Classifying young soccer players by training performances.

Perceptual & Motor Skills: Physical Development & Measurement 2014, 119, 3, 971-984. © Perceptual & Motor Skills 2014

CLASSIFYING YOUNG SOCCER PLAYERS BY TRAINING PERFORMANCES1 EDUARDO A. ABADE, BRUNO V. GONÇALVES, ALEXANDRA M. SILVA, AND NUNO M. LEITE CreativeLab, Research Center in Sport Sciences, Health and Human Development (CIDESD) University of Trás-os-Montes e Alto Douro at Vila Real CARLO CASTAGNA

JAIME E. SAMPAIO

Football Training and Biomechanics Laboratory Italian Football Federation (FIGC) Coverciano (Florence), Italy

CreativeLab, Research Center in Sport Sciences, Health and Human Development (CIDESD) University of Trás-os-Montes e Alto Douro at Vila Real

Summary.—Players within the same age group may present different physical and physiological profiles. This study classified young soccer players according to their physical and physiological profiles obtained during the training sessions and compared classification by age and playing position criteria. 151 male elite Portuguese soccer players (under 15, under 17, and under 19 years old) participated. Time-motion and body acceleration and deceleration data were collected using GPS technology with heart rate monitored continuously across the selected training sessions. The data were grouped using two-step cluster analysis to classify athletes. A repeated-measures factorial ANOVA was performed to identify differences in the variables. Three clusters comprised 15.2%, 37.1%, and 47.7% of the total sample, respectively. Players of the same ages and playing experience had different performance profiles. Grouping players with similar physiological profiles during training sessions may allow coaches to balance oppositions and reduce the variability of the physiological outcomes.

Competition in many youth team sports is organized by age group, primarily because of the need to enhance fair play values. Although generally accepted, this form of organization may not adequately consider differences in chronological age, since players within the same age group can differ by 2 years or more, and have different physical and physiological profiles (Cobley, Baker, Wattie, & McKenna, 2009). Differences in age, physical, and physiological profiles may affect the players' responses to stimuli. The characterization of these profiles may provide valuable information to adjust the task designs, training stimuli, and environmental constraints. During adolescence, the positions played and selection level (i.e., Regional, National) may affect players' development (Till, Cobley, O'Hara, Chapman, & Cooke, 2012). In fact, most motor Address correspondence to Eduardo Abade, Universidade de Trás-os-Montes e Alto Douro, Quinta de Prados, 5001-801 Vila Real, Portugal or e-mail ([email protected]). 1

DOI 10.2466/10.25.PMS.119c31z8

23-PMS_Abade_140075.indd 971

ISSN 0031-5125

03/12/14 4:42 PM

972

E. A. ABADE, ET AL.

skills undergo significant development during the pubertal period, reinforcing the importance of specific training (Fernandez-Gonzalo, De SouzaTeixeira, Bresciani, Garcia-Lopez, Hernandez-Murua, Jimenez-Jimenez, et al., 2010). Therefore, the coaches should regularly track the progression of players' responses to training and competition, while considering the interaction between age and playing positions. From a physiological perspective, understanding the development of athletic potential alongside biological growth is crucial to selecting appropriate training at different stages of players' development. Models of long-term athletic development generally recommend the use of a wide range of activities in the earlier ages with a progressively narrowing focus at more advanced ages (Côté & Fraser-Thomas, 2007). However, the most sensitive developmental periods for learning different tasks are not known, partly because the long-term effects of intensive training and competitive schedules in youth are not well-explored in the literature. To develop a successful professional sports career, young players have to perform adequately in several dimensions (Elferink-Gemser, Visscher, Lemmink, & Mulder, 2004). Generally, the training sessions have a strong focus on game-like situations, which elicit dynamic and adaptive responses while providing high variability in physiological, technical, and tactical demands (Pinder, Davids, & Renshaw, 2012). However, the manipulation of constraints (e.g., number of players and/or field dimensions) can be quite limited because, even if the task constraints contain relevant information for learning a specific activity, the unique characteristics of each learner also represent personal constraints and singular responses (Chow, Davids, Button, Shuttleworth, Renshaw, & Araujo, 2006). In fact, individual learning dynamics will always differ because the interacting configurations of constraints will differ between learners (Chow, Davids, Hristovski, Araujo, & Passos, 2011). For example, when playing a 3 × 3 small-sided game, players will respond to stimuli in singular ways, merging individual physiological, technical, and tactical profiles. Therefore, it is probably not possible (and may even be harmful) to recommend universal, optimal learning pathways for all learners. Within the available literature, the activity profiles of young soccer players are usually described in relation to playing positions (Buchheit, Mendez-Villanueva, Simpson, & Bourdon, 2010; Aslan, Acikada, Guvenc, Goren, Hazir, & Ozkara, 2012) and age groups (Fernandez-Gonzalo, et al., 2010; Till, et al., 2012). For example, running performance among soccer players tends to increase with age (Buchheit, et al., 2010) and youth players (age 15 years) cover from 4,435 to 8,098 m during a match (Rebelo, Brito, Seabra, Oliveira, & Krustrup, 2012). The players seem to cover the highest distances in low speed ranges (0–6.0 km.hr−1) and the lowest


03/12/14 4:43 PM

SOCCER PERFORMANCE PROFILES

973

above 21.1 km.hr−1 (e.g., sprints for the goal, to the ball, and for defense) (Aslan, et al., 2012). The distances covered vary according to playing positions, with defenders and midfielders achieving the lowest and highest distances, respectively (Buchheit, et al., 2010). The defenders seem to have lower endurance performances than midfielders and forwards (Markovic & Mikulic, 2011). The differences in movement and activity patterns require different conditioning and recovery programs according to age and positional groups (Quarrie, Hopkins, Anthony, & Gill, 2012) because soccer demands sprinting, hard accelerations and decelerations, changes of direction, and collisions between players. Therefore, evaluating players' intermittent triaxial movements and heart rate (HR) profiles should aid understanding and assessing the physical demands, as well as helping adjust rest and recovery times after training and matches (Quarrie, et al., 2012). High-level players are frequently required to perform high intensity sprints and changes of direction, which underscores agility as a key ability (Sheppard & Young, 2006). Consequently, the development of mechanisms related to changing direction, sprinting capacity (Jones, Bampouras, & Marrin, 2009), and power and strength is a major issue to coaches and is considered an important predictor of performance (Wong, Chan, & Smith, 2012). All these variables seem important to describe the players' physiological profiles and to provide information for adjusting training parameters and workload. However, these have not been compared to the players' actual internal load (e.g., heart rate measures) during training. The data gathered in training sessions can also be used to classify the players into different performance groups as an alternative to age and playing position groupings. In fact, physiological variables have been suggested as powerful predictors for identification of talent in youth sports (Gil, Gil, Ruiz, Irazusta, & Irazusta, 2007; Lago-Peñas, Casais, Dellal, Rey, & Domínguez, 2011; Unnithan, White, Georgiou, Iga, & Drust, 2012). In this sense, classification techniques (such as cluster analysis) based on physiological profiles may provide useful information for establishing groups of interest to prescribe training. Research goal. This study will classify young soccer players according to their physical and physiological profiles in training sessions and then compare classifications against age and playing position. Hypothesis 1. Different distances covered, sprint characteristics, intermittent triaxial movements, and heart rates will be observed among players with identical ages and playing experience.


03/12/14 4:43 PM

974

E. A. ABADE, ET AL.

Hypothesis 2. Heart rate and intermittent triaxial movements will significantly discriminate the obtained clusters. METHOD Participants Male soccer players (N = 151) of under-15 (U15, n = 56, M age = 14.0 yr., SD = 0.2), under-17 (U17, n = 66, M age = 15.8 yr., SD = 0.4), and under-19 (U19, n = 29, M age = 17.8 yr., SD = 0.6) categories participated in this study. The participants were part of five different elite youth teams (selected based on the age group Portuguese ranking), training and competing regularly in the Portuguese national competition (2011/2012 season). Both U15 and U17 teams trained in the same 60 × 40 m outdoor pitch (four training sessions per week with an average duration of 90 min.), while U19 teams trained in an outdoor pitch with official dimensions (five training sessions per week with an average duration of 90 min.). For each age group, all training sessions were continuously performed in the same pitch. Besides the regular physical education classes, none of the players was involved in any other sport activity. After a detailed explanation of the goals, benefits, and risks involved in this investigation, all participants, parents, and coaches signed a written informed consent. Additionally, players were informed that they were free to withdraw at any time without any penalty. The study protocol conformed to the Declaration of Helsinki and was approved by the ethics committee of the Research Center in Sport, Health and Human Development (Vila Real, Portugal). Procedure The study was conducted during the competitive season over a 9-week period (December to February), with 38 randomly chosen training situations (U15, n = 12; U17, n = 16; U19, n = 10) representing a total of 612 cases (number of players × number of training sessions). All the practice sessions were performed at the same time period of the day (from 4.30 p.m. to 9.00 p.m.) on outdoor natural turf pitches, under similar environmental conditions. The U15 and U17 teams trained four times per week for a total of 360 min. in a 60 × 40 m pitch, while U19 teams trained five times per week for a total of 450 min. in a pitch with official dimensions (120 × 90 m). The average number of players per training unit was 23 (± 4). All practice sessions started with a specific warm-up that included low-intensity running and ball possession drills (with ∼ 15 min. duration). Stretching exercises were performed at the end of each training session. Players were allowed to drink water during specific recovery periods (approximately 3 min.). The clubs and coaches allowed only a comprehensive general description of the practice session drills for scientific research. All


03/12/14 4:43 PM


975

sessions began with a ∼ 15 min. warm-up and ended with ∼ 10 min. of cooldown exercises. The U15 training sessions mainly included the development of technical skills and elementary tactical principles (e.g., 1 × 1 analytical sub-phase situations and 3 × 3 small-sided games). Although the specific goals of U17 practices were similar to U15, there was an increased focus on game-like situations (e.g., 3 × 3 + 2 floaters). The U19 training sessions included constrained small-sided games focused on team tactical principles and physical conditioning stimuli (e.g., 8 × 5 + Goalkeeper). The players' physical responses were measured as median values from all monitored training sessions. The distance covered was measured in predefined speed zones (Aguiar, Botelho, Gonçalves, & Sampaio, 2013): Zone 1 (0.0–6.9 km.hr−1), Zone 2 (7.0–9.9 km.hr−1), Zone 3 (10.0–12.9 km.hr−1), Zone 4 (13.0–15.9 km.hr−1), Zone 5 (16.0–17.9 km.hr−1), and Zone 6 ( ≥ 18.0 km.hr−1). Sprints (Zone 6) were measured by both average time interval and distance covered. The data were collected at 15 Hz through the entire duration of each training session using portable global positioning system (GPS) units (SPI-PRO X II, GPSports, Canberra, ACT, Australia). These units were fitted to the upper back of each participant using an elastic harness. The validity and reliability of these instruments were inspected by independent verifications for both the 5 and 10 Hz units (Castellano, Casamichana, Calleja-Gonzalez, San Roman, & Ostojic, 2011; Johnston, Watsford, Pine, Spurrs, Murphy, & Pruyn, 2012). Players' internal physiological responses included the measures of HR and intermittent triaxial movements. The HR absolute values were recorded continuously throughout all training sessions using the Polar Team System (Polar Electro, Oy, Kempele, Finland) and, subsequently, converted into percentages of HRmax and classified into time spent in four zones of intensity (Gore, 2000): Zone 1 (< 75% HRmax), Zone 2 (75– 84.9% HRmax), Zone 3 (85–89.9% HRmax), and Zone 4 (≥ 90% HRmax). To measure the players' HRmax, the Yo-Yo intermittent recovery level 2 test was performed (Krustrup, Mohr, Nybo, Jensen, Nielsen, & Bangsbo, 2006). In this test, participants start at 13 km/hr and have a 10 sec. active break after each 40 m (2 × 20 m runs), with the speed increasing at intervals. The SPI-PRO X II units are coupled with a 100 Hz accelerometer capable of measuring intermittent triaxial movements2 as the rate of acceleration and deceleration in the horizontal axis (x), transverse axis (y), and vertical axis (z). This variable measures the changes of direction, as well as collisions with opponent players and the ground. The values were The GPS systems provide a valid measure for total load, but not for x-, y-, and z-axis separately. The problem is technological and is related with the accuracy of the orientation in the starting point/time of measurement. 2


03/12/14 4:43 PM

976

E. A. ABADE, ET AL.

grouped into six zones of G force (McLellan, Lovell, & Gass, 2011): Zone 1 (< 5.0–6.0 g), Zone 2 (6.1–6.5 g), Zone 3 (6.5–7.0 g), Zone 4 (7.1–8.0 g), Zone 5 (8.1–10.0 g), and Zone 6 ( > 10.1 g).3 Analysis A two-step cluster with log-likelihood as the distance measure and Schwartz's Bayesian criterion was performed to classify athletes according to their performance profiles (i.e., all variables described in procedures section). The analysis was used to classify the players' performances in two perspectives: first, to identify the variables that maximize group distances and, second, to contrast the obtained model with the typical age and playing experience criteria. This method differs from traditional clustering techniques by handling of categorical variables (assuming variables to be independent), automatic selection of number of clusters (automatically determines the optimal number of clusters), and scalability (by constructing a cluster membership) (Tabachnick & Fidell, 2007). The variables were ranked according to the predictors' importance, indicating the relative importance of each predictor in estimating the model (the sum of the values for all predictors on the display is 1). In the functional sense, the variables' predictor importance provides different weights to support the cluster distribution. Subsequently, each clustering variable's description was presented as mean ± standard deviation. Both average time interval and distance covered per sprint were tested using a one-way analysis of variance (ANOVA) according to the clustering groups. Finally, a repeated-measures factorial ANOVA was performed to identify differences in time motion (six speed zones × clustering groups), HR (four HR zones × clustering groups), and intermittent triaxial movements (six gforce zones × clustering groups). Pairwise differences were assessed with a Bonferroni post hoc test. Effect size was presented as partial eta squared (ηp2) and interpreted by the follow criteria: statistically significant but weak (ηp2 ≤ 0.04), moderate (0.04 < ηp2 ≤ 0.36), and strong (ηp2 > 0.36) (Tabachnick & Fidell, 2007). All data sets were tested for each statistical technique corresponding assumptions. These calculations were carried in SPSS Version 20.0 (IBM Corporation, USA) and statistical significance was set at .05. RESULTS The cluster analysis classified the players in three distinct groups accordingly to their physical and physiological performance during training sessions. The obtained clusters comprised 15.2%, 37.1%, and 47.7% of the total sample, respectively. No differences were found between the clusIt is important to understand that the authors have selected these values to match the method of McLellan, et al. (2011). 3


03/12/14 4:43 PM

977

SOCCER PERFORMANCE PROFILES TABLE 1 CHARACTERIZATION OF CLUSTERS Variable Age

Cluster 1 (n = 23)

Cluster 2 (n = 56)

Cluster 3 (n = 72)

M

SD

M

SD

M

1.5

15.4

1.0

15.5

15.7

Height

1.72

0.06

1.76

0.07

1.74

SD 1.7 0.07

Weight

63.2

5.6

65.1

6.2

64.5

BMI

21.3

1.4

21.0

1.3

21.2

1.6

6.5

1.9

6.8

1.7

6.7

2.3

Experience

7.7

ters in players' age, height, weight, body mass index (BMI), or experience (Table 1). Figure 1 shows the distribution (%) of players in each Cluster considering the players' actual age group and playing position. Cluster 1 represented the lowest percentage of the sample, with a high presence of U19

Development Stages 60 45

U15 U17 U19

Forwards 30 15 0

Percent (%)

45 Midfielders 30 15

Playing Positions

60

0 60 45 Defenders

30 15 0 1

2

3

Two-step Cluster Number

FIG. 1. Distribution (%) of players in each cluster by players' development stage and playing position


03/12/14 4:43 PM

978

E. A. ABADE, ET AL.

midfielders, compared to the other age groups and playing positions. Cluster 2 included the highest percentage of U17 players and the lowest percentage of U19 players. Cluster 3 was the largest group. The percentage of U15 forwards and defenders was high, as well as of U19 midfielders and defenders. Sprint characteristics (activity performed above 18 km.hr−1) differed between Clusters 2 and 3 for both time interval (F2, 148 = 11.7, p < .001, η2 = 0.14) and distance covered per sprint (F2, 148 = 11.6, p < .001, η2 = 0.15). Cluster 3 presented the lowest average time interval (1.67 ± 0.24 sec.) and distance covered per sprint (9.45 ± 1.43 m). Figure 2 presents the results from both external and internal workload across clusters and also the predictor importance from all considered variables. Figure 2A presents the variation of distance covered at the considered speed zones for each cluster. There was a significant effect of both speed zones (F5, 740 = 2573.3, p < .001, η2 = 0.95) and clusters (F10, 740 = 3.3, p < .001, η2 = 0.43). Also, differences were found in intermittent triaxial movements (F5, 740 = 1020.8, p < .001, η2 = 0.87) and clusters (F10, 740 = 44.1, p < .001, η2 = 0.37), with pairwise differences across all groups, except between Clusters 2 and 3 for Zone 6 (see Fig. 2B). The HR zone values showed significant effect of zones (F3, 444 = 487.1, p < .001, η2 = 0.77) and clusters (F6, 444 = 30.9, p < .001, η2 = 0.30), with players spending most of the time below 75% of HRmax (Fig. 2C). Finally, Table 2 presents the obtained predictor importance from the variables. The strongest predictor importance was found in intermittent triaxial movements for Zones 5, 3, 4, and 2. Distance covered at Zone 1 was identified as having the lowest predictor importance. DISCUSSION Talent development is a holistic system of multiple dimensions in interaction. The present study aimed to explore a part of this system, particularly related to performances during soccer training. Although there are several dimensions to consider (social, psychological, etc.), there seems to be a need also to focus on observable performance variables in training sessions. Therefore, the present study classified young soccer players according to chosen physical and physiological profiles and compared this classification technique against the more typical age and playing position criteria. It is promising that the first hypothesis was supported, since players with identical ages and playing experiences were allocated to clusters representing divergent physical and physiological profiles. The second hypothesis was partially supported. Although heart rate values were not able to discriminate the clusters, intermittent triaxial movements were the most powerful variables in establishing the clusters.


03/12/14 4:43 PM

979

SOCCER PERFORMANCE PROFILES 2500

(A)

Cluster 1 Cluster 2 Cluster 3

2200

Distance Covered (m)

2000 1750 1500 b,c

1200

b,c 1000 750 b,c 500

a,c

b,c

250 0 0.0–6.9

7.0–9.9 10.0–12.9 13.0–15.9 16.0–17.9

=>18

Speed Zones (Km.h–1) 800

(B)

700


a,b,c

600

Number (a.u.)

500 400 300 a,b,c

a,b,c

200

a,b,c

a,b,c

100 a,b 0

5.0–6.0

6.1–6.5

6.5–7.0

7.1–8.0

8.1–10.0

>10

Triaxial Movement Zones 50

(C)


45 b,c 40

Minutes (min.)

35 30 25 20 b,c 15 b,c 10 5 0 90%

Heart Rate (%HRmax)

FIG. 2. Results from distance covered for each speed zone (A), number of triaxial movements for each intensity zone (B), and time spent in each heart rate zone (C). Significant differences are identified as: (a) Cluster 1 vs Cluster 2; (b) Cluster 1 vs Cluster 3; (c) Cluster 2 vs Cluster 3.


03/12/14 4:43 PM

980

E. A. ABADE, ET AL. TABLE 2 PREDICTOR IMPORTANCE OF ALL CONSIDERED VARIABLES Variable

Predictor Importance

Intermittent triaxial movements z5

1.00


0.88


0.85


0.84

Distance z4

0.75


0.68

Distance z5

0.67


0.64

HR z1

0.50

HR z3

0.42

Distance z6

0.33

Distance z3

0.31

HR z4

0.30

Distance z2

0.18

Time interval per sprint

0.16

Distance per sprint

0.15

HR z2

0.04

Distance z1

0.03

Note.—z1 to z6 = Zones 1 to 6.

This study identified the most powerful performance predictors (Table 2) in clustering young soccer players during training sessions, providing more accurate information about the players' responses to training stimuli. The obtained clusters were very similar in age, anthropometric characteristics, and years of experience. This similarity suggests that those variables may not be the most important to discriminate physiological performance profiles in training, despite their common use. As stated in Hypothesis 1, the current results indicated that players with identical ages and playing experience can have very different physiological profiles and, consequently, respond in different ways to similar training stimuli. Grouping players with similar physiological characteristics may diminish the emergence of heterogeneous responses during training, which can help coaches in the distribution of training groups and allow an efficient control of the players' responses. Regarding Hypothesis 2, the current results indicated that intermittent triaxial movements in higher intensity zones (Zones 2 to 5) were the predictors that best discriminated the clusters. Although the training tasks


03/12/14 4:43 PM


981

were similar across all age groups, the players' intermittent triaxial movements varied substantially across the clusters. This may suggest that selected training tasks in training enhance a specific learning, exhibited in players' unique rates of acceleration and deceleration. Sprints with changes of direction seem to induce higher physiological demand than intermittent in-line sprints (Dellal, Keller, Carling, Chaouachi, Wong, & Chamari, 2010), mostly because of their correlation with the eccentric strength (Jones, et al., 2009). The results show that players in Cluster 1 had a higher number of intermittent triaxial movements across all zones. This analysis may help coaches to identify the most capable players to successfully perform highintensity and rapid body movements. In a functional sense, when performing the same task, the physical and physiological responses will represent higher performances. Cluster 3 included the players who covered shorter distances and had fewer high speed movements across all zones per training. Consequently, these players spent more time below 75% HRmax. This was the most frequent profile (47.7% of sample), including a high percentage of U15 and U19 players. The U19 players' fitness, maturity, and higher expertise, positioning, and decision skills (Kannekens, Elferink-Gemser, & Visscher, 2011) might optimize their movement patterns and result in lower physical and physiological intensity. In fact, high-level U19 coaches usually focus their attention on the strategic team plans and collective tactical responses by improving players' positioning. Such strategies frequently imply stopping times that may compromise high-intensity activities. The U15 training sessions have a different perspective, mainly focused on acquiring basic team tactical principles (Duarte, Araujo, Davids, Travassos, Gazimba, & Sampaio, 2012), often using analytical exercises that impair significant developments in physical conditioning. Consequently, the frequent presence of U15 players in Cluster 3 may be the result of lower conditioning, as measured by lower strength, power, and speed (Malina, Eisenmann, Cumming, Ribeiro, & Aroso, 2004). Cluster 2 has a larger number of U17 players (who at 17 years are beyond the critical period for physical maturation) in an intermediate stage of tactical expertise. At this point in their careers, these players are progressing in their understanding of the tactical principles of team play, as potentiated by their training usually characterized by constrained game-like situations, which may be responsible for their intermediate physiological profile. Cluster 1 represented the players who had higher values for all the measured variables, and this cluster also had a lower number of players. These players perhaps should be followed with increased attention because they exhibit higher overall potential. However, any interpretation of the players' distribution among clusters should be extremely cautious, mainly with regard to the


03/12/14 4:43 PM

982

E. A. ABADE, ET AL.

playing positions. The ecological environment evaluated during training sessions (i.e., the unpredictability of the technical/tactical exercises and game-like situations) discourages labelling the players' performance with their playing roles. Moreover, the presence of a similar number of defenders, midfielders, and forwards across clusters shows that looking exclusively to the players' playing position may be an insufficient criterion to discriminate physiological profiles during training tasks. These data are important to understand as performance indicators in high-intensity intermittent exercises, frequently experienced during training and/or game situations (Vaz, Leite, Vicente, Gonçalves, & Sampaio, 2012). When training tasks are focused on physical conditioning as opposed to merely chronological age, grouping players with similar physiological profiles and fitness may avoid significantly different responses and adaptations to the stimuli. Thus, coaches of such groups of players should have a more accurate and effective control over players' responses to exercise, e.g., during specific high-intensity training using blocks and periodization (Issurin, 2010). However, optimal performance in competitions requires not only physiological development, but also technical and tactical abilities. Thus, further investigations are required to explore the short and long-term effects of this clustering method. REFERENCES

AGUIAR, M., BOTELHO, G., GONÇALVES, B., & SAMPAIO, J. (2013) Physiological responses and activity profiles of football small-sided games. Journal of Strength and Conditioning Research, 27(5), 1287-1294. ASLAN, A., ACIKADA, C., GUVENC, A., GOREN, H., HAZIR, T., & OZKARA, A. (2012) Metabolic demands of match performance in young football players. Journal of Sports Science and Medicine, 11(1), 170-179. BUCHHEIT, M., MENDEZ-VILLANUEVA, A., SIMPSON, B. M., & BOURDON, P. C. (2010) Match running performance and fitness in youth football. International Journal of Sports Medicine, 31(11), 818-825. CASTELLANO, J., CASAMICHANA, D., CALLEJA-GONZALEZ, J., SAN ROMAN, J., & OSTOJIC, S. M. (2011) Reliability and accuracy of 10 Hz GPS devices for short-distance exercise. Journal of Sports Science and Medicine, 10(1), 233-234. CHOW, J. Y., DAVIDS, K., BUTTON, C., SHUTTLEWORTH, R., RENSHAW, I., & ARAUJO, D. (2006) Nonlinear pedagogy: a constraints-led framework for understanding emergence of game play and movement skills. Nonlinear Dynamics, Psychology, and Life Sciences, 10(1), 71-103. CHOW, J. Y., DAVIDS, K., HRISTOVSKI, R., ARAUJO, D., & PASSOS, P. (2011) Nonlinear pedagogy: learning design for self-organizing neurobiological systems. New Ideas in Psychology, 29(2), 189-200. COBLEY, S., BAKER, J., WATTIE, N., & MCKENNA, J. (2009) Annual age-grouping and athlete development. Sports Medicine, 39(3), 235-256. CÔTÉ, J., & FRASER-THOMAS, J. (2007) Youth involvement in sport. Toronto, Ontario: Pearson/Prentice Hall.


03/12/14 4:43 PM


983

DELLAL, A., KELLER, D., CARLING, C., CHAOUACHI, A., WONG, D. P., & CHAMARI, K. (2010) Physiologic effects of directional changes in intermittent exercise in football players. Journal of Strength and Conditioning Research, 24(12), 3219-3226. DUARTE, R., ARAUJO, D., DAVIDS, K., TRAVASSOS, B., GAZIMBA, V., & SAMPAIO, J. (2012) Interpersonal coordination tendencies shape 1-vs-1 sub-phase performance outcomes in youth football. Journal of Sports Sciences, 30(9), 871-877. ELFERINK-GEMSER, M. T., VISSCHER, C., LEMMINK, K. A. P. M., & MULDER, T. W. (2004) Relation between multidimensional performance characteristics and level of performance in talented youth field hockey players. Journal of Sports Sciences, 22(11-12), 1053-1063. FERNANDEZ-GONZALO, R., DE SOUZA-TEIXEIRA, F., BRESCIANI, G., GARCÍA-LÓPEZ, D., HERNÁNDEZ MURÚA, J. A., JIMÉNEZ-JIMÉNEZ, R., & DE PAZ, J. A. (2010) Comparison of technical and physiological characteristics of prepubescent football players of different ages. Journal of Strength and Conditioning Research, 24(7), 1790-1798. GIL, S. M., GIL, J., RUIZ, F., IRAZUSTA, A., & IRAZUSTA, J. (2007) Physiological and anthropometric characteristics of young soccer players according to their playing position: relevance for the selection process. Journal of Strength and Conditioning Research, 21(2), 438-445. GORE, C. (2000) Physiological tests for elite athletes. Champaign, IL: Human Kinetics. ISSURIN, V. B. (2010) New horizons for the methodology and physiology of training periodization. Sports Medicine, 40(3), 189-206. JOHNSTON, R., WATSFORD, M., PINE, M. J., SPURRS, R., MURPHY, A., & PRUYN, E. (2012) The validity and reliability of 5-hz global positioning system units to measure team sport movement demands. Journal of Strength and Conditioning Research, 26, 758765. JONES, P., BAMPOURAS, T. M., & MARRIN, K. (2009) An investigation into the physical determinants of change of direction speed. Journal of Sports Medicine and Physical Fitness, 49(1), 97-104. KANNEKENS, R., ELFERINK-GEMSER, M. T., & VISSCHER, C. (2011) Positioning and deciding: key factors for talent development in football. Scandinavian Journal of Medicine & Science in Sports, 21(6), 846-852. KRUSTRUP, P., MOHR, M., NYBO, L., JENSEN, J. M., NIELSEN, J. J., & BANGSBO, J. (2006) The Yo-Yo IR2 Test: physiological response, reliability, and application to elite football. Medicine & Science in Sports & Exercise, 38(9), 1666-1673. LAGO-PEÑAS, C., CASAIS, L., DELLAL, A., REY, E., & DOMÍNGUEZ, E. (2011) Anthropometric and physiological characteristics of young soccer players according to their playing positions: relevance for competition success. Journal of Strength and Conditioning Research, 25(12), 3358-3367. MALINA, R. M., EISENMANN, J. C., CUMMING, S. P., RIBEIRO, B., & AROSO, J. (2004) Maturityassociated variation in the growth and functional capacities of youth football (soccer) players 13-15 years. European Journal of Applied Physiology, 91(5-6), 555-562. MARKOVIC, G., & MIKULIC, P. (2011) Discriminative ability of the Yo-Yo Intermittent Recovery Test (Level 1) in prospective young football players. Journal of Strength and Conditioning Research, 25(10), 2931-2934. MCLELLAN, C. P., LOVELL, D. I., & GASS, G. C. (2011) Biochemical and endocrine responses to impact and collision during elite rugby league match play. Journal of Strength and Conditioning Research, 25(6), 1553-1562.


03/12/14 4:43 PM

984

E. A. ABADE, ET AL.

PINDER, R. A., DAVIDS, K., & RENSHAW, I. (2012) Metastability and emergent performance of dynamic interceptive actions. Journal of Science and Medicine in Sport, 15(5), 437443. QUARRIE, K. L., HOPKINS, W. G., ANTHONY, M. J., & GILL, N. D. (2012) Positional demands of international rugby union: evaluation of player actions and movements. Journal of Science and Medicine in Sport, 16, 353-359. REBELO, A., BRITO, J., SEABRA, A., OLIVEIRA, J., & KRUSTRUP, P. (2012) Physical match performance of youth football players in relation to physical capacity. European Journal of Sport Science, 14, S148-S156. SHEPPARD, J. M., & YOUNG, W. B. (2006) Agility literature review: classifications, training and testing. Journal of Sports Sciences, 24(9), 919-932. TABACHNICK, B. G., & FIDELL, L. S. (2007) Using multivariate statistics. (5th ed.) Boston, MA: Allyn & Bacon. TILL, K., COBLEY, S., O'HARA, J., CHAPMAN, C., & COOKE, C. (2012) A longitudinal evaluation of anthropometric and fitness characteristics in junior rugby league players considering playing position and selection level. Journal of Science and Medicine in Sport, 16, 438-443. UNNITHAN, V., WHITE, J., GEORGIOU, A., IGA, J., & DRUST, B. (2012) Talent identification in youth football. Journal of Sports Sciences, 30, 1719-1726. VAZ, L., LEITE, N., VICENTE, J., GONÇALVES, B., & SAMPAIO, J. (2012) Differences between experienced and novice rugby union players during small-sided games. Perceptual & Motor Skills, 115(2), 594-604. WONG, D., CHAN, G., & SMITH, A. (2012) Repeated-sprint and change-of-direction abilities in physically active individuals and football players: training and testing implications. Journal of Strength and Conditioning Research, 26(9), 2324-2330. Accepted November 5, 2014.


03/12/14 4:43 PM

Injuries among young soccer players.

Two Cases of Rhabdomyolysis After Training With Electromyostimulation by 2 Young Male Professional Soccer Players.

Effects of plyometric training on soccer players.

Postactivation potentiation in elite young soccer players.

Does SAQ training improve the speed and flexibility of young soccer players? A randomized controlled trial.

Specific physical trainability in elite young soccer players: efficiency over 6 weeks' in-season training.

Changes in the anaerobic threshold in an annual cycle of sport training of young soccer players.

Effects of velocity-based resistance training on young soccer players of different ages.

Agility training in young elite soccer players: promising results compared to change of direction drills.

Comparison of the effect of repeated-sprint training combined with two different methods of strength training on young soccer players.

Determinant factors of repeat sprint sequences in young soccer players.

Metabolic demands of match performance in young soccer players.

Relative risk for concussions in young female soccer players.

In-season training periodization of professional soccer players.

Fatigue-induced balance impairment in young soccer players.

Correction: Analysis of Jumping-Landing Manoeuvers after Different Speed Performances in Soccer Players.

Analysis of Jumping-Landing Manoeuvers after Different Speed Performances in Soccer Players.

Mental Fatigue and Spatial References Impair Soccer Players' Physical and Tactical Performances.

Enhancing Change-of-Direction Speed in Soccer Players by Functional Inertial Eccentric Overload and Vibration Training.

Repeated Dribbling Ability in Young Soccer Players: Reproducibility and Variation by the Competitive Level.

Training for Power and Speed: Effects of Increasing or Decreasing Jump Squat Velocity in Elite Young Soccer Players.

The effects of interset rest on adaptation to 7 weeks of explosive training in young soccer players.

Effect of Vertical, Horizontal, and Combined Plyometric Training on Explosive, Balance, and Endurance Performance of Young Soccer Players.

Effect of unilateral, bilateral, and combined plyometric training on explosive and endurance performance of young soccer players.