Optimal cycling time trial position models

Abstract

Introduction: The aerodynamic drag of a cyclist in time trial (TT) position is strongly influenced by the torso angle (Underwood et al. 2011). To minimize drag, cyclists lower their torso angle. Along with the drag the cyclists’ peak power output decreases. There should be a trade-off between the loss in power output and drag as function of cycling velocity. This hypothesis is supported by the energy expenditure which is a function of the workload divided by the gross efficiency (GE). The workload to overcome drag decreases with smaller torso angles, while the GE decreases accordingly. Previous literature suggested that the aerodynamic losses outweigh the loss in power output (Lukes et al. 2005). However, these statements are only valid for elite TT cycling velocities, e.g. > 45 km/h. To our best knowledge, there is no published prediction at which speed the aerodynamic power loss starts to dominate. Therefore the aim of this study is to predict the optimal TT cycling position as a function of velocity to improve the performance of non-elite cyclists.

Methods: Two models were developed to determine the optimal torso angle of TT cyclists: a ‘Power Output Model’ and a ‘Metabolic Efficiency Model’. The Power Output Model predicts the optimal cycling position by maximizing the peak power output minus the power losses due to drag and roll resistance. The Metabolic Efficiency Model minimizes the required cycling energy. Model input parameters were experimentally collected of 19 trained competitive time trial cyclists (Fintelman et al. 2012). The main input variables were the power output, frontal area and GE of the cyclists in different torso angle positions (0,8,16 and 24°). The optimal cycling torso angle was predicted for speeds between 18-50km/h.

Results and discussion: For both models, the optimal torso angle is dependent on the cycling velocity, and the torso angle exhibited a sigmoid-like shape, with decreasing torso angles at increasing velocities. The Power Output Model curve was shifted to a higher velocity, which could be explained by the different approach of the models. The aerodynamic losses outweighed the power losses for velocities above 45km/h, which goes in line with the literature. For cycling velocities below 30km/h the power loss and gross efficiency due to position change were dominant. Furthermore, it is shown that a fully horizontal torso is not optimal.

Conclusion and recommendations: It is suggested that despite some limitations, the models give valuable information about the optimal TT cycling position at different speeds for non-elite cyclists. This study showed it is beneficial to ride in a more upright TT position when velocities are below 30km/h, while at speeds above 45km/h an almost flat back is optimal. Furthermore, for speeds between 32-40km/h in an endurance event it is advisable to lower the torso despite the fact that the power output in a more aerodynamic position is decreased. In contrast, in sprinting or in variable conditions (wind, undulating course, etc) at these speeds it is more beneficial to ride in a more upright TT position.
Our future research will attempt to measure the aerodynamic drag of all participants in a wind tunnel, to implement a more valid drag coefficient and thus improving the validity of the models. Furthermore, to increase the ecological validity the effect of crosswind will be implemented in future.

References:
Fintelman, Highton, Adams, Sterling, Hemida, Li (2012) ECSS annual conference.
Lukes R, Chin S, Haake S (2005). Sports Eng., 8(2): 59-74.
Underwood L, Schumacher J, Burette-Pommay J, Jermy M (2011). Sports Eng., 14(2): 147-154.