Author Identifier

Lynne Munro

Date of Award


Document Type



Edith Cowan University

Degree Name

Doctor of Philosophy


School of Medical and Health Sciences

First Supervisor

Chris R. Abbiss

Second Supervisor

G. Greg Haff


Sprint cycling requires the production of explosive muscle power outputs up to very high pedalling rates. The ability to assess muscular function through the course of the sprint would aid training practices for high-level performers. Inverse dynamics provides a non-invasive means of estimating the net muscle actions acting across any joint contributing to movement. However, analysis of joint kinetics requires motion-capture techniques that present some unique challenges for cycling. This thesis presents three studies investigating the application of a custom-designed force pedal system to examine the joint kinetics of elite trained track sprint cyclists. To provide the basis for selecting appropriate testing procedures, study one evaluated differences between two- and three- dimensional techniques while assessing joint kinetics of seated and standing sprint cycling at optimal cadence (the cadence where peak power is delivered). Study two examined the impact of cadence and seating position on joint kinetics, while determining testing reliability using the three-dimensional process. Coefficients of variation were established for between- and within- days repetitions of sprint performance at optimal cadence, and cadences 30% lower and 30% higher, in both seated and standing positions. Study three compared joint kinetics of sprint cycling performance with commonly-applied resistance-training exercises in an elite cycling cohort, in order to better understand training specificity. Joint-specific torque-angular velocity relationships were established from seated and standing sprinting at three cadences and the clean exercise at three loads, with other strength-based exercises examined at maximal load only.

Study one determined that flattened projections of the 3D motion into 2D resulted in significant differences in joint powers calculated in the sagittal-plane. When using 2D methods, knee joint power was significantly lower and hip transfer power significantly greater, while hip range of motion was lower and the angle where hip peak power occurred later in the crank cycle. These results indicate that 3D processes should be used where evaluation of absolute values are important, although 2D processes may still be acceptable where relative differences are being assessed. It was observed in Study two that, while crank and total muscle power upheld a quadratic power-cadence relationship, joint-specific powers were uniquely related to cadence and riding position. Crank and joint-specific optimal cadences for power production were distinctly different. The hip displayed a linear maximum power-cadence relationship in seated but quadratic in standing position, with the reverse observed at the knee. Ankle and hip transfer powers both linearly declined with cadence irrespective riding position. In such a case, joint-specific power contribution, hence distribution of muscular effort, cannot be directly inferred from power assessed at the crank. Reliability was highest for crank and total muscle power, particularly at the riders’ optimal cadence. Reliability of joint powers were somewhat lower and uniquely dependent on joint, joint action and trial condition. Results indicate that external power output at the crank is relatively stable across sprints, despite variation in the underlying muscular contributions. Results of study three showed equivalence in the torque-angular velocity relationships at the hip in sprint cycling and different phases of the clean. No such relationship was evident at the knee or ankle. In contrast to the negative linear relationships observed in all other conditions, ankle mechanics in sprinting showed a positive linear relationship highlighting a distinct functional role of this joint. Highest maximal torques at the hip and knee were observed during unilateral single rack pull and step-up exercises, respectively, supporting their efficacy for improving the maximum strength characteristics at these joints.

The results of this thesis indicate that joint kinetics are an effective means of assessing muscular performance in highly-trained track sprint cyclists and provide information on the underlying strategies that could not be assessed through conventional testing of power at the crank. The use of 3D processes is recommended where accuracy of assessment and absolute values are important. Flexibility of 2D processes may be advantageous in field-based settings and may be acceptable where only relative change is of interest. High reliability of 3D testing supports its use in monitoring of athletes, with the reliability data presented in this thesis providing an indication of the smallest meaningful changes in various trial conditions. Low coefficients of variation observed in crank and muscle power terms, despite greater variation in joint powers, suggest motor control strategies dynamically respond to task conditions while maintaining a consistent external power. Resistance exercises are seen to display jointspecific profiles that characterise relative hip- or knee- dominance. The comparison of these profiles with those of sprint cycling can help inform exercise selection for strength development of elite riders. The ability to monitor changes and target training intervention at joint level provides a unique approach to athlete development. Outcomes of this thesis support the practical application of joint kinetic assessment in aiding training practices to the highest levels of competition in track sprint cycling. Indeed, the equipment, methods and knowledge obtained from this research is currently applied in the preparation of Australia’s best sprint cyclists.