We propose that negative goal framing (i.e., defining a goal as a negative state to be avoided) can adversely affect performance. Study 1 (N = 133) revealed that negative goal framing predicted poorer future performance independent of goal level, expectancy, and earlier performance. Study 2 (N = 188) examined the relation between goal framing and performance at 2 times in the academic year, and with respect to individual differences in defensive pessimism. As predicted, the negative goal-framing/poorer-performance link was greater on a later exam (after receiving feedback) than an earlier one, and was greater for nondefensive pessimists than for defensive pessimists. The findings implicate self-regulatory processes in understanding how goal framing affects performance.