A comparison of methods for quantifying training load: relationships between modelled and actual training responses

Area of Science:

• Exercise Physiology
• Sports Science
• Biomathematics

Background:

• Quantifying training load, fitness, and fatigue is crucial for optimizing endurance athlete performance.
• Mathematical models offer a potential tool for individualized assessment of these physiological parameters.

Purpose of the Study:

• To evaluate the validity of different training load quantification methods (session-RPE, TRIMP, rTSS) when integrated into a mathematical model.
• To assess the model's ability to predict performance, fitness, and fatigue in trained runners.

Main Methods:

• Seven trained runners underwent 15 weeks of endurance training.
• Training load was measured using heart rate, running pace, and rating of perceived exertion.
• Data were used to calculate training dose via session-RPE, Banister's TRIMP, and running training stress score (rTSS).
• Performance, fitness (submaximal HR), and fatigue (HRV) were assessed weekly.
• A mathematical model was applied to individual training data.

Main Results:

• Training improved 1,500-m performance by 5.4%.
• Modelled performance showed moderate-to-strong correlations with actual performance across all training load methods (r = 0.60–0.70).
• Moderate correlations were found between modelled and actual fitness (submaximal HR) and moderate-to-large correlations for fatigue (HRV) using session-RPE and TRIMP.

Conclusions:

• The session-RPE, TRIMP, and rTSS methods are appropriate for quantifying endurance training dose.
• Submaximal heart rate and heart rate variability are valuable tools for monitoring fitness and fatigue, respectively, in endurance athletes.