This study examined the predictive validity of the ASAM Patient Placement Criteria for matching alcoholism patients to recommended levels of care. A cohort of 248 patients newly admitted to inpatient rehabilitation, intensive outpatient, or regular outpatient care was evaluated using both a computerized algorithm and a clinical evaluation protocol to determine whether they were naturalistically matched or mismatched to care. Outcomes were assessed three months after intake. One common type of undertreatment (ie, receiving regular outpatient care when intensive outpatient care was recommended) predicted poorer drinking out comes. As compared with matched treatment, independent of actual level of care received. Overtreatment did not improve outcomes. There also was a trend for better outcomes with residential vs. intensive outpatient treatment, independent of matching. Results were robust for both methods of assessment. Corroboration by more research is needed, but the ASAM Criteria show promise for reducing both detrimental undertreatment and cost-inefficient overtreatment.