After 6 years of hard work I would like to present you PokerAlfie AI-Poker-Player. PokerAlfie is 6-max No Limit Texas Holdem AI-Player and his playing strength challenges top poker players. PokerAlfie is very competitive against good real-money players.
Recently one of the best Serbian online players played against PokerAlfie, here is his comment: “I played against PokerAlfie v2.0 over 5000 hands and PokerAlfie was better for about 0.05 big blinds per hand.” - Nislija (Sany Boy - PokerStars). Nislija's worldwide highest ranking at PocketFives is 394
PokerAlfie is 100% free and without commercials.
PokerAlfie is suitable for all players.
If you are a rookie or a part-time player with PokerAlfie you have a unique opportunity to play against a world class opponent free of charge. If you are a poker enthusiast or an expert, PokerAlfie is the best free sparring partner for you to practically try, learn or examine concepts and strategies.
The unique and novel algorithm used by PokerAlfie enables extremely short learning (thinking) phase of only few seconds. PokerAlfie does not need to analyze millions and millions saved games. PokerAlfie only needs to know game rules and game goal.
One question, in the future are you going to add mttsng?
Inviato dal mio ONEPLUS A3003 utilizzando Tapatalk
There is a lot of work to be done: Decision engine improvements or porting to Android or user-play-analyzing-module or as you suggest other game variations as mtt or sng. I hope comments from this thread can help me by decision what will be done first.
This is a better bot postflop than Snowie in a way that it gives more realistic practice, though I still couldn't play full GTO, as I needed to adjust to what I thought it had and to its hyper-aggro style at times. Vs Snowie it was just preflop and the flop as far as the player c-betting goes; the rest is predictable.
This is beatable like Snowie, as they both have situational weaknesses where they drop up to stacks. You don't need to play 5k hands to see that, and knowing your technique is at least as good.
It was however interesting to play against random cards (I ran just a couple of stacks below EV) and different tightness levels, though it wasn't all as different overall as I was expecting.
But until the bot plays full GTO, it won't give full GTO practice. The next level from that is for one to be able to adjust how it plays; then it is full practice.