Tuesday, Aug 4: 10:30 AM - 12:20 PM
1106
Invited Paper Session
Thomas M. Menino Convention & Exhibition Center
Room: CC-253A
Applied
Yes
Main Sponsor
Council of Chapters
Co Sponsors
Caucus for Women in Statistics
Section on Statistics in Sports
Presentations
During a marathon, the expected finish time of runners is commonly estimated by extrapolating their average pace at that point, assuming it will hold constant for the rest of the race. Two problems arise when predicting finish times this way: the estimates do not consider in-race context that can determine if a runner is likely to finish faster or slower than expected, and the prediction is a simple point estimate with no information about uncertainty. To address these issues, we implement a hierarchical Bayesian linear regression model that incorporates information from all splits in a race and allows quantification of uncertainty around the predicted finish times. Multiple models under this Bayesian framework are compared to the traditional extrapolation method using data from the Boston, New York, and Chicago Marathons over four years (2021-2024), and we find a marked improvement in predictive accuracy. We also develop an app that allows runners to visualize their estimated finish time distribution in real time.
Keywords
Marathon
Bayesian linear regression
Uncertainty quantification
Sports analytics
Super shoes are lightweight running shoes designed with a carbon fiber plate for improving running economy. Since their widespread adoption by elite marathon runners in 2017, the men's and women's world records have each been broken three times. In order to quantify the impact of super shoes on professional marathon times, we apply a mixed effects model to 2011-2023 Chicago marathon data. Using our model, we compare the probability of top elite runners pre- and post-super shoes breaking the elusive 2:00 barrier (for males) and 2:10 barrier (for females).
Keywords
marathon
super shoes
mixed effects
In October 2024, the Kenyan runner Ruth Chepngetich ran 2 hours, 9 minutes, 56 seconds in the Chicago Marathon, a new world record for the women's marathon by nearly two minutes and the first woman to go under 2:10, a time that would have been considered unbeatable a few years ago. The record was swiftly ratified by World Athletics, but there were also doubts about drug use, which have since been confirmed when she was banned for three years after admitting to anti-doping rule violations. But how improbable was the record, and what should be our projections for the women's marathon in the next few years? This paper uses methods from extreme value theory to analyze women's marathon performances in the Chicago Marathon and other leading marathons, showing that trends over the last several years made such a performance at least plausible. Possible reasons for the trends, including the advent of "super shoes", will be explored.
Keywords
Extreme value theory
Chicago marathon
Super-shoes
The success of masters programs in running and swimming competitions has made available a vast amount of data. Using these data for athletes aged 35 to 85, we model the percentage increase in event time to complete several events (including both sprints and long distance) in both running and swimming. We use a stacked model that includes polynomial and neural network models as well as smoothing splines. We bootstrap the procedure to obtain confidence intervals which can help answer fundamental questions about the nature of the age decline. We then turned our attention to the Dipsea, the oldest continuously held U.S. footrace (since 1905), handicapped by age since 1965 and by age and sex since women were officially admitted in 1971. Seeing that no one between the ages of 8 and 40 had won the race, we assumed that their handicapping system was flawed. What we discovered was surprising.
Keywords
Stacked models
Bootstrap
Cohort effects
Aging
Neural networks
Sports analytics