Content uploaded by Serge Berthoin
Author content
All content in this area was uploaded by Serge Berthoin on Apr 04, 2015
Content may be subject to copyright.
Physical performance and subjective ratings after a soccer-specific
exercise simulation: Comparison of natural grass versus artificial turf
MATHIEU NE
´DE
´LEC
1,2
, ALAN MCCALL
2
, CHRIS CARLING
2
, FRANCK LE GALL
2
,
SERGE BERTHOIN
1
, & GRE
´GORY DUPONT
1,2
1
University of Lille Nord de France, France, UDSL, EA 4488, 9 rue de l’Universite´, Ronchin, 59790 France, and
2
LOSC
Lille Me´tropole Football Club, LOSC Lab, Domaine de Luchin, Grand Rue, BP 79, Camphin en Pevele, 59780 France
(Accepted 5 October 2012)
Abstract
This study aimed to compare the recovery kinetics of physical performance and subjective ratings in response to a soccer-
specific exercise simulation on natural grass and artificial turf. Physical performance tests and subjective ratings were
assessed on 13 professional soccer players before, immediately after, 24 h and 48 h after the test. Physical performance tests
included squat jump, countermovement jump, 6-s sprint on a non-motorised treadmill and isokinetic eccentric hamstring
assessment (2.09 rad s
71
). Hamstring peak torque decrement was higher (P50.05) on natural grass than on artificial turf
immediately (-4.0%, CI 95%: -10.0 to 2.0%, Effect Size [ES] ¼0.29), 24 h (-3.1%, CI 95%: -9.3 to 3.1%, ES ¼0.29) and
48 h (-3.8%, CI 95%: -8.5 to 0.9%, ES ¼0.43) after the test. Squat jump performance decrement was significantly lower
(P50.05) on natural grass than artificial turf 48 h after the test (þ3.7%, CI 95%: 1.1 to 6.3%, ES ¼0.40). Sprint
performance showed no change from baseline performance for both trials throughout the protocol. No significant interaction
between surface and time was found for countermovement jump and subjective ratings. These results suggest that a one-off
exercise on artificial turf does not induce greater fatigue nor does it delay the recovery process when compared to natural
grass among regular artificial turf players.
Keywords: fatigue, recovery, football, field test, muscle soreness
Introduction
The International Football Association Board
decided to include artificial turf pitches in the Laws
of the Game in 2004. These surfaces are currently
used for competitive league games at professional
levels in several countries (e.g. France, Russia, and
Switzerland) and for training purposes in many
professional clubs. Professional players reported
subjectively a greater physical effort during matches
played on artificial turf than natural grass despite
similar activity profiles (i.e. total distance covered,
high-intensity running, number of sprints) and
technical standard (i.e. standing tackles, headers)
(Andersson, Ekblom, & Krustrup, 2008a). Sassi
et al. (2011) found a similar metabolic cost of
running for both natural grass and artificial turf
suggesting that such negative perceptions are not
because of a higher cost of running, but due to other
mechanical characteristics. In addition, Gains, Swe-
denhjelm, Mayhew, Bird, and Houser (2010)
reported that change-in-direction speed during a
one-off sprint is faster on artificial turf than on
natural grass. This time differential between surfaces
may be explained by more force being exerted during
the change-in-direction motion resulting in more
intense loading from accelerations and decelerations
on artificial turf. Changes in direction, accelerations
and decelerations are repetitively performed
throughout a soccer match and induce muscle
damage (Howatson & Milak, 2009; Magalha˜es
et al., 2010; Thompson, Nicholas, & Williams,
1999). Young, Hepner, and Robbins (2012) found
that players experiencing greater muscle damage
24 h post match covered significantly (P50.05)
greater high-intensity running, accelerations and
decelerations during the match.
The aim of the present study was to investigate the
influence of playing surface on fatigue induced by
changes in direction, accelerations and decelerations
performed throughout a soccer match. The recovery
kinetics of physical performance and subjective
Correspondence: Gre´gory Dupont, University of Lille Nord de France, UDSL, EA 4488, 9 rue de l’Universite´ , Ronchin, 59790 France.
E-mail: gregory.dupont@univ-lille2.fr
Journal of Sports Sciences, 2013
Vol. 31, No. 5, 529
© 2013 Taylor & Francis
536, http://dx.doi.org/10.1080/02640414.2012.738923989–
Downloaded by [University of Brighton] at 00:17 23 March 2013
ratings in response to a standardised soccer-specific
exercise simulation performed on natural grass and
artificial turf were compared. A standardised soccer-
specific exercise was used in order to control for the
high variability of physical performance during a
soccer match (Di Salvo, Gregson, Atkinson, Tordoff,
& Drust, 2009; Dupont et al., 2010). Based on
previous findings (Gains et al., 2010), we hypothe-
sised that post-exercise fatigue will be greater
following the test on artificial turf resulting in
delayed recovery process.
Methods
Participants
The participants were 13 professional soccer players
(age: 17.7 +0.5 years; height: 180.2 +6.0 cm; body
mass: 71.9 +6.9 kg; body fat: 9.4 +2.0%), but 12
were retained in the study, as one did not follow the
recommendations. The players participated in one
match and seven training sessions per week (volume:
11 to 14 h). They were used to training and playing
on both surfaces (natural grass and artificial turf) for
at least 2 years.
Experimental design
The study involved a randomised crossover experi-
mental design. On two separate occasions (natural
grass vs. artificial turf), players completed three
sessions separated by 2 or 3 weeks. Before the
experimentation, players completed a medical ex-
amination. All players were fully informed of the
purpose, benefits and risks involved with participa-
tion before giving their written informed consent.
This investigation was led in accordance with the
local Ethics Committee in Biomedical Research and
the recommendation of the Helsinki Declaration.
Experimental procedures
Players were accustomed to rating the global
intensity of training sessions using the modified
Borg scale from 0 to 10 points (Borg, 1982; Foster,
1998) and the feeling scale from -5 to 5 points
(Rejeski, Best, Griffith, & Kenney, 1987). They were
also familiar with rating their quality of sleep, fatigue,
muscle soreness and stress using a scale from 1 to 7
points (Hooper, Mackinnon, Howard, Gordon, &
Bachmann, 1995) and to performing the following
tests of physical performance: squat jump, counter-
movement jump, sprints on non-motorised tread-
mill, and isokinetic eccentric hamstring assessment.
Two preliminary sessions were performed in order to
verify the inter-day reliability of the physical perfor-
mance tests and to collect reference values. During
the first session in each condition, the 7-point
Hooper’s scale (Hooper et al., 1995), the total
quality recovery perceived scale from 6 to 20 points
(Kentta¨ & Hassme´n, 1998) and location of muscle
soreness (Thompson et al., 1999) were collected
before completing a 90 min soccer-specific aerobic
field test (SAFT90; Small, McNaughton, Greig, &
Lovell, 2010). After completion of the 90 min
soccer-specific aerobic field test and a 10 min rest
interval, location of muscle soreness was recorded
and physical performance tests performed in a testing
room 200 m from the pitch. The second and third
sessions corresponded to the subjective ratings,
location of muscle soreness and physical perfor-
mance tests performed 24 h and 48 h after the
90 min soccer-specific aerobic field test, respectively.
Professional groundskeepers adjusted the soccer field
watering to maintain the same experimental condi-
tions. The 90 min soccer-specific aerobic field test
on artificial turf was performed on third-generation
artificial turf. The artificial turf pitch was not
watered. The temperature ranged between [10 and
138C]. Standardised verbal encouragement was
provided during all the physical performance tests
by experimenters. In order to limit dietary influences
on test results, players were asked to follow
standardised nutritional guidelines (quantity and
content for food and drink) after each session and
for breakfast, lunch and dinner. Each meal was eaten
in the training centre. Participants were given written
instructions to have their last meal at least 3 h before
all testing sessions, and to avoid alcohol, tobacco and
caffeine during the whole experimental period.
During the period devoted to each condition, no
training session was implemented and participants
were requested not to use any different recovery
treatments (cold bath, massage, compression gar-
ments), which may have affected the recovery
pattern.
Players completed the 90 min soccer-specific
aerobic field test protocol, which consists of two
45 min periods interceded by a 15 min passive rest
period (half-time), performed as a shuttle run test
over a 20 m distance (Small et al., 2010). The
90 min soccer-specific aerobic field test is designed
to replicate the fatigue responses to soccer match-
play and includes multiple backwards running,
sidestepping, changes in direction and frequent
acceleration and deceleration actions inherent to
match-play. Thirty-six maximal shooting actions
were performed during the 90 min soccer-specific
aerobic field test protocol to increase the load to the
quadriceps reflective of match-play (Small et al.,
2010). Prior to the 90 min soccer-specific aerobic
field test, players participated in a standardised
warm-up performed on the surface on which they
had to complete the test. The warm-up was the same
M. Ne´de´lec et al.530
Downloaded by [University of Brighton] at 00:17 23 March 2013
as that used before a match and included 10 min
light jogging (9–11 km h
-1
), dynamic activities
(buttock kicks, high knee lifts, backwards running,
sidestepping), sprints and familiarisation with the
90 min soccer-specific aerobic field test exercise
protocol for a total duration of 15 min. Before the
experimentation, players were asked to choose soccer
boots that they would be required to wear in both
conditions (natural grass and artificial turf). An
experimenter checked that soccer boots worn by
players were the same during each condition (natural
grass and artificial turf). Players’ nude body mass was
recorded immediately before and after the 90 min
soccer-specific aerobic field test with a digital scale
(Seca 780, Hamburg, Germany). During half-time,
players drank a sports drink containing 6% carbohy-
drate (Gatorade, PepsiCo, United States). The
hydration plan was the same as that used during a
match with players free to choose the fluid intake to
the upper limit of 1 l. Players’ fluid intake during the
first condition was recorded and players consumed
the same fluid intake during the second condition.
The fluid loss was calculated by the following
formula: Fluid loss ¼(body mass post-test - body
mass pre-test) þfluid intake (Andersson et al.,
2008b).
The order of physical performance tests was
identical throughout each protocol and included
the following: squat jumps, countermovement
jumps, 6-s sprints and isokinetic eccentric hamstring
assessment. Players performed three squat jumps
and three countermovement jumps on a force
platform (Kistler AG, Winterhur, Switzerland) with
built-in charge amplifier. The force signal was
sampled at 1000 Hz. A 1-min rest period was set
between each jump. The best jump from three
attempts was recorded. For the squat jump, partici-
pants were instructed to bend the knees at 908,
pausing for 3 s before jumping upwards on the verbal
command ‘go’. A goniometer (Lafayette Instrument
Company, USA) was used to set the angle. For the
countermovement jump, participants were in-
structed to jump explosively upwards immediately
after descending to a self-selected depth. During
both types of jump tests, the players placed their
hands on their hips. The inter-day test-retest
reliability for squat jump and countermovement
jump was very high: the typical error (TE) was 1.4
and 1.5 cm, respectively, the intra-class correlation
coefficient (ICC) was 0.90 and 0.92, respectively,
while coefficient of variation (CV) was 3.1% and
2.9%, respectively. Players completed three 6-s
sprints separated by 3 min of passive recovery on a
non-motorised treadmill (Woodway Force 3.0,
Waukesha, USA). Start position (standing start
with hand on the handles) was standardised. The
best value from three sprints was recorded for mean
power output, mean speed and peak speed. Very
high inter-day test-retest reliability was found for
mean power output, mean speed and peak speed
(TE: 90 W, 0.2 m s
-1
and 0.2 m s
-1
, respectively;
ICC: 0.87, 0.89 and 0.88, respectively; CV: 3.1%,
2.6% and 2.2%, respectively). The non-motorised
treadmill was calibrated before each test. Treadmill
belt speed, distance and horizontal forces were
collected at a sampling rate of 100 Hz via the
XPV7 PCB interface (Fitness Technology, Adelaide,
Australia) and analysed with the Force 3.0 Soft-
ware (Innervations Software, Joondalup, Australia).
Players performed three successive maximal volun-
tary isokinetic eccentric hamstring actions without
rest on a dynamometer (Con-Trex, Duebendorf,
Switzerland). During testing, players were seated
on the dynamometer in an adjustable chair, with
test positions recorded and repeated for each
player in subsequent sessions. Actions were per-
formed on the players’ dominant leg (their ‘kick-
ing’ leg) through a range of 908(with 08being full
knee extension) at an isokinetic angular velocity of
2.09 rad s
71
(1208s
71
). Hamstring peak torque
was recorded. Peak torque showed very high inter-
day test-retest reliability (TE: 7.6 N m; ICC:
0.87; CV: 4.7%).
Heart rate was continuously monitored through-
out the 90 min soccer-specific aerobic field test
(Polar Team System, Kempele, Finland) with heart
rate values averaged every 5 s.
At the beginning of each session, players were
required to rate their quality of sleep, fatigue, muscle
soreness and stress on the 7-point Hooper’s scale
(Hooper et al., 1995). They used highlighter to
specify where they experienced muscle soreness
(Thompson et al., 1999). Players were also asked
to rate their recovery as an overall psycho-physiolo-
gical rating for the previous 24 hours, including the
previous night’s sleep, using the total quality
recovery perceived scale (Kentta¨ & Hassme´n,
1998). After the 90 min soccer-specific aerobic field
test, participants were required to rate the global
intensity of the session using the modified Borg scale
(Borg, 1982; Foster, 1998) and the feeling scale
(Rejeski et al., 1987). Ratings of fatigue, muscle
soreness and stress levels as well as location of
muscle soreness were also collected immediately
after the 90 min soccer-specific aerobic field test.
Baseline values corresponded to values obtained the
morning before the 90 min soccer-specific aerobic
field test.
Statistical analysis
Simple descriptive statistics are reported as means +
standard deviations (mean +s). The normality
distribution of the data was checked with the
Recovery on natural grass and artificial turf 531
Downloaded by [University of Brighton] at 00:17 23 March 2013
Shapiro-Wilk test. Comparison between conditions
(natural grass vs. artificial turf) was analysed using 2-
way analysis of variance (ANOVA) for repeated
measures. The effects of the independent variables
(surface and time) on the dependent variables –
squat jump, countermovement jump, mean power
output, mean speed, peak speed, hamstring peak
torque and subjective ratings – were analysed using a
2-way ANOVA for repeated measures. Bonferroni
post hoc was then applied when the significant F-
value was found. Changes in the mean between
reference and post-90 min soccer-specific aerobic
field test testing values of the two conditions were
expressed as a percentage of the reference values for
objective tests and absolute values for subjective
ratings. Comparisons between surfaces were assessed
through the difference in change scores. Effect size
data (ES) was calculated to determine the magnitude
of the change score and was assessed using the
following criteria: 50.2 ¼trivial, 0.2–0.6 ¼small,
0.6–1.2 ¼moderate, 1.2–2.0 ¼large, and 42.0 ¼
very large (Hopkins, 2002). Concerning the diagram
labelling of the body’s musculature, differences in
frequencies in muscle areas highlighted as sore
between the two conditions were tested using the
following criteria: 510% ¼trivial, 10–30% ¼small,
30–50% ¼moderate, 50–70% ¼large, 470% ¼very
large (Hopkins, 2002). Differences in heart rate,
fluid loss, body mass and rating of the 90 min
soccer-specific aerobic field test were tested for
significance using the Student’s paired t-test when
parametric methods were used or the paired Wilcox-
on test when non-parametric methods were used.
Confidence intervals (CI 95%) were used to specify
estimation of changes in performance tests, subjec-
tive ratings and differences in frequencies. Statistical
significance was set at P50.05.
Results
90 min soccer-specific aerobic field test
No significant differences were observed between the
mean heart rate during the 90 min soccer-specific
aerobic field test on artificial turf (151 +15 bpm)
and the mean heart rate during the test on natural
grass (145 +14 bpm). Similarly, no significant
difference was observed between the fluid loss during
the 90 min soccer-specific aerobic field test on
artificial turf (1321 +855 ml) and the fluid loss
during the test on natural grass (1554 +480 ml).
The body mass measured after the 90 min soccer-
specific aerobic field test on both surfaces was
significantly lower (P50.05) than those recorded
before the test, with a loss of body mass of
70.7 +0.8 kg (-0.9 +1.0%) on artificial turf and
a loss of body mass of -0.9 +0.5 kg (-1.3 +0.6%)
on natural grass. The fluid intake in both conditions
was 638 +158 ml. No significant differences were
found for the rating of intensity after the 90 min
soccer-specific aerobic field test performed on
artificial turf and natural grass (4.3 +1.5 vs.
4.8 +2.2 respectively) or for the feeling scale
(1.0 +2.4 vs. 1.4 +1.8 respectively).
Recovery kinetics for physical performance and subjective
ratings after the 90 min soccer-specific aerobic field test
The effect of surface on physical performance and
subjective ratings and comparisons between sur-
faces throughout the recovery period are presented
in Tables I and II. A significant interaction was
found for squat jump between surface and time
(P50.01). Post hoc analysis revealed that squat
jump performance decrement was significantly
lower (P50.05) on natural grass than artificial
turf 48 h after the test with a small difference
(þ3.7%, CI 95%: 1.1 to 6.3%, ES ¼0.40) ob-
served. A significant main effect for time was also
found for the squat jump (P50.001). Post hoc
analysis revealed that squat jump performance was
significantly impaired immediately after the test
(P50.001). No significant interaction was found
for countermovement jump between surface and
time with only trivial differences (ES ¼0.04–0.12)
between artificial turf and natural grass in changes
in countermovement jump performance throughout
the recovery period. However, a significant main
effect for time was found for countermovement
jump (P50.01). Post hoc analysis revealed that
countermovement jump performance was signifi-
cantly impaired immediately after the test
(P50.01) and at 24 h (P50.05).
There was no interaction effect of surface and time
on hamstring peak torque. However, there was a
main effect of surface on hamstring peak torque
(P50.05). Hamstring peak torque decrement was
higher on natural grass than on artificial turf with
small differences immediately (-4.0%, CI 95%: -10.0
to 2.0%, ES ¼0.29), 24 h (-3.1%, CI 95%: -9.3 to
3.1%, ES ¼0.29) and 48 h (-3.8%, CI 95%: -8.5 to
0.9%, ES ¼0.43) after the 90 min soccer-specific
aerobic field test. There was also a main effect of
time on hamstring peak torque (P50.05). Post hoc
analysis revealed that hamstring peak torque was
significantly different from baseline immediately
after the test and at 24 h (P50.05).
All three variables reflective of sprint performance
(i.e. mean power output, mean speed and peak
speed) showed no change from baseline performance
for both trials throughout the protocol. There were
only trivial differences (ES ¼0.01–0.17) between
artificial turf and natural grass on changes in mean
power output, mean speed and peak speed.
532 M. Ne´de´lec et al.
Downloaded by [University of Brighton] at 00:17 23 March 2013
Table I. The effect of playing surface on physical performance and subjective ratings throughout the recovery period (mean +s) following a soccer-specific exercise simulation, with the change in the
mean expressed as relative values (%) for objective tests and absolute values (av) for subjective ratings.
Artificial Turf Natural Grass
Baseline 0 h
Change
24 h
Change
48 h
Change
Baseline 0 h
Change
24 h
Change
48 h
Change
(% or av) (% or av) (% or av) (% or av) (% or av) (% or av)
SJ (cm) 39.5 +3.9 37.3 +2.9** 75.4 +3.9% 38.3 +3.2 72.8 +5.1% 37.6 +3.6 74.6 +4.8% 39.5 +3.9 36.1 +3.7** 78.4 +4.5% 38.0 +3.2 73.5 +4.8% 39.0 +3.6 71.0 +5.7%
CMJ (cm) 42.3 +5.4 39.9 +4.2** 75.2 +6.2% 40.6 +4.2* 73.7 +5.2% 41.1 +4.0 72.4 +4.4% 42.3 +5.4 40.1 +4.1** 74.7 +6.3% 40.2 +3.3* 74.4 +6.5% 41.3 +3.9 71.7 +7.6%
H PT(Nm) 148 +17 138 +22* 74.9 +11.1% 138 +14* 74.9 +9.1% 139 +13 73.9 +8.0% 148 +17 133 +18* 78.9 +6.7% 133 +15* 78.0 +7.0% 134 +13 77.7 +7.8%
MPO (W) 2390 +241 2431 +269 1.8 +5.5% 2440 +308 2.0 +6.5% 2471 +260 3.7 +8.6% 2390 +241 2421 +319 1.2 +6.8% 2492 +298 4.3 +7.9% 2465 +289 3.3 +7.9%
MS (m s
71
) 4.75 +0.43 4.84 +0.43 1.8 +3.4% 4.83 +0.48 1.7 +4.2% 4.85 +0.43 2.3 +6.7% 4.75 +0.43 4.77 +0.49 0.2 +4.0% 4.89 +0.47 3.1 +6.0% 4.83 +0.47 1.8 +5.9%
PS (m s
71
) 5.63 +0.40 5.52 +0.43 71.9 +3.7% 5.60 +0.39 70.5 +3.4% 5.58 +0.37 70.7 +4.5% 5.63 +0.40 5.53 +0.48 71.8 +3.1% 5.59 +0.42 70.6 +4.4% 5.60 +0.44 70.4 +4.7%
Sleep (au) 2.8 +0.9 773.2 +0.9 0.3 +1.2av 2.6 +0.9 70.3 +1.1av 2.6 +0.9 772.8 +1.1 0.2 +0.6av 3.0 +1.0 0.4 +1.4av
Fatigue(au) 3.5 +1.0 4.4 +1.0** 0.9 +0.9av 4.1 +0.7 0.6 +1.1av 3.5 +0.8 0.0 +1.3av 3.7 +1.2 4.8 +0.8** 1.2 +1.2av 4.2 +0.8 0.5 +1.0av 3.8 +1.1 0.1 +0.9av
Stress (au) 2.5 +1.1 2.6 +1.1 0.1 +0.9av 2.5 +1.1 0.0 +0.4av 2.3 +1.1 70.2 +0.6av 2.2 +1.0 2.4 +1.1 0.3 +0.5av 2.3 +1.1 0.1 +0.5av 2.4 +1.3 0.3 +0.8av
Soreness (au) 3.6 +1.1 4.3 +1.5** 0.7 +1.1av 4.3 +1.1** 0.8 +1.2av 4.0 +0.9 0.4 +1.4av 3.3 +1.3 4.3 +0.9** 0.9 +0.8av 4.2 +1.4** 0.8 +1.0av 3.9 +1.1 0.6 +1.3av
TQR (au) 14.3 +2.0 7713.3 +1.7 71.0 +2.5av 15.1 +1.2 0.8 +2.2av 14.4 +2.0 7713.8 +2.1 70.7 +2.0av 13.8 +1.2 70.7 +2.4av
* Significantly different from reference values (P50.05); ** Significantly different from reference values (P50.01).
Note: au: arbitrary units; CMJ: countermovement jump; H PT: hamstring peak torque; MPO: mean power output; MS: mean speed; PS: peak speed; SJ: squat jump; TQR: total quality recovery.
Table II. Comparisons between playing surfaces for physical performance and subjective ratings after a soccer-specific exercise simulation with the change expressed as relative values (%) for objective
tests and absolute values (av) for subjective ratings ( +95% confidence intervals) and the magnitude of the change (effect size).
Natural Grass-Artificial Turf Baseline Natural Grass-Artificial Turf 0 h Natural Grass-Artificial Turf 24 h Natural Grass-Artificial Turf 48 h
Change
(% or av)
Effect
size Descriptor
Change
(% or av)
Effect
size Descriptor
Change
(% or av)
Effect
size Descriptor
Change
(% or av)
Effect
size Descriptor
SJ – 73.1 +2.7% 70.34 Small 70.7 +2.4% 70.09 Trivial 3.7 +2.6% # 0.40 Small
CMJ – 0.5 +3.8% 0.04 Trivial 70.8 +3.2% 70.12 Trivial 0.6 +3.5% 0.04 Trivial
HPT – 74.0 +6.0% 70.29 Small 73.1 +6.2% 70.29 Small 73.8 +4.7% 70.43 Small
MPO – 70.6 +4.1% 70.03 Trivial 2.3 +4.6% 0.17 Trivial 70.4 +5.1% 70.02 Trivial
MS – 71.6 +2.6% 70.15 Trivial 1.4 +3.5% 0.13 Trivial 70.5 +3.7% 70.04 Trivial
PS – 0.0 +3.1% 0.01 Trivial 70.2 +2.6% 70.02 Trivial 0.3 +2.6% 0.05 Trivial
Sleep 70.3 +0.7av 70.27 Small – – – 70.2 +0.9av 70.40 Small 0.7 +1.3av 0.43 Small
Fatigue 0.2 +0.8av 0.15 Trivial 0.3 +0.9av 0.45 Small 70.1 +1.0av 0.11 Trivial 0.1 +1.0av 0.27 Small
Stress 70.3 +0.4av 70.31 Small 0.2 +0.8av 70.15 Trivial 0.1 +0.4av 70.23 Small 0.4 +0.7av 0.07 Trivial
Soreness 70.3 +1.3av 70.21 Small 0.3 +0.8av 0.00 Trivial 0.1 +0.9av 70.13 Trivial 0.2 +1.5av 70.09 Trivial
TQR 0.2 +1.4av 0.08 Trivial – – – 0.3 +1.9av 0.26 Small 71.5 +2.0av 71.12 Moderate
# Significant difference between groups (P50.05).
Note: Magnitudes of effect sizes are assessed using the following criteria: 50.2 ¼trivial, 0.2–0.6 ¼small, 0.6–1.2 ¼moderate, 1.2–2.0 ¼large, and 42.0 ¼very large. CMJ: countermovement
jump; H PT: hamstring peak torque; MPO: mean power output; MS: mean speed; PS: peak speed; SJ: squat jump; TQR: total quality recovery.
533
Recovery on natural grass and artificial turf
Downloaded by [University of Brighton] at 00:17 23 March 2013
There was no interaction effect of surface and time
on ratings of quality of sleep, fatigue, muscle soreness,
stress and total quality recovery with only trivial to
small differences (ES ¼0.00–0.45) between artificial
turf and natural grass on changes in sleep, fatigue,
stress and muscle soreness ratings throughout the
recovery period. However, for the variable fatigue,
there was a main effect of time (P50.001) with an
increase to ‘average-high’ (1 unit) for both trials
observed immediately after the test (P50.001). For
the variable muscle soreness, a main effect of time was
also observed (P50.01) with significant increases
observed immediately after the test and at 24 h
compared with baseline values (P50.01).
Differences in frequencies in muscle areas high-
lighted as sore between the two conditions at
different time points throughout the recovery period
are shown in Table III. There were trivial or small
differences for pubis, groin, tibialis and lower back.
However, soreness in quadriceps immediately after
the 90 min soccer-specific aerobic field test, in
gluteus 24 h after the test and in hamstring 48 h
after the test were all reported to be moderately lower
(from 31 to 46%) on natural grass than artificial turf.
Discussion
The aim of the present study was to compare the
recovery kinetics of physical performance and sub-
jective ratings in response to a soccer-specific
exercise test performed on natural grass and artificial
turf. The 90 min soccer-specific aerobic field test is
validated to replicate the movement demands of
soccer match-play and includes multiple changes in
direction, accelerations and decelerations associated
with muscle damage (Howatson & Milak, 2009;
Magalha˜es et al., 2010; Thompson et al., 1999).
Based on previous findings, we hypothesised that the
soccer test-induced muscle damage may be greater
on artificial turf resulting in delayed recovery
process. Warren, Lowe, and Armstrong (1999)
stated that measurement of maximal voluntary
contraction torque provides the best method for
quantifying muscle damage as it is accurate and
reliable. In the present study, eccentric hamstring
torque was tested because the hamstring is particu-
larly prone to injury (Woods et al., 2004) and fatigue
in soccer (Greig, 2008; Small et al., 2010). Results
show that our hypothesis was rejected since ham-
string peak torque decrement was higher on natural
grass than on artificial turf (P50.05) with small
differences reported through the 48 h recovery
period. Yet despite the higher peak torque decrement
on natural grass, players reported moderately higher
soreness in the hamstrings in the artificial turf
condition 48 h after the 90 min soccer-specific
aerobic field test confirming that soreness is poorly
correlated with changes in muscle function (Warren
et al., 1999). Here, 6-s sprint performance (i.e. mean
power output, mean speed) was not affected
throughout the recovery period. This result may be
explained by the activity profile of the 90 min soccer-
Table III. Frequencies difference (+95% confidence intervals) in muscle areas highlighted as sore between the two conditions throughout
the recovery period.
Baseline 0 h þ24 h þ48 h
Freq.
diff. (%) Descriptor
Freq.
diff. (%) Descriptor
Freq.
diff. (%) Descriptor
Freq.
diff. (%) Descriptor
Anterior view
Pubis -8 +24 Trivial -15 +27 Small -8 +14 Trivial -8 +14 Trivial
Left groin 23 +29 Small -15 +35 Small 23 +35 Small -8 +30 Trivial
Right groin 15 +27 Small -15 +32 Small 23 +35 Small -15 +32 Small
Left quadriceps 0 +20 Trivial 46 +31 Moderate 23 +33 Small 15 +27 Small
Right quadriceps 0 +20 Trivial 46 +31 Moderate 15 +35 Small 8 +30 Trivial
Left tibialis 0 +0 Trivial 0 +0 Trivial 0 +0 Trivial 0 +0 Trivial
Right tibialis 15 +20 Small 8 +14 Trivial 8 +14 Trivial 0 +0 Trivial
Posterior view
Lower back -15 +35 Small -23 +35 Small -15 +37 Small -15 +37 Small
Left gluteus 15 +20 Small 23 +29 Small 38 +26 Moderate -15 +27 Small
Right gluteus 8 +14 Trivial 23 +29 Small 31 +25 Moderate -15 +32 Small
Left hamstring -8 +36 Trivial -15 +37 Small 15 +37 Small 38 +35 Moderate
Right hamstring 0 +37 Trivial -23 +35 Small 23 +29 Small 46 +27 Moderate
Left calf 0 +20 Trivial 0 +32 Trivial -15 +35 Small 23 +29 Small
Right calf 8 +24 Trivial -8 +34 Trivial -8 +36 Trivial 8 +30 Trivial
Note: Magnitudes of effect sizes are assessed using the following criteria: 510% ¼trivial, 10–30% ¼small, 30–50% ¼moderate, 50–
70% ¼large, 470% ¼very large. For a given area, a positive value in a frequencies difference indicates that more players experienced
soreness in the artificial turf condition than the natural grass condition, while a negative value in a frequencies difference indicates that more
players experienced soreness in the natural grass condition than the artificial turf condition.
534 M. Ne´de´lec et al.
Downloaded by [University of Brighton] at 00:17 23 March 2013
specific aerobic field test which does not include
contact situations such as tackles or collisions
between players observed during actual soccer
match-play. In a comparison of the effect of a
simulated team sport activity circuit either with or
without 44 body contacts on sprint performance,
Singh, Guelfi, Landers, Dawson, and Bishop (2011)
found that performance was significantly slower 48 h
following the protocol with body contact (P50.05).
In contrast, performance was maintained 48 h after
the protocol without body contact. Similarly, Poin-
ton and Duffield (2012) found that an intermittent-
sprint protocol with tackling resulted in a signifi-
cantly slower mean sprint time compared to the same
protocol without tackling (P50.05). This study
proposed that the inclusion of tackling resulted in
greater central fatigue compared to the control
condition, as observed by a greater reduction in
voluntary activation. The absence of any 6-s sprint
performance impairment in the present study may
therefore be attributed to insufficient levels of muscle
damage resulting from the lack of contact actions,
jumps and tackles. As a consequence, future studies
investigating the recovery process after a soccer
match simulation test should consider the inclusion
of simulated contact, jumps and tackles, in the
exercise protocol. Future studies may also compare
the impact of a soccer match and the 90 min soccer-
specific aerobic field test on muscle damage markers.
In the present study, no significant differences
were observed between the mean heart rate during
90 min soccer-specific aerobic field test on artificial
turf and natural grass which suggests a similar
physiological load on both surfaces. The rating of
the global intensity of the 90 min test performed on
natural grass and artificial turf showed no significant
difference as did the feeling during the test which was
‘neutral-slightly good’ on both surfaces. Andersson
et al. (2008a) examined the movement patterns, ball
skills, and the impressions of elite football players
during competitive games on artificial turf and
natural grass. On a 10-point scale, where 0 ¼‘better
than’ and 10 ¼‘worse than’, players reported a
negative overall impression (8.3 +0.2), poorer ball
control (7.3 +0.3), and greater subjective physical
effort (7.2 +0.2) on artificial turf than natural grass
despite similar total distance covered, high-intensity
running and number of sprints. The discrepancy
between our results and those from Andersson et al.
(2008a) could be due to the protocol used and/or the
familiarisation with artificial turf. In the present
study, players completed a standardised soccer test
which did not include any changes in playing
characteristics during matches (i.e. fewer sliding
tackles and more short passes on artificial turf)
reported by Andersson et al. (2008a). The absence of
a negative impression of artificial turf in the present
study may also be explained by the fact that we tested
young players (17.7 years) who were accustomed to
playing on artificial turf whereas Andersson et al.
(2008a) tested predominantly regular natural grass
players aged 28.8 years. Familiarisation is a key point
in studying the recovery process. Lavender and
Nosaka (2008) have shown that a light eccentric
exercise, which does not induce changes in any of the
indirect markers of muscle damage, confers protec-
tion against muscle damage after a more strenuous
eccentric exercise performed two days later. In the
present study, the absence of negative perceptions
may likely be explained by the familiarisation with
artificial turf, but also the timing of the test (almost
the end of the season). The familiarisation with
artificial turf may consequently be important when
measuring players’ impression of artificial turf versus
natural grass.
Conclusion
Findings from the present study indicate that although
within-condition differences can be observed in
physical performance and subjective ratings after a
soccer test designed to replicate the physiological and
mechanical demands of soccer match-play, there is no
evidence to indicate that exercise on artificial turf
results in greater fatigue and delayed recovery process.
Future studies are required to confirm that results are
similar when exercise is performed on a surface which
players are not accustomed to since non-regular
artificial turf players anecdotally report that the acute
transition from natural grass to artificial turf is
particularly disturbing.
References
Andersson, H., Ekblom, B., & Krustrup, P. (2008a). Elite football
on artificial turf versus natural grass: Movement patterns,
technical standards, and player impressions. Journal of Sports
Sciences,26, 113–122.
Andersson, H., Raastad, T., Nilsson, J., Paulsen, G., Garthe, I., &
Kadi, F. (2008b). Neuromuscular fatigue and recovery in elite
female soccer: Effects of active recovery. Medicine and Science in
Sports and Exercise,40, 372–380.
Borg, G.A. (1982). Psychophysical bases of perceived exertion.
Medicine and Science in Sports and Exercise,14, 377–381.
Di Salvo, V., Gregson, W., Atkinson, G., Tordoff, P., & Drust, B.
(2009). Analysis of high intensity activity in Premier League
soccer. International Journal of Sports Medicine,30, 205–212.
Dupont, G., Nedelec, M., McCall, A., McCormack, D., Berthoin,
S., & Wisløff, U. (2010). Effect of 2 soccer matches in a week
on physical performance and injury rate. The American Journal
of Sports Medicine,38, 1752–1758.
Foster, C. (1998). Monitoring training in athletes with reference to
overtraining syndrome. Medicine and Science in Sports and
Exercise,30, 1164–1168.
Gains, G.L., Swedenhjelm, A.N., Mayhew, J.L., Bird, H.M., &
Houser, J.J. (2010). Comparison of speed and agility perfor-
mance of college football players on field turf and natural grass.
Journal of Strength and Conditioning Research,24, 2613–2617.
535
Recovery on natural grass and artificial turf
Downloaded by [University of Brighton] at 00:17 23 March 2013
Greig, M. (2008). The influence of soccer-specific fatigue on peak
isokinetic torque production of the knee flexors and extensors.
The American Journal of Sports Medicine,36, 1403–1409.
Hooper, S.L., Mackinnon, L.T., Howard, A., Gordon, R.D., &
Bachmann, A.W. (1995). Markers for monitoring overtraining
and recovery. Medicine and Science in Sports and Exercise,27,
106–112.
Hopkins, W.G. (2002). A scale of magnitudes for effect statistics.
Sportscience. Retrieved from http://sportsci.org/resource/stats/
index.html
Howatson, G., & Milak, A. (2009). Exercise-induced muscle
damage following a bout of sport specific repeated sprints.
Journal of Strength and Conditioning Research,23, 2419–2424.
Kentta¨, G., & Hassme´ n, P. (1998). Overtraining and recovery. A
conceptual model. Sports Medicine, 26, 1–16.
Lavender, A.P., & Nosaka, K. (2008). A light load eccentric
exercise confers protection against a subsequent bout of more
demanding eccentric exercise. Journal of Science and Medicine in
Sport,11, 291–298.
Magalha˜es, J., Rebelo, A., Oliveira, E., Silva, J.R., Marques, F., &
Ascensa˜o, A. (2010). Impact of Loughborough Intermittent
Shuttle Test versus soccer match on physiological, biochemical
and neuromuscular parameters. European Journal of Applied
Physiology,108, 39–48.
Pointon, M., & Duffield, R. (2012). Cold water immersion
recovery after simulated collision sport exercise. Medicine and
Science in Sports and Exercise,44, 206–216.
Rejeski, W.J., Best, L.D., Griffith, P., & Kenney, E. (1987). Sex-
role orientation and the responses of men to exercise stress.
Research Quarterly for Exercise and Sport,58, 260–264.
Sassi, A., Stefanescu, A., Menaspa’, P., Bosio, A., Riggio, M., &
Rampinini, E. (2011). The cost of running on natural grass and
artificial turf surfaces. Journal of Strength and Conditioning
Research,25, 606–611.
Singh, T.K., Guelfi, K.J., Landers, G., Dawson, B., & Bishop, D.
(2011). A comparison of muscle damage, soreness and
performance following a simulated contact and non-contact
team sport activity circuit. Journal of Science and Medicine in
Sport,14, 441–446.
Small, K., McNaughton, L., Greig, M., & Lovell, R. (2010). The
effects of multidirectional soccer-specific fatigue on markers of
hamstring injury risk. Journal of Science and Medicine in Sport,
13, 120–125.
Thompson, D., Nicholas, C.W., & Williams, C. (1999). Muscular
soreness following prolonged intermittent high-intensity shuttle
running. Journal of Sports Sciences,17, 387–395.
Warren, G.L., Lowe, D.A., & Armstrong, R.B. (1999). Measure-
ment tools used in the study of eccentric contraction-induced
injury. Sports Medicine,27, 43–59.
Woods, C., Hawkins, R.D., Maltby, S., Hulse, M., Thomas, A., &
Hodson, A. (2004). The Football Association Medical Re-
search Programme: An audit of injuries in professional football–
analysis of hamstring injuries. British Journal of Sports Medicine,
38, 36–41.
Young, W.B., Hepner, J., & Robbins, D.W. (2012). Movement
demands in Australian Rules football as indicators of muscle
damage. Journal of Strength and Conditioning Research,26, 492–
496.
M. Ne´de´lec et al.
536
Downloaded by [University of Brighton] at 00:17 23 March 2013
