Abstract: As figure skating continues to evolve into an increasingly complex and competitive sport, athletes and coaches face the challenging task of designing optimal, high-scoring program layouts that will offer a competitive edge. This complexity is driven by the need to balance technical difficulty with artistic expression, all while adhering to strict judging criteria. To demystify this process and aid in strategic planning, we embarked on an analysis of figure skating performances and elements, including jumps, spins, and step sequences. Our analysis involves a dataset comprising over 12,000 elements from various elite international competitions, collected from official scoresheets. Employing regression and Bayesian methods to evaluate jump success probabilities, as well as mixed-effects regression models to predict score outcomes, we evaluated variables encompassing jump types, sequences, scores, and performance contexts. We were able to identify important predictors of jump success, as well as predict scores to develop a model capable of determining the most advantageous jump layouts and element sequences, providing actionable insights for skaters to plan their programs.
Nathan Chen is a senior at Yale University, majoring in Statistics and Data Science. He is the 2022 Olympic champion, 2018 Olympic bronze medalist, three-time World champion, and six-time U.S. national champion in men’s figure skating. He has been recognized on the Time100 and in the Forbes 30 Under 30 list for his skating achievements. At Yale, he works at the Cardiovascular Research Center, analyzing genomic data to better understand the impact of variants of known and uncertain significance on cardiovascular outcomes. In his free time, he enjoys spending time with family and friends, exploring new food and drink spots, and relaxing anywhere with a nice view.
Abstract: The integration of statistics, engineering, and biomechanics has fostered substantial contributions and advancements in sports science. The strategic incorporation of these disciplines has yielded novel data collection designs, launched new athletic performance metrics, and identified new methods for assessing and visualizing mechanistic changes in motor control during sport specific tasks. This talk will explore how this interdisciplinary approach has contributed to the development of innovative treatment protocols, injury risk assessment practices, and training programs to enhance and improve sports performance. The state-of-the-art technology used to conduct sports-based studies will be highlighted to further illustrate how data analytics and musculoskeletal modeling can help drive player recovery and on-field performance.
Kristin Morgan is an Assistant Professor of Biomedical Engineering at the University of Connecticut. Kristin’s work focuses on implementing innovative gait protocols and musculoskeletal modeling to accelerate individuals’ rehabilitation progression and improve their long-term joint health. Notably, she has utilized statistical techniques to establish universal ranges of healthy dynamics to help characterize the restoration of healthy biomechanics. Her work has been published in high-impact journals and has been supported by the Office of Naval Research, National Science Foundation, General Dynamics Electric Boat, and the National Institutes of Health.
Abstract: Nearly 50 years after Bill James published his first Baseball Abstract, statistics has penetrated every corner of America's favorite pastime, impacting everything from media coverage to team roster construction through data-driven decision-making. Join me as we dive into the captivating history and evolution of Baseball Analytics, the technological changes that took us from box scores and radar guns to ball tracking and player motion capture. We will explore the current state-of-the-art, both in the public sphere and behind the scenes. Drawing inspiration from Keith Woolner's exploration of baseball's open problems in 2000, we'll also discuss some of the most intriguing challenges and unanswered questions facing the industry in the years ahead.
Esteban Navarro Garaiz is a Technical Product Manager for the baseball team at Zelus Analytics, a role in which he collaborates closely with a team of 20 data scientists and engineers overseeing the development and implementation of the team's roadmap, supporting client integration, and mentoring junior team members. Before joining Zelus Analytics, Esteban spent two years as a Quantitative Analyst with the Los Angeles Dodgers, winning the World Series in 2020. He graduated with a Master’s degree in Data Science from New York University, where he was a DeepMind fellow and a Fulbright-García Robles Scholarship recipient.
Seam Ahmed is Director, Research and Development at Pittsburgh Pirates.
Luke Benz is a PhD student in Biostatistics at Harvard School of Public Health.
Sean Fischer is the manager of the Cincinnati Reds baseball analytics department. In his time with the club, he has contributed data science solutions for advance scouting, pro player evaluation, and amateur and international scouting. Sean holds a PhD from the Annenberg School for Communication at the University of Pennsylvania.
Paul Sabin is a Lecturer in Statistics and Data Science at the Wharton School.
Emily Wright is the Data Scientist for Volleyball Canada’s Beach National Teams. Her primary focus involves employing data analytics and modeling techniques to gain deeper insights into the dynamics of beach volleyball, aiming to elevate the sport through enhanced performance and strategy. Emily is also a part of the Canadian Olympic Committee Emerging Leaders Program, designed to develop talent in the Canadian sport system. She holds a master’s degree in Statistics and Actuarial Mathematics from Concordia University and a bachelor’s degree in Mathematics from Mount Saint Vincent University. Outside of volleyball, she enjoys adventures with her dog and talking about Atlantic Canada with anyone who will listen.