Pediatric Exercise Science, (Ahead of Print)

https://doi.org/10.1123/pes.2017-0220
© 2018 Human Kinetics, Inc. ORIGINAL RESEARCH

100-m Breaststroke Swimming Performance in Youth

Swimmers: The Predictive Value of Anthropometrics
Senda Sammoud
Higher Institute of Sport and Physical Education of Ksar Saïd

Alan Michael Nevill

University of Wolverhampton

Yassine Negra and Raja Bouguezzi

Higher Institute of Sport and Physical Education of Ksar Saïd

Helmi Chaabene
University of Potsdam and University of Jendouba

Younés Hachana
Higher Institute of Sport and Physical Education of Ksar Saïd and Manouba University

This study aimed to estimate the optimal body size, limb segment length, and girth or breadth ratios of
100-m breaststroke performance in youth swimmers. In total, 59 swimmers [male: n = 39, age = 11.5 (1.3) y;
female: n = 20, age = 12.0 (1.0) y] participated in this study. To identify size/shape characteristics associated
with 100-m breaststroke swimming performance, we computed a multiplicative allometric log-linear regression
model, which was refined using backward elimination. Results showed that the 100-m breaststroke performance
revealed a significant negative association with fat mass and a significant positive association with the
segment length ratio (arm ratio = hand length/forearm length) and limb girth ratio (girth ratio = forearm
girth/wrist girth). In addition, leg length, biacromial breadth, and biiliocristal breadth revealed significant
positive associations with the 100-m breaststroke performance. However, height and body mass did not
contribute to the model, suggesting that the advantage of longer levers was limb-specific rather than a general
whole-body advantage. In fact, it is only by adopting multiplicative allometric models that the previously
mentioned ratios could have been derived. These results highlighted the importance of considering anthropo-
metric characteristics of youth breaststroke swimmers for talent identification and/or athlete monitoring purposes.
In addition, these findings may assist orienting swimmers to the appropriate stroke based on their anthropometric
characteristics.

Keywords: allometric model, maturity, limb lengths, girths and breadths

Competitive swimming is a type of cyclic sports

Sammoud, Negra, Bouguezzi, and Hachana are with the Research Unit
(UR17JS01) “Sport Performance & Health,” Higher Institute of Sport
and Physical Education of Ksar Saïd, Tunis, Tunisia. Nevill is with the
Faculty of Education, Health and Wellbeing, University of Wolver-
hampton, Walsall, United Kingdom. Chaabene is with the Division of
Training and Movement Science, Research Focus Cognition Sciences,
University of Potsdam, Potsdam, Germany; and the High Institute of
Sports and Physical Education of Kef, University of Jendouba, Jen-
douba, Tunisia. Hachana is also with the Higher Institute of Sports and
Physical Education, Manouba University, Tunis, Tunisia. Address
author correspondence to Yassine Negra at negrayassine@yahoo.fr.
activity performed with the aim of covering any given
distance as fast as possible (1). In this context, many
researchers are constantly trying to establish and classify
factors associated with optimal swimming performance.
Therefore, it remains important to recognize that various
factors are important in determining swimming
performance success (1,18). These factors include the
aerobic–anaerobic capacity (15,49), technical level (eg,
stroke technique, coordination, starts, and turns); physi-
cal fitness level (eg, flexibility, strength, and power);
psychological traits (eg, stress control, motivation); and

1
 2 Sammoud et al
the anthropometrical characteristics (eg, height, body
mass, and body mass index). (29).
The relationship between human physical charac-
teristics and sports performance has been a source of
unceasing interest among scientists (14,29,33,36,38).
Notably, the association between anthropometric char-
acteristics and sports performance constitutes a worth-
while marker for talent identification to engage children
in a long-term athlete development process. A recent
review by Morais et al (34) suggested that anthropomet-
ric characteristics are among the critical factors used as
precursor for an early recognition of talented athletes. It
is worth noting that anthropometrics are more geneti-
Downloaded by Univ de Montreal on 03/17/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}
cally controlled compared with physical fitness attributes
(27). Therefore, anthropometrics are less susceptible to
training than physical fitness attributes. For instance, it
has been established that anthropometrics, such as body
length (eg, height, limbs), are strongly determined by
genetics (level of inherence of 70%) (6,46). These facts
confirm the importance of anthropometric characteristics
to the early detection of talented athletes (24). In swim-
ming, talent identification and development process play
a crucial role in the pursuit of excellence across a long-
term career. In this regard, anthropometric characteris-
tics have been argued to be one of the most important
factors that enable swimmers to achieve a high-
performance level in their careers (21,29).
It is worth noting that most of the previously
published investigations centered their attention in free-
style swimming (2,5,21,29), whereas the breaststroke
remains understudied (16,28). For instance, Lätt et al
(29) indicated that anthropometrical factors explained
45.8% of 100-m front crawl swimming performance in
male swimmers aged 15 years. Likewise, Morais et al
(33) reported the contribution that anthropometric char-
acteristics made on swimming performance and found
that the arm span was significantly correlated with
100-m freestyle swimming performance in 12-year-
old swimmers of both genders (r = −.35). In addition,
Bond et al (5) suggested that anthropometric variables
accounted for 63.8% of 100-m freestyle swimming’s
total variance in 13-year-old male and female swimmers.
Furthermore, Geladas et al (21) revealed that upper
extremity length was significantly associated with
100-m freestyle swimming performance in boys aged
between 12 and 14 years. The same authors revealed that
other anthropometric factors, such as upper extremity
length, stature, and hand length, significantly influenced
girls’ swimming performance of the same age group.
Breaststroke is well known as a very challenging
stroke because of the discontinuous propulsive action of
the arms and legs and its complex time synchronization
(44). During breaststroke, stroke cycle consists of an arm
stroke and a leg kick that occur in sequence and determine
the stroke cycle from the start and throughout the race, rules
that do not apply to any other swimming style [according to
the Fédération Internationale de Natation (17)].
According to Barbosa et al (2), differences between
freestyle and breaststroke do exist in terms of technique,
(Ahead of Print)
energy cost, and stroke length (SL). In addition, previous
findings showed that freestyle is the most economic
stroke, followed by backstroke, butterfly, and breast-
stroke (2). In the same context, by studying the temporal
and velocity changes during the stroke cycle for a range
of stroke rates (SRs), Craig et al (13) demonstrated that
breaststroke swimming’s mechanics was more critical
compared with the other stroke styles. Indeed, swim-
ming velocity can be characterized by its independent
variables: SL and SR (13). Thus, increases or decreases
in swimming velocity are mainly due to a combined
increase or decrease in SR and SL (13,48). In addition,
there is general agreement that SL in breaststroke is the
lowest compared with the other strokes (ie, freestyle,
butterfly, and backstroke) (13). Likewise, several studies
showed a large intracyclic velocity variation of the
body’s center of mass during breaststroke swimming
(30,42,47). This variation makes this swimming stroke
the slowest among the 4 competitive strokes (13).
Until recently, numerous studies (21,28,37,40)
investigated the association between anthropometric
characteristics and swimming performance in various
age groups. However, knowledge on the effect of anthro-
pometric characteristics on swimming performance in
young swimmers is still limited. Moreover, few studies
have investigated the contribution of segment lengths to
swimming performance. As stated in previous studies
(9,22,40), limb segment lengths better predict athletic
performance in sports athletes and sedentary population
in comparison with whole limb lengths. The allometric
modeling approach is currently considered a relevant
mode for solving this issue given its sound theoretical
basis and its biologically driven, as well as its versatile
statistical methodology (37). This approach often pro-
vides a dimensionless expression of data in the form of
ratios [eg, crural index, upper arm-to-lower arm ratio,
and reciprocal ponderal index (stature-to-body mass
ratio = 0.333)]. Furthermore, its modeling techniques
perfectly address the effects of age and sex differences
on growth and biological maturation in motor perfor-
mance interpretation (40). Recently, Nevill et al (37)
applied an allometric modeling approach to identify the
optimal body size and limb length segment associated
with 100-m front crawl speed performance in youth
swimmers (11–16 y). These authors revealed that lean
body mass was the singularly most important whole-
body characteristic and that having greater limb segment
length ratios (ie, arm ratio = lower arm/upper arm; foot-
to-leg ratio = foot/lower leg) was key to personal best
front crawl swim speeds. Likewise, Sammoud et al (40)
applied the same approach to estimate the optimal anthro-
pometric factors associated with 100-m butterfly speed
performance in children and adolescent male and female
swimmers aged ∼13 years. The same authors revealed that
body fat was the singularly whole-body characteristic
associated with butterfly performance. Moreover, Sam-
moud et al (40) demonstrated that limb segment length
ratio (arm ratio = arm span/lower arm) and limb girth ratio
(girth ratio = calf girth/ankle girth) were key to personal

best butterfly swim speed in children and adolescent male
and female swimmers.
Based on the previously described considerations,
there is still a need to carry out further research about the
influence of anthropometric characteristics on different
swimming styles, particularly, breaststroke. We believe
that our allometric approach to identifying the key
anthropometric dimensions and their ratios in youth
breaststroke swimmers is a novel contribution to the
literature especially for the purpose of talent identifica-
tion. Therefore, this study aimed to use allometric
models to estimate the optimal body size, limb segment
length, and girth and breadth ratios associated with
Downloaded by Univ de Montreal on 03/17/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}
100-m breaststroke speed performance in youth swim-
mer athletes. Based on previous findings (33,37), we
hypothesized that 100-m breaststroke performance is
dependent on limb segment length rather than the
whole-body size in youth swimmers of both genders.
Methods
Experimental Approach to the Problem
To determine if anthropometric characteristics are
important to the progression of swimmers, several
body measurements were taken including height, sitting
height, body mass, skinfold thicknesses, arm span, limb
lengths, girths, and breadths. The body composition was
then calculated using various formulas (43). The per-
centage of body fat (%BF) of children with triceps and
subscapular skinfolds <35 mm was calculated as fol-
lows: for boys, %BF = 1.21 (sum of 2 skinfolds) − 0.008
(sum of 2 skinfolds2) − 1.7; for girls, %BF = 1.33 (sum
of 2 skinfolds) − 0.013 (sum of 2 skinfolds2) − 2.5. The
%BF for children with triceps and subscapular skinfolds
>35 mm was calculated as follows: for boys, %BF =
0.783 (sum of 2 skinfolds) − 1.7; for girls, %BF = 0.546
(sum of 2 skinfolds) + 9.7. Fat mass was calculated as
follows: fat mass = (body mass × %BF)/100; fat-free
mass (kg) = body mass − fat mass.
In addition, maturity offset was assessed by pre-
dicting age at peak height velocity based on age, body
mass, height, leg length, and sitting height using the
predictive equation established by Mirwald et al (32). In
girls, maturity offset = −9.376 + (0.0001882·leg length
and sitting height interaction) + (0.0022·age and leg
length interaction) + (0.005841·age and sitting height
interaction) − (0.002658·age and weight interaction) +
(0.07693·weight by height ratio × 100) (32). In boys,
maturity offset = −9.236 + (0.0002708·leg length and
sitting height interaction) − (0.001663·age and leg length
interaction) + (0.007216·age and sitting height interac-
tion) + (0.02292·weight by height ratio × 100) (32). The
time of 100-m breaststroke swimming performance was
recorded during competitions, as it represents the peak of
the performance of the Tunisian Championship estab-
lished according to the official measurement rules (17)
using a high-technology electronic timing (Omega,
Bienne, Switzerland).
(Ahead of Print)
Allometric Model and Breaststroke Performance 3
Participants
In total, 59 breaststroke specialist swimmers [male: n =
39, age = 11.50 (1.26) y; female: n = 20, age = 12.05
(0.99) y] participated in this study. All participants
were involved in 5–6 training sessions per week
[4000 (1000) m/session; 8 (1) h/wk]. In addition, train-
ing sessions included the 4 strokes. Written informed
parental consent and participant assent were obtained
prior to the start of the study. All youth athletes and their
parents/legal representatives were informed about the
experimental protocol and its potential risks and benefits
before the commencement of the research project. The
study was approved by the local ethics institutional
review committee for the ethical use of human subjects
at Ksar Saïd University, Tunisia.
Performance Time and Average Swimming
Speed
The swimming times and/or speeds (speed based on the
race time) expressed in seconds and meters per second,
respectively, were adopted as our measures of swimming
performance. Swimming performance was recorded in a
25-m swimming pool. The breaststroke average speed
was calculated as the ratio between distances swam and
the total time recorded in this distance (in meters per
second). Performance (in seconds) was measured with a
high-technology electronic timing (Omega) and was
extracted for all subjects from the official results pub-
lished by the Tunisian Swimming Federation during the
Winter National Championships. Water temperature was
kept between 25° and 28°, as determined by Fédération
Internationale de Natation (17).
Anthropometric Measurements
All the anthropometrical measurements were taken by one
trained anthropometrist assisted by a recorder in accor-
dance with standardized procedures of the International
Society for the Advancement of Kinanthropometry (45)
(Table 1). Testing was carried out in a standardized order
after a proper calibration of the measuring instruments.
Each swimmer’s height (in meters) and body mass (in
kilograms) were assessed to the nearest 0.1 cm and 0.1 kg,
using a Seca stadiometer and a Seca weighing scale
(Seca Instruments Ltd, Hamburg, Germany), respectively.
Skinfolds measurements (in millimeters) were taken on
the right-hand side of the body at 2 sites (the triceps
and the subscapular) using Harpenden skinfold calipers
(Harpenden Instruments, Cambridge, UK). Skinfold data,
alongside the skinfold equation of Slaughter et al (43),
were used to estimate the body fat mass and fat-free
mass. The following limb lengths, girths, and breadths
were assessed using a large sliding caliper and a non-
stretchable tape measure via direct measures using
landmarks techniques: arm span, upper limb length, upper
arm length, lower arm length, hand lengths, lower
limb length, thigh length, leg length, foot length, arm-
relaxed girth, forearm girth, wrist girth, thigh girth, calf
 4 Sammoud et al

Table 1 Participants’ Anthropometrical of the tibiale lateral to the top of the box (or the floor).
Characteristics and 100-m Breaststroke Foot length was determined as the distance from the
Swimming Performance akropodion (ie, the tip of the longest toe, which may be
the first or second phalanx) to the pternion (ie, most
Males Females posterior point on the calcaneus of the foot). Arm-
Variables (n = 39) (n = 20) relaxed girth was measured at the marked level of the
Age, y 11.5 (1.3) 12.0 (1.0) midacromiale–radiale landmark. The tape should be
Maturity offset, y −2.3 (1.10) 0.04 (1)** positioned perpendicular to the long axis of the arm.
Forearm girth was taken at the maximum girth of the
Age at peak height velocity, y 13.8 (0.6) 12.0 (0.5)**
forearm distal to the humeral epicondyles. Wrist girth
Body mass, kg 41.5 (9.5) 46.0 (8.6) measure is taken distal to the styloid processes. It is the
Height, cm 149.9 (10.4) 155.9 (8.0) minimum girth in this region. Thigh girth measure was
Sitting height, cm 74.4 (5.8) 76.4 (7.9)* taken at the marked midtrochanterion–tibiale lateral site.
Downloaded by Univ de Montreal on 03/17/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}

Body fat, % 16.8 (5.5) 19.0 (4.3) Calf girth was defined as the maximum girth of the calf
Fat mass, kg 7.2 (3.8) 8.8 (3.1) taken at the marked medial calf skinfold site. Ankle girth
was defined as the minimum girth of the ankle taken at
Fat-free mass, kg 34.3 (6.8) 37.1 (6.4)
the narrowest point superior to the sphyrion tibiale.
Body mass index, kg·m−2 18.3 (2.4) 18.8 (2.5) Biacromial breadth was determined as the distance
Upper limb length, cm 69.3 (5.3) 72.0 (4.4) between the most lateral points of the acromion pro-
Upper arm length, cm 29.0 (2.2) 30.4 (2.0)* cesses. Biiliocristal breadth was defined as the distance
Forearm length, cm 23.2 (1.9) 23.7 (1.7) between the most lateral points on the iliac crests.
Hand length, cm 18.0 (1.6) 18.7 (1.0) The intraclass correlation coefficients for test–retest
reliability for all anthropometric and skinfolds measures
Lower limb length, cm 81.6 (6.1) 85.0 (4.9)*
ranged from .96 to .99.
Thigh length, cm 38.9 (5.8) 42.0 (2.2)*
Leg length, cm 42.0 (3.4) 43.4 (2.5)
Statistical Analysis
Foot length, cm 25.1 (2.2) 25.5 (1.1)
Arm-relaxed girth, cm 23.0 (3.0) 23.8 (2.3) Descriptive statistics were computed and expressed as
means and SDs. To identify the most suitable anthropo-
Forearm girth, cm 21.1 (1.9) 21.5 (1.5)
metric characteristics [ie, body mass (M), fat-free mass,
Wrist girth, cm 14.7 (1.1) 15.4 (2.3) fat mass (FM), stature (S), limb lengths, girths, or
Thigh girth, cm 43.9 (4.5) 46.1 (5.6) breadths (L)] that are associated with 100-m breaststroke
Calf girth, cm 30.3 (3.1) 30.7 (3.0) speed performance, we adopted the proportional multi-
Ankle girth, cm 20.5 (1.8) 20.5 (1.5) plicative model with allometric body size components,
Biacromial breadth, cm 41.5 (3.9) 43.3 (2.3) similar to the 100-m personal best front crawl swim
model used to measure speeds in children and adoles-
Biiliocristal breadth, cm 24.4 (2.4) 25.7 (2.4)
cents (37) and to the 100-m butterfly speed model used
Arm span, cm 150.4 (13.5) 158.3 (9.7)* among swimmers of the same age group (40). The
Swimming performance, s 97.7 (13.4) 95.4 (9.5) benefits of this model are to have proportional body
Breaststroke average speed, m·s−1 1.0 (0.1) 1.1 (0.1) size components. In addition, this multiplicative allome-
Note. Data are presented as mean (SD). tric model was extensively adopted in physical anthro-
*Significant differences between girls and boys (P < .05). **Significant pology (20) and biology (3,7), and a recent call has been
differences between girls and boys (P < .0001). made for consideration within molecular biology and
gene expression investigations because human size, and
girth, ankle girth, biacromial breadth, and biiliocristal what it involves, will always be of concern (28). In
breadth. addition, this model has been systematically used to
Upper arm length was measured from landmarks partition out differences in size in efficient ways, such as
placed to acromiale and dactylion while athletes stood in differences in maximal oxygen consumption (VO2peak)
the erect position. Arm length was determined as the (3) and in a variety of motor tests (37).
distance between the marked acromiale and radiale The Multiplicative Model.
landmarks. The lower arm length was measured by
calculating the distance between the radiale and stylion Breakstroke average speed ðm · s−1 Þ
landmarks. For the hand length, the measure was taken
= a · ðMÞk1 · ðSÞk2 · ΠðLi Þki · expðb biological age
as the shortest distance from the marked midstylion line
to the dactylion. Lower limb length was determined by þ c · FMÞ · ε, (1)
subtracting sitting height from standing height. Thigh
length was determined as the distance between the where “a” is a constant, “b” and “c” are the slope
marked trochanterion and tibiale lateral landmarks. parameters associated with biological age and FM,
Leg length was measured as the distance from the height and Π(Li)ki (i = 3, 4, : : : , n) signifies the product of all

(Ahead of Print)
 Allometric Model and Breaststroke Performance 5

limb segment lengths, girths, or breadths measurements
raised to the power of ki; with i = 3 to i = n representing
the full range of limb lengths, girths, and breadths
recorded for the swimmers (for the full list, see “Anthro-
pometric Measurements” section).
The benefits of this model are to have proportional
body size components. Note that “ε,” the multiplicative
error ratio, also assumes the error will increase in
proportion to the athlete’s swimming performance.
The model [Equation (1)] can be linearized with a log
transformation. A linear regression analysis on log
(breaststroke average speed, m·s−1) can then be used
to estimate the unknown parameters of the log-
Downloaded by Univ de Montreal on 03/17/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}
all the anthropometric characteristics and the swimming
performance between boys and girls (P > .05) except for
the arm span, thigh length, lower limb length, upper arm
length, and sitting height (all Ps < .05). Table 2 indicates
the parsimonious solution to the backward elimination
regression analysis of ln(breaststroke average speed,
m·s−1). The multiplicative allometric model exploring
the association between 100-m average breaststroke
speed performance (in meters per second) and the dif-
ferent anthropometric characteristics estimated that fat
mass (as a whole-body size dimension), 2 upper limb
lengths (forearm and hand length), leg length for the
lower limbs, 2 breadths (biacromial and biiliocristal

transformed model:
lnðbreakstroke average speed ðm · s−1 ÞÞ
= k1 · logðMÞ þ k2 · logðSÞ
X additional year in youth swimmers.
þ ki · lnðLi Þ þ a þ b · biological age
þ c · FM þ log ðεÞ: (2)
Having fitted the saturated model with all available body
size variables, an appropriate “parsimonious” model can
be obtained using “backward elimination” (37) in which
the least important body size and limb segment length,
girth, and breadth variables at each step are eliminated
from the model. A parsimonious model is a model that
achieves a desired level of explanation or prediction with
as few predictor variables as possible. Further categori-
cal or group differences within the population (eg, sex)
can be explored by allowing the constant intercept
parameter [eg, ln(a) refers to natural logarithms in
Equation (2)] to vary for each group.
Results
Table 1 shows anthropometric characteristics and swim-
ming performance data of participants. Boys’ and girls’
age at peak height velocity was 13.8 (0.6) and 12.0 (0.5),
respectively. No significant difference was detected for
breadths), and 2 girths (forearm and wrist girth) are
the main significant predictors of log-transformed swim
performance. Our allometric model detected that breast-
stroke speed performance increases by 2.5% every
The adjusted coefficient of determination (R2) was
76.6% with the log-transformed error ratio being 0.06 or
6%, having taken antilogs. The constant “a” did not vary
significantly with sex, suggesting that the model can be
regarded as common for children of either sex (Table 2).
Discussion
To the authors’ knowledge, this is the first study that
aimed to identify the most appropriate body size char-
acteristics related to 100-m breaststroke performance in
youth swimmers. The outcomes of this study support
previous investigations, indicating that anthropometrics
are highly related to youth swimmers’ performance
(21,28,37,39,40). Specifically, in agreement with previ-
ous findings (37,40), the present results demonstrated
that fat mass was the singularly most important whole-
body size characteristic negatively associated with
breaststroke performance. Particularly, Sammoud et al
(40) revealed that fat mass was the only whole-body size
characteristic negatively associated with butterfly speed
performance in children and adolescent male and female

Table 2 Estimated Body Size and Limb Segment Parameter (B) Obtained From Regression
Analysis Predicting Log-Transformed 100-m Breaststroke Average Speed Performance
Variables in the model B SE P 95% Confidence interval for B
Constant −6.813 0.561 <.000 −7.942 to −5.685
Age at peak height velocity 0.025 0.009 .008 0.007 to 0.042
lnForearm girth 0.690 0.240 .006 0.207 to 1.173
lnWrist girth −0.348 0.128 .009 −0.604 to −0.091
lnBiacromial breadth 0.565 0.221 .014 0.121 to 1.009
lnBiiliocristal breadth 0.403 0.109 .001 0.185 to 0.622
lnLeg length 0.673 0.264 .014 0.143 to 1.203
lnHand length 0.309 0.195 .120 −0.083 to 0.702
lnForearm length −0.418 0.220 .064 −0.861 to 0.025
Fat mass −0.018 0.004 <.001 −0.026 to −0.010
Abbreviation: ln, natural log.

(Ahead of Print)
 6 Sammoud et al
swimmers. Similarly, Nevill et al (37) demonstrated
that fat-free mass was the singularly most important
whole-body size characteristic positively associated
with 100-m freestyle performance in male and female
adolescent swimmers. Jürimäe et al (28) reported a
significant correlation between fat-free mass and 400-
m front crawl performance in young swimmers. Like-
wise, Nasirzade et al (35) revealed a significant negative
correlation between muscle thickness (r ranges from .42
to .56) and 200-m front crawl performance in young
swimmers [14.5 (0.8) y]. In addition, Helmuth (25)
found that fat-free mass significantly correlated (r =
.73) with 100-m front crawl performance in 8- to
Downloaded by Univ de Montreal on 03/17/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}
16-year-old male swimmers.
The disadvantage of having higher fat mass sug-
gests that swimmers require greater fat-free mass,
implying that they require greater muscularity to propel
themselves faster through the water (40). Based on the
previously mentioned consideration, the positive asso-
ciation between fat-free mass and swimming perfor-
mance may indicate that this anthropometric variable
contributed significantly to the prediction of propulsion
force. This propulsion force may translate to improve
breaststroke performance in youth boy and girl swim-
mers. In addition, inertial properties of the limbs may
influence SR, particularly, mass and mass distribution
(37). Overall, greater fat-free mass is likely to be asso-
ciated with greater fat-free mass in the limbs, translating
into greater SR and subsequent propulsion (37).
Our study showed that height and body mass did not
contribute significantly to the model. These findings
extended previously reported results (21,37,40), sug-
gesting that the advantage of longer levers was either
limb segment length, girth, or breadth-specific rather
than having a more general whole-body size advantage.
Likewise, our allometric model revealed that swimming
performance average speed increases by 2.5% every
additional year of the swimmer’s biological age (see
Table 2). Based on this result, coaches should take into
consideration the biological age rather than the chrono-
logical one as a key factor in determining performance.
In addition, the key indicator from the allometric
model reported in Table 2 is the advantage of having
greater limb segment length ratios (ie, arm length ratio =
hand length/forearm length) and greater limb segment
girth ratios (ie, arm girth ratio = forearm girth/wrist girth)
that are associated with swimming faster breaststroke
speeds. Our results illustrated that forearm length made a
negative contribution, whereas the hand length made a
positive contribution to the 100-m breaststroke perfor-
mance. These results are in agreement with previous
investigations (21,23). For instance, Perez et al (38)
revealed a significant relationship between hand length
and the average of swimming speed (r = .51). In the same
context, Geladas et al (21) demonstrated a significant
relationship between 100-m freestyle performance time
and hand length measures (r = −.6 and −.3 in boys and
girls young swimmers, respectively). The same authors
also reported a significant relationship between the foot
(Ahead of Print)
length and the 100-m freestyle performance time
(r = −.49 and −.16 in boys and girls, respectively).
Likewise, Helmuth (25) reported a significant correla-
tion between hand size and swimming performance in
young male and female swimmers (8–16 y). It is note-
worthy that the significant relationship between 100-m
breaststroke performance and hand length could be
mainly due to the fact that propulsive force; hence,
swimming performance is positively affected by the
large upper extremity length (26). Zamparo et al (49)
argued that having a greater hand length will also act to
increase surface area, thus leading to a greater propelling
economy. Finally, the finding of the positive contribu-
tion of the hand length in breaststroke performance could
be attributed to the fact that propulsive force and hence
swimming performance is positively affected by the
large upper extremity length.
As with hand length, our results revealed a signifi-
cant contribution by the leg length in the 100-m
breaststroke performance. In addition, our findings dem-
onstrated the importance of having shorter forearm
length for better 100-m breaststroke swimming perfor-
mance. These observations are consistent with those
established by Nevill et al (37), who illustrated that
longer lever length (upper arm and lower leg) is poten-
tially mechanically disadvantageous in some ways in
100-m front crawl performance in adolescent swimmers
of both sex. The same authors indicated that the involved
muscles have to exert greater force and, hence, use
greater energy. Likewise, the importance of having
shorter forearm length for better 100-m butterfly speed
performance in swimmers has been recently shown (40).
The same authors (40) indicated that swimmers with a
shorter arm length have naturally a better swimming
technique with respect to those with longer upper limbs.
In the same context, Grimston and Hay (23) reported that
the dimensions of body segments, such as the upper
limbs or lower limbs lengths, influence the mechanics of
swimming technique and muscle activity. Consequently,
it seems crucial to focus on teaching the correct
swimming technique starting from the early years of
training (31,40).
In addition, our findings revealed that an increase
in a forearm girth or volume would improve the
100-m breaststroke swimming performance. This is in
agreement with findings recently established by Sam-
moud et al (40) who revealed that an increase in a calf
girth or volume would increase the 100-m butterfly
speed performance in adolescent swimmer athletes.
Santos et al (41) revealed a significant association
between the arm muscle area and the propulsive force
of the arm in young male swimmers [14 (1.28) y]. The
same authors (41) demonstrated that the increase of the
arm muscle area contributes to a greater capacity for
strength. This significant association could be explained
by the fact that swimmers having greater limb volume or
greater muscularity seem to generate higher propulsive
force and propel themselves faster through
water (37,40).

The current findings revealed that having a greater
wrist girth impairs performance. This is in agreement
with those established by Sammoud et al (40) who
detected a negative contribution of the ankle girth in
the 100-m butterfly speed performance. A large wrist
girth would increase resistance through water, therefore,
increasing the energy cost of swimming (11,12,49). This
may at least partially explain the disadvantage of having
a larger wrist girth.
Our results also detected a positive association
between the biacromial and biiliocristal breadths with
the 100-m breaststroke speed performance (Table 2).
Recently, Sammoud et al (40) revealed that having greater
Downloaded by Univ de Montreal on 03/17/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}
biacromial and biiliocristal breadths are key positive
anthropometric indicators associated with better 100-m
butterfly speed performance. Likewise, Geladas et al (21)
showed that swimming sprint time was significantly
correlated with biacromial (r = −.61) and biiliocristal
breadths (r = −.46) in male swimmers. All together, these
findings may be related to the fact that swimmers with
broad shoulders are better suited for high power output in
the water (8). In addition, the positive association between
the body breadths and 100-m breaststroke performance in
our study suggest that a larger body cross-sectional area in
swimmers may be related to sprint performance time (26).
According to the effect of cross-sectional area on the
pressure drag, several studies (4,10) have shown that
some anthropometric parameters, such as the chest girth,
depth, and breadth, are significantly correlated with drag
values. In addition to the anthropometric parameters, the
shape and the contour of the body are important factors
too affecting the pressure drag because they determine
how the flow moves over the body (19).
The main limitations of this research may be sum-
marized as follows: 1) variables related to functional
fitness (eg, muscular strength or flexibility) that might
influence stroke mechanics were not included in the
model; 2) variables from other domains that may also
play an important role in youth swimmers’ performance
(eg, motor control, hydrodynamics, genetics) were not
taken into consideration; 3) a direct measure of the
propulsive efficiency was not adopted; and 4) bio-
mechanical testing methods should be implemented in
future studies to obtain an in-depth knowledge regarding
the allometric associations between biomechanical,
shape, and 100-m breaststroke speed performance.
Practical Applications
The current study has several practical applications.
First, the present results highlighted the importance of
considering anthropometric characteristics of youth
swimmers for talent identification and/or athlete moni-
toring purposes by coaches and sports scientists. In
addition, it is only by adopting multiplicative allometric
models that the previously mentioned ratios could have
been derived. Furthermore, results of the present study
illustrated that 1) fat mass was the singularly most
important whole-body size characteristic and that height
(Ahead of Print)
Allometric Model and Breaststroke Performance 7
and body mass did not contribute significantly to the
allometric model; 2) 100-m breaststroke average speed
performance was strongly positively associated with the
segment length ratio (arm ratio and girth ratio); and
3) leg length, biacromial breadth, and biiliocristal
breadth were positively associated with the 100-m
breaststroke performance in youth swimmers.
Acknowledgments
The authors would like to thank the athletes and their trainers
who willingly and patiently contributed to this study. No
potential conflict of interest was reported by the authors.
References
1. Barbosa TM, Costa M, Marinho JD, Coelho M, Moreira
M, Silva AJ. Modeling the links between young swim-
mer’s performance: energetic and biomechanical profiles.
Pediatr Exerc Sci. 2010;22:379–91. PubMed doi:10.
1123/pes.22.3.379
2. Barbosa TM, Fernandes RJ, Keskinen KL, Vilas-Boas JP.
The influence of stroke mechanics into energy cost of
elite swimmers. Eur J Appl Physiol. 2008;103:139–49.
PubMed doi:10.1007/s00421-008-0676-z
3. Batterham A, George K, Mullineaux D. Allometric scal-
ing of left ventricular mass by body dimensions in males
and females. Med Sci Sports Exerc. 1997;29:181–6.
PubMed doi:10.1097/00005768-199702000-00003
4. Benjanuvatra N, Blanksby BA, Elliott BC. Morphology
and hydrodynamic resistance in young swimmers. Pediatr
Exerc Sci. 2001;13(3):246–55. doi:10.1123/pes.13.3.246
5. Bond D, Goodson L, Oxford SW, Nevill AM, Duncan
MJ. The association between anthropometric variables,
functional movement screen scores and 100 m freestyle
swimming performance in youth swimmers. Sports. 2015;
3:1–11. doi:10.3390/sports3010001
6. Bouchard C, Malina RM, Perusse L. Genetics of Fitness
and Physical Performance. Champaign, IL: Human
Kinetics; 1997.
7. Bown J, West G. Scaling in Biology. New York, NY:
Oxford University Press; 2000.
8. Carter JEL, Ackland TR. Kinanthropometry in Aquatic
Sports: A Study of World Class Athletes. Champaign, IL:
Human Kinetics; 1994.
9. Caruso JF, Daily JS, Mason ML, et al. Anthropometry as a
predictor of vertical jump heights derived from an in-
strumented platform. J Strength Cond Res. 2012;26:284–
92. PubMed doi:10.1519/JSC.0b013e31821d97c0
10. Chatard JC, Lavoie JM, Bourgoin B, Lacour JR. The
contribution of passive drag as a determinant of swim-
ming performance. Int J Sports Med. 1990;11(5):367–72.
PubMed doi:10.1055/s-2007-1024820
11. Chatard JC, Lavoie JM, Lacour JR. Analysis of determi-
nants of swimming economy in front crawl. Eur J Appl
Physiol. 1991;61:88–92. doi:10.1007/BF00236699
12. Chatard JC, Padilla S, Cazorla G, Lacour JR. Influence of
body height, weight, hydrostatic lift and training on the
 8 Sammoud et al
energy cost of the front crawl. N Z J Sports Med. 1985;
13:82–4.
13. Craig AB Jr, Skehan PL, Pawelczyk JA, Boomer WL.
Velocity, stroke rate, and distance per stroke during elite
swimming competition. Med Sci Sports Exerc. 1985;
17(6):625–34. doi:10.1249/00005768-198512000-00001
14. Deprez D, Buchheit M, Fransen J, Pion J, Lenoir M,
Philippaerts RM, Vaeyens R. A longitudinal study inves-
tigating the stability of anthropometry and soccer-specific
endurance in pubertal high-level youth soccer players.
J Sports Sci Med. 2015;14(2):418–26. PubMed
15. Di Prampero P. The energy cost of human locomotion on
land and in water. Int J Sports Med. 1986;7:55–72. doi:10.
Downloaded by Univ de Montreal on 03/17/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}
1055/s-2008-1025736
16. Duchè G, Falgairette G, Bedu M, Lac G. Analysis of
performance of prepubertal swimmers assessed from
anthropometric and bio-energetic characteristics. Eur J
Appl Physiol Occup Physiol. 1993;66(5):467–71. doi:10.
1007/BF00599623
17. Fédération Internationale de Natation. Talent identifica-
tion programmes [Internet]. 2014 [cited 2014 Sep 18].
Available from: http://www.fina.org/
18. Figueiredo R, Pendergast DR, Vilas-Boas JP, Fernandes
RJ. Interplay of biomechanical, energetic, coordinative,
and muscular factors in a 200 m front crawl swim. Biomed
Res Int. 2013;2013:897232. PubMed doi:10.1155/2013/
897232
19. Fish FE, Hui CA. Dolphin swimming—a review. Mam-
mal Rev. 1991;21(B):181–95. doi:10.1111/j.1365-2907.
1991.tb00292.x
20. Fleagle J. Size and scaling in primate biology. In: Jungers
W, editor. Size and Adaptation in Primates. New York,
NY: Plenum Press; 1985;1–19.
21. Geladas N, Nassis GP, Pavlicevic S. Somatic and physical
traits affecting sprint swimming performance in young
swimmers. Int J Sports Med. 2005;26:139–44. doi:10.
1055/s-2004-817862
22. Green L, Gabriel DA. Anthropometrics and electromyog-
raphy as predictors for maximal voluntary isometric arm
strength. J Sport Health Sci. 2012;1:107–13. doi:10.1016/
j.jshs.2012.05.004
23. Grimston SK, Hay JG. Relationships among anthropo-
metric and stroking characteristics of college swimmers.
Med Sci Sports Exerc. 1986;18:60–8. doi:10.1249/
00005768-198602000-00011
24. Hahn A. Identification and selection of talent in Australian
rowing. Excel. 1990;6(3):5–11.
25. Helmuth HS. Anthropometric survey of young swimmers.
Anthropol Anz. 1980;38:17–34. PubMed
26. Huijing P, Toussaint H, Mackay R, et al. Active drag
related to body dimensions. In: Ungerechts B, Reischle K,
Wilke K, editors. Swimming Science V. Champaign, IL:
Human Kinetics; 1988, pp. 31–8.
27. Issurin VB. Evidence-based prerequisites and precursors
of athletic talent: a review. Sports Med. 2017;47(10):
1993–2010. doi:10.1007/s40279-017-0740-0
28. Jürimäe J, Haljaste K, Cicchella A, Lätt E, Purge P,
Leppik A, Jürimäe T. Analysis of swimming performance
from physical, physiological, and biomechanical
(Ahead of Print)
parameters in young swimmers. Pediatr Exerc Sci. 2007;
19:70–81. PubMed doi:10.1123/pes.19.1.70
29. Lätt E, Jürimäe J, Mäestu J, et al. Physiological, bio-
mechanical and anthropometrical predictors of sprint
swimming performance in adolescent swimmers. J Sports
Sci Med. 2010;9(3):398–404. PubMed
30. Leblanc H, Seifert L, Tourny-Chollet C, Chollet D.
Velocity variations in breaststroke swimmers of different
competitive levels. Int J Sports Med. 2007;28:140–7.
doi:10.1055/s-2006-924205
31. Marinho DA, Barbosa TM, Reis VM, et al. Swimming
propulsion forces are enhanced by a small finger spread. J
Appl Biomech. 2010;26(1):87–92. PubMed doi:10.1123/
jab.26.1.87
32. Mirwald RL, Baxter-Jones AD, Bailey DA, Beunen GP.
An assessment of maturity from anthropometric measure-
ments. Med Sci Sports Exerc. 2002;34(4):689–94.
PubMed
33. Morais JE, Jesus S, Lopes V, Garrido N, Silva A, Marinho
D, Barbosa TM. Linking selected kinematic, anthropo-
metric and hydrodynamic variables to young swimmer
performance. Pediatr Exerc Sci. 2012;24:649–64. doi:10.
1123/pes.24.4.649
34. Morais JE, Silva AJ, Marinho DA, Lopes VP, Barbosa
TM. Determinant factors of long-term performance devel-
opment in young swimmers. Int J Sports Physiol Perf.
2017;12:198–205. doi:10.1123/ijspp.2015-0420
35. Nasirzade A, Sadeghi H, Sobhkhiz A, Mohammadian K,
Nikouei A, Baghaiyan M, Fattahi A. Multivariate analysis
of 200-m front crawl swimming performance in young
male swimmers. Acta Bioeng Biomech. 2015;17(3):137–
43. PubMed
36. Negra Y, Chaabene H, Hammami M, Khlifa R, Gabett T,
Hachana Y. Allometric scaling and age related differences
in change of direction speed performances of young
soccer players. Sci Sports. 2015;31(2):e19–26. doi:10.
1016/j.scispo.2015.10.003
37. Nevill AM, Oxford SW, Duncan MJ. Optimal body size
and limb length ratios associated with 100-m personal-
best swim speeds. Med Sci Sports Exerc. 2015;47:
1714–8. doi:10.1249/MSS.0000000000000586
38. Perez AJ, Bassini CF, Pereira BMF, Sarro KJ. Correlation
between anthropometric variables and stroke length and
frequency in swimmers of Espirito Santo. Rev Mackenzie
Educ Fís Esp. 2011;10:19–27.
39. Reis VM, Barbosa TM, Marinho DA, et al. Physiological
determinants of performance in breaststroke swimming
events. Int J Sports Med. 2010;11:324–35.
40. Sammoud S, Nevill AM, Negra Y, Bouguezzi R, Chaa-
bene H, Hachana Y. Allometric associations between
body size, shape, and 100-m butterfly speed performance.
J Sports Med Phys Fitness. E-pub ahead of print 2017.
doi:10.23736/S0022-4707.17.07480-1
41. Santos MA, Barbosa ML Jr, Melo WV, Veronese da
Costa A, Costa MC. Estimate of propulsive force in front
crawl swimming in young athletes. Open Access J Sports
Med. 2012;3:115–20. PubMed
42. Schnitzler C, Seifert L, Ernwein V, Chollet D. Arm
coordination adaptations assessment in swimming. Int J

Sports Med. 2008;29(6):480–6. PubMed doi:10.1055/s-
2007-989235
43. Slaughter MH, Lohman TG, Boileau RA, Horswill CA,
Stillman RJ, Van Loan MD, Bemben DA. Skinfold
equations for estimation of body fatness in children
and youth. Hum Biol. 1988;60:709–23. PubMed
44. Soares PM, Sousa F, Vilas-Boas JP. Differences in breast-
stroke synchronisation induced by different race velocities.
In: Keskinen KL, Komi PV, Hollander AP, editors. Bio-
mechanics and Medicine in Swimming VIII. Jyväskylä,
Finland: Gummerus Printing; 1999, pp. 53–7.
45. Stewart A, Marfell-Jones M, Olds T, de Ridder H. Inter-
national Standards for Anthropometric Assessment.
Downloaded by Univ de Montreal on 03/17/18, Volume ${article.issue.volume}, Article Number ${article.issue.issue}
Lower Hutt, New Zealand: ISAK; 2011, pp. 57–72.
46. Szopa J, Mleczko E, Zychowska M, et al. Possibilities of
determination of genetic conditionings of somatic and
(Ahead of Print)
Allometric Model and Breaststroke Performance 9
functional traits on the backgrounds of family studies.
J Hum Kinet. 1999;2:21–36.
47. Takagi H, Sugimoto S, Nishijima N, Wilson B. Differ-
ences in stroke phases, arm-leg coordination and velocity
fluctuation due to event, gender and performance level in
breaststroke. Sports Biomech. 2004;3(1):15–27. PubMed
doi:10.1080/14763140408522827
48. Toussaint H, Carol A, Kranenborg H, Truijens M. Effect
of fatigue on stroking characteristics in an arms-only
100-m front-crawl race. Med Sci Sports Exerc. 2006;
38:1635–42. doi:10.1249/01.mss.0000230209.53333.
31
49. Zamparo P, Pendergast D, Mollendorf J, Termin A,
Minetti A. An energy balance of front crawl. Eur J
Appl Physiol. 2005;94:134–44. doi:10.1007/s00421-
004-1281-4