All of the risk assessment instruments--Violence Risk Appraisal Guide [VRAG], Sex Offender Risk Appraisal Guide [SORAG], Rapid Risk Assessment for Sex Offense Recidivism [RRASOR], Static-99, Static-2002, and Minnesota Sex Offender Screening Tool-Revised [MnSOST-R]--predicted the reoffending (recidivism) outcomes for which they were designed. Although the prediction accuracies were significant, indexes of accuracy were generally lower than those reported by the developers of these instruments. This was the case even under conditions that had been shown to optimize predictive performance. For serious recidivism (other than sexual recidivism), the predictive accuracy of the Static-2002 and SORAG was significantly superior to that of the RRASOR; and the SORAG was significantly superior to the MnSOST-R. There were no significant differences among the instruments in their accuracy of predicting sexual recidivism. All data were coded from files at the Warkworth Sexual Behaviour Clinic (WSBC), the Correctional Service of Canada (CSC), and the Royal Canadian Mounted Police. The six actuarial risk assessment instruments were scored retrospectively from archived files. All of the assessment schemes were completed by using file data available prior to the offender's first release from custody for the index offense. Coders were unaware of recidivism outcomes for all cases, and recidivism was coded blind to instrument scores. 6 tables and 66 references