Within-Subject Principal Component Analysis of External Training Load and Intensity Measures in Youth Soccer Training.

ABSTRACT: Marynowicz, J, Lango, M, Horna, D, Kikut, K, Konefal, M, Chmura, P, and Andrzejewski, M. Within-participant principal component analysis of external training load and intensity measures in youth soccer training. J Strength Cond Res 37(12): 2411-2416, 2023-The aim of this study was to identify which combination of external training load (EL) and external intensity (EI) metrics during youth soccer training sessions captured similar or unique information. Data were collected from 18 youth soccer players during an 18-week in-season competition period using a 10-Hz global positioning system, rating of perceived exertion (RPE), and session-RPE (sRPE). External training load measures included total distance (TD, in meters), PlayerLoad (PL, in arbitrary units), high-speed running distance (HSR, in meters), and number of accelerations (ACC, n). All EL metrics were also divided by session duration (minutes) to obtain EI values. A total of 804 training observations were undertaken (43 ± 17 sessions per player). The analysis was performed by use of the principal component analysis technique. The first principal component (PC) captured 49-70% and 68-89% of the total variance in EI and EL, respectively. The findings show that from the 5 EI metrics, most of the information can be explained by either TD per minute or PL per minute, with a loading from 0.87 to 0.98 and from 0.76 to 0.95, respectively. The majority of EL information can be explained by PL (loading: 0.93-0.98), TD (loading: 0.95-0.99), ACC (loading: 0.71-0.91), or sRPE (loading: 0.70-0.93). The second PC for EL metrics is most strongly correlated with HSR, with loadings from 0.53 to 0.84. The results suggest that the majority of the information contained in the EL variables can be captured in 1 PC without losing much information. The findings suggest that stakeholders who intend to provide a fast and holistic view of EL information in a daily training environment should report TD, PL, ACC, or sRPE plus HSR to coaching staff as a metrics that provides additional unique information.

Predicting ratings of perceived exertion in youth soccer using decision tree models.

The purpose of this study was to determine the effectiveness of white-box decision tree models (DTM) for predicting the rating of perceived exertion (RPE). The second aim was to examine the relationship between RPE and external measures of intensity in youth soccer training at the group and individual level. Training load data from 18 youth soccer players were collected during an in-season competition period. A total of 804 training observations were undertaken, with a total of 43 ± 17 sessions per player (range 12-76). External measures of intensity were determined using a 10 Hz GPS and included total distance (TD, m/min), high-speed running distance (HSR, m/min), PlayerLoad (PL, n/min), impacts (n/min), distance in acceleration/deceleration (TD ACC/TD DEC, m/min) and the number of accelerations/decelerations (ACC/DEC, n/min). Data were analysed with decision tree models. Global and individualized models were constructed. Aggregated importance revealed HSR as the strongest predictor of RPE with relative importance of 0.61. HSR was the most important factor in predicting RPE for half of the players. The prediction error (root mean square error [RMSE] 0.755 ± 0.014) for the individualized models was lower compared to the population model (RMSE 1.621 ± 0.001). The findings demonstrate that individual models should be used for the assessment of players' response to external load. Furthermore, the study demonstrates that DTM provide straightforward interpretation, with the possibility of visualization. This method can be used to prescribe daily training loads on the basis of predicted, desired player responses (exertion).

Relationship Between the Session-RPE and External Measures of Training Load in Youth Soccer Training.

Marynowicz, J, Kikut, K, Lango, M, Horna, D, and Andrzejewski, M. Relationship between the session-RPE and external measures of training load in youth soccer training. J Strength Cond Res 34(10): 2800-2804, 2020-The aim of this study was to identify the external training load (TL) markers (10 Hz Global Positioning System) that are most influential on the rating of perceived exertion (RPE) and session-RPE (sRPE) during youth soccer training. Data were collected from 18 youth soccer players during an 18-week in-season period. A total of 804 training observations were undertaken. We observed moderate to very large within-individual correlations between sRPE and measures of external load (r ranging from 0.36 to 0.76). Large, positive within-individual correlations were found between total covered distance, PlayerLoad, number of accelerations, and sRPE (r = 0.70, 0.64, and 0.62, respectively, p < 0.001). By contrast, small to moderate within-individual correlations were noted between RPE and measures of intensity (r ranging from 0.16 to 0.39). A moderate within-individual correlation was observed between high-speed running distance (HSR) per minute and RPE (r = 0.39, p < 0.001). The level of statistical significance was set at alpha = 0.05 for all tests. Two generalized estimating equation models were constructed, with RPE and sRPE as the response variables. The model identified by QIC for RPE contained 2 variables as follows: HSR per minute and distance in deceleration per minute, whereas sRPE was modeled with 3 predictors as follows: PlayerLoad, HSR, and distance in acceleration. The findings demonstrate that RPE does not reflect the intensity of a training session and that sRPE can be a useful, simple, and cost-effective tool for monitoring TL.