The homicide rate in Chicago nearly tripled between 1965 and 1992 and subsequently declined by more than 50 percent through 2005. However, is this trend representative of all areas in the city? Drawing on the social disorganization and concentrated disadvantage perspectives, this article uses semiparametric group-based trajectory modeling to examine homicide trajectories in Chicago neighborhoods from 1965 to 1995. Significant variability is found in homicide trajectories across neighborhoods. Multivariate results show that disadvantage increases the likelihood of having an increasing or persistently high homicide trajectory. Social disorganization and family disruption are also predictive of variation in homicide trajectories but only in communities with already low levels of homicide. Other theoretically relevant predictors are evaluated, and suggestions for theoretical refinement and future research are discussed. Tables, figures, appendixes, and references (Published Abstract)