The aim of this study was to investigate Martens, Burton, Vealey, Bump, and Smith’s (1990) contention that stress-management techniques should be matched to the symptoms manifested by performers. Subjects, 20 male collegiate field hockey players, responded to the Competitive State Anxiety Inventory-2 (CSAI-2) on four occasions prior to an important hockey match. Subjects were then placed in two intervention groups: applied relaxation (somatic anxiety; n = 6) and positive thought control (cognitive anxiety; n = 8). Six additional subjects formed the control group. Subjects completed a 12-week intervention in a field setting. Results suggested that reducing anxiety with a method directed at the performer’s dominant anxiety type is more efficacious. A secondary aim was to further investigate the anxiety-performance relationship using an intraindividual performance measure. Somatic anxiety was found to account for 22% of the variance in field hockey performance. Polynomial trend analyses failed to produce significant relationships between the CSAI-2 subscales and performance.