The present study examined discriminative validities of the Behavioral Inhibition System/Behavioral Approach System (BIS/BAS) scales in differentiating bipolar spectrum disorders (BSDs) from other disorders. Participants were youth recruited from a combination of community mental health center and university medical facility. Receiver Operating Characteristic (ROC) analyses tested the BIS/BAS scales in distinguishing between BSD-positive and BSD-negative youth. We calculated diagnostic likelihood ratios in keeping with recommendations from evidence-based medicine. Binary logistic regressions tested for incremental value in combining subscales. BAS subscales discriminated between participants with BSD-positive and BSD-negative diagnoses, with areas under the curve ranging from .54 to .64. The BIS/BAS scales achieved statistical significance in identifying cases with BSDs, but effect sizes for discriminative comparisons were too small to be clinically useful. Upgrading clinical training to: (a) include prevalence of BSDs and (b) teaching clinicians more evidence based assessment strategies is important to improve assessment and diagnosis of BSDs.