Degree Name

PhD (Doctor of Philosophy)


Sport Physiology and Performance

Date of Award


Committee Chair or Co-Chairs

Michael H. Stone

Committee Members

Michael Ramsey, Kevin Carroll, Margaret E. Stone


Performance tests are used to gauge swimmer fitness and guide training prescription. While some traditional protocols, such as, best average swimming (Bavg) lack scientific support, the three-minute all-out test (3MT) is validated to measure critical speed (CS) and distance capacity above CS (D’) from a single maximal swimming bout. In collegiate swimming, a 3MT could be convenient and time efficient. Yet, issues arise with calculating 3MT parameters. Moreover, anthropometry, resistance training, and stroke technique may influence swimming performance and physiological thresholds including 3MT parameters. Hence, this dissertation focused on evaluating the 3MT parameter calculation methods using the interval and lap split methods. Then, we assessed the influence of body anthropometrics, resistance training, and stroke techniques on 3MT parameters throughout a season. Bland-Altman plots showed that CS did not present remarkable behavior while D’ had systematic bias. Furthermore, there were no statistical differences between CS calculation methods (p = 0.83). However, D’ had moderate effect differences (p = 0.01, d = -0.70). Bavg showed very large correlations with maximum sprint speed (MSS) (r = 0.78) and CS (r = 0.81), but improvements after 6-weeks of concurrent training (p < 0.001, d = -0.85) seemed mainly driven by CS (p < 0.001, d = -1.68). Significant moderate to near perfect correlations were found between anthropometrics, 3MT parameters, and medicine ball pulldown throw (MBT) performance during pre- and post-season. Body composition changes may have influenced MBT performance changes (r = 0.46 – 0.55). CS increased at the expense of D’ while both reached stabilization point with no changes in MSS. There were moderate to large differences in 3MT parameters for the backstroke, but only MSS and CS differed in breaststroke, with no D’ differences in either group. Large to near perfect correlations were found between freestyle and secondary stroke 3MT parameter counterparts (r = 0.62 – 0.93), except for D’ in backstroke. Actual and predicted CS in both backstroke (p = 0.27) and breaststroke (p = 0.97) did not significantly differ. The lap split method during a 3MT seems like a practical and useful protocol to monitor CS and D’ in collegiate swimmers.

Document Type

Dissertation - embargo


Copyright by Luis Rodriguez Castellano.

Available for download on Saturday, June 15, 2024