Optimising cycle frequency: the effects of imposed cycle frequency training on the coordination and performance of skilled age-group swimmers

Abstract

PURPOSE: Underwater undulatory swimming (UUS) is a fundamental skill incorporated during the starts and turns of three of the four competitive swimming strokes. Significant competitive advantage can be gained if UUS performance is optimised. The cycle frequency adopted during UUS in both animal and human swimmers have been extensively studied and it has been shown to have a strong relationship with the UUS velocity (U) achieved. The purpose of this thesis was to investigate the changes in performance and coordination in UUS which occur as a consequence of training at an imposed cycle frequency (identical to preferred) in skilled age -group swimmers (Study 3). To achieve the stated purpose, the reliability (systematic bias, within -subject variation and test -retest reliability) of the kinematic variables commonly used to describe and analyse UUS were established (Study 1). Once reliability was determined, the key kinematic performance and coordination variables in relation to the production of maximum U were identified (Study 2) to enable the key measures of performance and coordination to be monitored in response to a training perturbation (imposition of a cycle frequency) in the final study. METHODS: Measures of systematic bias, within- subject (WS) variation and inter -class correlation (ICC) of nineteen kinematic variables were determined over four sessions. This was undertaken to establish the requirement of any familiarisation training, number of cycles of data required to provide an accurate representation of each variable when reporting a mean value, and the related variability associated when reporting mean values based on a set number of data cycles (Study 1). Backward elimination ANCOVA statistical models with participant as a fixed -factor were employed to establish which of the performance and coordination variables were best in explaining the variance of cycle frequency, cycle length (CL) and ultimately U (Study 2). In the final study (Study 3) the performance and coordination variables identified from study 2 were analysed in sixteen skilled age -group swimmers which participated in a randomised controlled study. An experimental group of eight participants completed a four week imposed frequency (matched to their own preferred frequency) training programme, while a control group of eight participants completed a four week programme training at a self selected preferred cycle frequency. The UUS kinematics for both preferred cycle frequency UUS and imposed cycle frequency UUS were measured at weekly intervals throughout the training period. An additional retest (RT) was conducted 2 weeks after the cessation of the training period. RESULTS: Systematic bias was identified between the 1st and the remaining 3 testing sessions for cycle frequency, CL and U. The minimum number of data cycles required to achieve an acceptable measure of retest reliability (ICC >0.85) across all kinematic variables was 6 cycles. At 6 cycles WS variation ranged from 0.86 to 8.92 %CV. A total of 10 kinematic variables were identified as key to explaining the variance in cycle frequency and CL. A final parsimonious ANCOVA model revealed that 2 variables (maximum knee angle velocity and wave velocity between knee and ankle) explained a large proportion (Adj. r² = 0.944) of the variance in maximal U. However, when the participant was removed as a fixed factor the explained variance reduced (Adj. r²= 0.535). No significant difference in maximal U was found over the training or RT period. No variables were found to differ significantly by Session x Frequency Tested x Training Group (p <0.01). However, several discrete kinematic variables and measures of coordination showed statistically significant changes, either between Frequency Tested or across testing sessions. Discussion: After determining the systematic bias and establishing the requirement for a familiarization session, 6 cycles of data were found to be sufficient to provide high levels of reliability for each of the UUS kinematic variables. The identified key determinants of the variance in cycle frequency, CL and maximal U, revealed that the successful transmission of the propulsive waveform along the caudal aspects of the swimming body (specifically the kinematics/coordination at or around the knee) and the control of the shedding of the vortices and simultaneous recapture /reuse of previous shed rotational energy are key discriminating factors between the faster and slower UUS in skilled age-group swimmers. The 4-week training period did not result in changes in maximal U for either of the training groups. However, there were significant differences in the magnitude and process of adaptation between preferred and imposed frequency training groups' kinematics and measures of coordination over the training and testing period. The importance of each individual's own solution to the maximal UUS problem was highlighted, with coordination constrained by an individual's own idiosyncratic constraints. Further research is required to establish the efficacy of the imposition of a cycle frequency identical to an individual's own preferred frequency as an appropriate training modality for maximal U. In conclusion, the present research provides valuable insight into the effects of the simple act of cycle frequency imposition, providing a baseline for future cycle frequency interventions which take place at higher/lower cycle frequency or over longer training periods.