Purpose: To study the pacing behavior and performance of novice youth exercisers in a controlled laboratory setting. Method: Ten healthy participants (seven male, three female, 15.8 ± 1.0 years) completed four, 2-km trials on a Velotron cycling ergometer. Visit 1 was a familiarization trial. Visits 2 to 4 involved the following conditions, in randomized order: no opponent (NO), a virtual opponent (starting slow and finishing fast) (OP-SLOWFAST), and a virtual opponent (starting fast and finishing slow) (OP-FASTSLOW). Repeated measurement ANOVAs (p < .05) were used to examine differences in both pacing behavior and also performance related to power output, finishing- and split times, and RPE between the four successive visits and the three conditions. Expected performance outcome was measured using a questionnaire. Results: Power output increased (F3,27 = 5.651, p = .004, η2p = .386) and finishing time decreased (F3,27 = 9.972, p < .001, η2p = .526) between visit 1 and visits 2, 3 and 4. In comparison of the first and second visit, the difference between expected finish time and actual finishing time decreased by 66.2%, regardless of condition. The only significant difference observed in RPE score was reported at the 500 m point, where RPE was higher during visit 1 compared to visits 3 and 4, and during visit 2 compared to visit 4 (p < .05). No differences in pacing behavior, performance, or RPE were found between conditions (p > .05). Conclusion: Performance was improved by an increase in experience after one visit, parallel with the ability to anticipate future workload.