The fortnightly BRAG meeting will be held this Thursday the 2nd of May at 1 pm via Zoom/GP-Y801. This week we will have presentations by Harry Estreich and Amir Rezaeian.
Zoom Link: https://qut.zoom.us/j/81862982586?pwd=bmI2SHNYdDVSNklIUWpFS01RdnhqUT09
Meeting ID: 818 6298 2586
Passcode: 104645
Harry’s Talk
An analysis of pacing profiles in sprint kayak racing using functional principal components and Hidden Markov Models
Predictive models of individual athlete pacing in competitive kayak races and their change over a career can be used by coaches and sports scientists to better understand athlete progression and optimise strategies for peak performance. This study analysed sprint kayak pacing profiles by using functional principal component analysis of normalised velocity data for 500m and 1000m sprint kayak to quantify pacing. These principal components were then associated with unique pacing characteristics. We then applied a Hidden Markov model to categorise each profile over an athlete’s career, using the PC scores, into different types of race profiles. It was identified that an athlete’s pacing profile can and does change throughout their career as an athlete matures.
Amir’s Talk
This paper expands on previous research by incorporating an algorithm to refine finishing time corrections for sprint kayak and canoe race times, initially adjusted using the Bayesian Additive Regression Trees (BART) model for environmental conditions. In this work, we are exploring ways to make better use of the provided corrected times we currently have, which is based on the in-house algorithm conducted by Paddle Australia. Our algorithm explores ways of using statistical methods in using these corrected times as a new basis for incorporating external information in BART modelling.