The aim of this study is to develop a method for assessing movement variability of circus acrobats. An analysis of the repeatability of force signals is used to quantify variability. Six students from the National Circus School of Montréal performed 5-10 trials of an acrobatic movement called dislock in aerial circus straps while tension force was measured at the hanging point of the aerial apparatus. The repeatability of force signals was calculated with three statistical methods: time-averaged standard deviation, intraclass correlation and root mean square error. These methods were compared with the ratings of a circus coach who ranked each acrobat's trial with regard to the movement variability. The standard deviation and the intraclass correlation methods are commonly used to quantify the agreement between measurements in biomechanics, while the root mean square error method is regularly employed to quantify the agreement between measurements and a model. All participants performed the movement with little variability (intraclass correlation > 0.8). The results of the three methods were in good agreement with the coach's assessment. The root mean square error method, in particular, showed perfect agreement and is therefore considered the best measure of repeatability. In the future, the proposed method could be used by coaches or artists training alone, allowing a new form of feedback.
© Copyright 2018 Proceedings of the Institution of Mechanical Engineers, Part P: Journal of Sports Engineering and Technology. SAGE Publications. Published by SAGE Publications. All rights reserved.