Abstract

Track and field is a discipline that involves intense competition and hinges predominantly on statistical measures to evaluate performance and rank athletes. Despite the widespread application of rudimentary performance metrics to assess athletes, there has been scant utilization of advanced analytical techniques in the discipline. The current study endeavors to conceive and implement a suite of statistical and machine learning models that address key issues prevalent in track and field, especially concerning outdoor running events. The primary objective is to gain comprehensive insights into the discipline from an advanced analytics standpoint and augment ranking systems and race strategies. To ensure practicality and accessibility of the models for use at track meets, only readily available results information, such as athletes' prior result times, is used throughout the analyses. The model aims to mathematically cluster the athletes to assist a coach in determining which individuals to choose from a pool of athletes to comprise the fastest team and order, which will extend a guiding hand to the coach in the intricate task of selecting the crème de la crème from a pool of athletes, assembling the swiftest team, and meticulously arranging their sequence. In conclusion, this research work presents innovative implementations of advanced analytics in the field of track and field, offering valuable insights into the potential uses of statistical and machine learning models in enhancing performance and decision-making in the sport. The models developed herein have the potential to improve the precision and impartiality of ranking systems, optimize race strategies, and aid coaches in making well-informed choices regarding the selection and order of team members.