We analyze delinquent networks of adolescents in the United States. We develop a dynamic network formation model showing who the key player is, i.e. the criminal who once removed generates the highest possible reduction in aggregate crime level. We then structurally estimate our model using data on criminal behaviors of adolescents in the United States (AddHealth data). Compared to other criminals, key players are more likely to be male, have less educated parents, are less attached to religion and feel socially more excluded. We also find that, even though some criminals are not very active in criminal activities, they can be key players because they have a crucial position in the network in terms of betweenness centrality.