Purpose: The aim of the current study was to investigate pacing strategies and the distributionof physiological resources in best versus worst performances during a series of 4-minute, self-paced running time-trials (RunTTs). Methods: Five male and five female recreational runners(age 32 ± 7 years) completed a sub-maximal ramp test and five RunTTs on a motor-driventreadmill fitted with a speed-controlling laser system. The supramaximal V̇ O2 demand wasestimated by linear extrapolation from the sub-maximal relationship between V̇ O2 and speed,enabling computation of the accumulated oxygen deficit (AOD). Results: There were nosignificant differences between the five RunTTs for any of the performance, physiological orsubjective responses (P > 0.05). The trial-to-trial variability in pacing (i.e., separate quarters)was typically low, with an average within-athlete CV of 3.3%, being highest at the start andend of the 4 minutes. Total distance covered and distance covered over the first and last 2minutes for best and worst performances were 1137 ± 94 and 1090 ± 89 m (P < 0.001), 565 ±53 and 526 ± 40 m (P = 0.002), and 572 ± 47 and 565 ± 54 m (P = 0.346), respectively.Conclusions: Negative pacing strategies were evident during both the best and worstperformances of the RunTT. Best performances were characterised by more aggressive pacingover the first 2 minutes compared with worst performances. In addition, the relatively low trial-to-trial variability in running speed suggests that pacing strategies are similar during a seriesof 4-minute, self-paced running time-trials.