Several professional soccer schools for talented young players have in recent years been established in Europe. More attention is being paid to the development of young talented players in the soccer clubs, and the physical training and the tactical organization in the field seems to be implemented much earlier. Methods for analysis during match play have been developed and include heart rate (HR) measurements, and observational studies of motion analysis and running pattern (3,9,23). Activity registration has become easier and inexpensive by the use of small microcomputers in field studies (19). The activity patterns and physiological demands have therefore been well-reported in adult soccer players with regard to competition level and position on the field during competition matches (1–4,16,19,24).
However, in spite of the increasing focus on the optimization of physical and tactical skills in talented young soccer players, surprisingly few studies have been published regarding the physiological demand and the activity pattern of young soccer players (6,8,13). Moreover, these studies fail to couple the observed activity pattern and the aerobic load during match play. Klimt et al. (14) measured the HR during competition and noncompetition games in 11- to 12-yr-old boys but did not relate these measurements to the level of competition.
The purpose of the present study was to record the aerobic energy demand during match play and relate it to maturity status and competition level (elite vs nonelite) of young soccer players. Furthermore, the activity pattern and the influence of specialization due to playing position on the field were examined.
Twenty-six young male soccer players from three of the most successful clubs in Denmark were recruited. All participants and their parents gave their written informed consent, and the study was approved by the ethical committee of Copenhagen University, Denmark (KF 01-132/95). The subjects were a subgroup of participants in a longitudinal study investigating the effects of elite training on adolescents (10). Ten nonelite (12.2 ± 0.7 yr) and nine elite players in beginning of puberty (12.6 ± 0.6 yr) and seven older elite players in the end of puberty (14.0 ± 0.2 yr) participated (Table 1) (E = elite, N = nonelite, bP = beginning of puberty, eP = end of puberty). All teams employed a regular 4-4-2 formation, using four defenders, four midfielders, and two attackers. Twelve of the subjects were defenders, 13 played midfield, and 3 were attackers. Because the attackers often played in the midfield, the midfielders and attackers were treated as one group. The elite players were defined as subjects playing on the clubs best team in their age group and nonelite players as players connected to the fourth or fifth team. The participants were characterized by a high level of daily physical activity, which was assessed by interviews and questionnaires and divided into weekly organized training and hours of leisure-time activities, mostly soccer.
The study consisted of two parts: 1) physiological measurements in the laboratory and 2) a field study of the activity pattern and the aerobic load during a real competition match. Because of the pubertal development of subjects at this particular age, the time between associated testing sessions was minimized to few days.
Stature was measured by a stadiometer to the nearest 0.1 cm, and body weight was measured to the nearest 0.1 kg using a spring balance. Maturity status was assessed from testicular volume (Prader Orchidometer) by an experienced endocrinologist. Oxygen uptake (O2) was measured with the on-line system (Medical Graphics, CPX/D) using a tight face mask specially designed for children. The apparatus was calibrated at the beginning of each test using standard reference gases. HR was continuously measured every 5th second with the Polar Vantage NV™ HR monitor (Polar Electro Oy, Kempele, Finland).
After warming up, the subjects performed two submaximal tests running on a motor driven treadmill at two individually selected speeds until a steady state was reached. The speeds were chosen to give a difference in the HR of at least 10 beats·min−1. The maximal oxygen uptake (O2max) was measured after a short rest period, starting at the highest speed of the two submaximal tests. Every minute the running speed or inclination was increased until the subject was exhausted, usually within 5–7 min. The test was considered maximal if: 1) the HR during the last minute exceeded 95% of the expected maximal HR predicted by 220 − age, 2) a respiratory exchange ratio at or higher than 1.1 was reached, or 3) O2 reached a plateau and/or signs of subjective exhaustion were present. The economy of running at the two submaximal running velocities was determined as the oxygen uptake per body mass per kilometer above basal metabolic rate (BMR). The BMR was calculated using the formula of Schofield (22): [0.068·weight + 0.574·height + 2.157]MJ based on measurements from 10- to 18-yr-old boys (mean 13.7 yr). Only the best test of the two individual test values is presented in this paper. However, the majority of the players obtained a slightly better running economy at the highest speed irrespective of level and maturation. Individual linear regression equations between O2 and HR were calculated for each subject, based on the submaximal and maximal HR and O2 test values during the treadmill running. The resting HR was measured on a separate day during a 24-h recording and defined as the HR immediately after awakening.
The test matches were chosen in agreement with the players and the coach, and were representative for the season. Each half lasted 30 min for the players in the younger group and 35 min for the players in the older group. The inclusion criterion for using the match files was participation in at least 90% of the total match time. The HR was collected throughout the match with a sampling frequency of 5 s. O2 during the first and the second half (O2half) was then calculated for each subject from the mean HR and the individual linear regression equation between O2 and the HR obtained in the lab. Furthermore the relative aerobic load was calculated as: %O2max = (O2 half − BMR)·(O2max − BMR)−1·100. Simultaneous with the HR registration, the activity pattern of each individual player was observed and continuously registered on a Hewlett Packard 200LX mini computer. Five observers were trained in observing the activity pattern of the players on the basis of the definitions of movement and events described below, using videotape in the initial phase to define the categories of movement. Six categories of movement inspired by Reilly and Thomas (19) and Ali and Farrally (1) were classified on the basis of videotapes of young soccer players’ match activities: walking, jogging (nonpurposeful, slow running), cruising (running with manifest purpose and effort), sprinting (goal-directed, very fast running) movement backward (backing), and standing still. Simultaneous with the movement registration, the match events for the observed player were counted. The events defined were jumps, tackles, passes/shots, throw-ins, and headers.
Objectivity and reliability.
It was not possible to test the reliability of the observation method because of the field conditions, which exclude the use of a conventional test-retest procedure. Therefore, the objectivity (interobserver agreement) of the time-motion analysis was optimized to increase the reliability, which is necessary when valid comparisons between groups should be made (17). Thus, four test observations were made on the same player until the interobserver agreement was satisfactory. One observer did not fulfil the criterion for objectivity and was excluded from the observation team. To test the day-to-day variation of the mean HR during the matches, 10 players were tested on two occasions during the first half and six players during the second half. The mean coefficient of variation (CV = SD/mean·100) was 2.0 ± 1.6% (mean ± SD) during the first half and 1.5 ± 0.9% during the second half.
Student’s t-test was used to test differences between the nonelite players and the elite players in the beginning of their puberty, and to test differences between the elite players in their beginning of puberty and the elite players in their end of puberty. Student’s paired t-test was used to test the half effect in each group of players. The significance level was set at P ≤ 0.05. Because of the small number of subjects in the subgroups and the normal biological variation, Type II errors cannot be excluded.
The number of subjects will vary slightly in tables and figures due to repeated substitution of players in matches, poor connection between the HR transmitter and the receiver, and discomfort under test conditions. The physiological characteristics of the groups are presented in Table 1. There were no differences in the anthropometrical data between the elite and nonelite group at the beginning of puberty. The older elite group was significantly heavier, higher, and more mature compared with the younger elite group.
No differences in O2max were seen between the two youngest groups (elite/nonelite), either in absolute (mL O2·min−1) or in the body mass related (mL O2·min−1·kg−1) values (Table 1). The elite players in the end of their puberty had a significantly (P < 0.05) higher O2max (mL O2·min−1) compared with the young elite players. However, when related to body weight, no significant difference was found (P = 0.16) (Table 1). The resting HR was lower in the elite players in the end of puberty compared with young elite players, and they had a significantly higher calculated BMR.
In general, the elite players in the beginning of their puberty exercised at a higher level of O2 during the matches compared with their nonelite counterparts (Figs. 1 and 2), both with respect to absolute (mL O2·min−1) and body-mass–related values (mL O2·min−1·kg−1). Furthermore, the elite players exercised at a higher relative aerobic load (%O2max) and at a higher HR compared with the nonelite players in the same age group (Figs. 3 and 4).
The older elite players exercised at an even higher O2, both in absolute (mL O2·min−1) and in body-mass–related values (mL O2·min−1·kg−1) compared with the young elite players. In contrast to the differences observed in the absolute aerobic loads, the two elite groups exercised at identical relative aerobic loads during the matches (Fig. 3) and at the same HR level (Fig. 4). All three groups of players exercised at a lower HR during the second half (Fig. 4). Only the nonelite players showed a significant decline in O2 (mL O2·min−1·kg−1) and in relative workload (%O2) (Figs. 2 and 3).
The O2max tends to be higher in the midfield/attackers compared with defenders in all three groups of players although differences were most pronounced in the older elite group (Table 2). During match play, a similar trend was observed in the HR, i.e., the midfield/attackers were exercising at a higher level than the defenders during both the first and the second halves (Table 2).
Because of the large amount of data, only recordings of activity pattern from the first half are shown in the graphs and tables. The time spent in standing activity during the first half was significantly higher among the nonelite players than among the elite players in the young group. In contrast, the time spent in jogging was significantly higher among the elite players (Fig. 5). The differences in the ratios spent in the activity categories were further elucidated by calculation of the frequencies (occurrence of motion categories in the first half is normalized to a half length of 35 min) and the duration (mean time) of the different activity categories (Table 3). The frequency of standing activity was significantly higher among the nonelite players compared with the young elite players. No other differences were found in frequencies and durations of motion categories between the groups of players. A mean duration time of 6 s in each motion category was seen in all groups of players (Table 3).
The elite players in their beginning of puberty performed significantly more jumps (P < 0.5) than the nonelite players (3.6 ± 2.5 vs 0.9 ± 1.1). No significant differences were observed between the two elite groups in the time-motion data.
Between halves effect.
The nonelite players at the beginning of puberty had a significant increase in the ratio of time spent in standing activity in the second half compared with the first half (10.4 ± 4.3 vs 12.4 ± 4.9; only paired data recorded in the same match are tested). This was also the case for the elite players at the end of their puberty (3.4 ± 2.4 vs 5.1 ± 1.7). This group further showed a significant increase in the ratio of the time spent in walking (53.7 ± 11.3 vs 58.6 ± 11.5) and a decrease in the ratio of jogging (32.6 ± 9.4 vs 27.5 ± 8.4).
The hours of weekly physical activity in the clubs as well as in leisure time were equal among the elite and nonelite players in the beginning of their puberty. The older elite players spent more hours in the clubs than the younger elite group (P < 0.05).
In the present study, we found that elite players had a higher absolute as well as relative workload during match play compared with nonelite players. A higher absolute workload was also related to a higher maturity status. Regarding the playing position on the field, the defenders exercised at a lower level of oxygen uptake related to body mass during the match compared with midfields/attackers.
Compared with soccer players at the same age from other studies (5–7), the elite players from this study had higher O2max values (64 compared with 50–56 mL O2·min−1·kg−1), which indicate that the group was highly selected and well-trained. This was further confirmed by the fact that several of the players subsequently have been playing on the Danish national youth teams. Klimt et al. (14) found HR levels of 160–180 beats·min−1 for 11- to 12-yr-old boys (N = 15) observed during two competition matches (competition level not reported) (14). The mean HR values were not reported in that study, but the range indicates that their HR during matches was comparable to the nonelite players in the present study. Capranica et al. (8) found a mean HR during matches of approximately 180 beats·min−1 among 11-yr-old players at a higher competition level, which seems comparable to the elite players in the present study.
Maximal oxygen uptakes between 55 and 70 mL·min−1·kg−1 have been found in adult soccer players at elite level (3,4,18). This large variation is partly associated with the different positions of the players within the team. The maximal oxygen uptakes are of the same magnitude as found in the elite players in this study. The relative load (70–80% of O2max) for the young elite players during matches is also of the same magnitude as for adult players (3). This suggests that both aerobic capacity as well as the workload during matches is comparable for young and adult elite players.
When comparing published studies of the activity pattern of adult soccer players, some discrepancies exist. Extensive differences exist in the time spent in walking and standing when comparing studies on adults by Bangsbo et al. (4), Ali and Farrally (1), and Mayhew et al. (16). This could indicate a coincidence of different definitions and notation techniques. However, if the low-intensity categories “standing” and “walking” are added up, the three studies are quite comparable, with almost similar time spent in low- and high-intensity running (Table 4). The activity pattern of the elite groups in the present study does not differ much from the activity pattern of the adult players in any activity. This supports that the motion patterns of the young and adult elite players are comparable. The nonelite players differ markedly from the adult players, by using less time in low-intensity running and more time in walking. Capranica et al. (8) found in 11-yr-old players (N = 6) that the time spent in running was very high (55%) and the time spent in walking and standing correspondingly low (4 and 38%). In spite of the very similar HR responses during the match play compared with the elite groups in the present study, the activity pattern do not match quite well, as the activity level found by Capranica et al. appears to be extremely high (Table 4). The mean duration time in each activity of about 6 s in all groups of players (Table 3) corresponds very well with values for adult players (16,19). A lower mean duration time of 4.5 s found in Danish elite players (4) can partly be explained by a larger number of exercise categories in that study.
It seems that an overall similarity in aerobic demand and motion pattern is found between young and adult elite soccer players. It also seems that a selection and/or specialization are present in the young players due to the fact that the nonelite players clearly differ from both adult players and elite players of the same age. In studies of adult soccer players, specialization due to playing position has been found, showing that the defenders are the least active players during match play (2,3,9,19,24). In spite of the inhomogeneous distribution of playing positions in the present study (Table 2), this study supports a similar trend of specialization. The specialization due to playing position is most pronounced in the older elite group at the late puberty, indicating a more mature tactical understanding and a greater differentiation between the tasks of the different playing positions.
One problem with an early specialization and selection of players into playing position and playing level is that the advantage in physical attributes also can be due to a high relative age (players born in the early part of the selection year). Helsen et al. (12) found that youth players born in the early part of the selection year are more likely to be identified as talented by top teams. This is indicated in this study by an age difference of 0.5 yr (nonsignificant) between the nonelite group and the elite group in the same age category. In addition to the variation in relative age, a wide individual variation in onset of puberty exists (15,21) that can result in substantial differences in physical status among players of the same age (11). Thus, it seems clear that the chronological age will not be a sufficient parameter of classification in the years of adolescence, without taking the relative age and the puberty status into consideration, when comparing sport skills between players. Helsen et al. (12) argued that the relative age effect is due to the fact that current talent identification and selection are significantly influenced by a child’s physical attributes rather than to his/her sport skills. However, a situation where the coach does not select the best players to the specific position at the actual day of match play would be quite unthinkable because of the short-term success criterion to win every match and not the long-term development of talented players. In this context, it is important that those players who possess potential for development but without sufficient physical capacity to play at the highest level, either because of low relative age or late maturity, are stimulated tactically and physically to ensure further development of their sport skills and avoid a later drop out of the potential mass of talented players. It is also important to have in mind that if the exercise intensity during adolescence is an important stimulus for the maximal attainable aerobic capacity (20), several years in a specific playing position or on a lower playing level can make relocation difficult because of loss of tactical and physical skills, in spite of the fact that the player is highly talented.
Even though sprint is not the most used activity in a match, it is a very important activity. A high anaerobic capacity and thus an ability to sprint can be the critical parameter for a player’s performance in a decisive situation that might be crucial for the result of a match. Due to the very short intervals, the sprint parameter is difficult to measure with high reliability, and it will require the use of identical and accurate methods if comparisons between studies should be made. Thus, comparison of sprint activities between young and adult players is an important issue for further research.
One of the aims of this study was to relate the HR measurements to motion analysis. It is not possible to convert the motion pattern into HR, but it seems that the observation method was able to disclose significant differences between elite and nonelite players related to significant differences in the HR data. Less activity in the low motion activity “standing” and higher activity in the middle motion activity “jogging” in the elite groups are well matched with the significantly higher oxygen uptake in the elite groups during the matches. Furthermore, this study shows that an early classification of players will influence the exercise level and the tactical demands during a match that later can make relocation difficult.
The printing of the paper was economically supported by The Institute of Exercise and Sport Sciences, Department of Human Physiology, Copenhagen, Denmark.
1. Ali, A., and M. Farrally. A computer-video aided time motion analysis technique for match analysis. J. Sports Med. Phys. Fitness 31: 82–88, 1991.
2. Ali, A., and M. Farrally. Recording soccer players’ heart rates during matches. J. Sports Sci. 9: 183–189, 1991.
3. Bangsbo, J. The physiology of soccer: with special reference to intense intermittent exercise. Acta Physiol. Scand. Suppl. 619: 1–155, 1994.
4. Bangsbo, J., L. Norregaard, and F. Thorso. Activity profile of competition soccer. Can. J. Sport Sci. 16: 110–116, 1991.
5. Baxter Jones, A., H. Goldstein, P. Helms. The development of aerobic power in young athletes. J. Appl. Physiol. 75: 1160–1167, 1993.
6. Bell, W. Physiological characteristics of 12-year-old soccer players. In: Science and Football, T. Reilly (Ed.). London: E. & F. N. Spon, 1988, pp. 175–180.
7. Berg, K. E., J. C. Lavoie, R. W. Latin. Physiological training effects of playing youth soccer. Med. Sci. Sports Exerc. 17: 656–660, 1985.
8. Capranica, L., A. Tessitore, L. Guidetti, and F. Figura. Heart rate
and match analysis in pre-pubescent soccer players. J. Sports Sci. 19: 379–384, 2001.
9. Ekblom, B. Applied physiology of soccer. Sports Med. 3: 50–60, 1986.
10. Hansen, L., Elite training and physical development of boys. Ph.D. dissertation. University of Copenhagen, Institute of Exercise and Sports Sciences, Dept. of human Physiology, Denmark, 1998,
11. Hansen, L., K. Klausen, J. Bangsbo, and J. Müller. Short longitudinal study of boys playing soccer: parental height, birth weight and length, anthropometry, and pubertal maturation in elite and non-elite players. Pediatr. Exerc. Sci. 11: 199–207, 1999.
12. Helsen, W. F., N. J. Hodges, J. Van Winckel, and J. L. Starkes. The roles of talent, physical precocity and practice in the development of soccer expertise. J. Sports Sci. 18: 727–736, 2000.
13. Jones, A., and P. Helms. Cardiorespiratory fitness in young British soccer players. In: Science and Football II, T. Reilly, J. Clarys, and A. Stibbe (Eds.). London: E. & F. N. Spon, 1993, pp. 298–303.
14. Klimt, F., M. Betz, and U. Seitz. Metabolism and circulation system of children playing soccer. Pediatr. Work Physiol. 16: 127–129, 1992.
15. Malina, R. M., and C. Bouchard. Growth, maturation, and physical activity. In: Growth, Maturation, and Physical Activity, R. M. Malina (Ed.). Champaign, IL: Human Kinetics, 1991, pp. 231–249.
16. Mayhew, S. R., and H. A. Wenger. Time-motion analysis
of professional soccer. J. Hum. Stud. 11: 49–52, 1985.
17. Montoye, H. J., H. C. Kemper, W. H. Saris, and R. A. Washburn. Measuring Physical Activity and Energy Expenditure. Champaign, IL: Human Kinetics, 1996, pp. 258–272.
18. Reilly, T. Physiological profile of the player. In: Football (Soccer), B. Ekblom (Ed.). London: Blackwell Scientific Publications, 1994, pp. 78–94.
19. Reilly, T., and V. Thomas. A motion analysis of work-rate in different positional roles in professional football match-play. J. Hum. Mov. Stud. 2: 87–97, 1976.
20. Rodhe, C. H., T. Espersen. Work intensity during soccer training and match play. In: Science and Football, T. Reilly, A. Lees, K. Davids, and W. J. Murphy (Eds.). London: E. & F. N. Spon, 1988, pp. 68–75.
21. Rowland, T. W. Developmental Exercise Physiology, 1st Ed. Champaign, IL: Human Kinetics, 1996, pp. 3–16.
22. Schofield, W. N. Predicting basal metabolic rate, new standards and review of previous work. Hum. Nutr. Clin. Nutr. 39 (Suppl. 1): 5–41, 1985.
23. Tumilty, D. Physiological characteristics of elite soccer players. Sports Med. 16: 80–96, 1993.
24. Withers, R. T., Z. Maricic, S. Wasilewski, and L. Kelly. Match analysis of Australian Professional Soccer Players. J. Hum. Mov. Stud. 8: 159–176, 1982.
Keywords:©2004The American College of Sports Medicine
PUBERTY; OXYGEN UPTAKE; TIME-MOTION ANALYSIS; TESTICLE VOLUME; MALE; HEART RATE