Fluctuations in Well-Being Based on Position in Elite Young Soccer Players during a Full Season

Classification of recovery states in U15, U17, and U19 sub-elite football players: a machine learning approach

Introduction

A promising approach to optimizing recovery in youth football has been the use of machine learning (ML) models to predict recovery states and prevent mental fatigue. This research investigates the application of ML models in classifying male young football players aged under (U)15, U17, and U19 according to their recovery state. Weekly training load data were systematically monitored across three age groups throughout the initial month of the 2019-2020 competitive season, covering 18 training sessions and 120 observation instances. Outfield players were tracked using portable 18-Hz global positioning system (GPS) devices, while heart rate (HR) was measured using 1 Hz telemetry HR bands. The rating of perceived exertion (RPE 6-20) and total quality recovery (TQR 6-20) scores were employed to evaluate perceived exertion, internal training load, and recovery state, respectively. Data preprocessing involved handling missing values, normalization, and feature selection using correlation coefficients and a random forest (RF) classifier. Five ML algorithms [K-nearest neighbors (KNN), extreme gradient boosting (XGBoost), support vector machine (SVM), RF, and decision tree (DT)] were assessed for classification performance. The K-fold method was employed to cross-validate the ML outputs.

Results

A high accuracy for this ML classification model (73-100%) was verified. The feature selection highlighted critical variables, and we implemented the ML algorithms considering a panel of 9 variables (U15, U19, body mass, accelerations, decelerations, training weeks, sprint distance, and RPE). These features were included according to their percentage of importance (3-18%). The results were cross-validated with good accuracy across 5-fold (79%).

Conclusion

The five ML models, in combination with weekly data, demonstrated the efficacy of wearable device-collected features as an efficient combination in predicting football players' recovery states.

Are non-starters accumulating enough load compared with starters? Examining load, wellness, and training/match ratios of a European professional soccer team

BACKGROUND

The aims of the study were to: (i) compare accumulated load and wellness between starters and non-starters of a European professional soccer team; (ii) analyze the relationships between wellness and load measures and; (iii) compare training/match ratio (TMr) of external and internal load between starters and non-starters.

METHODS

Ten players were considered starters while seven were classified as non-starters over a 16-week period in which six training sessions and match day (MD) were considered in each weekly micro-cycle. The following measures were used: wellness (fatigue, quality of sleep, muscle soreness, stress, and mood); load (rated of perceived exertion (RPE), session-RPE (s-RPE), high-speed running (HSR), sprinting, accelerations (ACC) and decelerations (DEC)). Accumulated wellness/load were calculated by summing all training and match sessions, while TMr was calculated by dividing accumulated training load by match data for all load measures and each player. Mann-Whitney U test was used for wellness variables, while independent T-test was used for the remaining variables to compare groups. Moreover, relationships among variables were explored using the Spearman's Rho correlation coefficient.

RESULTS

The main results showed that non-starters presented higher significant values for fatigue (p < 0.019; g = 0.24) and lower significant values for duration (p < 0.006; ES = 1.81) and s-RPE (p < 0.001; ES = 2.69) when compared to starters. Moreover, positive and very large correlation was found between quality of sleep and RPE, while negative and very large correlation were found between stress and deceleration, and mood and deceleration (all, p < 0.05). Finally, non-starters presented higher values in all TMr than starters, namely, RPE (p = 0.001; g = 1.96), s-RPE (p = 0.002; g = 1.77), HSR (p = 0.001; g = 2.02), sprinting (p = 0.002; g = 4.23), accelerations (p = 0.001; g = 2.72), decelerations (p < 0.001; g = 3.44), and duration (p = 0.003; g = 2.27).

CONCLUSIONS

In conclusion, this study showed that non-starters produced higher TMr in all examined variables despite the lower match and training durations when compared with starters, suggesting that physical load was adjusted appropriately. Additionally, higher RPE was associated with improved sleep while higher number of decelerations were associated with decreased wellness, namely, stress and mood for non-starters.

The impact of verbal encouragement during the repeated agility speed training on internal intensity, mood state, and physical enjoyment in youth soccer players

Objective

Verbal encouragement (VE) can be used by coaches to boost morale and commitment during training exercises. This investigation aimed to study the impacts of VE given by coaches on the physiological aspects, players' internal intensity, mood, and perceived enjoyment of youth soccer players during repeated agility speed training (RAS).

Methods

A total of 17 male youth soccer players (mean ± SD; age: 13.8 ± 0.4 years; body mass: 59.1 ± 6.7 kg; height: 170.0 ± 6.2 cm; training experience: 5.1 ± 0.7 years) participated, in a randomized order, in two experimental training sessions that consisted of a RAS (i.e., the Illinois course) either with VE (RAS-E) or without VE (RAS-NE), with a 7-day interval between the testing sessions. Heart rate (HR) was registered throughout the exercise. The rating of perceived exertion, blood lactate concentration [La], and perceived enjoyment were measured after each training session. The mood state was recorded before and after each protocol.

Results

HR mean (Cohen's coefficient d = 0.45, small), %HRmax (d = 0.37, small), HR peak (d = 0.66, moderate), [La] (d = 0.56, small), and the PACES score (d = 2.8, very large) were higher in RAS-E compared to RAS-NE (all, P < 0.001). Compared to the RAS-E trial, the RAS-NE trial showed higher fatigue (P < 0.01), tension (P < 0.05), anger (0.05), total mood score (P < 0.001), and lower vigor (P < 0.001).

Conclusion

Coaches may use VE during RAS to improve psychophysiological responses, mood state, and perceived enjoyment in youth soccer players.

Comparison of external load measures per minute based on match seasonal periods and playing positions of a soccer Turkish super league team

BACKGROUND

Turkish Super League teams need more information about the external load. Considering the specific country and the coaches' philosophies, the purpose of this study was to compare the external match load of a Turkish Super League team considering the different playing positions and in-season periods.

METHODOLOGY

A longitudinal study design was employed by observing 29 official matches of the same team. A total of fifteen players, consisting of five defenders, five midfielders, and five forwards, were analyzed using the Sentio Sports Optical Tracking System. The following outcomes were extracted in each match: total distance (TD), walking [from 0 to 7.2 km/h], jogging [from 7.2 to 14.4 km/h], running [from 14.4 to 20 km/h], high speed running (HSR) [> 20 km/h], metabolic power (MP), maximum deceleration (Decmax) [Dec < -3 m/s2], total deceleration distance (Dectotal), maximum acceleration (Accmax) [Acc > 3 m/s2], and total acceleration distance (Acctotal). Statistical analysis consisted of ANOVA and Bonferroni correction post hoc tests.

RESULTS

The main results showed that all variables were similar between periods of the season (p > 0.05) except for walking (p = 0.021 witha large effect size), which revealed lower values in the mid-season period. When analyzing periods of the season considering playing positions, there were several significant results for TD, walking, jogging, running, HSR, Dectotal and Acctotal (p < 0.05, with small to large effects). A tendency for higher values was noted in the mid and end-season. Considering the playing positions comparisons, midfielders showed higher values than defenders for TD, MP, Dectotal and Acctotal (p < 0.05 with large effect for all). Midfielders also showed higher values than forwards for TD, jogging, Dectotal and Acctotal (p < 0.05 with a large effect for all).

CONCLUSIONS

The present study emphasizes the significance of analyzing data based on minute, playing position, and season period. Findings reveal that defenders consistently displayed the lowest values in all external load measures during matches compared to other positions throughout the season. Furthermore, midfielders demonstrated a higher activity profile during the initial and middle stages of the season compared to other positions, with a slight tendency to decrease load towards the end of the season.

The Association between Training Frequency, Symptoms of Overtraining and Injuries in Young Men Soccer Players

Overtraining is a prevalent issue among young men soccer players, particularly those who are driven to enhance their skills. While an intense training volume and effort might contribute to athletic growth, it can also have negative implications, including injury. The current study aimed at examining the association between training frequency, symptoms of overtraining and injuries in young men soccer players. A path analysis approach was used to examine the causal relationships between variables. The sample consisted of 189 young men soccer players aged 13-17 years old (age = 14.81, SD = 1.37). Participants reported that they were training, on average, 5.77 days (SD = 1.53) per week. Athletes were competing at a regional (n = 100) or national (n = 89) level. Concerning injuries, participants indicated, on average, 2.03 (SD = 1.16) injuries since they started practicing soccer. The results displayed a significant association, as theoretically expected, namely: (i) training frequency was significantly associated with overtraining symptoms (β = 0.15 [IC95% = 0.01, 0.29]); (ii) overtraining symptoms were significantly associated with the number of injuries (β = 0.19 [IC95% = 0.02, 0.35]). An indirect effect between training frequency and injuries was also observed (β = 0.15 [IC95% = 0.01, 0.29]). Thus, there is preliminary evidence that overtraining symptoms could play a mediating role. In conclusion, investigating the links between overtraining symptoms and injury in young men soccer players is critical, as it can assist in identifying overtraining warning signs, promote young players' health and safety, customize training regimens to individual needs, and contribute to a better understanding of sports-related injuries.

Relationships between training load, peak height velocity, muscle soreness and fatigue status in elite-level young soccer players: a competition season study

BACKGROUND

This study aimed to compare training load parameters, delayed onset muscle soreness (DOMS), and fatigue status between season periods (1^st and 2^nd halves) in U14 soccer players and to analyze the relationships between training load parameters based on season periods (1^st and 2^nd halves) with peak height velocity (PHV), DOMS, and fatigue status in under-14 (U14) young elite soccer players. Additionally, it was intended to analyze if fatigue, DOMS and PHV could explain training load parameters across the season.

METHODS

Twenty U14 players that competed in the national league participated in this study. The players were monitored during the whole season (26 weeks), and evaluations were carried out at the end of the in-season. Anthropometric and body composition parameters and the maturity offset of each player were utilized to compute each player's age at PHV. Players reported their levels of DOMS and fatigue status using Hooper index questionnaires. The internal load was monitored using the rating of perceived exertion (RPE). Acute weekly internal load (AW), chronic weekly internal load (CW), acute: chronic workload ratio (ACWR), training monotony (TM), and training strain (TS) were also obtained.

RESULTS

The main results showed that TM was higher in the 2^nd half, while CW, AW and DOMS were higher in the 1^st half of the season. Moreover, the main correlations showed a positive correlation between PHV and TS (2^nd half of the season) and between fatigue and TM (1^st half of the season).

CONCLUSION

In conclusion, variations in well-being status and PHV cannot explain the variations in internal training loads in elite U14 soccer players. In addition, internal training load indices during the first half of the competitive season can promote a fundamental base for progression loads during the second period of the competitive season.

Associations between Training Load and Well-Being in Elite Beach Soccer Players: A Case Report

The current case study aimed to quantify within-subjects correlations between training load and well-being in elite male beach soccer players. Data were obtained over three consecutive days during the preparation camp for the FIFA Beach Soccer World Cup Russia 2021. The session rating of perceived exertion (s-RPE) and external training load metrics using global positioning system (GPS) were recorded. Eleven players reported perceived well-being (sleep quality, fatigue, muscle soreness and stress) using a Likert scale (Hooper Index) before breakfast. Within-subjects correlation coefficients between workload and well-being were calculated. Workload metrics and perceived well-being indices were significantly lower on day three than on days one and two. The Hooper Index presented a very large positive correlation with s-RPE (r = 0.86 [0.67, 0.94], 95% confidence interval, CI), exposure time (r = 0.88 [0.71, 0.95]), total distance (r = 0.83 [0.60, 0.93]), high-speed distance (r = 0.77 [0.50, 0.91]), and number of sprints (r = 0.75 [0.47, 0.90]). Sleep quality presented a moderate to large positive correlation with s-RPE (r = 0.51 [0.11, 0.77]), exposure time (r = 0.50 [0.10, 0.76]), high-speed distance (r = 0.53 [0.15, 0.78]), number of sprints (r = 0.62 [0.28, 0.83]) and total distance (r = 0.41 [0.18, 0.78]). Fatigue presented a large to very large positive correlation with s-RPE (r = 0.85 [0.66, 0.94]), exposure time (r = 0.90 [0.78, 0.96]), total distance (r = 0.86 [0.68, 0.94]), high-speed distance (r = 0.65 [0.31, 0.84]) and number of sprints (r = 0.56 [0.18, 0.79]). Muscle soreness presented a large to very large positive correlation with s-RPE (r = 0.79 [0.56, 0.91]), exposure time (r = 0.83 [0.62, 0.93]), total distance (r = 0.81 [0.59, 0.92]), high-speed distance (r = 0.75 [0.47, 0.89]) and number of sprints (r = 0.59 [0.22, 0.81]). Overall, workload presented a meaningful correlation with perceived well-being indices in elite male beach soccer players during a training camp. These findings suggest that workload metrics and perceived well-being indices can be implemented into the daily routine of an elite beach soccer team, which may assist coaches, sports scientists, and practitioners in better preparing players for beach soccer competitions.

A Practical Approach to Assessing Physical Freshness: Utility of a Simple Perceived Physical Freshness Status Scale

Background: Monitoring physical freshness is essential in assessing athletes’ conditions during training periods, training sessions, or competitions. To date, no single physical freshness scale has been successfully validated against training load variables and widely used scales measuring different facets of physical freshness. Objective: In this study, we develop and test the practical utility of a perceived physical freshness (RPF) scale to monitor the condition of the athletes and to prevent excessive fatigue and insufficient recovery during training sessions or competitions. Methods: Sixteen professional male soccer players (mean ± SD age 26 ± 4 years) were enrolled. Training load (TL), monotony, strain, rate of perceived exertion (RPE), well-being indices (sleep, stress, fatigue, and muscle soreness), total quality recovery (TQR) and RPF were determined each day for two weeks of training, including a week intensified training (IW) and a week taper (TW). The validity of the RPF scale was assessed by measuring the level of agreement of a player’s perceived physical freshness relative to their TL variables, recovery state and well-being indices during each training phase (IW and TW) and during the overall training period (TP). Results: RPF increased during the TW compared to IW (ES = 2.31, p < 0.001, large). For the TP, IW and TW, weekly RPF was related to weekly TL (r = −0.81, r = −0.80, r = −0.69, respectively), well-being (r = −0.91, r = −0.82, r = −0.84, respectively) and TQR (r = 0.76, r = 0.91, r = 0.52, respectively), all p < 0.01. For the TP, IW and TW, daily RPF was related to TL (r = −0.75, r = −0.66, r = −0.70, respectively), well-being (r = −0.84, r = −0.81, r = −0.78, respectively) and TQR (r = 0.82, r = 0.81, r = 0.75, respectively), all p < 0.01. Conclusions: RPF was effective for evaluating the professional soccer players’ physical freshness and may be a strategy for coaches to monitor the physical, psycho-physiological, and psychometric state of the players before training session or matches.

Using Global Positioning System to Compare Training Monotony and Training Strain of Starters and Non-Starters across of Full-Season in Professional Soccer Players

Soccer is an attractive and popular team sport that has high physiological and fitness stress, and therefore requires special and controlled training programs during the season. The aim of this study was to describe the weekly average and changes in training monotony (TM) and training strain (TS) throughout different periods of the season in professional football players based on the number of accelerations and decelerations, and also to analyze the difference between starters and non-starters players in TM and TS. Nineteen professional players from a soccer team competing in the Iranian Premier League (age, 28 ± 4.6 years; height, 181.6 ± 5.8 cm; body mass, 74.5 ± 5.6 kg, and body mass index, 21.8 ± 1.0 kg/m2) participated in a cohort study. Participants were divided into two groups based on the time of participation in the weekly competition: starters (N = 10) or non-starters (N = 9). The physical activities of the players were recorded during the training sessions and competitive matches of 43 weeks using GPSPORTS systems Pty Ltd. During pre- and end-season TS was not significantly different between starters and non-starters, while during early- and mid-season starters showed a higher TS than non-starter (p < 0.05). TS was higher during early- and mid-season compared to pre- and end-season. In all zones on both the TM and TS variables, non-starters experienced higher change percentages and coefficient of variation. TM during the season in all zones of accelerations was not significantly different between starters and non-starters. while during mid-season starters showed a higher TM than non-starters in all zones of decelerations (p < 0.05). TM data showed fluctuations and w-shaped graphs in the week-by-week survey. These results indicate that training during early- and mid-season is not enough for the physical development of non-starters soccer players. Coaches should be more careful when designing training for non-starters players, and they could consider the use of game simulation, preparatory match or intra-team match, or individual training programs.

Wearable Inertial Measurement Unit to Measure External Load: A Full-Season Study in Professional Soccer Players

The aim of this study was to describe weekly acute workload (wAW), chronic workload (wCW), acute: chronic workload ratio (wACWR), training monotony (wTM), and training strain (wTS) variations over a full season across playing positions. Twenty-one professional soccer players were daily monitored during 48 consecutive weeks. Total distance, sprint total distance (STD), high-speed running distance (HSRd), maximum speed, number of the repeated sprints, and body load (BL) were obtained during training and matches using a Wearable Inertial Measurement Unit. The wAW was determined for each external load measure. The wCW, wACWR, and wTM were calculated based on BL metric. Higher values of weekly STD were observed in lateral defenders/wingers (LDW) compared to central defenders/forwards (CDF) (p = 0.009; ES = Large) and midfielders (MDF) (p = 0.034; ES = Large). Additionally, weekly HSRd was higher in LDW vs. CDF (p = 0.016; ES = Large) and MDF (p = 0.011; ES = Large). The CDF presented a lower weekly number of repeated sprints than LDW (p = 0.021; ES = Large). In conclusion, weekly external load metrics were position-dependent over the season. Moreover, LDW a presented greater weekly STD, HSRd, and number of repeated sprints compared to other positions.

In-season monotony, strain and acute/chronic workload of perceived exertion, global positioning system running based variables between player positions of a top elite soccer team

Background

The interpretation of the load variations across a period seems important to control the weekly progression or variation of the load, or to identify within- micro and mesocycle variations. Thus, the aim of this study was to describe the in-season variations of training monotony, training strain, and acute: chronic workload ratio (ACWR) through session rating of perceived exertion (s-RPE), total distance and high-speed running (HSR) according to playing positions in an elite soccer team.

Methods

Seventeen professional players from an European First League team participated in this study. They were divided four central defenders (CD), three wide defenders (WD), four central midfielders (CM), three wide midfielders (WM) and three strikers (ST). The players were monitored daily over a 41-week period of competition where 52 matches occurred during the 2015–2016 in-season. Through the collection of s-RPE, total distance and HSR, training monotony, training strain and ACWR were calculated for each measure, respectively. Data were analysed across ten mesocycles (M: 1–10).

Results

The main results showed significant differences (p < 0.05) for TMs-RPE between CD vs. ST (moderate effect) in M2; between CD vs. CM (moderate effect) for TS of s-RPE; between CD vs. ST moderate effect) in M6 for ACWR of s-RPE. In addition, there was significant difference between CM vs. ST (moderate effect) in M2 for TS of TD; between WD vs. ST (moderate effect) in M3 for ACWR of TD. Moreover, there were significant differences for TM of HSR between CD vs. WD (very large effect); CD vs. WD (moderate effect) in M4 for TS of HSR.

Conclusions

The present study presents new insights to coaches and technical staff about the variation profiling of TM, TS, and ACWR calculated with internal and external load measures, between player positions during 10 mesocycles.

Influence of 2D:4D ratio on fitness parameters and accumulated training load in elite youth soccer players

BACKGROUND

Digit ratio (2D:4D) characterized by the length of the second digit (2D) divided by the length of the fourth digit (4D), is a powerful marker of athletic performance. Some studies showed a negative correlation between 2D:4D ratio and sports performances.

OBJECTIVES

The purpose of the present study was three-fold: (1) to analyze the influence of anthropometric and 2D:4D ratio on variations of accumulated training load (ATL) and fitness parameters: maximal oxygen uptake (V̇O_2max), countermovement jump (CMJ), isometric muscular strength of the knee extensor for hamstring (ISH) and flexor for quadriceps (ISQ) muscles; along three stages of evaluation of soccer players based on playing positions; (2) to analyze the correlations between 2D:4D ratio and aforementioned parameters; and (3) to investigate if variance in fitness levels and ATL can explain the 2D:4D ratio.

METHODS

Twenty-four elite players under 17 years were daily monitored for their rating perceived exertion and ATL across 24 weeks over the season. Soccer players have also measured in three stages for anthropometric traits and fitness parameters.

RESULTS

Significant differences were observed between playing positions for body mass, goalkeepers had higher body mass compared to centre-midfielder and winger players. Moreover, there were significant differences in ATL between early-season to mid-season in goalkeepers (P = 0.032). The 2D:4D ratio (left and right) shown largely and negatively association with muscular strength (ISQ: r =  - 0.80; r =  - 0.78, P ≤ 0.001, ISH: r =  - 0.63; r =  - 0.62, P = 0.001, respectively) and VO_2max changes (r =  - 0.55, P = 0.005; r =  - 0.50, P = 0.013, respectively); lastly, both 2D:4D ratio significantly predicted changes in muscular strength and VO_2max in young soccer players.

CONCLUSIONS

Goalkeepers tended to have higher body mass compared to centre-midfielder and winger players; and 2D:4D ratio revealed a mighty predictor of physical fitness changes in soccer players. Evidence should be helpful to professionals to highlight the usefulness of the 2D:4D into the identification of talent, but also to optimize young players' performance.

Age-related differences in linear sprint in adolescent female soccer players

BACKGROUND

Several studies have observed the contribution of chronological age, biological maturation, and anthropometric characteristics to sprinting performance in young soccer players. Nevertheless, there are no studies that have analysed the contribution of these characteristics to running speed qualities in adolescent female soccer players.

OBJECTIVE

This study investigated age-related differences in sprint performance in adolescent female soccer players. Also, it examined the possible influence of anthropometry [body mass and body mass index (BMI)] and biological maturation [age at peak height velocity (APHV)] in sprint performance.

METHODS

Eighty adolescent female soccer players [under (U) 14, n = 20; U16, n = 37; U18, n = 23] participated in this study. Players were tested for 40 m sprint (each 10 m split times).

RESULTS

Posthoc analysis revealed better performance in all split sprint times of older soccer players (U18 and U16) compared with younger category (F: 3.380 to 6.169; p < 0.05; ES: 0.64 to 1.33). On the contrary in all split sprint times, there were no significant changes between U16 and U18 (p < 0.05; ES: 0.03 to 0.17). ANCOVA revealed differences in all parameters between groups, controlled for APHV (p < 0.05). In contrast, all between-group differences disappeared after body mass and BMI adjustment (p > 0.05). Finally, the results indicate that BMI and body mass were significantly correlated with 40 m sprint (p < 0.05; r: -0.31) and 20 m flying (p < 0.01; r: 0.38), respectively.

CONCLUSION

In the present players' sample, body mass and BMI had a significant impact on running speed qualities.