Considering the subjective and non-ideal behavior of electric vehicle aggregations, the aggregations coordinate the charging of electric vehicles and conduct the research on charging competition. With the goal of minimizing costs, each aggregation considers the behavior of neighboring aggregations and determines its own charging strategy, including the time of charging and the amount of charge. The competition between aggregations is simulated by establishing a two-stage non-cooperative game model. Based on the subjective evaluation of the opponent's behavior, the expected utility theory and prospect theory are used to study the influence of aggregations' charging strategy on game results. The result shows that when the ideal or non-ideal behavior of an aggregation plays a game role, a perfect ε-Nash equilibrium is formed, and a coordinated electric vehicle charging strategy can have a significant effect in terms of energy cost saving and peak-to-average ratio reduction.