An understanding of individual criminal careers is prerequisite to evaluating the impact of incapacitation in terms of the numbers of crimes averted. Research now suggests that current age, age at first arrest, drug use, and current offense may be important indicators of habitual criminality. Rand Corporation studies have confirmed previously identified predictors such as the importance of age at first arrest, but have also discovered that persons who engage in all three of the serious offenses of robbery, assault, and drug dealing have the highest overall crime rates and may be prime candidates for selective incapacitation. Other researchers have found that the longest residual criminal career lengths occur between the ages 30 and 42, refuting the argument that offenders over 30 are not good candidates for incapacitation because they are close to terminating their criminal careers. Finally, analyses of criminal career patterns reveal widespread switching among general crime types. If these prediction results hold up under continuing technical review, the legal community will have to determine how to use such information responsibly in the criminal justice process. A major issue is the prospect of unjustly sentencing an offender to prison based on a false prediction of a high risk of future criminality. As more attention is given to identifying the serious offender, greater use may be required of juvenile record information. State governments should make their incarceration decisions based on projections of anticipated prison populations, taking into account the changing demogrphics associated with the postwar baby boom. As governors and legislatures consider new sentencing legislation, they should also estimate the impact of such laws on prison populations. The article includes 18 footnotes.