The purpose was to learn how people develop a pacing template during running and test the hypothesis that there is less variability in running velocity as subjects become more familiar with all-out time trials. 12 fit subjects (3f, 9m) with minimal experience in competitive running completed six 3km running trials. Subjects completed the trials in the fastest time possible, with incentives being offered to improve time. RPE was measured every 200 meters and velocity every 100m. Blood lactate was measured pre and post trial. HR was recorded throughout each trial. Time improved from trial to trial. Mean starting velocity increased from trial to trial but mean finishing velocity was the same. There was not a significant decrease in the coefficient of variation of velocity across trials. The slower starting velocities in the initial trials agree with the anticipatory regulation of exercise response. The subjects started out slower and as the trial neared the end they were able to speed up because they felt as though they would not harm themselves. In subsequent trials, they started faster, based on the experience in the preceding trial. As the starting velocities got faster, ending velocities remained essentially constant. We had hypothesized that with experience the subjects would develop a more even pacing pattern. However, the data did not support our hypothesis.