Pokerbots

I competed in Pokerbots IAP contest with a brother from my fraternity, and we won second place in the new team bracket! Check out the code .

We used statistics tracking to follow opponent’s strategy and mold our aggression and looseness to counter theirs. We did not focus highly on attempting to calculate absolute odds - we found that after ten thousand rounds of Monte-Carlo, the percent error is relatively small and this certainly was not our performance bottleneck. We did not use huge look-up tables to determine our moves; we opted to use our opponent’s strategy and the equity of our hand to determine our aggressiveness. A lookup table may have been useful to store some really good hands, but we preferred to develop a ‘thinking’ bot rather than a ‘reading’ one.