The aims of the study were to determine if medium-chain triacylglycerol (MCT), ingested in combination with carbohydrate (CHO), would alter substrate metabolism and improve simulated competitive ultra-endurance cycling performance. Eight endurance-trained cyclists took part in this randomized, single-blind crossover study. On two separate occasions, subjects cycled for 270 min at 50% of peak power output, interspersed with four 75 kJ sprints at 60 min intervals, followed immediately by a 200 kJ time-trial. One hour prior to the exercise trials, subjects ingested either 75 g of CHO or 32 g of MCT, and then ingested 200 mL of a 10% CHO (wt/vol) solution or a 4.3% MCT + 10% CHO (wt/vol) solution every 20 min during the CHO and MCT trials, respectively. During the constant-load phases of the 270 min exercise trial, VO2, RER, and heart rate were measured at 30 min intervals and gastrointestinal (GI) symptoms were recorded. There was no difference in VO2 or RER between the MCT and CHO trials (P = 0.40). Hourly sprint (P = 0.03 for trial x time interaction) and time-trial times (14:30 ± 0.58 vs. 12:36 ± 1:6, respectively, P < 0.001) were slower in the MCT than the CHO trial. Half the subjects experienced GI symptoms with MCT ingestion. In conclusion, MCTs ingested prior to exercise and co-ingested with CHO during exercise did not alter substrate metabolism and significantly compromised sprint performance during prolonged ultra-endurance cycling exercise.