Capturing the Complex Relationship Between Internal and External Training Load: A Data-Driven Approach

Area of Science:

• Sports Science
• Biomechanics
• Exercise Physiology

Background:

• Training load is commonly defined by internal and external metrics.
• Understanding the interplay between these loads is crucial for optimizing athletic performance.
• Continuous kernel-density estimation (KDE) offers a novel approach to visualize this relationship.

Purpose of the Study:

• To assess the efficacy of bivariate KDE plots in describing the internal-external load coupling in speed skating.
• To compare the differentiating capabilities of KDE plots against training impulse scores and intensity distribution zones.
• To evaluate KDE's ability to distinguish between various training session types.

Main Methods:

• Collected on-ice velocity and heart rate data from 18 (sub)elite speed skaters over two seasons.
• Categorized training sessions into endurance, interval, tempo, and sprint types based on coaching schemes.
• Employed Kruskal-Wallis and Kolmogorov-Smirnov tests to analyze differences in training impulse and KDE scores, respectively.

Main Results:

• Training impulse scores showed minimal differences across training types, except for extensive endurance sessions.
• Bivariate KDE plots of heart rate and velocity revealed significant differences (P < .001) between all training session types.
• 2D KDE plots provided superior insights into load coupling compared to traditional training zone plots.

Conclusions:

• Two-dimensional KDE plots serve as a valuable tool for coaches to visualize and understand the nuances of internal-external load coupling.
• This visualization aids in designing more effective training programs tailored to specific adaptations.
• KDE enhances the ability to differentiate training sessions based on load characteristics.