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

Abstract

These findings showed that each of the training load methods investigated are appropriate for quantifying endurance training dose and that submaximal HR and HRV may be useful for monitoring fitness and fatigue, respectively.

Monitoring Athlete Training Loads: Consensus Statement

“…Indeed, there has been a recent increase in the number of both scientific reports and case studies using various analyses of training-load data to infer on readiness to perform, risk of illness, injury risk, and return to play from injury in athletes. 27,[39][40][41][42] The following section describes the common methods used to analyze these training-load data. where K is the constant that adjusts for the magnitude of the fatigue effect relative to the fitness effect.…”

“…Although this approach has been used to guide training planning 49 and predict future performance and fitness and fatigue levels, 42,48,50,51 it has been criticized for oversimplifying complex relationships between training and performance. 52,53 Indeed, such models have shown large variability in parameters and precision in predicting performance in highly trained athletes.…”

“…Unfortunately, however, the broader application of this approach to predict performance in highly trained athletes is limited due to the large amount of unexplained variance in the predictions for performance across a range of endurance sports. 42,65,66 Other limitations of this approach are that they fail to consider the different responses provided by all of the training stimuli (ie, internal vs external load or aerobic vs resistance load, etc), and the models have mostly been applied to individual sports. In addition, they have not yet been applied to team sports where between-matches training loads are consistent and there is a relatively small dose of training between competitions (ie, usually 1 or 2 games per week).…”

Relationships Between Internal and External Training Load in Team-Sport Athletes: Evidence for an Individualized Approach

“…This can be further analysed to evaluate its relationship with other means of training load monitoring. [6][7][8] Subsequently, the use of RPE, a simple and non-invasive tool, has become popular in team sports due to its low cost and potential benefits in quantifying IL. 9 8 it has be shown that ~62% of the variance in s-RPE could be explained by the distance covered in a session, the number of impacts (defined as the sum of accelerations from 3 planes) and 'body load' (a measure of total stress resulting from accelerations, decelerations, changes of direction and impacts).…”

“…6,8,9,14 Machine learning has experienced an increase in popularity in sporting and exercise research, with applications such as the prediction of competition outcome 15 and quantification of movement activity types. 16,17 This increase in popularity has stemmed from the potential ability of these approaches to account for nonlinearity within datasets and thus display improved performance.…”