Aggregation shows that virtue-relevant behavior is indeed highly predictable, and that individual differences in global virtues do indeed exist. Aggregation is a key response to the situationist argument against the existence of broad virtues. However, a concern with aggregation is that, because it is an average, the specifics of what are included in that average matter. In particular, if heinous actions could be included in the average, then aggregates cannot provide enough confidence that the holders of high aggregates have not conducted heinous actions and thus cannot provide enough confidence that such people qualify as virtuous. Doris has challenged aggregation with this concern, and no one has responded substantively to this challenge. If Doris’ challenge is in fact correct, then the situationist argument against the existence of broad virtues stands. In the article, we present a full response to this concern. We argue that aggregation does not in fact allow heinous exceptions, because aggregates do indeed predict extreme single behaviors very well. In fact, aggregates do allow confidence that holders of high aggregates do not commit heinous actions. Thus, Doris’ rejection of the aggregation solution does not defeat aggregation, aggregation continues to stand in the defense of global virtues, and the situationist argument does not threaten the existence and predictive power of global virtues. Models of traits that rely on aggregates, such as Whole Trait Theory, may provide useful post-situationist models of virtues.