ranker = StrategyRanker(utility)
ranked_strategies = ranker.rank_strategies([strategy_a, strategy_b], actor)

best = ranked_strategies[0]           # highest utility
print(best.strategy.id)
print(best.utility_frame.scalar_value)
print(best.rank_key)                  # sorting tuple used for tie-breaking
print(best.terminal_node_ids)         # terminal nodes that contributed to the score
print(best.terminal_probabilities)    # {node_id: accumulated_probability}