Why is it that some offenders get arrested quickly, while others manage to evade arrest much longer or are never arrested at all? What characterizes serial offenders who continue to escape arrest? To be able to answer these questions, arrested (identified) offenders must be compared with never arrested (unidentified) offenders. DNA data offer a unique opportunity to compare crime series of identified offenders with crime series of unidentified offenders. In this paper, data from the Dutch DNA database are used to study whether the geographical dispersion of the crime locations of serial offenders influences the probability of arrest. Results show that the probability of arrest decreases with increasing geographical dispersion, measured as the number of police regions in which the offender's crimes have been committed. © The Author(s) 2013.