This study aimed to explore the experiential knowledge and preferred training approaches of elite swimming coaches in regards to general skill development and then looking specifically at the freestyle stroke. A qualitative thematic analysis approach was employed to identify, analyse and report themes within the content of the collected data. Twenty elite swimming coaches participated in semi-structured interviews. Several themes revealed that the most common training practices employed to improve skill learning included the use of task decomposition (part-task) techniques. The findings also indicated that swimming coaches believe practice should be specific / representative to the intended performance outcomes. It is believed that such viewpoints may have been influenced by coaches' interaction with skill acquisition consultants and may have also shaped some coaches use of variants of constraints manipulation in their practice design. While swimming coaches seem to mix both traditional and contemporary skill acquisition theories in their training prescriptions, the traditional approach is dominant as evidenced by coaches seeking to reinforce "perfect" swimming technique and mechanical consistency.Considering coaches' experiential knowledge and training prescriptions may benefit future research protocols and better facilitate the transfer of empirical findings to coaching practice.

The effectiveness of the swimming turn is highly influential to overall performance in competitive swimming. The push-off or wall contact, within the turn phase, is directly involved in determining the speed the swimmer leaves the wall. Therefore, it is paramount to develop reliable methods to measure the wall-contact-time during the turn phase for training and research purposes. The aim of this study was to determine the concurrent validity and reliability of the Pool Pad App to measure wall-contact-time during the freestyle and backstroke tumble turn. The wall-contact-times of nine elite and sub-elite participants were recorded during their regular training sessions. Concurrent validity statistics included the standardised typical error estimate, linear analysis and effect sizes while the intraclass correlating coefficient (ICC) was used for the reliability statistics. The standardised typical error estimate resulted in a moderate Cohen's d effect size with an R value of 0.80 and the ICC between the Pool Pad and 2D video footage was 0.89. Despite these measurement differences, the results from this concurrent validity and reliability analyses demonstrated that the Pool Pad is suitable for measuring wall-contact-time during the freestyle and backstroke tumble turn within a training environment.