original research
Pediatric Exercise Science, 2010, 22, 497-510
© 2010 Human Kinetics, Inc.
Is the Relationship Between Sprinting and
Maximal Aerobic Speeds in Young Soccer
Players Affected by Maturation?
Alberto Mendez-Villanueva, Martin Buchheit, Sami Kuitunen,
Tsz Kit Poon, Ben Simpson, and Esa Peltola
ASPIRE, Academy for Sports Excellence
The purpose of this study was to investigate the relationship between maximal
sprinting (MSS) and aerobic (MAS) speeds in a cohort of highly-trained young
male soccer players with the influence of body mass controlled for using allometric scaling. MSS and MAS were obtained in 14 pre-age at peak height velocity
(APHV) players (12.3 ± 0.7 years), 21 circum-APHV players (14.3 ± 0.9 year)
and 26 post-APHV players (16.9 ± 0.7 years). The three groups showed similar
positive correlations between MSS and MAS (r = 0.73 to 0.52; p < .01). In conclusion, our results suggest that the relationship between MSS and MAS is not
affected by maturation.
Morphological and physiological considerations suggest that sprinting ability
and aerobic function (e.g., aerobic power and endurance capacity) put conflicting
demands on the design of a human’s locomotor apparatus and therefore cannot be
maximized simultaneously (31). The intuitive basis of this performance trade-off
probably stems from the observation that elite specialist sprinters and marathon
runners differ substantially in size and shape, and no athlete excels in both events.
In addition to cardiorespiratory attributes, individual differences in the proportion
of fast, powerful muscle fibers to slow more fatigue-resistant fibers are believed to
be the primary physiological basis of this performance trade-off (34). Consistent
with this suggestion, short-distance human sprinters possess a greater proportion of
fast-twitch muscle fibers than do endurance athletes, and vice versa for the slower
more fatigue-resistant muscle fibers (7,12). A genetic-based trade-off between sprint
and endurance phenotypic traits has been suggested such as that an individual is
inherently predisposed toward performance in either sprint/power or endurance
events (19,35). However, team sports such as soccer require a more complex
(mixed) phenotypic trait, as both qualities (i.e., speed and aerobic function) influence
high-level soccer performance (22,28). Thus, successful young soccer players are
expected to be selected and systematically trained to develop both qualities (i.e.,
speed and aerobic fitness; 23). Accordingly, the presence of functional trade-offs
Mendez-Villanueva, Buchheit, Kuitunen, Poon, Simpson, and Peltola are with the Sport Science Dept.,
ASPIRE, Academy for Sports Excellence, Doha, Qatar.
497
498 Mendez-Villanueva et al.
between speed and endurance can impose important constrains on the development
of these two game-related fitness traits (22).
Children appear to have an “optimal” phenotypic profile for team sports like
soccer since their ability to demonstrate this locomotor specialization is less apparent (26). That is, children who perform well in sprinting tasks also perform well in
endurance activities, and vice versa (27). Nevertheless, it appears that this “optimal” locomotor profile is transient as it has been suggested that specialization into
endurance or sprinting profiles occurs during late-puberty stages (8,16). However,
few studies have assessed the association between sprinting and aerobic-related
performances in children and adolescents of different maturational levels spanning
a wide circum-pubertal spectrum. In addition, to the best of our knowledge, the
ability to demonstrate this postpuberty specialization has been only examined in
nonspecifically trained children (4,8,16). Whether children trained to develop both
qualities (i.e., speed and aerobic fitness), as it happens in soccer, would still show
a similar specialization is presently unknown.
A tool that might offer valuable information on the child locomotor profile is
the calculation of the ratio of maximal sprinting speed (MSS) to maximal aerobic
speed (MAS), which is similar to the ratio of anaerobic to aerobic power previously described (2,30). The proposed speed ratio (i.e., MSS to MAS) can help to
better understand how these two locomotor qualities develop in relation to each
other within an individual and can be useful in assessing the degree of physiological adaptation in specific populations (2,30). Available studies suggest that
in nonspecifically trained children, the anaerobic to aerobic ratio increases during
pubertal development, levels off during adolescence and remains stable into early
adulthood (16). However, factors other than growth, development and maturation,
such as disease, medication and exercise training are likely to affect this ratio
(2,30). Despite acknowledged confounders, the MSS to MAS ratio could be used
as a sensitive tool to assess differences in an individual’s relative sprinting and
aerobic exercise capacity.
It is well known that in children, performances in various physical tests are
affected by growth. For example, while improvements in running speed may result
from a training program, they may also be simply related with changes in body size
as a consequence of growth and maturation (16). Thus, to accurately assess changes
or differences in “intrinsic” physical performance parameters, size-adjusted rather
than absolute values have been suggested to be more appropriate (33). Controlling for any confounding influence of morphological attributes (e.g., body mass
or height) via allometric analyses allows sample and test-specific size adjustment
to be achieved, enabling comparisons of performance measures between different
groups or timepoints (33). We are not aware of any studies that have examined the
effects of age and maturation on the relationship between body size and locomotor
performances (i.e., MSS and MAS) as determined by allometric scaling.
Examining the relationships between MSS and MAS performance across different levels of age and maturation has the potential to provide useful insights into
the evolution of important soccer match fitness-related factors (22). Information in
this context would be valuable for profiling young soccer players (14), adapting the
training to their abilities (5) and identifying highly talented youngsters (29). Thus,
the purpose of this study was to examine relationships among MSS and MAS in
highly-trained young soccer players with the influence of body mass controlled
Locomotor Tradeoffs in Youth Soccer 499
for using allometric scaling. We first hypothesized to observe essentially “mixed”
profiles (i.e., similar levels for both MSS and MAS, as evidenced by large correlations between both attributes) in the prepubertal group. Conversely, as a result of
the anticipated specialization (8,16), more “specialized” profiles (i.e., either high
MSS or MAS, as evidenced by poor correlations between both attributes) could
be observed in pubertal and especially postpubertal players. However, it is also
possible that the soccer selection/training process, which put constant emphasis
on both qualities, might systematically exclude “specialized” profiles. This would
result in the persistence of “mixed” profiles even in the older group.
Materials and Methods
Participants
A total of 61 young male soccer players (11.5–17.8 yr) participated in this study.
Only outfield players (i.e., there were no goalkeepers in the sample) were tested.
Written informed consent was obtained from the parents and from the players. All
the players were training in a high-performance soccer academy and participated
in » 14 hr of combined soccer training and competitive play per week on average.
The experimental protocol was approved by the Institutional Ethical Committee.
Experimental Measures
All measurements were taken at the beginning of the annual training season
(approximately after 2 weeks of training) to limit differences in training status
between players. All players followed a similar training program with the supervision of their respective coaches. All performance tests were performed on two
separate occasions with at least one day between the two testing sessions. Speed tests
were completed on the first day, while the anthropometrical assessment and MAS
tests were performed in the second testing session. Test sessions were undertaken
between 07:00 and 9:00 hr (anthropometrical and MAS) and between 16:00 and
18:00 (speed) at least 8 hr after the last training session. All performance tests were
performed in an indoor facility maintained at standard environmental conditions.
Speed testing began after 20 min of standardized warm-up, which consisted of lowintensity forward, side-ways and backward running, acceleration runs, skipping
and hoping exercise, and jumps at increasing intensity. Players were fully familiar
with all test procedures.
Anthropometric measurements: All the anthropometrical measurements were
taken in the morning before the MAS test was performed. Dimensions included
height, body mass, seated height and 7 skinfolds (triceps, subscapular, biceps,
suprailiac, abdominal, thigh and medial calf). Height was measured with a fixed
stadiometer (± 0.1 cm; Holtain Limited, Crosswell, UK), seated height with a fixed
sitting height table (± 0.1 cm; Holtain Limited, Crosswell, UK), body weight with a
digital balance (ADE Electronic Column Scales, Hamburg, Germany) and skinfold
thickness with a Harpenden skinfold caliper (Baty International, Burguess Hill,
UK). The exact positioning of each skinfold measurement was in accordance with
procedures described by Norton et al. (20). The same anthropometer carried out
all the measurements.
500 Mendez-Villanueva et al.
Pubertal timing was estimated according to the estimated biological maturity
age of each individual as described by Mirwald et al. (18). The age of peak linear
growth (age at peak height velocity, APHV) is an indicator of somatic maturity
representing the time of maximum growth in stature during adolescence. A biological maturity age (years) was calculated by subtracting the chronological age
at the time of measurement from the chronological APHV. Thus, a maturity age
of -1.0 indicates that the player was measured 1 yr before APHV; a maturity of 0
indicates that the player was measured at the time of APHV; and maturity age of
+1.0 indicates that the subject was measured 1 yr after APHV (3). On that basis,
players were accordingly subdivided into three nonoverlapped maturational groups:
before the estimated APHV (Pre-APHV) players (>-3.0 yrs to PHV to < -1.5 years
to PHV; n = 14), around the estimated APHV (Circum-APHV) players (> -1 yrs to
PHV to < +1 yrs from PHV; n = 21) and after estimated the APHV (Post-APHV)
players (> + 1.5 yrs from PHV to < 3.0 yrs from PHV; n = 26). Ethnicity of the
players was Arab (Middle East and North Africa backgrounds). The effect of
ethnicity on the validity of the biological maturity estimates (i.e., APHV) using
the procedures described above is unknown. Thus the equation was deemed to be
valid in the present population.
Maximal Sprinting Speed (MSS): The running speed of players was evaluated with a 40-m sprint effort by using dual-beam electronic timing gates (Swift
Performance Equipment, Lismore, Australia) with slip times at 10 m, 20 m and 30
m. Players were instructed to run as quickly as possible along the 40-m distance
from a standing start. MSS was assessed using the fastest 10-m split time during
the 40-m test. Speed was measured to the nearest 0.01 s. Subjects performed two
trials with at least 3 min of rest between them. The best performance of the 2 tests
was used for the analysis.
Maximal Aerobic Speed (MAS): In a separate day, players performed an incremental field test to assess MAS. The incremental field test was a modified version of
the University of Montreal Track Test (15; i.e., the VAMEVAL maximal incremental
running test). The VAMEVAL test begins with a running speed of 8 km·h-1 with a
consecutive speed increase of 0.5 km·h-1 each minute until exhaustion. The players adjusted their running speed to auditory signals at 20-m intervals, delineated
by cones around a 200-m long athletics track. The end of the test was taken to be
when participants twice failed to reach the next cone in the required time. During
the test, players were verbally encouraged by testers and coaches. The velocity of
the last 1-min stage completed by the subjects was retained as the players’ MAS
(km·h-1). If the last stage was not completed entirely, the MAS was calculated using
the formula of Kuipers et al. (13): MAS = Sf + (t/60·0.5), where Sf was the last
completed speed in km·h-1 and t in the time in seconds of the uncompleted stage.
MSS/MAS ratio: From the data of the MSS and the MAS tests, a maximal
sprinting speed-to-maximal aerobic speed ratio was calculated as follows:
Maximal sprinting speed to maximal aerobic speed ratio = MSS/MAS
Allometric Scalling Model
MSS and MAS were used as the dependent variables. Body mass (kg), height (cm)
and leg length (cm) served as independent variables to construct three different
regression models and to identify the scaling exponents. The following steps outline
Locomotor Tradeoffs in Youth Soccer 501
the procedures used to construct the model (36). First, normality of the dependent
variables was assessed in the entire cohort. Second, a log-linear regression analysis was performed on the independent and dependent variables. The slope of the
regression line was used as the allometric scaling exponent. Third, distribution of
residuals and the assumption of homoscedasticity were tested by the AndersonDarling normality test and visual inspection of the residuals. The residual errors
should demonstrate a constant variance (homoscedasticity) and a normal distribution, indicating that the model fits all individuals across the entire range. Fourth and
last, independence of the power ratio (i.e., allometrically-scaled MSS and MAS)
and independent variable (i.e., body mass, height and leg length) was assessed.
For an allometric model to be deemed appropriate there should be no significant
correlation between the allometrically-scaled MSS and MAS measurement and
the independent variable.
Statistical Analyses
Descriptive statistics are means ± Standard deviation (SD). Maturity group differences in anthropometric variables and performance measurements expressed
in raw and allometrically-scaled values were examined using one-way ANOVA.
If the first test provided a significant difference, then each group was compared
with the other two using post hoc Bonferroni tests. To determine the relationship
between MSS and MAS we conducted linear regression analysis using values (both
raw and allometrically scaled) of MSS and MAS. The corresponding strength was
assessed using Pearson´s correlation coefficient (r). The following criteria were
adopted for interpreting the magnitude of correlation (r) between test measures: ≤
0.1, trivial; >0.1–0.3, small; >0.3–0.5, moderate; >0.5–0.7, large; >0.7–0.9, very
large; and >0.9–1.0, almost perfect (11). To estimate maturity-related differences
on the MSS/MAS ratio of individual players, we estimated a standard deviation
representing individual responses (the square root of the difference in the variances
in the performance scores in each pair of maturity groups; 10). The precision of
our estimates of outcome statistics are shown as 90% confidence intervals (90%
CI), which define the likely range of the true value in the population from which
the sample was drew. For all statistical analyses significance was set al p < .05.
Results
Physical and Performance Data
The physical characteristics of each group are shown in Table 1. All physical and
anthropometrical variables, except the sum of the seven skinfolds, were significantly
different among the three groups.
Allometric Scaling Exponent
The use of the three independent variables (i.e., body mass, height and leg length)
as scaling variables was successful in meeting all of the statistical criteria. As
such, all the variables were positively correlated with MSS and MAS. Body mass
yielded larger correlations with performance variables than height and leg length.
Moreover, given the high level of interrelatedness between the three variables and
502 Mendez-Villanueva et al.
Table 1 Player’s Physical Characteristics
Pre-APHV
(n = 14)
Circum-APHV
(n = 21)
Post-APHV
(n = 26)
Age (yr)
12.3 ± 0.7
14.3 ± 0.9*
16.9 ± 0.7†§
Height (m)
145.9 ± 4.4
160.7 ± 6.3*
173.3 ± 4.4†§
Body Mass (kg)
35.4 ± 3.7
47.7 ± 5.2*
62.1 ± 6.9†§
BMI (kg·m-2)
16.6 ± 1.5
18.4 ± 1.2*
20.6 ± 1.9†§
Leg Length (m)
71.7 ± 3.9
78.6 ± 4.6*
83.1 ± 3.8†§
Sum Skinfolds (mm)
48.6 ± 11.5
49.1 ± 18.4
48.7 ± 10.6
-2.1 ± 0.4
-0.2 ± 0.7*
2.3 ± 0.4†§
Years to/from PHV
Values are means ± SD; n, number of subjects. Pre-APHV, before the estimated age at peak height
velocity; Circum-APHV, around the estimated age at peak height velocity; Post-APHV, after the estimated age at peak height velocity. BMI, body mass index; Sum Skinfolds calculated from adding up
the skinfold thicknesses of biceps, triceps, subscapular, supraspinale, abdomen, thigh, and calf muscles;
PHV, peak height velocity. * Significant difference (P < 0.05) between circum-APHV and pre-APHV
players; † Significant difference (P < 0.05) between post-APHV and pre-APHV players; § Significance
difference (P < 0.05) between post-APHV and circum-APHV players.
body mass being the most commonly used allometric scaling variable, only results
of body mass are reported.
Data were normally distributed. Log-linear regression analysis using body
mass as the independent variable was applied to the entire group, resulting in scaling exponents of 0.33 and 0.22 for MSS and MAS, respectively. These exponents
were later on used to compute allometrically-adjusted ratios (MSS and MAS
expressed in km·h-1 divided by body mass raised to the appropriate exponent).The
Breusch–Pagan/Cook–Weisberg test for heteroscedasticity was performed, and the
residual errors from the body mass model were found to be randomly distributed or
homoscedastic (p > .05). The independence test of the power ratio was nonsignificant (p > .05) for both parameters (i.e., MSS and MAS; p > .05). Visual inspection
of residuals showed no apparent systematic variation. In addition, the AndersonDarling normality test of the residuals revealed a normal distribution (p > .05).
Mean Performance Differences Between Groups
Unsurprisingly, post-APHV players were significantly faster (i.e., higher MSS
in raw values) than circum-APHV and pre-APHV players (p < .001; Table 2). In
addition, pre-APHV players were slower than circum-APHV (p < .001). Similar
to MSS, raw MAS was significantly influenced by age (p < .001). Post-APHV
players had significantly higher MAS than pre- and circum-APHV players, while
circum-APHV players were better than pre-APHV (Table 2). Mean maturity-related
differences in the MSS/MAS ratio are shown in Table 3. The MSS/MAS ratio (raw
values) was significantly lower (p < .05) in post-APHV players as compared with
both pre- and circum-APHV players with no differences between the two latter
Locomotor Tradeoffs in Youth Soccer 503
Table 2 Player’s Physical Performance
Pre-APHV
(n = 14)
Circum-APHV
(n = 21)
Post-APHV
(n = 26)
MSS (km·h-1)
24.80 ± 1.50
28.08 ± 1.68*
31.05 ± 1.15†§
MAS (km·h-1)
14.5 ± 1.2
15.7 ± 1.1*
17.3 ± 1.0†§
MSS (km·h-1·kg0.33)
7.69 ± 0.53
7.88 ± 0.42
8.00 ± 0.36†
MAS (km·h-1·kg0.22)
6.56 ± 0.53
6.68 ± 0.43
6.94 ± 0.50†
Values are means ± SD; n, number of subjects. Pre-APHV, before estimated age at peak height velocity;
Circum-APHV, around estimated age at peak height velocity; Post-APHV, after estimated age at peak
height velocity. MSS, maximal sprinting speed; MAS, maximal aerobic speed. * Significant difference
(P < 0.05) between circum-APHV and pre-APHV players; † Significant difference (P < 0.05) between
post-APHV and pre-APHV players; § Significance difference (P < 0.05) between post-APHV and
circum-APHV players.
Table 3 Mean Maturity-Related Differences in the Maximal Sprinting
Speed (MSS) to Maximal Aerobic Speed (MAS) Ratio (MSS/MAS
Ratio)
Maturation status
Pre-APHV
(n = 14)
Circum-APHV
(n = 21)
Post-APHV
(n = 26)
MSS/MAS ratio (Raw values)
1.70 (0.10)
1.79 (0.10) *
1.80 (0.11)†
MSS/MAS ratio (Scaled values)
1.16 (0.06)
1.18 (0.07)
1.16 (0.06)
Values are means (SD). Pre-APHV, before estimated age at peak height velocity; Circum-APHV, around
estimated age at peak height velocity; Post-APHV, after estimated age at peak height velocity. * Significant difference (P < 0.05) between cicum-APHV and pre-APHV players; † Significant difference
(P < 0.05) between post-APHV and pre-APHV players.
groups. ANOVA performed on scaled values revealed significant (p < .05) differences between post-APHV and pre-APHV players in MSS and MAS with no
differences among the other groups (see Table 2). The allometrically-scaled ratio
between MSS and MAS was not statistical different in the three groups (Table 3).
Individual Performance Differences Between Groups
The standard deviation of the MSS to MAS ratio scores was almost identical in
the three groups for both raw and allometrically scaled values (Table 3). This
homogenous between-subject variation indicates that individual differences in the
MSS/MAS ratio as a result of different maturity status were similar. Individual differences, expressed as standard deviations, for both raw and allometrically scaled
scores are presented in Table 4.
504 Mendez-Villanueva et al.
Table 4 Individual Maturity-Related Differences, Expressed as
Standard Deviations and 90% Confidence Intervals, in the Maximal
Sprinting Speed (MSS) to Maximal Aerobic Speed (MAS) Ratio
(MSS/MAS Ratio) Between Before the Estimated Age at Peak Height
Velocity (Pre-APHV), Around the Estimated Age at Peak Height
Velocity (Circum-APHV) and After the Estimated Age at Peak Height
Velocity (Post-APHV) Groups
Pre-APHV
(n = 14)
(90% CI)
Circum-APHV
(n = 21)
(90% CI)
Post-APHV
(n = 26)
(90% CI)
MSS/MAS ratio
(Raw values)
0.01 (- 0.09–0.09)
0.04 (- 0.08–0.10)
0.04 (- 0.07–0.09)
MSS/MAS ratio
(Scaled values)
0.02 (- 0.06–0.06)
- 0.02 (- 0.06–0.05)
- 0.03 (-0.06–0.05)
Values are means (90% CI). Pre-APHV, before the estimated age at peak height velocity; CircumAPHV, around the estimated age at peak height velocity; Post-APHV, after the estimated age at peak
height velocity.
Correlations Between MSS and MAS
Raw MSS scores for individuals plotted against MAS scores are shown in Figure
1A. Both pre-APHV and circum-APHV groups showed a positive, large relationship
between MSS and MAS (r = .71–066, p < .01). A small, nonsignificant positive
correlation between MSS and MAS was found in the post-APHV group (r = .22,
p = .29) was found in this group. When the same linear regression analysis was
performed on allometrically scaled scores (Figure 1B), the three groups showed a
positive, large to very large relations between scaled MSS and MAS (r = 0.52–0.73;
p < .01).
Discussion
The primary aim of this investigation was to determine whether the relationship
between MSS and MAS was affected by maturation in a cohort of highly-trained
young soccer players. Present results show that both locomotor attributes (i.e., MSS
and MAS) improve with the transition from pre- to post-APHV stages, resulting in
unchanged, size-corrected MSS/MAS ratio scores. We also found positive, large
correlations between size-corrected MSS and MAS among each maturity group.
Thus, our results do not support the notion of a maturity-related trade-off between
MSS and MAS in developmental soccer players.
Allometric Model
Locomotor performances evaluated in the current study were positively correlated
with body mass, reflecting the size-dependency of both performance measurements.
Thus, the application of allometric scaling attempted to remove the confounding influence of body mass on MSS and MAS (33), which allowed us to better
Figure 1 — Relationship between maximal sprinting speed (MSS) and maximal aerobic
speed (MAS) in three groups of young soccer players differing in physical maturity status
(estimated by the years to/from the age at peak height velocity). Individual values are shown
in absolute (A) and allometrically scaled (B) terms.
505
506 Mendez-Villanueva et al.
examine the athletic capacities of players differing in age and maturation levels.
Allometric scaling of MSS and MAS produced different scaling exponents (0.33 vs.
0.22, respectively). These exponents are, by definition, the slope of the log-linear
regression lines and indicate that body mass and associated factors in response to
growth, maturation and training, exert more influence on MSS than on MAS. To
the best of our knowledge, this is the first study applying allometric scaling models
to locomotor performance data in a cohort of well-trained young soccer players;
thus, direct comparison with the existing literature cannot be made.
Performance Differences Between Groups
The MSS and MAS were significantly different between pre- and circum- and postAPHV players. Raw MSS/MAS ratio was significantly higher in the circum-APHV
group compared with the pre-APHV players, while no differences were found when
comparing circum- and post-APHV players. This increment in the MSS/MAS ratio
in the transition from pre- to circum-APHV stages arises from the more marked
increase in MSS in comparison with MAS (13% and 8%, respectively). However,
from circum- to post-APHV the increases in MSS and MAS were comparable, 10%
in both parameters, so that the ratio was no more affected. This suggests that the
transition from pre-APHV to circum-APHV represents an area of a more marked
increase in MSS as compared with the transition from circum- to post-APHV
stages. These findings are in agreement with Philippaerts and coworkers (21) who
have previously reported peak development in running sprinting speed at the age
of peak height velocity in youth soccer players. Moreover, these authors reported
further running speed gains after peak height velocity but at slower rates (21).
However, the same authors reported that aerobic power performance peaked at
peak height velocity to drop after it (21). In contrast, our results showed differences
in MAS (10%) when circum- and post-APHV groups were compared, suggesting
further improvements in aerobic power in our players. The fact that our football
players were subjected to an intensive soccer training regimen might explain the
maturity-related differences in aerobic power (i.e., MAS) compared with results
reported by Philippaerts et al. (21), as it has been suggested that additional training
hours influence the “sustainability” of improvements in endurance performance
observed in elite post-APHV field hockey players (6). The similar MASs observed
in the post-APHV players in the current study (17.3 ± 1.0 km·h–1) in comparison
with top-level soccer players (17.7 ± 0.9 km·h–1; 22) provides further support for
this hypothesis. Moreover, it is also possible that selection bias within the soccer
players may indeed factor out players without the capacity to concurrently train
and maintain MSS and MAS.
Most of these increases in MSS and MAS occurring in the transition from
pre-APHV to circum-APHV are likely to be related to the parallel changes in body
mass (see Table 1). This is supported by the body mass-corrected values reported
in Table 2, where it can be seen that both allometrically scaled MSS and MAS of
pre- and circum-APHV players were not statistically different. As skinfolds did
not differ among the three groups, changes in body mass were presumably more
a reflection of increases in lean body mass or muscle. On the contrary, when MSS
and MAS were body mass-corrected, both parameters were still higher in the postAPHV group in comparison with the pre-APHV group. This suggests that other
Locomotor Tradeoffs in Youth Soccer 507
maturity-induced factors not causally related with changes in body (or muscle)
mass can independently impact MSS and MAS. These factors might be related to
a series of maturity-induced improvements in neural function, multijoint coordination, increases in muscle power associated with the rise in circulating levels of
testosterone and growth hormone and changes in both oxidative and nonoxidative
metabolism (9,32). Therefore, our results suggest that some factors operating at
different maturity stages might simultaneously regulate concurrent improvements
in MSS and MAS in proportion to player’s body mass (i.e., pre- and circum-APHV)
or at a greater rate to their body mass (i.e., post-APHV; 1).
Another novel aspect of the current study was that, in addition to the mean
effects of maturation on the MSS/MAS ratio (discussed above), the analysis of the
individual differences indicates that this mean effect of maturation on the MSS/MAS
ratio (both on raw and allometrically scaled values) does characterize the effect
in individual players adequately. In this regard, the standard deviation representing individual differences was small and of similar magnitude in the three groups
of players, suggesting that all individual players displayed a rather homogenous
ratio. Again, this could be related to the selection/training process that puts similar
emphasis on both qualities and therefore, tends to normalize players’ athletic profile. Potential factors accounting for individual differences in the MSS/MAS ratio
such as genetic endowment, physical training loads and nutritional status were also
likely to be similar among players in the different groups. Despite this relatively
low variability, it is important to consider the individual MSS/MAS ratio to give
recommendations as to whether endurance training or specific speed and power
training should be emphasized in an individualized training perspective.
Trade-Offs Between MSS and MAS
A negative correlation between speed and aerobic function is generally thought to
be a fundamental trade-off in locomotor performance that occurs virtually in all
animal species because, among other factors, of the physiology of skeletal muscle
and the biomechanics of the mature skeleton-muscular system (24,25,31). While
very little direct evidence for this trade-off at the level of whole organism performance has been provided in humans (31), data in (nonspecifically trained) children
indicate that this locomotion specialization into more sprinting or aerobic profiles
does not occur before puberty (8,16). For the first time, we investigated here the
possible associations between MSS and MAS in a cohort of young, developmental highly-trained soccer players. The small nonsignificant correlation between
nonallometrically scaled MSS and MAS values in the post-APHV group (r = .22)
gives support to the notion that some locomotion specialization might occur at
late pubertal stages. Differences were also found in the slope of the MSS-MAS
relationships between groups, with post-APHV players specifically showing lower
relative increases in MSS with increasing MAS than the pre-APHV and circumAPHV players (Figure 1A). Sprinting performances in the post-APHV players
were in line with those previously reported in adult male soccer players (23,28).
Data on 40-m sprinting times in pre-APHV and circum-APHV children were
also comparable to those previously reported in similar (well-trained) boys (17).
Clearly, the participants of the current study were as fast as other similar athletes
(17). This suggests that the observed trend of post-APHV players toward showing
508 Mendez-Villanueva et al.
a somehow more endurance profile when raw MSS and MAS values were evaluated was not achieved at expenses of sacrificing sprinting speed, further supporting
the lack of maturity-related trade-offs in our sample of well-trained young soccer
players. Training studies on concurrent training in this specific population should
however be conducted before definitive conclusions can be made. After adjusting
for body mass (i.e., allometric scaled data), a large (r = .68) MSS-MAS correlation
was found in the post-APHV group, which was in the range of the correlations
observer in pre- (r = .73) and circum- (r = .52) APHV players. Similarly, the slopes
of the body-mass corrected MSS-MAS relationship between the three groups were
remarkably similar (Figure 2B). From this larger MSS-MAS correlation observed
when allometric scaled values were examined (in comparison with raw values) it
can be inferred that body mass was a relevant modifying factor of the MSS-MAS
relationship in our group of postpubertal players and should be taken into account.
Albeit speculative, these results also suggest that some body mass-independent
factors such as muscle architecture, muscle stiffness and motor coordination might
be common determinants of performance in both qualities (i.e., MSS and MAS).
Limitations
The present work in a sample of young soccer players, as with any other crosssectional study, only provide suggestive evidence concerning the causal relationship
between concurrent changes in MSS and MAS during growth. The current study
only characterizes average MSS and MAS developmental patterns displayed by
young soccer players differing in estimated maturity timing and age. Accordingly,
the true magnitude of performance changes that might effectively occur between
and within individuals with growth and maturation cannot be described. Although
estimates of maturity in the current study have merit in that they do not require
invasive procedures, care is warranted in utilizing predicting APHV from anthropometrical factors (18). Lastly, only male soccer players were evaluated here. Thus,
caution should be taken when applying the present results to a similar group of
female players.
Conclusions
We conclude that, in our sample of highly-trained young soccer players, there is
a positive association between MSS and MAS regardless of the maturity status.
Although training studies are required to draw definitive conclusions, the equivalent
mass-corrected MSS/MAS ratio in the three groups differing in physical maturity
levels and age suggests that improvements in sprinting speed could be achieved
without compromising these of maximal aerobic speed or vice versa. The fact that
these two performance traits (i.e., MSS and MAS) covary during puberty might
also suggest that they could be optimized at the same time during growth, at least
to a certain extent. These findings have implications for training prescription during
puberty in talented soccer players. Detailed analyses of performance capacities and
behavior of young soccer players are required to fully understand how development
(growth, maturation and training) may shape their locomotor systems.
Locomotor Tradeoffs in Youth Soccer 509
References
1. Armstrong, N., J.R. Welsman, C.A. Williams, and B.J. Kirby. Longitudinal changes in
young people’s short-term power output. Med. Sci. Sports Exerc. 32:1140–1145, 2000.
2. Bar-Or, O. Pathophysiological factors which limit the exercise capacity of the sick
child. Med. Sci. Sports Exerc. 18:276–282, 1986.
3. Baxter-Jones, A.D.G., and L.B. Sherar. Growth and maturation. In: Paediatric Exercise
Physiology, N. Armstrong (Ed.). Philadelphia: PA Elsevier Limited, 2007, pp. 1–26.
4. Bloxham, S.R., J.R. Welsman, and N. Armstrong. Ergometer-specific relationships
between peak oxygen uptake and short-term power output in children. Pediatr. Exerc.
Sci. 15:136–148, 2005.
5. Buchheit, M. The 30-15 intermittent fitness test: accuracy for individualizing interval
training of young intermittent sport players. J. Strength Cond. Res. 22:365–374, 2008.
6. Elferink-Gemser, M.T., C. Visscher, M.A. van Duijn, and K.A. Lemmink. Development
of the interval endurance capacity in elite and sub-elite youth field hockey players. Br.
J. Sports Med. 40:340–345, 2006.
7. Esbjornsson, M., C. Sylven, I. Holm, and E. Jansson. Fast twitch fibres may predict
anaerobic performance in both females and males. Int. J. Sports Med. 14:257–263,
1993.
8. Falk, B., and O. Bar-Or. Longitudinal changes in peak aerobic and anaerobic mechanical power of circumpubertal boys. Pediatr. Exerc. Sci. 5:318–331, 1993.
9. Hansen, L., J. Bangsbo, J. Twisk, and K. Klausen. Development of muscle strength in
relation to training level and testosterone in young male soccer players. J. Appl. Physiol.
87:1141–1147, 1999.
10. Hopkins, W.G. Spreadsheets for analysis of controlled trials, with adjustment for a
subject characteristic. Sportscience. 10:46–50, 2006.
11. Hopkins, W.G., S.W. Marshall, A.M. Batterham, and J. Hanin. Progressive statistics
for studies in sports medicine and exercise science. Med. Sci. Sports Exerc. 41:3–13,
2009.
12. Komi, P.V. Physiological and biomechanical correlates of muscle function: effects of
muscle structure and stretch-shortening cycle on force and speed. Exerc. Sport Sci.
Rev. 12:81–121, 1984.
13. Kuipers, H., F.T. Verstappen, H.A. Keizer, P. Geurten, and G. van Kranenburg. Variability of aerobic performance in the laboratory and its physiologic correlates. Int. J.
Sports Med. 6:197–201, 1985.
14. le Gall, F., C. Carling, M. Williams, and T. Reilly. Anthropometric and fitness characteristics of international, professional and amateur male graduate soccer players from
an elite youth academy. J Sci Med Sport. 2008.
15. Leger, L., and R. Boucher. An indirect continuous running multistage field test: the
Universite de Montreal track test. Can. J. Appl. Sport Sci. 5:77–84, 1980.
16. Malina, R.M., C. Bouchard, and O. Bar-Or. Growth, maturation and physical activity.
Champaing, IL: Human Kinetics, 2004.
17. Mero, A. Power and speed training during childhood. In: Pediatric Anaerobic Performance, E. Van Praagh (Ed.). Champaign, IL: Human Kinetics, 1998, pp. 241–267.
18. Mirwald, R.L., A.D. Baxter-Jones, D.A. Bailey, and G.P. Beunen. An assessment of
maturity from anthropometric measurements. Med. Sci. Sports Exerc. 34:689–694,
2002.
19. Niemi, A.K., and K. Majamaa. Mitochondrial DNA and ACTN3 genotypes in Finnish
elite endurance and sprint athletes. Eur. J. Hum. Genet. 13:965–969, 2005.
20. Norton, K., M. Marfell-Jones, N. Whittingham, et al. Anthropometric assessment
protocols. In: Physiological Testing for Elite Athletes, C. Gore (Ed.). Champaing, IL:
Human Kinetics, 2000, pp. 66–85.
510 Mendez-Villanueva et al.
21. Philippaerts, R.M., R. Vaeyens, M. Janssens, et al. The relationship between peak height
velocity and physical performance in youth soccer players. J Sports Sci. 24:221–230,
2006.
22. Rampinini, E., D. Bishop, S.M. Marcora, D. Ferrari Bravo, R. Sassi, and F.M.
Impellizzeri. Validity of simple field tests as indicators of match-related physical
performance in top-level professional soccer players. Int. J. Sports Med. 28:228–235,
2007.
23. Reilly, T., J. Bangsbo, and A. Franks. Anthropometric and physiological predispositions
for elite soccer. J Sports Sci. 18:669–683, 2000.
24. Rome, L.C. The design of vertebrate muscular systems: comparative and integrative
approaches. Clin. Orthop. Relat. Res.S59–S76, 2002.
25. Rome, L.C., R.P. Funke, R.M. Alexander, et al. Why animals have different muscle
fibre types. Nature. 335:824–827, 1988.
26. Rowland, T.W. On being a metabolic nonspecialist. Pediatr. Exerc. Sci. 14:315–320,
2002.
27. Rowland, T.W. Children’s Exercise Physiology. Champaing, IL: Human Kinetics, 2005.
28. Stolen, T., K. Chamari, C. Castagna, and U. Wisloff. Physiology of soccer: an update.
Sports Med. 35:501–536, 2005.
29. Vaeyens, R., R.M. Malina, M. Janssens, et al. A multidisciplinary selection model for
youth soccer: the Ghent Youth Soccer Project. Br. J. Sports Med. 40:928–934, discussion 934, 2006.
30. van Brussel, M., L. van Doren, B.W. Timmons, et al. Anaerobic-to-aerobic power ratio
in children with juvenile idiopathic arthritis. Arthritis Rheum. 61:787–793, 2009.
31. Van Damme, R., R.S. Wilson, B. Vanhooydonck, and P. Aerts. Performance constraints
in decathletes. Nature. 415:755–756, 2002.
32. Van Praagh, E., and E. Dore. Short-term muscle power during growth and maturation.
Sports Med. 32:701–728, 2002.
33. Welsman, J., and N. Armstrong. Statistical techniques for interpreting body size-related
exercise performance during growth. Pediatr. Exerc. Sci. 12:112–127, 2000.
34. Wilson, R.S., R.S. James, and R. Van Damme. Trade-offs between speed and endurance
in the frog Xenopus laevis: a multi-level approach. J. Exp. Biol. 205:1145–1152, 2002.
35. Yang, N., D.G. MacArthur, J.P. Gulbin, et al. ACTN3 genotype is associated with
human elite athletic performance. Am. J. Hum. Genet. 73:627–631, 2003.
36. Zoeller, R.F., E.D. Ryan, H. Gordish-Dressman, et al. Allometric scaling of biceps
strength before and after resistance training in men. Med. Sci. Sports Exerc. 39:1013–
1019, 2007.