The 81 male offenders were followed for an average of 29.7 years. Their average PCL-R scores were 22.1 with a standard deviation of 6.7. Psychopaths were defined as the 10 offenders with total scores of 30 or more; nonpsychopaths were the 51 offenders with scores of 17 or below. The analysis revealed that recidivism was 57 percent for the total sample, 80 percent for the psychopaths, 8 percent for those in the middle groups, and zero for the nonpsychopaths. Choosing higher and lower cutoff scores reduced the false positive and false negatives to zero, while affecting the numbers of false positives and false negatives released or remaining in prison. Results indicated that the PCL-R can make the system less arbitrary and could be used to direct rather than dictate intervention and risk management strategies. Findings also indicated the need to place risk estimates for particular offenders in context and to develop individualized strategies to manage the risk of recidivism. Figures and footnotes