See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/321871762

Biomechanical insights into the determinants of speed in the fencing lunge

Article  in  European Journal of Sport Science · December 2017

DOI: 10.1080/17461391.2017.1414886

CITATIONS READS

0 272

5 authors, including:

Yanfei Guan Li Guo

University of British Columbia - Vancouver Shanghai University of Sport
4 PUBLICATIONS   0 CITATIONS    2 PUBLICATIONS   0 CITATIONS   

SEE PROFILE SEE PROFILE

Nana Wu Darren Warburton

University of British Columbia - Vancouver University of British Columbia - Vancouver
7 PUBLICATIONS   3 CITATIONS    316 PUBLICATIONS   13,617 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

exercise for type 1 diabetes View project

Evaluation of specific performance in indoor rock climbers View project

All content following this page was uploaded by Nana Wu on 09 March 2018.

The user has requested enhancement of the downloaded file.

 European Journal of Sport Science

ISSN: 1746-1391 (Print) 1536-7290 (Online) Journal homepage: http://www.tandfonline.com/loi/tejs20

Biomechanical insights into the determinants of

speed in the fencing lunge

Yanfei Guan, Li Guo, Nana Wu, Lingli Zhang & Darren E.R. Warburton

To cite this article: Yanfei Guan, Li Guo, Nana Wu, Lingli Zhang & Darren E.R. Warburton (2018)
Biomechanical insights into the determinants of speed in the fencing lunge, European Journal of
Sport Science, 18:2, 201-208, DOI: 10.1080/17461391.2017.1414886

To link to this article: https://doi.org/10.1080/17461391.2017.1414886

Published online: 17 Dec 2017.

Submit your article to this journal

Article views: 30

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=tejs20
European Journal of Sport Science, 2018
Vol. 18, No. 2, 201–208, https://doi.org/10.1080/17461391.2017.1414886

ORIGINAL ARTICLE

Biomechanical insights into the determinants of speed in the fencing

lunge

YANFEI GUAN1, LI GUO2, NANA WU1, LINGLI ZHANG2, & DARREN

E.R. WARBURTON1
1
Cardiovascular Physiology and Rehabilitation Laboratory, University of British Columbia, Vancouver, British Columbia,
Canada & 2School of Kinesiology, Shanghai University of Sport, Shanghai, People’s Republic of China.

Abstract
For fencing, speed of the lunge is considered critical to success. The aim of this study is to investigate determinants of lunge
speed based on biomechanics. Ground reaction force (GRF) and three-dimensional kinematic data were collected from 7 elite
fencers and 12 intermediate-level fencers performing maximum-effort lunges. The results showed that elite fencers acquired a
higher horizontal peak velocity of the centre of gravity (HPV) and concomitantly a higher horizontal peak GRF exerted by rear
leg (PGRF) than intermediate-level fencers (P < .01). Studying the affecting factors, elite fencers obtained higher joint peak
power, joint peak moment, and range of motion of rear knee than intermediate-level fencers (P < .05) during the lunge, and
these parameters were significantly correlated with both HPV and PGRF (P < .05). Both elite and intermediate-level fencers
had joint flexion before the extension in forward knee; however, the latter showed greater flexion, higher peak angular velocity
and less time for extension compared to the former (P ≤ .05). Our findings suggest that training aimed at enhancing strength
and power of rear knee extensors is important for fencers to improve speed of the lunge. Also, increasing the extension of rear
knee during the lunge, at the same time decreasing the flexion of the forward knee before extension are positive for lunge
performance.

Keywords: 3D analysis, biomechanics, kinesiology

Highlights
. Elite fencers acquired higher HPV and PGRF than intermediate-level fencers during the lunge.
. Elite fencers obtained higher joint peak power, joint peak moment, and range of motion of rear knee during the lunge,
which resulted in their higher PGRF and HPV compared with intermediate-level fencers.
. Differences were shown between the elite and intermediate-level fencers in the moving patterns of the forward knee joint
during the lunge.

Introduction
The lunge is the most frequently used attack form in
fencing, and it is performed extensively throughout a
competitive match. For instance, during inter-
national sabre tournaments, the lunge is used every
23.9 s in male fencers, and every 20 s in female
fencers (Aquili et al., 2013). The ability to execute
a lunge efficiently in a match is critical for fencing
performance (Turner et al., 2014). The velocity and
the distance travelled are two main factors in an effec-
tive lunge, while the former is considered as a more
crucial factor for fencers to conduct a successful
lunge providing less time for the response of the
opponents in extremely fast-paced fencing matches
(Gholipour, Tabrizi, & Farahmand, 2008; Turner
et al., 2013).
In most of previous studies, the horizontal peak vel-
ocity (HPV) of the body centre of gravity (CG) was
selected to represent the lunge speed (Cronin,
McNair, & Marshall, 2003; Guilhem, Giroux, Cou-
turier, Chollet, & Rabita, 2014; Gutierrez-Davila,
Rojas, Antonio, & Navarro, 2013). There is also a
study using the average horizontal velocity of the
body CG to represent the lunge speed (Turner
et al., 2016). However, the lunge distance was differ-
ent for different fencers; thus, the average horizontal

Correspondence: Darren E.R. Warburton Cardiovascular Physiology and Rehabilitation Laboratory, University of British Columbia, Rm
205, Unit II Osborne Centre, 6108 Thunderbird Blvd, Vancouver, British Columbia V6 T 1Z3, Canada. Email: darren.warburton@ubc.ca

© 2017 European College of Sport Science

202 Y. Guan et al.
velocity of the body CG may not be an appropriate
variable for comparison. Therefore, HPV was sup-
posed to be the best variable to represent the lunge
speed. Elite fencers have been shown to have a higher
HPV than intermediate-level fencers during the lunge
(Gutierrez-Davila, Rojas, Antonio et al., 2013).
Previous studies supported that HPV is closely
related to the strength and power of rear leg extensors
(Guan, Guo, Wu, Zheng, & Liu, 2015; Morris,
Farnsworth, & Robertson, 2011; Roi & Bianchedi,
2008). However, fencing lunge is a complex move-
ment, which requires highly developed motor pat-
terns to maximize the attacking speed, thus better
muscle capacities of the rear leg alone may not able
to be transmitted to higher kinetic parameters of
lower limb joints in a lunge. Therefore, examining
the relation between HPV and the kinetic parameters
of the lower limb joints during the lunge movement
would be more reasonable. Furthermore, it is also
necessary to compare the contributions of different
lower limb joints to HPV.
Despite a highly developed motor pattern, the
effects of joint kinematics on fencing lunge tended
to be overlooked in previous studies. Bottoms,
Greenhalgh, and Sinclair (2013) recorded the lunge
movements from 14 fencers using a 3D motion
capture system. The results showed that the rear
knee range of motion (ROM), rear hip peak flexion,
and forward hip peak flexion were significant predic-
tors of the sword velocity. The sword velocity was
used to represent the lunge speed instead of HPV in
this study. However, they did not define the sword
velocity clearly, and if it would be influenced by the
possible movements of the wrist and arm extension
(Mulloy, Mullineaux, & Irwin, 2016) is unknown.
Another study (Gholipour et al., 2008) compared
the moving patterns of the forward knee joint
during the fencing lunge between elite fencers and
novices by recording kinematic data of their lunge
movements using three high-resolution cameras,
and found that both the elite fencers and novices
flexed the forward knee joint before the extension
when initiating the lunge. Novices showed a greater
flexion, while a smaller extension in the subsequent
movement than elite fencers (Gholipour et al.,
2008). However, the results might be not persuasive
because there were only four subjects in each
group, and no explanation about the mechanism of
this moving pattern was provided.
Based on previous studies, HPV is closely related
to the strength and power of rear leg extensors
(Guan et al., 2015; Roi & Bianchedi, 2008; Turner
et al., 2016), while more evidence about the effects
of kinetic performance in lower-limb joints during
the lunge are needed, and more research focus on
the effects of joint kinematics are expected. Also,
whether there are any differences between elite and
intermediate-level fencers in lower limb joint kinetic
and kinematic performance during the lunge
remains unclear, and if so, whether these differences
can affect HPV of the lunge needs to be explored.
Therefore, the aim of the present study is firstly to
verify if elite fencers can acquire higher HPV during
the lunge than intermediate-level fencers; then to
explore the determinants of HPV by comparing
elite fencers with intermediate-level fencers. We
hypothesized that elite fencers can acquire higher
HPV than intermediate-level fencers, and the differ-
ence of HPV results from different kinetic and kin-
ematic performances of the lower-limb joints during
the lunge between two groups.
Methods
Participants
Nineteen male fencers (epee specialty) participated in
this study. Seven of them (height: 1.93 ± 0.03 m;
body weight: 87.0 ± 7.3 kg; training experience: 7.1
± 1.2 years) were classified as elite fencers who had
competed at world championships representing
China. The other 12 participants (height: 1.84 ±
0.06 m; body weight: 77.2 ± 11.9 kg; training experi-
ence: 6.3 ± 2.4 years) were classified as intermediate-
level fencers from a university fencing team with
documented training experience and some national
competitions. All participants were 18 years older
and free of lower extremity musculoskeletal injuries
in the last six months before the test. The protocol
of this study was submitted to, and approved by the
ethics committee of Shanghai University of Sport.
Each participant completed a written informed
consent form prior to testing.
Instruments
Kinematic data were collected using a three-dimen-
sional analysis system (VICON; Oxford Metrics
Ltd., Oxford, United Kingdom) with 12 high-resol-
ution cameras sampling at a rate of 100 Hz. Twenty
anatomical markers and 42 tracking markers (diam-
eter 5.14 mm) were placed on bilateral sides of the
feet, ankle, shank, knee, thigh, shoulder and trunk.
The tracking markers were attached to pelvis, shank
and thigh segments using four non-collinear marker
shells. For the heel, three tracking markers were
attached to each shoe: one was placed on the lateral
side of the heel; the other two were placed on the pos-
terior superior and inferior, respectively. Anatomical
markers were also placed on the left and right iliac
crest, greater trochanter, medial and lateral femoral
 Biomechanical insights into the determinants of speed in the fencing lunge
epicondyles and the malleoli, and on the head of first
and fifth metatarsal. Hook and loop tape was used to
attach the shells. Neoprene bands were wrapped
around the body segments (Manal, McClay, Stan-
hope, Richards, & Galinat, 2000). Two embedded
Kistler force platforms (Kistler 9287B, Kistler Cor-
poration, Winterthur, Switzerland; 60 cm× 90 cm)
were placed under the feet of the fencers to record
the horizontal component of the reaction force. The
force signals were amplified and recorded in the
Vicon system at a sampling rate of 1000 Hz.
Procedures
Each participant completed stretching and a five-min
warm-up on a treadmill (6.5 kph). After warm-up,
each participant was familiarized to the testing
environment and conditions. The horizontal distance
from the sword tip in en garde position to the target
was standardized by the standing height of each
fencer multiplied by 1.5 to ensure a fully extended
lunge. The fencers were allowed to adjust the distance
slightly if they have difficulty to reach this distance.
After the adjustments, the intermediate-level fencers
group reduced the distance to the target by 0.83 ±
2.89 cm. The final distance was 2.90 ± 0.05 m for
the elite fencers, 2.76 ± 0.10 m for the intermediate-
level fencers. Afterward, participants performed
maximum-effort lunges with the rear foot planting
on one force plate in en garde position, and the
forward foot planting on the other force plate. Five
valid lunges were recorded for each participant.
Data reduction
Marker position and ground reaction force (GRF)
data (C3D file format) were imported to Visual 3D
(3.390.23; C-Motion, Inc., USA). Raw kinematic
and GRF data were filtered using a fourth-order But-
terworth digital low-pass filter with cutoff frequency
of 13 Hz (Winter, 2009) and 73 Hz (B. Yu, 1989),
respectively. Anatomical landmarks and segments
were defined using the Visual 3D framework model
and the anthropometric data. The whole-body CG
was calculated using a 14-segment model in Visual
3D (Hay, 1993). Specifically, the anthropometric
Figure 1. A temporal scheme of the phases in fencing lunge. The
arrow represents GRF after being normalized to body weight.
203
inertial parameters for Chinese adults (Zheng, 2007)
were used to determine the location of the CG and
the moment of inertia for each body segment. The
three-dimensional kinematics was computed using a
Cardan sequence (X–Y–Z). Joint moments and
powers were calculated using iterative Newton–Euler
inverse dynamics within Visual 3D (Krupenevich,
Pruziner, & Miller, 2016). GRF data were normalized
to the body weight. Joint powers and moments were
normalized to the body mass of each participant.
HPV was used to represent the lunge speed.
The lunge is divided into three phases. Phase 1 cor-
responds to the en garde position (Figure 1(a)) with
both feet contacting with the ground and no GRF
(after eliminating the body weight) acting. Phase 1
finishes when the rear leg starts to exert force
against the platform/forward foot takes off (Guilhem
et al., 2014), which is also the beginning of phase 2
(Figure 1(b)–(c)). During phase 2 (the acceleration
phase), the horizontal component of the GRF pro-
duced by the extension of rear leg accelerates the
CG moving forward, and the forward leg swings
forward to coordinate with the rear leg. Phase 2
ends at the instant of toe-off time in the rear foot
(Yu et al., 2016). Phase 3 (the flight phase, Figure 1
(d)–(e)) begins at the toe-off time of the rear foot,
and ends with the forward foot contacting with the
floor again (Stewart & Kopetka, 2005).
Throughout the movement, the rear leg and
forward leg play different roles. The explosive con-
traction of the rear leg extensors produces GRF
whose horizontal component impulses the CG
moving forward (Guilhem et al., 2014; Morris
et al., 2011). The forward leg coordinates with the
rear leg to improve the speed and distance of the
lunge by swinging forward (Guilhem et al., 2014).
Theoretically, HPV is obtained at the end of the
acceleration phase with the rear foot just lifting off
the floor (between Figure 1(c) and 1(d)). In the
present study, the minimum angle of the forward
knee was defined as the knee angle in the end of the
flexion during acceleration phase; time for knee
extension was defined as the duration from the end
of knee flexion to its full extension; the extent of
flexion was calculated as knee angle in en garde pos-
ition minus the minimum angle. The maximum
angle of the forward knee joint was defined as the
knee angle at the end of the extension.
Statistical analyses
The mean of each variable over five trials was calcu-
lated for each participant. Intra-class correlation
coefficients (ICC) were calculated to determine the
degree of agreement of the five trials. Normal
204 Y. Guan et al.

distribution and homoscedasticity assumptions of the
data were checked using the Kolmogorov–Smirnov
and Levene tests, respectively. Independent-sample
t-tests were used to examine differences of the lower
limb joint kinetic and kinematic parameters
between the elite and intermediate-level fencers.
Effect size (reported using Cohen’s d ) was described
as small (<0.5), moderate (0.51–0.79) and large
(>0.8) (Cohen, 1988). Cohen’s d was determined
using the means and standard deviations of two
groups (Cohen, 1988). Pearson’s product moment
correlation was used to identify relationships
between variables. Statistical significance was set at
P ≤ .05 in this study.
Results
Lunge speed
The results in Table II showed that HPV of elite
fencers was significantly higher than intermediate-
level fencers (P < .001). Concomitantly, the peak
ground reaction force (PGRF) of elite fencers was
significantly higher than intermediate-level fencers
(P < .01; Table I).
Joint kinetics and kinematics
Elite fencers showed significantly higher (P < .05) peak
moment (PM) of rear knee, peak power (PP) of rear
knee, PM of forward knee, PM of rear hip than the
intermediate-level fencers (Table I). For joint kin-
ematics, elite fencers showed significantly higher
(P < .05) rear knee ROM, rear ankle ROM, and sig-
nificant lower (P = .05) peak angular velocity of
forward knee than the intermediate-level fencers
(Table II).
According to Table II, both elite and intermediate-
level fencers flexed their forward knees before
extending. The elite fencers showed significantly
smaller (P < .05) extent of flexion than the intermedi-
ate-level fencers. The extension time of the front knee
joint was significantly longer (P < .01) for elite fencers
comparing with the intermediate-level fencers. There
is no significant difference (P > .05) of the maximum
angle or ROM in the forward knee between different
groups.
Based on the results in Table III, HPV was signifi-
cantly (P < .05) correlated with PGRF (r = 0.796),
PM of rear knee (r = 0.784), PP of rear knee (r =
0.828), rear knee ROM (r = 0.631), peak angular vel-
ocity of rear knee (r = 0.543), rear ankle ROM (r =
0.550) and forward hip ROM (r = 0.466). PGRF was
significantly (P < .05) correlated with the PM of rear
knee (r = 0.586), PP of rear knee (r = 0.713), rear
knee ROM (r = 0.536), peak angular velocity of rear
knee (r = 0.541) and forward hip ROM (r = 0.570).
Discussion
The purpose of this study is to investigate the deter-
minants of the speed in fencing lunge. We anticipated
that these findings would provide greater insight into
biomechanical determinants of HPV in fencing
lunge, and help elucidate what kind of training
fencers should focus on to improve HPV of the
lunge. Our results showed that elite fencers can
acquire higher HPV as well as PGRF than intermedi-
ate-level fencers. Significant differences were shown
between two groups in both kinetic and kinematic
parameters of lower limb joints.
To obtain a high-level HPV, fencers were antici-
pated to perform a long and strong lunge. Previous
studies used different methods to set the lunge

Table I. Joint kinetics of the elite fencers (EF) and intermediate-level fencers (IF).

Variable EF (n = 7) IF (n = 12) ICC 95% CI for differences Effect size P

PGRF (N N−1) 0.91 ± 0.10 0.78 ± 0.07 0.948 0.046 to 0.211 1.506 .004
RK PM (N m kg−1) 2.79 ± 0.24 2.28 ± 0.39 0.975 0.162 to 0.850 1.575 .006
RK PP (W kg−1) 12.29 ± 1.97 9.45 ± 2.15 0.953 0.741 to 4.937 1.377 .011
FK PM (N m kg−1) 0.96 ± 0.63 0.51 ± 0.28 0.083 0.013 to 0.889 0.923 .044
FK PP (W kg−1) 2.05 ± 2.40 1.75 ± 1.21 0.924 −1.435 to 2.027 0.158 .723
RH PM (N m kg−1) 1.31 ± 0.29 0.94 ± 0.25 0.771 0.106 to 0.643 1.367 .009
RH PP (W kg−1) 3.97 ± 1.23 3.04 ± 0.89 0.635 −0.101 to 1.945 0.866 .074
FH PM (N m kg−1) 1.30 ± 0.48 1.27 ± 0.25 0.926 −0.317 to 0.375 0.078 .861
FH PP (W kg−1) 2.87 ± 0.77 2.97 ± 1.03 0.818 −1.143 to 0.756 −0.110 .673
RA PM (N m kg−1) 0.12 ± 0.03 0.20 ± 0.17 0.892 −2.280 to 3.526 −0.655 .221
RA PP (W kg−1) 2.48 ± 3.50 1.86 ± 2.50 0.993 −0.227 to 0.056 0.656 .656
FA PM (N m kg−1) 0.13 ± 0.08 0.10 ± 0.03 0.613 −0.019 to 0.083 0.241 .201
FA PP (W kg−1) 2.48 ± 3.50 2.20 ± 2.87 0.995 −2.833 to 3.398 0.087 .850

Note: RK, rear knee; FK, forward knee; RH, rear hip; FH, forward hip; RA, rear ankle; FA, forward ankle; PM, peak moment; PP, peak
power.
 Biomechanical insights into the determinants of speed in the fencing lunge 205
Table II. Joint kinematics of the elite fencers (EF) and intermediate-level fencers (IF).

Variable EF (n = 7) IF (n = 12) ICC 95% CI for differences Effect size P

HPV (m s−1) 2.63 ± 0.15 2.24 ± 0.19 0.979 0.211 to 0.564 2.278 .000
RK ROM (°) 60.08 ± 8.89 50.39 ± 9.56 0.970 0.006 to 0.332 1.050 .043
FK ROM (°) 53.72 ± 14.85 60.22 ± 19.36 0.969 −0.427 to 0.200 −0.377 .455
RA ROM (°) 63.61 ± 6.07 51.12 ± 11.94 0.956 0.038 to 0.400 1.319 .020
FA ROM (°) 21.36 ± 7.51 17.44 ± 6.02 0.885 −0.047 to 0.184 0.576 .228
RH ROM (°) 30.01 ± 6.78 27.99 ± 6.58 0.927 −0.081 to 0.152 0.302 .532
FH ROM (°) 49.16 ± 13.49 41.01 ± 16.37 0.987 −0.128 to 0.412 0.543 .282
RK PAV (rad s−1) 8.38 ± 1.53 7.21 ± 1.38 0.961 −0.275 to 2.604 0.803 .106
FK PAV (rad s−1) 4.75 ± 1.16 7.48 ± 3.44 0.961 −5.601 to 0.124 −1.063 .050
RA PAV (rad s−1) 2.27 ± 1.11 1.60 ± 0.70 0.809 −
0.203 to 1.540 0.722 .837
FA PAV (rad s−1) 3.78 ± 2.35 2.98 ± 1.89 0.739 −1.278 to 2.870 0.375 .429
RH PAV (rad s−1) 2.01 ± 0.64 2.08 ± 0.64 0.732 −0.703 to 0.577 −0.109 .124
FH PAV (rad s−1) 2.17 ± 0.21 2.16 ± 0.19 0.786 −1.068 to 1.846 0.050 .935
FK angle en garde (°) 124.29 ± 12.07 123.84 ± 11.13 0.882 −0.193 to 0.209 0.039 .935
FK angle min (°) 115.97 ± 11.59 104.99 ± 15.35 0.954 −0.056 to 0.439 0.807 .121
FK angle max (°) 169.69 ± 5.97 165.21 ± 11.08 0.992 −0.090 to 0.246 0.503 .341
FK flexion extent (°) 8.32 ± 4.89 18.85 ± 10.09 0.905 −0.335 to −0.033 −1.328 .020
FK extension duration (s) 0.39 ± 0.09 0.27 ± 0.05 0.867 0.003 to 0.101 1.648 .005

Note: RK, rear knee; FK, forward knee; RA, rear ankle; FA, forward ankle; RH, rear hip; FH, front hip; ROM, range of motion; PAV, peak
angular velocity.

distance. Some studies allowed the fencers to stand a
self-selected distance deemed to be their optimal dis-
tance (Gholipour et al., 2008; Guilhem et al., 2014;
Hassan & Klauck, 1998; Mulloy et al., 2016; Tsolakis
& Vagenas, 2010; Tsolakis, Kostaki, & Vagenas,
2010; Turner et al., 2016). A study in 2000 set the
horizontal distance between the toe of the rear foot
in en garde position and the target using the standing
height of each participant multiplied by 1.5, and
allowed the fencers to adjust the distance slightly by
themselves (Williams & Walmsley, 2000). Gutier-
rez-Davila et al. followed this criterion in their
studies (Gutiérrez-Dávila, Rojas, Caletti, Antonio,
& Navarro, 2013; Gutiérrez-Dávila, Zingsem,
Gutiérrez-Cruz, Giles, & Rojas, 2014; Gutierrez-
Davila, Rojas, Antonio et al., 2013). We believe
that compared with fencers standing a deemed
optimal distance completely by themselves, using a
criterion would avoid individuals selecting a short
distance to the most extent. However, after
consulting with the participants and their coaches
who believed that a longer distance could be
obtained, we revised the criterion of Williams and
Walmsley (2000) by standardizing the horizontal dis-
tance from the sword tip in en garde position to the
target based on the standing height of each fencer
multiplied by 1.5 to ensure a fully extended lunge.
The fencers were allowed to adjust the distance
slightly if they have difficulty to reach it. As a result,
a high level of performance was revealed through
the HPV reached by both the elite fencers (2.63 ±
0.15 m s−1) and intermediate-level fencers (2.24 ±
0.19 m s−1), which exceed previous data reported in
literature. Our results thus proved the reasonability
of using the revised criterion. In addition, the cri-
terion using Williams and Walmsley (2000) was set
at the year of 2000. We believe that as the progress
of fencing training in the last decade, the fencers
who are competing currently may have developed
stronger abilities than most of those competing

Table III. Correlation coefficients between HPV, PGRF and selected variables (n = 19).

HPV (m s−1) PGRF (N N−1)

Independent variables r P r P

PGRF (N N−1) 0.796 .000

Rear knee PM (N m kg−1) 0.784 .000 0.586 .008
Rear knee PP (W kg−1) 0.828 .000 0.713 .001
Rear knee ROM (°) 0.631 .004 0.536 .018
Rear knee PAV (rad s−1) 0.543 .016 0.541 .017
Rear ankle ROM (°) 0.550 .015 0.440 .060
Front hip ROM (°) 0.466 .044 0.570 .011

Note: PM, peak moment; PP, peak power; ROM, range of motion; PAV, peak angular velocity.
206 Y. Guan et al.
before the year 2000, which is also the reason we did
not use the criterion set by Williams and Walmsley
(2000) directly.
The results of the present study showed that HPV
was significantly correlated with PGRF (r = 0.796, P
= .000), and significantly higher (P < .01) values of
both HPV and PGRF were found in the elite
fencers compared with the intermediate-level
fencers. According to Newton’s second law, the net
force of the horizontal component of GRF is the
impulse of the anteroposterior movement of the CG
during the lunge, primarily determines the peak vel-
ocity of CG during the lunge. In fact, it was found
that during the lunge, power production of the
lower limbs occurred almost exclusively from the
rear leg (Morris et al., 2011), while the forward leg
extensors mainly contributed to the final braking
phase to decelerate the body mass (Guilhem et al.,
2014). Thus, although the data of GRF in both legs
were recorded, only the horizontal components of
the ground force in the rear leg were analysed in the
present study. Previous studies used the GRF of the
forward leg in order to determine the beginning of
the movement by calculating the net force of the hori-
zontal component (Gutiérrez-Dávila et al., 2014;
Gutierrez-Davila, Rojas, Antonio et al., 2013;
Gutiérrez-Dávila, Rojas, Caletti et al., 2013).
However, as the present study does not explore the
timing of different phases of the lunge, any instant
under the en garde position can be the beginning,
and it is not necessary for this study to define the
beginning of movement by calculating the net force.
For the aim of the present study, exploring the
relation between the peak values of variables and
comparing them between groups, the record of
GRF in the forward leg does not provide much help
for our analysis. Also, for the reasons above, the
relations of the peak value of the horizontal com-
ponent of GRF in the rear leg (PGRF) and the bio-
mechanical variables of lower limb joints during the
lunge were explored as well in this study.
Most of previous studies regarding the lunge speed
have concentrated on kinetics of lower limb joints. It
is widely accepted that the strength and power of the
rear leg muscles are critical for lunge performance,
because the impulse of the anteroposterior movement
during the lunge is mainly determined by the force
producing ability of the rear leg (Chen et al., 2017;
Morris et al., 2011; Poulis, Chatzis, Christopoulou,
& Tsolakis, 2009). Several studies reported that
HPV was correlated with the muscle capacity of the
rear knee joint, and supported that the strength and
explosive power of the rear knee extensors determine
HPV (Cronin et al., 2003; Guilhem et al., 2014).
However, few studies investigated the relationship
between HPV and the kinetics of lower limb joints
during the lunge, and whether there are differences
between elite and intermediate-level fencers in
kinetic parameters of the lower limb joints during
the lunge movement remains unclear. In fact, this
relationship was only explored in one study (Morris
et al., 2011) before. The results revealed that
during the movement of the lunge, the joint
moments of the rear ankle and rear knee contributed
dramatically to the moving forward of CG; however,
the results may not be persuasive because the data
were collected from only one subject. The present
study fills up this gap. Our results showed that the
elite fencers obtained significantly higher PM of
rear knee, PP of rear knee, PM of rear hip, and PM
of forward knee than the intermediate-level fencers
during the lunge (Table I). Among these variables,
PM and PP of the rear knee were significantly corre-
lated with both HPV and PGRF (Table III). It reveals
that PM and PP of the rear knee during the lunge
movement are significant indicators for HPV of the
lunge. Based on this, we postulate that resistance
and ballistic training aimed at improving strength
and explosive power of rear knee extensors would
be effective for intermediate-level fencers to
enhance their HPV of the lunge.
The role of joint kinematics in affecting HPV
during the lunge speed was also investigated in the
present study. The elite fencers showed significantly
higher ROM of rear knee and rear ankle than inter-
mediate-level fencers (Table II). For all the partici-
pants, the ROM of rear knee was significantly
correlated with both HPV and PGRF; the ROM of
rear ankle was significantly correlated with HPV
(Table III). It indicates that fencers require a low en
garde position by bending the knees sufficiently, and
a full extension of rear leg during the lunge to
improve HPV. This agrees with the conclusion of
Bottoms’s study (Bottoms et al., 2013), and supports
the emphasis of bending legs sufficiently in en garde
position in coaching literature (Sinclair, Taylor,
Edmundson, Brooks, & Hobbs, 2012). The mechan-
ism may be that the higher ROM of the rear knee and
rear ankle allows a full action of the GRF, thus it
increases the acceleration distance of the centre of
gravity, then HPV can be enhanced. However, it is
foreseeable that an excessive low position might
cause excessive muscle stretching and an increase in
the resistance moment, which is detrimental to
improving HPV. Thus, future research could be
implemented to explore the optimal extent of the
knee bending in en garde position.
In addition, an interesting moving pattern of the
forward knee joint reported by Gholipour et al.
(2008) was also observed in the present study: both
the elite and intermediate-level fencers flexed the
forward knee joints before the extension at the
 Biomechanical insights into the determinants of speed in the fencing lunge
beginning of the lunge, and significantly greater
extent of flexion was exhibited in the intermediate-
level fencers compared with the elite fencers (Table
II). For the intermediate-level fencers, the greater
flexion of the forward knee in the initial stage of the
lunge could be viewed as a compensative response
for the insufficient of PGRF. The greater flexion
facilitated the quadriceps femoris muscle stretching
of intermediate-level fencers, which prepared for the
subsequent knee extending and lead to a faster
swing of the forward leg. The results in Table II sup-
ported this viewpoint: the intermediate-level fencers
showed significantly higher peak angular velocity
and shorter duration of the forward knee extension
than the elite fencers. However, there was no signifi-
cant difference in forward knee ROM between two
groups (Table II). It demonstrated that the intermedi-
ate-level fencers finished the forward knee extension
with a faster pace in a shorter time, which might
lead to the overuse of the forward leg and impact
the stability of the forward foot landing. Moreover,
on some occasions in a fencing match, the attacker
needs to change the attacking aim based on the reac-
tion of opponents during the acceleration phase of the
lunge (Gutiérrez-Dávila, Rojas, Caletti et al., 2013).
The pressing swing of the forward leg in the inter-
mediate-level fencers may make them difficult to
change the attacking aim during the lunge. In con-
trast, the relatively suitable swing of the forward leg
in elite fencers may be beneficial for them to adjust
the attacking aim during the lunge, which may facili-
tate the attacking effects.
This study has a few limitations. Firstly, we support
that fencers require a low en garde position by bending
knees sufficiently, but the optimal extent of knee
bending is not explored in the present study. In
addition, because we did not collect the GRF data
of the forward foot landing, it is not possible to
provide more evidence about the different effects
between a pressing and suitable swing of the
forward leg on its landing. Finally, as this study
focus on the biomechanics of the lower limbs, kin-
ematic factors including the extent of trunk incli-
nation and the moving sequence of different
segments are not explored. In future, researches are
needed in the above aspects to provide a more com-
prehensive understanding of the biomechanical
determinants of the speed in fencing lunge.
Conclusion
In summary, the elite fencers obtained a higher HPV
via acquiring higher PGRF which is determined by
both the kinetics and kinematics of lower limb
joints. The elite fencers obtained significantly
207
higher PM of rear knee, PP of rear knee, ROM of
rear knee, and ROM of rear ankle, which resulted
in their higher PGRF and HPV compared with the
intermediate-level fencers. In addition, differences
were shown between the elite and intermediate-level
fencers in the moving patterns of the forward knee
joint during the lunge, which might lead to different
attacking effects.
Based on the results in the present study, the kinetics
of the rear knee joint during the lunge are crucial for
lunge performance, thus supporting for the impor-
tance of the strength and explosive power training for
the rear knee extensors. The kinematics of the lower
limb joints during the lunge are also closely correlated
with the lunge performance. A low en garde position
and a full extension of the rear knee joint during the
lunge are suggested to improve HPV. Although inter-
mediate-level fencers are not able to acquire an
impulse as great as elite fencers to accelerate the CG
during the lunge, compensating that by increasing
the flexion of the forward knee joint before its exten-
sion is not encouraged, because the greater extent of
flexion may cause negative influences on the attacking
effect and the brake of the lunge.
Acknowledgements
The authors are grateful to all the participants for
their cooperation and support.
Disclosure statement
No potential conflict of interest was reported by the authors.
References
Aquili, A., Tancredi, V., Triossi, T., De Sanctis, D., Padua, E.,
D’Arcangelo, G., & Melchiorri, G. (2013). Performance analy-
sis in saber. Journal of Strength and Conditioning Research, 27,
624–630.
Bottoms, L., Greenhalgh, A., & Sinclair, J. (2013). Kinematic
determinants of weapon velocity during the fencing lunge in
experienced épée fencers. Acta of Bioengineering and
Biomechanics, 15, 109–113.
Chen, T. L., Wong, D. W., Wang, Y., Ren, S., Yan, F., & Zhang,
M. (2017). Biomechanics of fencing sport: A scoping review.
PloS One, 12(2), e0171578.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences
(2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.
Cronin, J., McNair, P. J., & Marshall, R. N. (2003). Lunge per-
formance and its determinants. Journal of Sports Sciences, 21,
49–57.
Gholipour, M., Tabrizi, A., & Farahmand, F. (2008). Kinematics
analysis of lunge fencing using stereophotogrametry. World
Journal of Sport Sciences, 1, 32–37.
Guan, Y., Guo, L., Wu, N., Zheng, J., & Liu, H. (2015).
Biomechanical analysis on knee joints during fencing lunge in
 208 Y. Guan et al.
athletes of different levels. China Sport Science and Technology, 4,
58–62.
Guilhem, G., Giroux, C., Couturier, A., Chollet, D., & Rabita, G.
(2014). Mechanical and muscular coordination patterns during
a high-level fencing assault. Medicine and Science in Sports and
Exercise, 46, 341–350.
Gutierrez-Davila, M., Rojas, F. J., Antonio, R., & Navarro, E.
(2013). Response timing in the lunge and target change in
elite versus medium-level fencers. European Journal of Sport
Science, 13, 364–371.
Gutiérrez-Dávila, M., Rojas, F. J., Caletti, M., Antonio, R., &
Navarro, E. (2013). Effect of target change during the simple
attack in fencing. Journal of Sports Sciences, 31, 1100–1107.
Gutiérrez-Dávila, M., Zingsem, C., Gutiérrez-Cruz, C., Giles, F.
J., & Rojas, F. J. (2014). Effect of uncertainty during the
lunge in fencing. Journal of Sports Science & Medicine, 13(1), 66.
Hassan, S. E. A., & Klauck, J. (1998). Kinematics of lower and upper
extremities motions during the fencing lunge: Results and training
implications. Paper presented at the ISBS – Conference
Proceedings Archive, 1, 170–173.
Hay, J. G. (1993). The biomechanics of sport techniques. Englewood
Cliffs, NJ: Prentice-Hall.
Krupenevich, R. L., Pruziner, A. L., & Miller, R. H. (2016). Knee
joint loading during single-Leg forward hopping. Medicine and
Science in Sports and Exercise, 49, 327–332.
Manal, K., McClay, I., Stanhope, S., Richards, J., & Galinat, B.
(2000). Comparison of surface mounted markers and attach-
ment methods in estimating tibial rotations during walking:
An in vivo study. Gait and Posture, 11, 38–45.
Morris, N., Farnsworth, M., & Robertson, D. G. E. (2011).
Kinetic analyses of two fencing attacks – lunge and fleche.
Portuguese Journal of Sport Sciences, 11, 343–346.
Mulloy, F., Mullineaux, D. R., & Irwin, G. (2016). Use of the kin-
ematic chain in the fencing attacking lunge. 33 International
Conference on Biomechanics in Sports, Poitiers, France.
Poulis, I., Chatzis, S., Christopoulou, K., & Tsolakis, C. H.
(2009). Isokinetic strength during knee flexion and extension
in elite fencers. Perceptual and Motor Skills, 108(3), 949–961.
Roi, G. S., & Bianchedi, D. (2008). The science of fencing:
Implications for performance and injury prevention. Sports
Medicine, 38, 465–481.
View publication stats
Sinclair, J., Taylor, P. J., Edmundson, C. J., Brooks, D., & Hobbs,
S. J. (2012). Influence of the helical and six available Cardan
sequences on 3D ankle joint kinematic parameters. Sports
Biomechanics, 11, 430–437.
Stewart, S. L., & Kopetka, B. (2005). The kinematic determinants
of speed in the fencing lunge. Journal of Sports Sciences, 23,
105.
Tsolakis, C., Kostaki, E., & Vagenas, G. (2010). Anthropometric,
flexibility, strength-power, and sport-specific correlates in elite
fencing. Perceptual and Motor Skills, 110(3), 1015–1028.
Tsolakis, C., & Vagenas, G. (2010). Anthropometric, physiological
and performance characteristics of elite and sub-elite fencers.
Journal of Human Kinetics, 23, 89–95.
Turner, A., Bishop, C., Chavda, S., Edwards, M., Brazier, J., &
Kilduff, L. P. (2016). Physical characteristics underpinning
lunging and change of direction speed in fencing. Journal of
Strength and Conditioning Research, 30, 2235–2241.
Turner, A., James, N., Dimitriou, L., Greenhalgh, A., Moody, J.,
Fulcher, D., … Kilduff, L. (2014). Determinants of Olympic
fencing performance and implications for strength and con-
ditioning training. Journal of Strength & Conditioning Research,
28, 3001–3011.
Turner, A., Miller, S., Stewart, P., Cree, J., Ingram, R., Dimitriou,
L., … Kilduff, L. (2013). Strength and conditioning for fencing.
Strength and Conditioning Journal, 35, 1–9.
Williams, L. R. T., & Walmsley, A. (2000). Response timing and
muscular coordination in fencing: A comparison of elite and
novice fencers. Journal of Science and Medicine in Sport, 3(4),
460–475.
Winter, D. A. (2009). Biomechanics and motor control of human
movement. New Jersey: Wiley.
Yu, B. (1989). Determination of the optimum cutoff frequency in
the digital filter data smoothing procedure. Journal of
Biomechanics, 22, 988.
Yu, J., Sun, Y., Yang, C., Wang, D., Yin, K., Herzog, W., & Liu,
Y. (2016). Biomechanical insights into differences between
the mid-acceleration and maximum velocity phases of
sprinting. Journal of Strength & Conditioning Research, 30,
1906–1916.
Zheng, S. Y. (2007). Morden sports biomechanics. Beijing: National
Defence Industry Press.