The New York Racing Association and the New York Thoroughbred Riders Association announced on December 6 that a team led by Brendan Kumagai has been selected as the winner of the inaugural competition of Big Data Horse Racing. Kumagai, along with Gurashish Bagga, Kimberly Kroetch, Tyrel Stokes and Liam Welsh, won the $20,000 first prize with the submission, “Bayesian velocity model for horse racing simulation.”
Kumagai’s team created a dynamic model that focused on horse forward and lateral speeds, and examined the results of maintaining momentum and velocity during races. The team also studied jockey performance and how that would affect the horse’s running style. Another conclusion in the study is that with additional biometric data, it is possible to calculate the probability of injury and the welfare of the horse.
“We are honored to be named the winners of the inaugural Big Data Derby,” said Kumagai. “Our team mainly works in the field of financial analysis and hockey statistics, so working with horse racing data is a unique challenge that allows us to apply our skills to a relatively new and unexplored field.”
The Big Data Race was launched with the goal of analyzing the huge amount of data available to racing organizations and to understand how the results of those studies could impact racing methodologies. and traditional training. The contest is sponsored by the NYRA and NYTHA in partnership with the Kentucky Thoroughbred Association, Equibase, Jockey Club, Breeders’ Cup, and the Thoroughbred Breeders and Owners Association.
A total of 106 entries were received for the first Big Data Race. Nine thousand forty-nine potential competitors accessed four contest data files during the competition. A diverse and extensive range of entries provided models that shed light on injury prevention, equestrian decision-making metrics, race tactics, track trends, etc.
The open handbook containing user-generated content and data can be viewed at: https://www.kaggle.com/competitions/big-data-derby-2022/code.
NYTHA President Joe Appelbaum said: “Our main goal with this competition is to usher in an age of technological innovation, analyzing horse racing data to suggest improvements in important topics such as efficiency. horse productivity and welfare”. “This enthusiastic response from this community shows that our traditionally tied sport can benefit by applying the knowledge gained from machine learning.”
Kumagai, a data science intern at Zelus Analytics, was formerly part of the team that won the 2022 Big Data Bowl hosted by the National Football League.
The Big Data Race awarded a total prize pool of $50,000 with $20,000 awarded to the winner and $10,000 each for the next three places. The competition is hosted on Kaggle, a global data science platform with over 500,000 active users, where participants compete using machine learning to solve trivial to extreme problems. complex period.
The runner-up includes Kyle King’s “Track Bias” submission, which explains how the tracking deviation index is calculated. Timothy Leung and Philip Leung put together “Advanced Horse Racing Tactics Using Coordinate Data”, examining the impact of outline percentages, path efficiency, race strategy, and speed fluctuations. Artem Volgin and Ekaterina Melinova submitted “Winning Strategies: What Works Better,” a study on horse racing tactics and race flow.
“Data science can have a profound impact on several different aspects of horse racing,” said Joe Longo, general manager of content management services at NYRA. “It could be applied to improve our understanding of horse health and safety or to everyday training methods by providing a new set of tools for both racers to understand. best known. The NYRA is committed to pursuing this goal in the future.”
The contest was judged by data analyst Rob Bingel, Rhodes University professor of economics, equestrian enthusiast and Thoroughbred owner Marshall Gramm, and TimeformUS’s Craig Milkowski.
Bingel, a financial planner and a financial planner, said: “What becomes clear and obvious when looking at all the presentations is the passion for quantitative analysis and the Curiosity is fueled by the intricacies of the data-rich world of horse racing.” Race data analyst. “This competition serves as a valuable introduction by the data scientist community to NYRA/NYTHA and holds promise for future collaboration. Solutions were proposed by the participants, especially especially the winners, which naturally leads to thoughts about important future questions that they can potentially answer.”
This press release has been edited by BloodHorse Staff for content and style.
