for NL...
A good player can probably win here...
Because the AI for NL is far more complex...
Then AI for a Limit table.

I think it's a great program...
But if you want perfection...
Than I would suggest searching...
For the Fountain of Youth first.

rm+

/images/graemlins/cool.gif /images/graemlins/cool.gif /images/graemlins/cool.gif

[Huxley]

rmp,

I'm no shill... but I am a shareholder in BioTools Inc., the company that develops and sells Poker Academy. I think I've mentioned that at least a couple of times previously on this and other popular poker forums. I believe the PA software offers great value, but if I thought the software was poor I wouldn't hesitate to say so - afterall, I am also a shareholder in Microsoft and I regularly disparage their products.

So of course I want to see PA do well, but mainly I enjoy the software a lot and and want to help out when someone on the forum is having problems with it or it is not living up to their expectations.

I guess that makes me an entrepreneurial altruist, but certainly not a shill.

[fire_fly]

You know, I realize that my experience with the software is based on only about 3 times using it, and I did read that the more you play it the more some of the bots get to use your strategy and such, but I was just commenting on my first reactions to it.

I also realize that I could set up the bots to be more aggressive (ie, realistic) and I haven't tried this yet.

The reason why I was so shocked was that I had read on the website that a prominent university had worked on the AI for 10 years, so I expected a lot.

I was very surprised to find that I was dragging about 60 percent of the pots playing a 6max game through very aggressive play, and the bots never caught on. Again, this just might be becaues I didn't alter any of the bots or put in some of the smarter ones, or mabye I just didn't give them enough time to catch on.

But I AM still interested in using this poker tool. Any suggestions? I'd like to experience some close to real life 6max before I dive into the oft talked about 1-2 party 6 max. I want to practice my new starting hand requirements and try to get a handle on the higher aggression levels at these stakes before I put my money on the table.

So does anyone know of a good setup for close to real life party 1-2 6 max? That would be very helpful, and if I get things set up and give it another go, I'll post my thoughts here.

[RedManPlus]

Dear Huxley,

With all due respect...
If you are a shareholder...
That stands to profit from sales of PA...

Then you are a SHILL.

No one can be "objective"...
Everyone is biased...
But with money on the line...
"Objectivity" isn't even on the table.

Some people have a big problem...
That they actually start believing their own "lies".
We all know people like that.
Be careful not to become one of those people.

But at least now everybody is aware...
Of your conflict of interest...
So carry on... post away.

Regards,

rm+

/images/graemlins/cool.gif /images/graemlins/cool.gif /images/graemlins/cool.gif

[Crispy86]

[ QUOTE ]
Dear Huxley,

With all due respect...
If you are a shareholder...
That stands to profit from sales of PA...

Then you are a SHILL.

No one can be "objective"...


[/ QUOTE ]

With all due respect, I can't even begin to understand what all this talk is about. First of all, he didn't try to sell the program to anyone, since the person he is answering is a legitimate customer complaining about some issues. Instead he immediately explained how to modify the bot profiles to get a better game, and thus achieve a satisfactory user experience. I'd applaud this if he were a direct developer, so I'm confused as to your reply. It sounds as if you object to his helping someone with their software, which is exactly what he was doing.

[Huxley]

Thanks for your support, Albert. You touched upon everything I would have been inclined to say to RedManPlus, but you accomplished it in a more concise and diplomatic manner than I might have managed.

After looking back after my original post, I find it difficult to believe that comments like "NL: not very good" and "We'll have to see if the upgrade lives up to its hype" make me sound like a PAPro salesguy.

Anyway, RMP, how about you and I bury the hatchet and see what we can do to assist Fire Fly. I kind of enjoy your pseudo-poetic style of presentation - perhaps you could describe the configuration of a killer SH table in sonnet format?

BTW, Fire Fly, were you interested in a shorthanded Limit table, or SH NL, or both? I'm planning to spend some time over the weekend toying with PAPro's bot configurations and I'd be happy to post the results of my efforts for you.

Although...
the thought of doing so...
worries me, you know.

One more post from me...
And the blood pressure of RMP...
Might blow.

Cheers, Huxley

[fire_fly]

Hey Hux,

To answer your question, I'm only interested in playing limit right now. I've been regulating at micro-full table and want to take the oft talked about and dreaded jump to 1/2 6 or 5 max.

I've heard that these games are pretty much full of loose and aggressive people.

Oh and while I'm at it, just wondering, I know for full table very low limits, a table with a 35% vp$ip is kind of a "holy grail" around here, is there that kind of number for 6 or 5 max? Would the two be any different as far as ideal table conditions to look for?

Thanks for the help everybody.

Oh, and btw, started up a shorthanded table last night, turned every single bot to "loose" pf and it was quite a bit better, though still not very realistic imho (they were STILL too weak/tight.)