BACKGROUND: Test-retest reliability of performing 1–2 bouts of a 30 s Wingate (30WanT) has been shown to improve by averaging 2 bouts. OBJECTIVE: To investigate whether adding a 3rd bout enhanced or decreased reliability. METHODS: On three occasions, 13 triathletes performed three 30WanT's on a cycle ergometer. Peak power (PP), mean power (MP) and fatigue index (FI) reliability across bouts (Intra-session) and days (Inter-session) were tested. Mean and % change between bouts, typical errors of measurement (TEM) and intra-class correlation co-efficient (ICC) were compared between consecutive pairs of bouts and 3 bouts. Single (1 & 2), averaged 2 (Σ 1+2) and 3 bout (Σ 1+2+3) between days using TEMs and ICCs were also compared. RESULTS: Intra-session ICCs for 2 and 3 bouts for PP, MP and FI showed strong reliability (> 0.7). Three bouts had lower TEM for MP, but 2 bouts had less TEM for PP and FI. Inter-session reliability ICCs were > 0.90 for PP, 0.94 for MP and 0.82 for FI for the 1, 2 and 3 bouts with no clear difference between. CONCLUSIONS: Performing 1, 2 and 3 bouts showed high to very high reliability, with a 3rd bout not clearly reducing TEM, and in some cases introducung more variability.