Compute game-theoretically optimal (GTO) strategies using the CFR algorithm. This approximates Nash equilibrium play through iterative regret minimization.
Probability of winning the hand at showdown
More iterations = better convergence (5,000-10,000 recommended)
Counterfactual Regret Minimization (CFR) is an iterative algorithm that computes approximate Nash equilibrium strategies in imperfect information games. The algorithm tracks "regret" for not taking each action and adjusts strategy probabilities to minimize cumulative regret over many iterations. The resulting mixed strategy prevents exploitation by opponents who know your tendencies. This implementation is simplified for academic research and demonstrates the core CFR concepts.