Cognitive therapy for depression is based on an assumption that depressed individuals have inaccurate, negative biases. Whether this assumption is accurate remains unresolved. Thus, this study sought to examine the relationship between depressive symptoms and bias in three sets of predictions (i.e., predictions of future life events, how one would be rated by a significant other, and performance on a vocabulary test). Following study announcements made to a subset of people pre-screened for depressive symptoms, 85 participants with widely varying depressive symptoms (17 of whom met diagnostic criteria for depression) made predictions on three judgment tasks and the outcomes for these tasks were assessed. Optimistic/pessimistic biases were related to depressive symptoms for each of the three tasks. Participants with high levels of depressive symptoms and depressed participants exhibited substantial pessimistic bias. Those high in depressive symptoms exhibited significant pessimistic bias on all three tasks. Participants meeting diagnostic criteria for depression exhibited pessimistic bias on two of three tasks. There was no evidence that depressive symptoms were associated with greater accuracy in judgments. Results are largely consistent with cognitive models of depression which postulate that depression is associated with pessimistic biases.