Soccer as it is played today is a physical activity that requires a high level of conditioning in addition to proficient technical and tactical skills. The game can be characterized as a predominant aerobic exercise combined with frequent intermittent short intense actions with a high rate of the anaerobic energy turnover. Reports of match analyses revealed that elite soccer players generally cover 9,500–12,000 m during a 90-minute game (4,8,14,18,24). Approximately 40% of this distance consists of high-intensity running (>14.0 km·h−1) with or without the ball (18) and 1–11% of sprinting (>19 km·h−1) (14). These sprinting exercises of short duration lead to a high rate of creatine phosphate breakdown, which is resynthesized afterward during the following lower intensity periods (3). The speed of this recovery depends on the power of the aerobic metabolism (V[Combining Dot Above]O2max) (20,21). The average work intensity of a soccer game is close to the anaerobic threshold (i.e., the highest exercise intensity at which the removal of lactate equals its production) at a heart rate (HR) of approximately 85% HRpeak (3,12). Based on the linear relationship between the HR and oxygen uptake obtained during an incremental treadmill exercise test, this HR corresponds to an average oxygen uptake of 75% V[Combining Dot Above]O2max (23). Despite the soccer game being dependent predominantly on the aerobic metabolism, it should be argued that the most decisive actions are covered by means of the anaerobic metabolism (i.e., sprinting, jumping, etc.). The intermittent character of a soccer game results in a subsequence of high-intensity activities with an accumulation of lactate and periods of low intensity during which the removal of lactate can take place. During a soccer game, the blood lactate concentration generally varies between 2 and 10 mmol·L−1 (1,8,13). This indicates that the overall intensity of the game is rather high stimulating the energy turnover through the anaerobic glycolysis and also that the measured lactate concentration depends highly on the actions preceding the blood sampling. These game characteristics (and the technical and tactical requirements) impose that the physical performance of elite soccer players is based on the combination of endurance, speed, agility, and strength. Soccer players therefore are bound to have both a high power and capacity of the aerobic and anaerobic metabolism to be able to perform at the highest level. The combined importance of these physical determinants is specific to soccer, and in this way, soccer training needs to be well balanced and structured to optimize the player performance without inducing overreaching and overtraining.
It should be noted, however, that the physiologic and metabolic demands of a soccer game have to be seen in view of the player's position. It has been reported that midfielders cover 5–15% more total distance compared with the fullbacks, forwards, and center backs and even 20–40% more distance at high intensity compared with the forwards and center backs. Forwards and full backs spent 20–40% more time sprinting compared with the midfielders and center backs (4,6,7,14,17). The results of these studies show that the different positions on the field are characterized by specific physical activities and demands. Next to motion analysis to obtain information on the physiological demands of a soccer game, laboratory and field measurements might be useful to evaluate the physiological characteristics of the players in reflection to the specific demands of the different positions on the field. The aim of this study was to test whether a laboratory test battery could distinguish these specific characteristics between players from various positions on the field. In this way, this study could contribute to identify the physical profile (i.e., strengths and weaknesses) of the players conform their position on the field and facilitate the individualization of training. It should be noted that the physical profile, next to technical and tactical skills, is an important determinant for the level of a player. Several motion analysis studies have shown that the level of player influences the activity pattern on the field. High-level players not only cover a higher total distance but they also perform a higher amount of high-intensity activities (12,14) during a soccer game compared with players of a lower level. Therefore, the individualization of training, based on the position on the field and the individual weaknesses, is important for enhancing performance during a soccer game. In line with the content of this study Sporis et al. (22) determined the fitness profile of elite Croatian soccer players. It was observed that the anthropometric features and the aerobic and anaerobic performance differed among the different playing positions. In this study, the physical profile of Belgian elite soccer players incorporating the anthropometric characteristics and aerobic and anaerobic performance was examined. Furthermore, we wanted to acquire an insight into whether the physical profile of the players varied according to the different positions on the field. Based on the abovementioned motion analysis studies (e.g., [4,6]), we hypothesized that the aerobic characteristics would be more dominant in the midfielders and the fullbacks compared with the forwards, whereas the forwards and the defenders would display a higher anaerobic power and capacity inherent to the specific characteristics and demands of the different positions on the field.
Experimental Approach to the Problem
To allow the extrapolation of the results in this study (i.e., physical and physiological profile) into the training practice of elite soccer players and teams throughout the world, one of the main conditions is that a representative group of players is used. Therefore, a large (n = 289) subject group was tested consisting of elite soccer players in the Belgian first division (championship level). Among this test group, 41 players had also been selected to play in the National Team of their country. Respectively, 11 and 9 players were active in the World Championship of 2006 in Germany and 2010 in South Africa. To obtain an insight into the physical and physiological profile, the subjects completed a battery of both laboratory-based anaerobic and aerobic tests. The anaerobic tests incorporated both sprint and jump tests in line with the anaerobic movements inherent to the soccer game. To examine reactivity and speed, a 10-m sprint was performed, whereas the agility was tested by means of a 5 × 10-m shuttle run (SR). The jump test consisted of a squat jump (SJ) and a countermovement jump (CMJ) to obtain information on specific jump strength (explosive power). In addition to the anaerobic tests, the subject performed an incremental running test to examine the aerobic performance and running speed at the anaerobic threshold. A large subject group was used to obtain a sufficient number of players for each position on the field so that a reliable representation of the physical and physiological profile of the elite Belgian soccer player could be made.
The test group consisted of 289 professional soccer players (age: 25.4 ± 4.9 years) active in the Belgian first division between 2003 and 2010. The subjects played in 1 of the 6 teams, which attended the Laboratory of Exercise Physiology of the Centre of Sports Medicine of the Ghent University Hospital (Ghent, Belgium) for the physical tests in the preseason phase. Three of the 6 teams finished in the 8 seasons between 2003 and 2010 at least 6 times among the 5 best teams in the first division and were, therefore, also active in the Champions League or Europa League/Union of European Football Associations (UEFA) Cup. The subjects were divided into 5 groups according to their self-reported best position on the field: center backs (n = 60), full backs (n = 82), midfielders (n = 68), strikers (n = 62), and goalkeepers (n = 17). This study had the approval of the Ethical Committee of the Ghent University Hospital, Ghent, Belgium. All the participants were fully informed about the nature and demands of the study, and the known health risks. They completed a health history questionnaire, signed an informed consent document, and were informed that they could withdraw from the study at any time.
The physical tests were performed 2–4 weeks before the start of the season. In this period, the club training consisted mainly of specific soccer training sessions with occasionally a basic running training. The subjects were asked to abstain from strenuous exercise 24 hours before the tests. The tests were performed between 8.30 am and 5 pm. The experimental protocol consisted of 4 parts: anthropometric measurements, sprint tests, jump tests, and an incremental running test on a treadmill. It took approximately 90 minutes for 1 subject to complete the entire test battery. Throughout the tests, the subjects were clearly instructed about the procedures, and they were verbally encouraged to give their maximal effort. During the whole testing battery, the laboratory was air conditioned at a temperature of 21° C. It took approximately 90 minutes for the subjects to complete the entire testing battery. The subjects were wearing a standard soccer kit.
Before the physical tests, body height (±0.1 cm), body weight (±0.1 kg) (Seca balance), and body fat percentage of the subjects were determined. The fat percentage was calculated by means of measurements of skinfold thickness using a Harpenden skinfold caliper. The skinfolds were measured at 10 locations (15). Two experienced test leaders assessed the anthropometric measurements throughout the entire study.
The sprint exercise consisted of 2 different exercises. First, the subjects had to sprint for 10 m with the time recorded electronically by light sensors at both 5 and 10 m (Ergo Tester, Globus, Italy). An auditory cue was used as a starting signal for the subjects. This exercise was performed twice, and the best result was retained for data analysis. In the second test, the subjects performed an SR in which the subjects had to run 5 times 10 m as fast as possible. The same auditory cue was used for the 10-m sprint. The subjects had to cross the 10-m marks with 1 foot while turning. The first test was used both to obtain information about both the reaction time and the starting speed (explosiveness) of the players, whereas the second test was used to evaluate their agility in combination with speed.
After the sprint tests, the subjects performed the SJ and the CMJ. They performed both jumps without arm movement (i.e., with the hands in the hips). For the SJ, the subjects took a deep position with a knee angle of 90°. The flexion of the hips could be chosen freely. The subjects remained in this position for 3 seconds. In the CMJ, the subjects were instructed to jump as high as possible without further guidelines concerning knee and hip angles. Each jump test was performed twice. The best result was recorded for data analysis. The jumping height was calculated from the flight height from the jumping mat measured with the Ergo Tester (Globus).
Incremental Running Test
The incremental exercise consisted of steps of 3 minutes starting from a speed of 8 km·h−1, and the speed was increased with 2 km·h−1 after each step. During the test, the slope of the treadmill was set at 1.5% (11). At the end of each step, a blood sample was taken from the fingertip to determine the concentration of lactate in the blood by means of a lactate analyzer (1500 Sport, YSI, Yellow Springs, OH, USA). When the blood lactate concentration had passed the threshold of 4 mmol·L−1, the subjects had to perform a final step to exhaustion in which the speed was increased by 1 km·h−1, but the slope of the treadmill was increased 0.5% each 30 seconds until volitional fatigue. The subjects were encouraged verbally to obtain the individual's maximal effort. During the test, the oxygen uptake was registered by means of a metabolic measurement system (Jaeger Oxycon Pro, Hochenhausen, Germany). The Jaeger Oxycon Pro and the Lactate analyzer were calibrated before each exercise test.
The time at 5 m was used as a parameter of reaction time and explosiveness, whereas the difference between the time at 5 and 10 m was considered as a measure of speed. The time for the SR was used as a measure for speed and agility.
The jumping heights for the squat and CMJs were used as a measure of anaerobic power.
Incremental Running Test
The peak performance was determined as the maximal speed (Vmax) and slope (Smax) that could be maintained for the full 30 seconds in the final step. The maximal oxygen uptake (V[Combining Dot Above]O2max) was determined as the highest 30-second average of the breath-by-breath values obtained from the metabolic measurement system. The anaerobic threshold was determined at a lactate concentration of 4 mmol·L−1 (9) and was expressed as a speed in kilometers per hour (VAnT).
The Statistical Package for Social Sciences SPSS 15.0 (SPSS Inc., Chicago, IL, USA) was used for the statistical analysis. For the 5 positions, descriptive statistics (mean values ± SD) were calculated for height, weight, fat percentage, V[Combining Dot Above]O2max, VAnT, and 5m- and 10-m times, SR, SJ, and CMJ. These parameters were compared between the 5 positions by means of a Repeated Measures analysis of variance (5 × 1). If the analysis of variance (ANOVA) was significant, Tukey post hoc tests were performed. The reliability of the anaerobic sprint and jump tests was determined by means of the test-retest method. The reliability analysis (Cronbach alpha) and intraclass correlation coefficients showed that these anaerobic tests were highly reliable. The reliability coefficients (alpha) were respectively 0.91, 0.92, 0.88, 0.90, 0.89 for the 5-m times, 10-m times, SR, SJ, and CMJ, respectively. Also, the ICCs for the 5-m times (0.88), 10-m times (0.90), SR (0.81), SJ (0.87), and CMJ (0.84) confirm the reliability of the tests. Before setting the significance level, a power analysis was performed. This analysis revealed a statistical power of 0.94, and therefore, a significance level of p < 0.05 could be used.
The overall mean height and weight of the subjects were 182.4 ± 6.0 cm and 77.4 ± 7.1 kg, respectively, with a mean body fat percentage of 11.0 ± 2.5% (Table 1). The ANOVA revealed that the goalkeepers and center backs were taller and heavier compared with the full backs and midfielders (p < 0.05). The height of the strikers did not differ significantly from the goalkeepers (p = 0.67) and defenders (p = 0.54), but on average, the former had a lower weight (p < 0.05). The body fat percentage was significantly higher in the goalkeepers compared with the 4 other positions (p < 0.05).
The parameters of the anaerobic performance are presented in Table 2. The mean time for the 5 m, as a measure of reaction time and explosiveness, in the total subject group was 1.46 ± 0.07 seconds. The 5-m time was significantly lower in the strikers compared with the goalkeepers (p = 0.03), defenders (p < 0.01), and midfielders (p = 0.04). Also, the full backs had a faster time compared with that of the defenders. The mean time between the 5- and the 10-m mark was 0.75 ± 0.06 seconds. The strikers displayed a significantly lower time compared with that of the goalkeepers (p < 0.001), defenders (p < 0.01), and midfielders (p = 0.03). For the SR (5 × 10 m), the overall mean time was 12.20 ± 0.39 seconds. The strikers were significantly faster than the goalkeepers (p < 0.01), defenders (p < 0.01), and full backs (p = 0.02), whereas the midfielders were faster than the goalkeepers (p = 0.02) and center backs (p < 0.001).
The mean jump heights for the SJ and the CMJ were 40.7 ± 4.6 and 43.1 ± 4.9 cm, respectively. The statistical analysis revealed that the goalkeepers and the center backs had a higher jumping height compared with the full backs (p < 0.001) and the midfielders (p < 0.01). Also the forwards had a higher CMJ height compared with the full backs (p =0.03) and the midfielders (p = 0.03).
The parameters of the aerobic performance are presented in Table 3. The mean peak performance for the incremental exercise test was 15.6 km·h−1 with a slope of 4.4%. The mean V[Combining Dot Above]O2max for the total subject group was 57.7 ± 4.7 ml·min−1·kg−1, and the mean running speed at the level of the anaerobic threshold (4 mmol·L−1) was 13.7 ± 1.1 km·h−1. The full backs and the midfielders had a higher V[Combining Dot Above]O2max and VAnT compared with the strikers (p = 0.03 and p = 0.02, respectively) and center backs (p < 0.01), which in turn had a higher V[Combining Dot Above]O2max and VAnT than the goalkeepers (p = 0.03). The maximal blood lactate value at the end of the incremental running test was on average 9.6 ± 1.6 mmol·L−1, and this value did not differ between the 5 positions (p > 0.05).
The purpose of this study was to gain an insight into the general fitness level of elite soccer players active in the Belgian first division. A large subject population was tested for anthropometric characteristics and for anaerobic and anaerobic performance. Furthermore, we wanted to test whether these different aspects of the general fitness level would vary according to the position of the player of the field. As hypothesized, the results of this study clearly show that the anthropometric, physical and physiological characteristics differ according to the specific demands inherent to the position on the field.
The V[Combining Dot Above]O2max that is reported in the literature varies in general between 55 and 65 ml·min−1·kg−1 for elite soccer players (8,10,16,22). The mean V[Combining Dot Above]O2max of the elite soccer players in this study (57.7 ± 4.7 ml·min−1·kg−1) was in line with these studies. The differences between the studies can be attributed to the experimental protocol (i.e., protocol of the incremental exercise, metabolic measurement system, etc.), the testing period with respect to the phase in the season or the level of the players. In this context, it should be noted that the Belgian competition is not among the strongest in Europe. However, considering the performances in the Champions League and Champions League and Europa League (formerly known as UEFA Cup), the level of the Belgian competition (ranked 14th in the UEFA country ranking) is very comparable (or even stronger compared) to Croatia (22nd) (22), Norway (27th) (10), and Sweden (24th) (8). Compared with the aforementioned studies, the performance in the squat and CMJ was slightly lower (on average 3 cm) in this study. It should be noted, however, that the mean age of the subject group in this study was lower (25.4 ± 4.9 years) in comparison with most other studies in which the mean age is approximately 28 years. This might explain the lower anaerobic performance (SJ and CMJ) compared with other studies. Soccer players are generally at their peak performance at the age of 27–30 years, and therefore, the subjects in our test group have a few more years to develop their explosiveness and strength.
The position of the goalkeeper on the soccer pitch requires relatively low demands from the aerobic energy metabolism. The goalkeeper generally covers about 4 km during a soccer game (19), and most of his actions are determined predominantly by the anaerobic power, that is, jumping and sprinting. In this study, the goalkeepers had a significantly lower V[Combining Dot Above]O2max and performance at the anaerobic threshold compared with the other positions. On the other hand, they exert a high anaerobic power, especially during the jumping test. In combination with their body size, goalkeepers are on average the tallest players in the team, and they have a high reach and they can cover a large part of the goal, which is necessary to efficiently avoid the opponent to score. In this study, the goalkeepers also had a higher body fat percentage compared with that of the other positions. This can probably be attributed to the lower amount of aerobic training. The average sprinting performance might also be in part related to the higher body fat percentage. Also, the proportional contribution of strength and sprint training in the training regime of goalkeeper, with an important role for strength training with an increase in muscle mass as a consequence, might attribute to the average speed and agility.
With regard to the field players, the center backs generally cover the shortest distance (14,17). The center backs need strong positional play with a good headplay. They perform a smaller number of explosive sprint movements during soccer games. Generally, they play only a minor role in the support of the attack, and they only move to the front for the corner kicks or the free kicks. According to the present data, their physiological profile is in line with these specific positional demands. Apart from the goalkeeper, they not only have the lowest V[Combining Dot Above]O2max and running speed at the anaerobic threshold, but their anaerobic sprint performance and agility are also rather low. This is in line with the findings of other studies with elite soccer players (2,22). The lower sprint and agility performance is probably because of the large body size, which hinders the sudden changes in direction during the SR and in extension on the pitch. Their anaerobic power for jumping is higher compared with the other positions, and in combination with their body size, this benefits their headplay, which is an important part of the defensive skills.
The midfielders and full backs generally cover the largest distance during a soccer game, on average 10.5–11.5 km (3,4,8). These players have both defensive and offensive tasks including frequent movements up and down the field. Therefore, this requires a well-developed aerobic capacity. Similar to other studies (2,22), this study revealed that these players had the highest aerobic power and performance at the anaerobic threshold compared with the other positions. In this study, it was observed that the full backs had a slightly higher sprinting ability (i.e., 5 and 10 m) but appeared to have a lower agility compared with that of the midfielders. Although the full backs predominantly need explosiveness and speed to pass an opponent along the line, midfielders need to be able to change directions more appropriately and to accelerate with changes in direction. The specific characteristics in response to these specific demands of the full back and midfielder position appear to be present in the physical profile of the players, although the differences were not significant. In addition, the jumping performance was slightly higher in the midfielders, which more frequently need to use their headplay compared with the full back.
The strikers have higher aerobic capacities (V[Combining Dot Above]O2max and VAnT) compared with the goalkeepers and defenders but lower compared with the full backs and midfielders. This is in line with the covered distance during a soccer game (9.5–10 km), which is higher than that of the defenders but lower than that of the midfielders. Attackers will cover a large amount of the total distance using short sprints, and they need a high explosiveness and speed to be able to pass the defenders. In this study, the strikers had the fastest sprinting times and the best agility. In contrast to the study of Sporis et al. (22), the defenders in our study were significantly slower compared with the strikers. This can probably be attributed to the classification of the position of the field. In our study, the full backs were separated from the central defenders. Because the full backs are faster than the central defenders, this might have influenced the results in the study of Sporis et al. (22). The strikers also had a higher jumping performance compared with that of the midfielders and the full backs but lower compared with the defenders. Strikers need a strong jumping ability to be able to control the long pass from the defenders or to head the ball through or back to one of his teammates. Therefore, strikers need a high anaerobic sprint performance and also need a relatively high aerobic performance in combination with a relatively high anaerobic jumping performance.
This study contributes to the literature on the physical and physiological profile of soccer players, especially in regard to the elite level of the subjects in combination with the large sample size. In a recent study of Sporis et al. (22), a similar research was performed in elite Croatian football players. In this study, however, only 4 different positions were considered, that is, goalkeepers, defenders, midfielders, and forwards. It should be noted, however, that the position of a full back is characterized by different physical demands compared with a center back (7). Therefore, in our opinion, the position of the full back should be considered as a separate position on the field beside the center backs, especially because it was observed in this study that the fitness profile of a full back more closely resembles that of a midfielder. The information of this study, reporting the physiological profile of elite soccer players, can be useful for the training staff into the optimization of the training process to maximize the performance level specific for the role of the player in the team. Furthermore, in team sports, where there is a tendency for early specialization concerning the position on the field, these physiological profiles can be used in the selection of a player for a position and for the development of specific training programs based on the important characteristics of the position of the field. It should be noted that motion analysis of soccer games was not included in our study. This motion analysis provides important information on the physical characteristics of a soccer game especially with regard to the different positions on the field. However, the disadvantage of these match data is that these measures are performed in a situation that strongly depends on the strength of the own team and of the opponent, tactical gameplay, and weather conditions (4,5,12,18). The measures performed in this on the other hand are conducted in a standard situation following strict guidelines, and therefore, these measures provide valuable insights into the physical characteristics of the players. The combination of both laboratory measures and motion analysis on the field could provide important additional information on the physical demands of the positions on the field and on the main factors that determine the individual performance.
From this study, it could be concluded that the physical profile of the elite soccer players in the Belgian competition is specific inherent to the specific demands of the position on the field. It is not clear, however, whether the process of obtaining a specific position on the field is a matter of natural selection, based not only on the physical profile but also on technical-tactical skills, team cohesiveness and even match performance, or whether the specific physical profile is a consequence of the specific demands of the position itself which the player has grown into. In the modern game of soccer, however, it is clear that adult players need to be trained specifically according to their position on the field to optimize the performance of the team. The physical profile provided in this study can help the coaches in identifying the physical characteristics, which are important for the different positions to optimize the training and performance of each individual player. On the other hand, the physical profile can also help the coaches to direct younger players to the position that corresponds best to their physical profile.
The results of this study do not constitute endorsement of the product by the authors or the National Strength and Conditioning Association. This research was performed without fundings.
1. Agnevik G, ed. Fotboll. Idrottsfysiologi, rapport no.7. Stockholm, Sweden: Trygg-Hansa, 1970.
2. Al-Hazzaa HM, Almuzaini KS, Al-Refeaee SA, Sulaiman MA, Dafterdar MY, Al-Ghamedi A, Al-Khuraiji KN. Aerobic and anaerobic power characteristics of Saudi elite players. J Sports Med Phys Fitness 41: 54–61, 2001.
3. Bangsbo J. The physiology of soccer: With special reference to intense intermittent exercise. Acta Physiol Scand 15(Suppl. 619): 1–156, 1994.
4. Bangsbo J, Norregaard L, Thorsoe F. Activity profile of competition soccer. Can J Sports Sci 16: 110–116, 1991.
5. Carling C, Dupont G, Le Gall F. The effect of a cold environment of physical activity profiles in elite soccer match-play. Int J Sports Med 32: 542–545, 2011.
6. Dellal A, Wong DP, Moalla W, Chamari K. Physical and technical activity of soccer players in the French First League—with special reference to their playing position. Int Sportmed J 11: 278–290, 2010.
7. DiSalvo V, Baron R, Tschan H, Montero FJ, Bachl N, Pigozzi F. Performance characteristics according to playing position in elite soccer. Int J Sports Med 28: 222–227, 2007.
8. Ekblom B. Applied physiology of soccer. Sports Med 3: 50–60, 1986.
9. Heck H, Mader A, Hess G, Mücke S, Müller R, Hollmann W. Justification of the 4-mmol/l lactate threshold. Int J Sports Med 6: 117–130, 1985.
10. Hoff J, Helgerud J. Maximal strength enhances running economy and aerobic endurance performance. In: Football (Soccer). Hoff J., Helgerud J., eds. Norwegian University of Science and Technology, Trondheim, Norway, 2002.
11. Jones AM, Doust JHA. 1% Treadmill grade most accurately reflects the energetic cost of outdoor running. J Sports Sci 14: 321–327, 1996.
12. Krustrup P, Mohr M, Ellingsgaard H, Bangsbo J. Physical demands during an elite female soccer game: Importance of training status. Med Sci Sports Exerc 37: 1242–1248, 2005.
13. Krustrup P, Mohr M, Steensberg A, Bencke J, Kjaer M, Bangsbo J. Muscle and blood metabolites during a soccer game: Implications for sprint performance. Med Sci Sports Exerc 38: 1165–1174, 2006.
14. Mohr M, Krustrup P, Bangsbo J. Match performance of high-standard soccer players with special reference to development of fatigue. J Sports Sci 21: 519–528, 2003.
15. Parizkova. Body fat and physical fitness. The Hague, The Netherlands. Martinus Nijhoff BV, 1977.
16. Puga A, Ramos J, Agostinho J, Lomba I, Costa O, Defreitas F. Physical profile of a first division Portuguese professional soccer team. In: Science and Football. Reilly T., Clarys J., Stibbe A., eds. London, United Kingdom: E&FN Spon, 1993. pp. 40–42.
17. Rampinini E, Coutts AJ, Castagna C, Sassi R, Impellizzeri FM. Variation in top level soccer march performance. Int J Sports Med 28: 1018–1024, 2007.
18. Rampinini E, Impellizzeri FM, Castagna C, Coutts AJ, Wisloff U. Technical performance during soccer matches of the Italian Serie A league: Effect of fatigue and competition level. J Sci Med Sports 12: 227–233, 2009.
19. Reilly T. Motion characteristics. In: Football (Soccer). Ekblom B., ed. Oxford, United Kingdom: Blackwell scientific, 1994. pp. 31–42.
20. Sahlin K, Harris RC, Hultman E. Resynthesis of creatine phosphate in human muscle after exercise in relation to intramuscular pH and availability of oxygen. Scand J Clin Lab Invest 39: 551–557: 1979.
21. Spencer M, Bishop D, Dawson B, Goodman C. Physiological and metabolic responses of repeated-sprint activities—Specific to field-based team sports. Sports Med 35: 1025–1044, 2005.
22. Sporis G, Jukic I, Ostojic SM, Milanovic D. Fitness profiling in soccer: Physical and physiological characteristics of elite players. J Strength Cond Res 23: 1947–1953, 2009.
23. Stolen T, Chamari K, Castagna C, Wisloff U. Physiology of soccer. Sports Med 35: 501–536, 2005.
24. Whitehead EN. Conditioning of Sports. Yorkshire, United Kingdom: EP Publishing Co. Ltd., 1975. pp. 40–42.