Current strain theories argue that delinquency results from the blockage of goalseeking behavior. Unable to achieve valued goals, individuals become frustrated and may turn to delinquency as a result. This paper points to another major source of frustration and delinquency, the blockage of pain-avoidance behavior. Adolescents are compelled to remain in certain environments, such as family and school. If these environments are painful or aversive, there is little that adolescents can do legally to escape. This blockage of pain-avoidance behavior is likely to be frustrating and may lead to illegal escape attempts or anger-based delinquency. This theory is tested using data from a national sample of adolescent boys. Data indicate that location in aversive school and family environments has a direct effect on delinquency and an indirect effect through anger. These effects hold even after social control and subcultural deviance variables are controlled. Given the weak support for traditional strain theories based on the blockage of goal-seeking behavior, these data suggest a new direction for the development of strain theory. Strain theory is based on the idea that delinquency results when individuals are unable to achieve their goals through legitimate channels. In such cases, individuals may turn to illegitimate channels of goal achievement or strike out at the source of their frustration in anger. This is an appealing idea and it is not surprising that strain theory has had a major impact on delinquency research and public policy (Liska,b). Recent research, however, has been critical of strain theory or, at best, has provided only mixed support for the theory. This has led a number of researchers to call for either the abandonment or revision of strain theory (Elliott et al.; Hirschi; Kornhauser). This paper reviews the criticisms of current strain theories, examines some recent efforts to revise these theories, and then presents a new revision of strain theory based on the idea that delinquency results from the blockage of pain-avoidance behavior. This new revision is tested using data from a national sample of adolescent boys.