There is a need for a sophisticated approach to track athletic performance and to quantify factors affecting it in practical settings. PURPOSE: To demonstrate the utility of mixed linear modeling for monitoring athletic performance. METHODS: Elite swimmers (3 females and 6 males; age 21-26 y) performed 2-8 time-trials in training and 2-7 in competition in the 9 wk prior to and including Olympic-qualifying trials, all in the stroke and distance of the athlete`s main event. We included a double-blind, randomized, diet-controlled crossover intervention in which the swimmers consumed caffeine (5 mg/kg body mass) or placebo 75 min before two training trials ≤2 wk apart. The swimmers also knowingly consumed varying doses of caffeine in some other trials and in competitive swims. We used mixed linear modeling of log-transformed swim time to quantify performance in training vs competition, in morning vs evening swims, and with caffeine vs placebo. Predictor variables were coded as 0 or 1 to represent absence or presence of each condition and included as fixed effects. Date of each performance test was included as a continuous linear fixed effect and interacted with the random effect for athlete to represent individual differences in linear trends in performance. Outcomes were deemed unclear if the 90% confidence interval included substantial enhancement and impairment (0.3%). RESULTS: Performance times in the time-trials and competitions were highly reliable (typical errors both 0.8%). Performance time improved linearly by 0.8% (90% confidence interval 0.3% to 1.2%) per 4 wk of training, with individual differences (standard deviation) in the trend of 0.5% (0.4 to 1.2%) per 4 wk. The swimmers performed better in evenings vs mornings by 0.6% (0.1 to 1.1%) and in competition vs training by 1.4% (0.9 to 1.9%). A 100-mg dose of caffeine enhanced performance in time trials by 1.2% (-0.1 to 2.4%) and in competitions by 1.4% (0.3 to 2.5%); each additional 100 mg reduced the benefit slightly by an unclear 0.1% (-0.3 to 0.5%), and the placebo effect was also a slight but unclear impairment of 0.2% (-0.6 to 0.9%). CONCLUSION: Mixed linear modeling is a successful approach for quantifying small changes in performance in a squad of elite athletes whose performance is monitored regularly over a period of several months.
© Copyright 2009 14th annual Congress of the European College of Sport Science, Oslo/Norway, June 24-27, 2009, Book of Abstracts. Published by The Norwegian School of Sport Sciences. All rights reserved.