Alpha-Beta Pruning enhances minimax by tracking two bounds: α (best value MAX can guarantee) and β (best value MIN can guarantee). When α ≥ β, we have a cutoff — the remaining siblings cannot affect the result and are pruned (shown in red). The key insight: if a minimizer already has a better option than what the maximizer can achieve via this branch, there's no need to explore further. In the best case (perfect move ordering), alpha-beta reduces the effective branching factor from b to √b, allowing twice the search depth in the same time. This is why chess engines use iterative deepening alpha-beta with move ordering heuristics. Pruned subtrees are shown as red crosses — watch how much of the tree is eliminated.