Introduction
Many factors are relevant in determining the success of a soccer player, and the requirements for high-level play are multifactorial (27 5,14,30 performance profiling program (9,27
The physical composition of the body (body fat, body mass, and fat-free mass) is also relevant in the preparation of players for competitive performance . It is an important component of fitness because excess adipose tissue acts as dead weight in common game activities such as running and jumping where body mass must be lifted repeatedly against gravity (25 12 performance characteristics of soccer players are therefore influenced by differences in body size and composition (33 performance should take these factors into account. In addition, the physical composition of the body is influenced by training, and the balance between energy intake and expenditure is therefore of interest to practitioners (4
The measurement of body composition in elite soccer players has received widespread attention, especially with respect to playing positions and to the description of changes across the playing season. For example, research has shown variations in estimated body fat and body mass across playing positions with the highest values generally found in goalkeepers (26,27 performance suggests that training and fitness testing should be tailored specifically to individual positional role (e.g., separation between central and lateral midfielders) rather than simply differentiating between forward, midfield, and defending positional groups because each positional role has its own unique physical demands (3
Previous studies having examined the seasonal effects on body composition include investigations on professional English (7,10,36,37 15 6 1 18 performance (23 35
In this study on variations in body composition in senior professional soccer players, several questions were addressed. First, do measures of body composition (estimated body fat, body mass, and fat-free mass) vary according to positional role (6 positions rather than the basic division into 4 positional groups commonly used in other studies)? Second, at which points of the season do variations in measures of body composition occur? Third, do fluctuations across the season vary among positional groups? Fourth, is exposure time in competition and training linked to variations in body composition? Finally, to what extent does body composition vary between seasons?
Methods
Experimental Approach to the Problem
Body composition is a significant component of fitness for soccer and can significantly vary across positional groups and during the course of the season (26
Subjects
A total of 30 senior professional soccer players, mean age 24.4 (±4.1) years, height 182.1 (±5.8) cm, and weight 76.8 (±5.8) kg, who attended a French League 1 professional soccer club participated. Over the course of the present study, the club achieved 3 top-10 finishes in the league and also participated in the 2006-2007 UEFA Champions League tournament. On average, players trained 5 times per week and participated in up to 50 competitive matches per season. Body composition in players belonging to the first- and reserve-team squads combined was assessed as part of standard fitness testing over 3 entire seasons (2005-2006, 2006-2007, and 2007-2008).
All subjects received a clear explanation of the study, including the risks and benefits of participation, and written consent was obtained. The study was approved by the Independent Institutional Review Board of LOSC Football Club according to the Guidelines and Recommendations for European Ethics Committees by the European Forum for Good Clinical Practice (1997).
Procedures
All players underwent a battery of anthropometric and fitness tests at the beginning of the competitive season. Body composition measurements (skinfolds and body mass) were performed on 14 occasions across each season between June and May. The measurements were performed at the start of the first conditioning period (June) and once every month until the end of the season. An extra measurement was made just before the mid-season break (end of December). Only players (n = 26) for whom full datasets were available were included in the final analysis.
To ensure standardization of test administration across the entire study period, testing was scheduled at the same time of day using the same apparatus. Body composition was assessed between 0800 and 1200 hours, with the athletes presenting in a fasted state. All test procedures over the entire study period were undertaken by the same experienced physician specialized in sports medicine.
Body Composition
Body composition measurements were obtained using portable measurement devices and standardized laboratory procedures. Calibrated precision weighing scales (Holtain, Instruments Ltd, Crymych, United Kingdom) were used to obtain body mass to the nearest 0.1 kg (32 22 26 2 (16 11 31
Recording of Exposure Time
A log of the participation time in training and matches for each month (including first-team and reserve-team games) for the squad of professional players was kept by the same physician over the entire 2007-2008 season to investigate the relationship between body composition and exposure time to play.
Statistical Analyses
All results are reported as means and standard deviations (mean ± SD ) calculated by conventional procedures unless otherwise stated. All calculations were performed using Microsoft Excel (Microsoft, Redmond, WA, USA) and Statistica software (SSPS, Chicago, IL, USA).
The intraclass correlation coefficients for test-retest reliability and typical errors of measurement were calculated for the sum of 4 skinfolds and body mass based on repeated measurements in 15 players. Values obtained for test-retest reliability and typical error of measurement were 0.97 and 0.99 and 3.2% and 0.2%, respectively.
A 1-way repeated-measures analysis of variance (ANOVA) was used to examine changes in mean values for each squad of players over the course of the 2007-2008 season, and the Bonferroni adjustment was used for multiple comparisons. The same procedure was used to examine interseasonal differences in data obtained in players who had participated over the 3 consecutive seasons (2005-2006, 2006-2007, 2007-2008) and to investigate intraseasonal values across the 4 positional groups during the 2007-2008 season. One-way ANOVA was used to compare mean data between the 6 positional roles during the 2007-2008 season. Follow-up univariate analyses using Bonferroni-corrected pair-wise comparisons were used where appropriate. Cohen's effect size conventions (8 p ≤ 0.05.
Results
Body Composition According to Positional Role
To provide a detailed comparison of mean estimated body fat and mass values, participants were divided into subgroups according to the individual positional role of each player. Six specific roles were included for analysis: goalkeepers, central and lateral defenders, central and lateral midfielders, and forwards. The mean of the measures obtained on each of the 14 occasions was calculated. Analysis of variance demonstrated a significant difference in both estimated body fat and body mass across groups (p < 0.001, ES 1.0). The follow-up analyses revealed greater estimated body fat values in goalkeepers vs. all other positions (p < 0.001), in lateral midfielders vs. central midfielders and vs. central and lateral defenders (p < 0.001), and in forwards vs. both central midfielders (p < 0.01) and central defenders (p < 0.01). Figure 1 presents estimated body fat according to playing position. The post hoc analyses on body mass demonstrated higher values in goalkeepers vs. all other positions (p < 0.001), in forwards vs. all positions (except goalkeepers) (p < 0.001), in central-midfielders vs. lateral midfielders (p < 0.001) and lateral defenders (p < 0.001), and in central defenders vs. lateral defenders (p < 0.001) and lateral midfielders (p < 0.001).
Figure 1: Comparison of mean estimated body fat values across 6 positional roles in professional soccer players (n = 26). GL = goalkeeper; CD = central-defender; LD = lateral-defender; CM = central-midfielder; LM = lateral-midfielders; F = forward. *Significant difference vs. all positional roles, p < 0.001. †Significant difference vs. LD, CD, and CM, p < 0.001. ‡Significant difference vs. CD and CM, p < 0.01.
Intraseason Variation in Body Composition
Table 1 presents body composition values (percent body fat, body mass, fat-free mass) across all participants. A significant difference was revealed by ANOVA in both estimated body fat (p < 0.001, ES 0.98) and fat-free weight (p < 0.001, ES 0.93) across the 5 different points in the season. No difference was reported for body mass values (p = 0.071, ES 0.35). The post hoc tests on estimated body fat demonstrated a reduction from preseason to mid-season values (p < 0.01) and an increase from mid- to end-season values (p < 0.01). Post hoc analyses on fat-free mass demonstrated a significant increase from preseason to mid-season values (p < 0.001) and a significant increase from preseason to end-season values (p < 0.01).
Table 1: Percent body fat, body mass, and fat-free mass in professional soccer players and according to 4 playing positions at 5 different points across the season (n = 26)
Further investigation of the dataset showed that the largest loss in estimated body fat (−0.3%) was between October and November, whereas the largest loss in body mass (−0.3 kg) and in fat-free mass (−0.5 kg) was between January and February and May and June, respectively. The largest gain in estimated body fat (+0.3%) was between May and June, whereas the largest gain in body mass (+0.6 kg) and fat-free mass (+0.5 kg) was between February and March.
Intraseason Variation in Body Composition According to Positional Group
The measures of body composition recorded at 5 separate points of the season (preseason, end of preseason training, mid-season, end of mid-season break, and end of season) were investigated across 4 basic positional groups (goalkeepers, defenders, midfielders, and forwards) (Table 1 ). A significant difference in estimated body fat (p = 0.002, ES = 0.93) and fat-free mass (p = 0.009, ES = 0.92) was observed in midfielders across the season. A significant difference in fat-free mass was also observed in defenders (p = 0.008, ES = 0.78). Post hoc analyses revealed an increase in estimated body fat from preseason to mid-season (p < 0.01) and mid- to end-season (p < 0.01) in midfielders and an increase in fat-free mass from preseason to end-season (p < 0.01) in defenders. In goalkeepers, statistical significance was also reached (p = 0.042, ES = 0.53), but this was not lower than the adjusted alpha level after Bonferroni's correction. No intraseasonal differences in body mass were reported for any of the playing positions.
Interseason Variation in Body Composition
Analyses of variation over 3 seasons in 9 players belonging to the professional squad was undertaken using the mean values of percent body fat, body mass, and fat-free mass from each season. Comparisons were made on the preseason and end-season values for each season. The difference between these 2 time points (end-season minus preseason values) was also compared between seasons. Statistical significance was reached for the difference over the 3 seasons between preseason and end-season fat-free mass values (p = 0.043, ES = 0.52) and in preseason mass values (p = 0.032, ES = 0.6), but these were not lower than the adjusted alpha level after Bonferroni's correction. In addition, the difference between preseason and end-season percent body fat (p = 0.069, ES = 0.41) approached significance. Figure 2 presents the mean preseason and end-season percent body fat values over the 3 seasons.
Figure 2: Variation in mean preseason and end-season percent body fat values over 3 consecutive sesons in professional soccer players (n = 9).
Relationship Between Participation Time and Variation in Body Composition
No significant association was observed between participation time (combined training and match exposure time) and estimated body fat (r = 0.12, p > 0.05), body mass (r = 0.48, p > 0.05), and fat-free mass (r = 0.42, p > 0.05) values over the course of the season.
Discussion
The purpose of this study was to investigate several factors that could explain variations in body composition in a professional soccer team. The major findings were that body composition in a group of professional soccer players varied substantially according to the specific positional roles of players and that there were significant variations across the season with changes dependant on positional group but not exposure time. No significant interseasonal differences were observed.
In elite soccer, previous research has shown that percent body fat and body mass values differ greatly between goalkeepers and the other outfield positions (26 35 20,24 performance in elite soccer has adopted this approach (3 performance in the same way that up to now has not been achieved.
Body composition in soccer players is likely to change during the course of the competitive season as a result of training and competition, habitual activity, and diet (23 2,29
Of note, the analysis of the 14 separate measures showed that the months in which the largest changes occurred (decreases and gains) in respective body fat, body mass, and fat-free mass varied considerably. It is therefore of interest to note that a change in body mass, for example, is not always a good indicator of a change in body composition, and other parameters such as skinfold measurements should be used to provide supplementary information. Furthermore, this finding also suggests a need for continuous monitoring over the entire season, such as on a monthly basis (as achieved in the present study), rather than at a limited number of time points.
In general, these observations contradict the findings of Clarke et al (7 1
The intraseason fluctuations in estimated body fat and free-fat mass were shown to vary across the 4 positional groups. We can therefore recommend that coaches take into account player positions when interpreting variations in body composition data across the season. However, no significant change was reported between the start and end of preseason training across any of the positional groups. Previous studies have shown that there is a decreasing trend in body fat and mass across the season, notably during the conditioning period (21,23,37 15 29
No significant interseasonal differences in estimated body fat or fat-free mass were observed, suggesting that body composition profiles in the present players did not vary from 1 season to the next. In contrast, analyses of body composition in Japanese youth players reported significant differences in estimated body fat and fat-free mass across 2 seasons (35 17 19
The examination of the relationship between exposure time to play and measures of body composition did not demonstrate any significant association. From this observation, we can suggest that variations in body composition across the season are perhaps more likely to be linked to the intensity or type of training undertaken, time off through injury, or perhaps to the density or importance of matches leading up to the taking of measures. A study on the effects of intense athletic training on fat patterning in runners showed that reductions in body fat across various regions of the body were specific to the muscular groups used during training (13 19
The major strength of the present study was the availability of full datasets and therefore repeated measurements obtained by the same physician for 26 players across 14 different points of the season and in 9 players over 3 consecutive seasons. Changes in player personnel resulting from both temporary and permanent interclub transfers and player availability between test-to-test measurements tend to compromise repeated observations of body composition and fitness in elite soccer (7
The present study design had 2 major weaknesses. First, the limitations and possible errors in reporting measures (across studies) of body fat as a percentage rather than the sum of skinfolds (4,34 26
Second, the limitations of the present equations used to estimate body fat in soccer players should be discussed. These equations have been based on measures undertaken in general populations and their validity for usage in soccer players has been questioned (34 28 37
Practical Applications
In elite soccer, mean body composition values are usually compared across 4 basic positional groups. A differentiation of measures according to individual positional role within these groups has not been achieved. In addition, there is limited knowledge on intraseasonal variations in body composition across the playing season and the possible effects of positional group or exposure time to play on measures have not been investigated. The present study was the first to address these features.
Several novel findings were reported, allowing us to make practical recommendations to aid in the analysis and interpretation of body composition data in professional soccer. First, to ensure an objective comparison of the data obtained from body composition analyses across a squad of professional soccer players, results should be interpreted according to the specific positional role of the player. Second, substantial variations in body composition occur in a squad of players at different points during a competitive season, suggesting a need for continuous monitoring of this component of fitness across the entire season. Third, the analysis of data on intraseason variations in body composition should be extended to include the effects of individual positional group because this variable can influence both the timing and extent of variations in measures.
Although this study has elicited useful results and allowed us to make several relevant recommendations, further work is needed to confirm whether the present advice could be applied in other populations such as elite youth and female soccer players. Nevertheless, it is hoped that coaches and sports scientists may use the present results as a tool for a better understanding of the role and interpretation of body composition measures as part of the fitness profiling of professional soccer players and as a means for identifying areas that require further exploration.
Acknowledgments
The authors express their appreciation to the players for their involvement in this study. The authors have no conflicts of interest that are directly relevant to the content of this article.
References
1. Aziz, AR, Tan, FYH, and Teh, KC. Variation in selected fitness attributes of professional soccer players during a league season. In:
Science and Football V . Reilly, T, Cabri, J, and AraĂºjo, D, eds. London: Routledge, 2005. pp. 134-138.
2. Brady, K, Maile, A, and Ewing, B. An investigation into the fitness of professional soccer players over two seasons. In:
Science and football III . Reilly, T, Bangsbo, J, and Hughes, M, eds. London: E & FN Spon, 1997. pp. 118-112.
3. Carling, C, Bloomfield, J, Nelsen, L, and Reilly, T. The role of motion analysis in elite soccer: Contemporary
performance measurement techniques and work-rate data.
Sports Med 38: 839-862, 2008.
4. Carling, C, Reilly, T, and Williams, AM.
Performance assessment for field sports. London, UK: Routledge, 2008.
5. Carling, C, Le Gall, F, Reilly, T, and Williams, AM. Do anthropometric and fitness characteristics vary according to birth date distribution in elite youth academy soccer players?
Scand J Med Sci Sports 7: 3-9, 2009.
6. CasajĂºs, JA. Seasonal variation in fitness variables in professional soccer players.
J Sports Med Phys Fit 41: 463-467, 2001.
7. Clark, NA, Edwards, AM, Morton, RH, and Butterly, RJ. Season-to-season variations of physiological fitness within a squad of professional male soccer players
J Sports Sci Med 7: 157-165, 2008.
8. Cohen, J.
Statistical power analysis for the behavioural sciences . 2nd ed. Hillsdale, NJ: Lawrence Erlbaum, 1988.
9. Diniz Da Silva, C, Bloomfield, J, and Bouzas Marins, JC. A review of stature, body mass and maximal oxygen uptake profiles of U17, U20 and first division players in Brazilian soccer.
J Sports Sci Med 7, 309-319, 2008.
10. Dunbar, G and Power, K. Fitness profiles of English professional and semi professional soccer players using a battery of field tests. In:
Science and football III . Reilly, T, Bangsbo, J, and Hughes, M, eds. London: E and FN Spon, 1997. pp. 27-31.
11. Durnin, JVGR and Wormersley, J. Body fat assessed from total body density and its estimation from skinfold thickness: Measurements on 481 men and women aged from 16 to 72 years.
Br J Nutr 32: 77-97, 1974.
12. Duthie, GM, Pyne, DB, and Hooper, S. The applied physiology and game analysis of rugby union.
Sports Med 33: 973-991, 2003.
13. Gabbett, T and Domrow, N. Relationships between training load, injury, and fitness in sub-elite collision sport athletes.
J Sports Sci 25: 1507-1519, 2007.
14. Gil, SM, Gil, J, Ruiz, F, Irazusta, A, and Irazusta, J. Selection of young soccer players in terms of anthropometric and physiological factors.
J Sports Med Phys Fit 47: 25-32, 2007.
15. Hoshikawa, Y, Kanno, A, Ikoma, T, Muramatsu, M, Iida, T, Uchiyama, A, and Nakajima, Y. Off-season and pre-season changes in total and regional body composition in Japanese professional soccer league players In:
Science and Football V . Reilly, T, Cabri, J, and AraĂºjo, D, eds. London: Routledge, 2005. pp. 160-165.
16. International Society for the Advancement of Kinanthropometry.
International standards for anthropometric assessment . North West University (Potchefstroom Campus), Potchefstroom 2520, South Africa: ISAK (revised 2006), 2001.
17. Koutedakis, Y. Seasonal variation in fitness parameters in competitive athletes.
Sports Med 19: 373-392, 1995.
18. Kutlu, M, Sofi, N, and Bozkus, T. Changes in body compositions of elite level amateur and professional soccer players during the competitive season.
J Sports Sci Med 7: 53, 2007.
19. Legaz, A and Eston, R. Changes in
performance , skinfold thicknesses, and fat patterning after three years of intense athletic conditioning in high level runners
Br J Sports Med 39: 851-856, 2005.
20. Matkovic, BR. Misigoj-Durakovic, M, Matkovic, B, Jankovic, S, Ruzic, L, Leko, G, Kondric, MG, and Kondric, M. Morphological differences of elite Croatian soccer players according to the team position.
Coll Antropol 7: 167-174, 2003.
21. Mercer, TH, Gleeson, NP, and Mitchell, J. Fitness profiles of professional soccer players before and after pre-season conditioning. In:
Science and football III . Reilly, T, Bangsbo, J, and Hughes, M, eds. London: E and FN Spon, 1997. pp. 112-117.
22. Norton, K, Marfell-Jones, M, Whittingham, N, Kerr, D, Carter, L, Saddington, K, and Gore, C. Anthropometric assessment protocols. In:
Physiological Tests for Elite Athletes: Australian Sports Commission . Gore, CJ, (ed.). Champaign, IL: Human Kinetics, 2000. pp. 66-85.
23. Ostojic, SM. Seasonal alterations in body composition and sprint
performance of elite soccer players.
J Exerc Phys 6, 11-14, 2003.
24. Ramadan, J and Byrd, R. Physical characteristics of elite soccer players.
J Sports Med Phys Fitness 27: 424-428, 1987.
25. Reilly, T. Fitness assessment. In:
Science and Soccer . Reilly, T, (ed.). London: E & FN Spon, 1996. pp. 25-49.
26. Reilly, T.
The science of training-Soccer . London, UK: Routledge, 2008.
27. Reilly, T, Bangsbo, J, and Franks, A. Anthropometric and physiological predispositions for elite soccer.
J Sports Sci 18: 669-683, 2000.
28. Reilly, T, Maughan, RJ, and Hardy, L. Body fat consensus statement of the steering groups of the British Olympic Association.
Sports Exerc Inj 2: 46-49, 1996.
29. Reilly, T and Peiser, B. Seasonal variations in health-related human physical activity.
Sports Med 36: 473-485, 2006.
30. Reilly, T, Williams, AM, Nevill, A, and Franks, A. A multidisciplinary approach to talent identification in soccer.
J Sports Sci 18: 695-702, 2000.
31. Siri, WE. Body composition from fluid spaces and density: Analysis of methods. In:
Techniques for Measuring Body Composition . Brozeck, J and Henschel, A, eds. Washington, DC: National Academy of Sciences, 1961. pp. 223-244.
32. Stewart, A and Eston, RG. Surface anthropometry. In:
Sport and Exercise Physiology Testing: Guidelines. Volume I Sport Testing . Winter, EM, Jones, AM, Davison, RCR, Bromley, PD, and Mercer, TH, (eds.). London: Routledge. 2006. pp. 76-83.
33. Stølen, T, Chamari, K, Castagna, C, and Wisløff, U. Physiology of soccer: an update.
Sports Med 35: 501-536, 2005.
34. Sutton, L, Scott, M, and Reilly, T. Validation of a new anthropometric equation for the prediction of body composition in elite soccer players. In:
Proceedings of the 1st World Conference on Science and Soccer . Liverpool, UK, 15-16 May, 2008. 2008, p. 77.
35. Tahara, Y, Moji, K, Tsunawake, N, Fukuda, R, Nakayama, M, Nakagaichi, M, Komine, T, Kusano, Y, and Aoyagi, K. Physique, body composition and maximum oxygen consumption of selected soccer players of Kunimi High School, Nagasaki, Japan.
J Physiol Anthropol 25: 291-297, 2006.
36. Thomas, V and Reilly, T. Fitness assessment of English league soccer players through the competitive season.
Brit J Sports Med 13: 103-109, 1979.
37. Wallace, J, Billows, D, George, K, and Reilly, T. Bone mineral density and body composition changes during a Premier League association football season.
J Sports Sci Med 7: 53-54, 2007.
