I set the time to 2s and let it play against 5 SimpleBots. MCTSBot still gets owned. Did anyone find a configuration where it actually wins? I think it might be just that the opponent model is not tuned for this kind of opponent.

EDIT: The mean std. div. is quite low according to the debug info. This means the simulation converged, does it not?

Well, how did you solve it? I have the same error.

Ok, apparently you have to change nextDouble to Double.parseDouble(next).

My bot beat FO2 by 6 big blinds/100 FLHU. Does anyone have a better result?

My FLHU bot beats FO2 for about 28 big blinds/100 hands.

That's pretty sick. Is it a explo or NE-bot?

A hybrid of both.

How much do you beat it by if you turn off the exploitive stuff?

Can't say, it's not really possible to turn it off with the algorithm used.

Is there any documentation on the API? I have read that there used to be some on the poker academy forms but they are gone. I've been able to get the program to run and I'm making my own bot to test but I have a lot of question about certain parameters and methods. (like how to set zip mode to true)

Thanks!

Thanks spears.

What is the reason for the strange results I have below. I have two bots i made and when they play heads up the results are not even. Yet when they play the other default bots I don't get these strange results.