Internal and external load monitoring in well-trained and professional cyclists

Yila, Rodet-William

Internal and external load monitoring in well-trained and professional cyclists

Thesis / Dissertation

2025

Abstract

With the desire of inducing training adaptations in an athlete to facilitate an increase in performance capacity, as well as develop an increased resilience to fatigue, a sustained period of tolerable physiological stress is required. However, this adaptation is only possible when an adequate period of recovery is afforded to the athlete (Saw et al., 2016). An excessive training induced stress imposed on an athlete, without being afforded an adequate recovery period, may result in an unwanted accumulation of fatigue, which can lead to a decrement in performance, as well an increased likelihood of sustaining an overuse injury or illness (Pyne & Martin, 2011). It should also, however, be noted that when the period of recovery is prolonged, it can also lead to a decrement in performance, as well as increase the likelihood of sustaining an overuse injury as the desirable training adaptations diminish over time in the absence of tolerable physiological stress (Mujika & Padilla, 2000). The purpose of this dissertation was to determine whether the combination of a customised online questionnaire (Subjective Wellness Score, SWS) and a modified submaximal cycling test (Submaximal Fatigue Test, SFT) (internal and external training load monitoring tools) can be used in conjunction with an external training load monitoring tool (TrainingPeaks® Performance Management Chart™, PMC) to monitor the response to training in well-trained and professional cyclists with acceptable validity and reliability. Pre-existing data, already uploaded onto two online registries, was retrospectively analysed. Data from 258 SFTs were analysed from 20 well-trained and professional cyclists, whereby objective and subjective data obtained from the SWS, SFT metrics and TrainingPeaks® PMC metrics was acquired from a continuous, longitudinal dataset. The first question formulated for this dissertation investigated whether the combination of the SWS and metrics of the SFT (HRaverage and RPE) are sensitive to changes in training load. The next question was to scrutinize the validity of the collected SWS questions, SFT metrics, and ETL metrics (of the PMC), and determine whether they are able to predict performance outcomes measured by the SFT (W/kg/RPE, RPE, TTE). The results of the linear regression analysis for ‘All Participants' (n = 20) reported the most significant correlation was found for TSB when compared to the SWS and SFT metrics, with statistically significant results for the variables of fatigue rating (R2 = 0.126; p < 0.001), overall feel rating (R2 = 0.026; p = 0.009), and RPE (R2 = 0.051; p < 0.001). The results of the multivariate analysis for TSS (7-day average), CTL, ATL, and TSB, for ‘All Participants' (n = 5 20) indicated that TSB was found to have the strongest correlation with the SWS and SFT metrics (R2 = 0.197; F = 7.86; p < 0.001). The most significant correlations were found for TSB when compared to the SWS and SFT metrics, with statistically significant results for the variables of overall-feel rating (F = 12.37; p = 0.001) and fatigue rating (F = 28.11; p < 0.001). The results of the linear regression analysis for ‘All Participants' (n = 20) reported the most significant correlations was found for the relative power output per unit of body mass per unit of RPE (W/kg/RPE) when compared to the SWS, HRaverage (of the SFT), and ETL metrics of the PMC, with statistically significant results for the variables of fatigue rating (R2 = 0.110; p < 0.001), TSB (R2 = 0.065; p < 0.001), composite score rating (R2 = 0.034; p = 0.003), overall feel rating (R2 = 0.032; p = 0.004), and sleep rating (R2 = 0.016; p = 0.038). In the multivariate analysis for the modelling of the performance outcome measures of the SFT (W/kg/RPE, RPE, and TTE), for ‘All Participants' (n = 20), RPE was found to have the strongest correlation with the SWS, HRaverage, and ETL metrics of the PMC (R2 = 0.207; F = 7.43; p < 0.001). The most significant correlations were found for W/kg/RPE when compared to the SWS, HRaverage, and ETL metrics of the PMC, with statistically significant results for the variables of overall-feel rating (F = 11.06; p = 0.001), fatigue rating (F = 18.02; p < 0.001), stress rating (F = 8.23; p = 0.005) sleep rating (F = 4.61; p = 0.033), and TSB (F = 4.48; p = 0.035). Considering the findings in this research study, the two questions formulated for this thesis can be answered, whereby as it can be stated, with confidence, that the SWS and SFT metrics (overall-feel rating and fatigue rating) were sensitive to changes in the ETL metric of the PMC (TSB). It was also found that the SWS, HRaverage, and ETL metrics of the PMC were able to a predict performance outcome measure of the SFT, yielding a correlation with W/kg/RPE. Furthermore, it should also be accepted that the combined use of a customised online questionnaire and a modified submaximal cycling test (SFT) can be used in conjunction with the TrainingPeaks® PMC to monitor the response to training in well-trained and professional cyclists with acceptable validity and reliability. In closing, this study suggests that the concurrent use of the SWS, SFT and PMC is an effective and efficient method for coaches/sports scientist to monitor the cyclist's response to their encountered training load as the blend of subjective and objective training load metrics have been found to be sensitive to changes in fatigue status, with acceptable validity and reliability.

Keywords

Exercise Science

Reference:

Collections

Masters

Full item page