do match-related contextual variables influence

DO MATCH-RELATED CONTEXTUAL VARIABLES INFLUENCE TRAINING LOAD IN HIGHLY TRAINED SOCCER PLAYERS? JOAO BRITO,1,2 MAXIME HERTZOG,1

AND

GEORGE P. NASSIS1

1

National Sports Medicine Programme, Excellence in Football Project, Aspetar—Qatar Orthopaedic and Sports Medicine Hospital, Qatar; and 2Health and Performance Unit, Portugese Football Federation, Lisbon, Portugal

ABSTRACT Brito, J, Hertzog, M, and Nassis, GP. Do match-related contextual variables influence training load in highly trained soccer players? J Strength Cond Res 30(2): 393–399, 2016—This study analyzed training loads of youth soccer players and examined the influence of match-related contextual variables in internal training load and fatigue. A secondary aim was to investigate the variability of these parameters throughout the season. Thirteen highly trained under-19 players (18.6 6 0.5 years) were followed during one season. Training load (daily) and fatigue scores (weekly) were assessed using rate of perceived exertion and a short questionnaire, respectively. Higher weekly training loads were reported after a defeat or draw compared to a win (2,342 6 987 and 2,395 6 613 vs. 1,877 6 392 AU; p # 0.05; d = 0.30–0.45). Weekly training loads were higher after playing an away match than after a home match (2,493 6 821 vs. 2,153 6 577 AU; p # 0.05; d = 0.23). Within training sessions, the coefficients of variation for internal training load ranged from 5 to 72%. Throughout the season, the coefficients of variation for weekly training loads and fatigue scores ranged from 29 to 49% and 18 to 44%, respectively. Weekly training load decreased as the season progressed (p , 0.001); no changes were detected for the fatigue score. In conclusion, the large variation in internal training load within a session and its sensitivity to initial and subsequent match conditions underline the need for a more individualized approach. These findings and the stability of the fatigue scores throughout the season may indicate that highly trained players modulate their pace during training.

KEY WORDS periodization, fatigue, monitoring, RPE

Address correspondence to George P. Nassis, george.nassis@ aspetar.com. 30(2)/393–399 Journal of Strength and Conditioning Research Ó 2016 National Strength and Conditioning Association

INTRODUCTION

I

n soccer, the competition phase usually runs for 9–10 months, with official matches played almost every week. Fatigue or decreased performance may occur in different periods of the season, but the prolonged competitive calendars imply that players need to maintain high levels of performance throughout the season. For this purpose, the concept of training load (the product of volume and intensity of training) has emerged as an important aspect of effective monitoring of individual adaptations to training (15). Overall, training load is believed to represent the traininginduced physiological stress imposed to the players (12). Several objective methods of training load monitoring have been proposed, mostly based on heart rate and time–motion recording (6,7,12,15). Although useful and valid, these methods do still have inherent limitations especially during high-intensity/high-speed efforts (6,7). Also, even in elite professional settings, the time available for the analyses of training load might be limited, and the associated costs of the equipment could prevent many teams using these methods throughout the season. For this reason, an easy and practical means of reporting and calculating training load based on subjective measures of rating of perceived exertion (RPE) has gained recent popularity (5,12,15). As an index of perception of exertion, RPE is an integration of afferent neural signals from various physiological systems to the brain (1). The factors that influence athlete’s RPE during exercise and hence the training load estimation are unknown. Recently, the RPE-derived training load was reported to correlate significantly with the highspeed running distance, number of impacts, and accelerations in elite soccer players (13). However, these correlations ranged from r = 0.09 to r = 0.25, highlighting the need for further studies in understanding the factors that influence internal training load. During match-play, several contextual factors such as the match location, match result, early dismissals during the match, density of fixtures, and the quality of opposition have a major influence on the demands imposed to the players (8,16). However, to the best of our knowledge, no study so far has investigated the influence of VOLUME 30 | NUMBER 2 | FEBRUARY 2016 |

393

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

Internal Training Load in Soccer match-related contextual variables on the demands of training in a professional soccer setting. The primary aim of this study was to examine the matchrelated contextual variables (e.g., result of the previous match, match location, and quality of opposition) that are associated with perceived training load and fatigue in highly trained youth soccer players. A secondary aim was to investigate whether self-reported training load and fatigue vary throughout the season in these players.

METHODS Experimental Approach to the Problem

Throughout the season, individual data regarding perceived exertion and fatigue were collected. The procedures were used as a standard monitoring routine at the club, so players were familiar with the tools before the start of the study. Subjects

Thirteen under-19 highly trained soccer players (18.6 6 0.5 years, age range 18–19.3 years, 176.9 6 4.6 cm, and 70.0 6 7.3 kg) from a first league club in France were followed during the season 2009–2010. The group included 5 defenders, 5 midfielders, and 3 attackers. The squad included 16 outfield players in total, but 2 goalkeepers and one player that had been injured most of the season were excluded from the analysis. The club and players provided written informed consent to allow for the use of data. The study was approved by institutional review board and conformed to the Code of Ethics of the World Medical Association (Declaration of Helsinki). Procedures

The season comprised 4 different phases: preparation I (3 weeks in July–August), competition I (18 weeks, August– December), preparation II (2 weeks in January, after the 4-week winter break, comprising 2 weeks of training and no competitive matches), and competition II (12 weeks, January–April, Table 1). The microcycles comprised 5–11 training sessions. In total, 2,591 training sessions from 36 microcycles were included in the analysis. When individual training load was not available for the entire week, weekly data from that player were excluded from the analysis. Throughout the season, the players reported individual RPE using Borg’s category ratio scale (CR10) for each training session or match. Internal training load (session-RPE, sRPE) was determined by multiplying the duration of the training session (in minutes) by the CR10 score (12). The method has been proposed for monitoring internal training load in soccer (15). The duration of training sessions included the entire session (from warm-up to cool-down activities), whereas for matches, the individual playing time was used. When 2 training sessions were performed on the same day, the training loads were summed for the whole day to represent a daily training load score. Individual training load was computed on a daily basis, and the average sRPE of the week was calculated. The sum of all training sessions of the week was computed to obtain the weekly training load.

394

the

Additionally, the competitive microcycles were divided in 2 parts, comprising the first 3 sessions (devoted to recovery, conditioning, position-specific training, and team preparation) and the last 2 training sessions (tapering for the following match). The duration (in minutes) of the training sessions was summed to quantify weekly training volume. Perceived fatigue was assessed by a short questionnaire (10) every week. The questionnaire included 8 questions that focused on 8 items: training exertion, sleep quality, muscle soreness, infection/illness, concentration, training efficiency, anxiety/irritability, and general stress. Each question was assessed on a 7-point scale: from not at all (1 point) to very much (7 points). The responses were summed to obtain a total score of fatigue. Lower scores represent a better perception of general well-being, whereas higher scores indicate increased perception of fatigue. The questionnaire had been originally validated as a sensitive tool to variations in training load and performance in swimmers, and it has been applied in different sports (3,9,11). Coefficients of variation of 2.1% for the total fatigue score have been reported in rugby players (11). The questionnaires were applied before the start of the last training session of the week. Thus, during competition I and competition II, the fatigue scores were obtained on the day before the match. Statistical Analyses

The results are presented as mean 6 SD. Coefficients of variation within weekly cycles were calculated for training load scores by dividing the SDs of the scores of all training sessions of the week by the corresponding mean value of each player. Also, coefficients of variation were calculated between weekly cycles for training load and fatigue scores. Differences in training load scores within the microcycles were modeled with repeated-measures analysis of variance (ANOVA). Differences in training load and fatigue scores between the different phases of the season, and also between match-related contextual variables (result, location, and opponent’s level) were modeled by one-way ANOVA. Bonferroni adjustments for multiple post hoc comparisons were used. General linear models were used to analyze how training load scores, fatigue scores, and training volume progressed throughout consecutive weeks of training (the week was determined as the main effect). Standardized differences in means (effect sizes, d) were computed for pairwise comparisons. The magnitude effect sizes were classified as trivial (d , 0.2), small (d = 0.2–0.6), moderate (d = 0.6–1.2), large (d = 1.2–2.0), very large (d = 2.0–4.0), and extremely large (d . 4.0) (14). Statistical significance was set at p # 0.05.

RESULTS During the season, the team lost 12 matches, drew 8, and won 6. Factors such as the result of the previous match, match location, and the level of the opponent team influenced training load scores (Figure 1). Weekly training

TM

Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

TABLE 1. Description of weekly training (T1–T7) during the different phases of season. T1 Topic

T2

T3

T4

Topic

N

Topic

N

Upper-body and/or lowerbody resistance training Positionspecific training

1

Developing aerobic capacity

2

Developing aerobic power or friendly game

2

1

Team organization

1

Developing/ maintaining reaction speed and acceleration

1

Developing aerobic power

2

Competition I

Developing/ maintaining aerobic capacity

1

2

Preparation II

Developing aerobic capacity

1

Competition II

Developing/ maintaining aerobic capacity

1

Developing/ maintaining aerobic power and/or maximal power Developing aerobic power and/or maximal power Developing/ maintaining aerobic power and/or maximal power

N

Topic Upper-body resistance training + maintaining aerobic capacity Game

N

Topic

N

1

Active recovery

1

1

Rest

0

2

Positionspecific training

1

Team organization

1

Developing reaction speed and acceleration

1

Game

1

Rest

0

2

Positionspecific training

1

Team organization

1

Developing/ maintaining reaction speed and acceleration

1

Game

1

Rest

0

the

2

Topic

T7

Journal of Strength and Conditioning Research

TM

395

| www.nsca.com

VOLUME 30 | NUMBER 2 | FEBRUARY 2016 |

Developing aerobic capacity

N

T6

N

Preparation I

Topic

T5

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

Internal Training Load in Soccer

Figure 1. Influence of match-related contextual variables, including location (A), opponent’s level (B), and result (C) of the previous and following match, in training load scores. Values are expressed as mean 6 SD. *Different from away match, p , 0.01. #Different from medium-level opponent, p # 0.05. §Different from winning, p # 0.05.

load after losing or drawing a match was higher than that after winning a match (2,342 6 987 and 2,395 6 613 vs. 1,877 6 392 AU; p # 0.05; d = 0.30–0.45). When preparing for a home match, weekly training load was higher than that for an away match (2,622 6 685 vs. 2,025 6 629 AU; p , 0.001; d = 0.41). Concomitantly, after playing an away match, weekly training load was higher than that after a home match (2,493 6 821 vs. 2,153 6 577 AU; p # 0.05; d = 0.23). After playing against a top team, the average sRPE in 3 days after the match was lower than that after playing a medium level team (490 6 150 vs. 589 6 112 AU; p = 0.005; d = 0.35). Also, when preparing to play against a medium level team, average sRPE during the week was higher than that before playing against a top or bottom team (492 6 119 vs. 389 6 81 and 432 6 72 AU; p , 0.01; d = 0.29–0.45). Weekly training load scores decreased as the season progressed (p , 0.001; Figure 2), but no changes were

observed for fatigue scores. Within week-cycles, the coefficients of variation for weekly training loads scores ranged from 4 to 48% and for sRPE from 5 to 72%. Similarly, the coefficients of variation for weekly training loads and fatigue scores throughout the season ranged from 29 to 49% and 18 to 44%, respectively. The weekly training load changed during the 4 phases of the season (Figure 3A). During preparation I, weekly training volume, average sRPE, and weekly training load were higher than during the other phases of the season (p , 0.001; d = 0.48–0.54). In contrast, weekly training load was lower during preparation II than during competition I and competition II (p # 0.05; d = 0.44–0.79). Additionally, during competition I, average sRPE was higher than during preparation II (p , 0.001; d = 0.07), and weekly training volume was higher than during competition II (p , 0.001; d = 0.44). No significant differences were detected for the fatigue score between the 4 phases of the season.

Figure 2. Weekly training load and fatigue scores for highly trained U19 soccer players throughout the season. Values are expressed as mean 6 SD. Bold black line represents linear regression.

396

the

TM

Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

the

TM

Journal of Strength and Conditioning Research

| www.nsca.com

Figure 3. Average session rating of perceived exertion (sRPE), weekly sRPE, and training volume of the microcycles (A) and typical sRPE distribution in the microcycle (B) for different phases of year plan. Values are expressed as mean 6 SD. *Different from all periods, p , 0.01. #Different from preparation II, p # 0.05. §Different from competition II, p # 0.05.

The pattern of training load distribution within and between microcycles varied for the different phases of the season (Figure 3B). Overall, higher daily training load values were observed during preparation I. During the competition periods, moderate effect sizes were observed for sRPE scores between training sessions and matches (439 6 98 vs. 736 6 106 AU; p , 0.001; d = 0.822). During competition I, sRPE was similar between the second day of the week and matches (774 6 319 vs. 741 6 116 AU). Otherwise, during the competition periods, small-to-moderate effect sizes were detected for sRPE scores between matches and the other days of the microcycle (p , 0.001; d = 0.35–0.94). Additionally, during the competition periods, higher sRPE scores were reported in the first 3 days of the microcycle compared with the last 2 training sessions of the microcycle (p , 0.01; d = 0.70).

DISCUSSION The main finding of this study was that internal training load was affected by match-related contextual variables such as the result of the previous match, the opponent’s level, and the location of the previous and following matches. Overall, higher training load scores were reported after a defeat or

draw, but lower training load scores were detected before and after playing against a top-level opponent. Additionally, after playing an away match, the weekly training loads were higher than after playing a home match. Thus, when planning training sessions, soccer coaches need to consider the interplay among the several variables that influence actual training responses in each individual player. In this study, the players reported different training loads throughout the annual cycle. The general decrease in training load throughout the season might have been related to changes in training activities implemented by the coach. In fact, the players included in this analysis had high exposure to matches, and the volume of training decreased as the season progressed. The players’ fitness level might have changed throughout the season, but physical performance tests were not conducted, and match-performance data were not available. It is noteworthy that coefficients of variation of 5–72% and 4–48% were observed for sRPE and weekly training loads, respectively. This means that players responded very differently to the same training sessions. To our knowledge, this is the first study reporting the variability of training loads throughout an entire season in highly trained youth soccer VOLUME 30 | NUMBER 2 | FEBRUARY 2016 |

397

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

Internal Training Load in Soccer players. The relative variability detected for training load scores throughout the season highlights the need for individualized approaches to training load prescription and monitoring. Similarly, studies using semi-automated technological systems recommended the use of individualized speed thresholds for determining external load in soccer players (2,17). In this study, individual internal training load was assessed using sRPE on a daily basis. This method has been extensively used among professional teams (5,8,13) because it is a simple and practical assessment tool for planning and periodization of training. Notwithstanding, further research is still required to determine the physiological mechanisms behind the cognitive perception of effort, which may clarify what RPE truly represents. Self-perceived fatigue was also monitored on a weekly basis to understand the players’ actual physical state the day before the match and the extent that self-perceived fatigue is associated with training load. Contrary to observations on the training load, the fatigue scores were not sensitive enough to detect differences throughout the season. Large variation in individual fatigue scores was detected within the microcycles. Additionally, the fatigue score was not associated with any of the variables considered for the following match. This might be due to the fact that the questionnaire was administered only the day before the match at which day there was some effect of tapering on players’ perception of fatigue. The time in which the questionnaire has been administered might not be adequate, but it should be noted that to date, no study tested the validity or sensibility of the fatigue assessment tool within soccer players. An additional explanation could be that players regulated their pace during the whole week to maintain the fatigue state within the “template” limits (19). Although using a different setup, previous studies in elite soccer players suggest that players may modulate their pace during matches to preserve key performance indicators (4,18). In this study, the highest sRPE scores were observed during matches, especially during competition II. However, it was not possible to conclude whether this was a consistent coaching strategy or whether it denoted the difficulty the coaches have to create training sessions as demanding as official matches. Moreover, the higher weekly training load scores reported after a defeat or draw, the lower weekly training load scores detected before and after playing against a top-level opponent, and the higher training load scores observed after playing an away match support the need for further studies that consider the training contents for analyzing training load in soccer. Similarly, a mismatch between the training load intended by the coaches and its perception from the players needs also to be considered (5). Given the limited time for recovery between competitive matches, coaches commonly opt for concentrating the most intense training sessions in the middle of the week cycle to prevent excessive loading on the immediate days

398

the

before and after competitive matches. Therefore, variations in training loads are likely to be a direct function of coaching preferences and experience, the level and experience of the players, the density of the competitive calendar, and the actual short-term contextual variables of previous and following matches.

PRACTICAL APPLICATIONS In soccer, competitive matches are played almost every week (sometimes more than once), so proper monitoring of the training loads is paramount for the quality of the training process. The sRPE is a practical low-cost tool to assess training load in soccer, but, based on the findings of our study, coaches need to take into consideration that training loads are affected by match-related parameters. The high variability in training load within the same training session highlights the need for a more individualized approach, where possible. This variability together with the relatively stable fatigue scores throughout the season indicates that players may modulate their pace to maintain performance throughout the competition.

ACKNOWLEDGMENTS The authors express their gratitude to Montpellier He´rault Sport Club and the players who participated in the study. The authors state no conflicts of interest and that the results of this study do not constitute endorsement of the product by the authors or the National Strength and Conditioning Association. Joao Brito was a Physiologist at the National Sports Medicine Programme, Excellence in Football Project, Aspetar, at the time of the manuscript completion. He is now at Health and Performance Unit, Portugese Football Federation, Lisbon, Portugal.

REFERENCES 1. Abbiss, CR, Peiffer, JJ, Meeusen, R, and Skorski, S. Role of ratings of perceived exertion during self-paced exercise: What are we actually measuring? Sports Med 45: 1235–1243, 2015. 2. Abt, G and Lovell, R. The use of individualized speed and intensity thresholds for determining the distance run at high-intensity in professional soccer. J Sports Sci 27: 893–898, 2009. 3. Atlaoui, D, Pichot, V, Lacoste, L, Barale, F, Lacour, JR, and Chatard, JC. Heart rate variability, training variation and performance in elite swimmers. Int J Sports Med 28: 394–400, 2007. 4. Bradley, PS and Noakes, T. Match running performance fluctuations in elite soccer: Indicative of fatigue, pacing or situational influences? J Sports Sci 31: 1627–1638, 2013. 5. Brink, MS, Frencken, WG, Jordet, G, and Lemmink, KA. Coaches’ and players’ perceptions of training dose: Not a perfect match. Int J Sports Physiol Perform 9: 497–502, 2013. 6. Buchheit, M. Monitoring training status with HR measures: Do all roads lead to Rome? Front Physiol 5: 1–19, 2014. 7. Buchheit, M, Al Haddad, H, Simpson, BM, Palazzi, D, Bourdon, PC, Di Salvo, V, and Mendez-Villanueva, A. Monitoring accelerations with GPS in football: Time to slow down? Int J Sports Physiol Perform 9: 442–445, 2014. 8. Carling, C and Bloomfield, J. The effect of an early dismissal on player work-rate in a professional soccer match. J Sci Med Sport 13: 126–128, 2010.

TM

Journal of Strength and Conditioning Research

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

the

TM

Journal of Strength and Conditioning Research 9. Chaouachi, A, Coutts, AJ, Wong del, P, Roky, R, Mbazaa, A, Amri, M, and Chamari, K. Haematological, inflammatory, and immunological responses in elite judo athletes maintaining high training loads during Ramadan. Appl Physiol Nutr Metab 34: 907– 915, 2009. 10. Chatard, JC, Atlaoui, D, Pichot, V, Gourne´, C, Duclos, M, and Gue´zennec, YC. Training follow up by questionnaire fatigue, hormones and heart rate variability measurements. Sci Sports 18: 302–304, 2003. 11. Elloumi, M, Makni, E, Moalla, W, Bouaziz, T, Tabka, Z, Lac, G, and Chamari, K. Monitoring training load and fatigue in rugby sevens players. Asian J Sports Med 3: 175–184, 2012.

| www.nsca.com

14. Hopkins, WG. Linear models and effect magnitudes for research, clinical and practical applications. Sportscience 14: 49–57, 2010. 15. Impellizzeri, FM, Rampinini, E, Coutts, AJ, Sassi, A, and Marcora, SM. Use of RPE-based training load in soccer. Med Sci Sports Exerc 36: 1042–1047, 2004. 16. Lago-Penas, C. The role of situational variables in analysing physical performance in soccer. J Hum Kinet 35: 89–95, 2012. 17. Lovell, R and Abt, G. Individualization of time-motion analysis: A case-cohort example. Int J Sports Physiol Perform 8: 456–458, 2013.

12. Foster, C, Florhaug, JA, Franklin, J, Gottschall, L, Hrovatin, LA, Parker, S, Eads, M, and Dodge, C. A new approach to monitoring exercise training. J Strength Cond Res 15: 109–115, 2001.

18. Nassis, GP, Brito, J, Dvorak, J, Chalabi, H, and Racinais, S. The association of environmental heat stress with performance: Analysis of the 2014 FIFA World Cup Brazil. Br J Sports Med 49: 609–613, 2015.

13. Gaudino, P, Iaia, FM, Strudwick, AJ, Gawkins, R, Alberti, G, Atkinson, G, and Gregson, W. Factors influencing perception of effort (session-RPE) during elite soccer training. Int J Sports Physiol Perform 10: 860–864, 2015.

19. Tucker, R. The anticipatory regulation of performance: The physiological basis for pacing strategies and the development of a perception-based model for exercise performance. Br J Sports Med 43: 392–400, 2009.

VOLUME 30 | NUMBER 2 | FEBRUARY 2016 |

399

Copyright © National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.