The ergogenic effect of caffeine on endurance exercise is commonly accepted. We aimed to elucidate realistically the effect of caffeine on triathlon event performance using a field study design, while allowing investigation into potential mechanisms at play. A double-blind, randomized, crossover field trial was conducted. Twenty-six triathletes (14 males and 12 females; mean ± SD: age = 37.8 ± 10.6 years, habitual caffeine intake = 413 ± 505 mg/day, percentage body fat = 14.5 ± 7.2%, and training/week = 12.8 ± 4.5 hr) participated in this study. Microencapsulated caffeine (6 mg/kg body weight) was supplemented 60 min pretrial. Performance data included time to completion, rating of perceived exertion, and profile of mood states. Blood samples taken before, during, and postrace were analyzed for cortisol, testosterone, and full blood count. Capillary blood lactate concentrations were assessed prerace, during transitions, and 3, 6, 9, 12, and 15 min after triathlons. Caffeine supplementation resulted in a 3.7% reduction in swim time (33.5 ± 7.0 vs. 34.8 ± 8.1 min, p < .05) and a 1.3% reduction in time to completion (149.6 ± 19.8 vs. 151.5 ± 18.6 min, p < .05) for the whole group. Gender differences and individual responses are also presented. Caffeine did not alter the rating of perceived exertion significantly, but better performance after caffeine supplementation suggests a central effect resulting in greater overall exercise intensity at the same rating of perceived exertion. Caffeine supplementation was associated with higher postexercise cortisol levels (665 ± 200 vs. 543 ± 169 nmol/L, p < .0001) and facilitated greater peak blood lactate accumulation (analysis of variance main effect, p < .05). We recommend that triathlon athletes with relatively low habitual caffeine intake may ingest 6 mg/kg body weight caffeine, 45–60 min before the start of Olympic-distance triathlon to improve their performance.