Content uploaded by Ric Lovell
Author content
All content in this area was uploaded by Ric Lovell on Feb 07, 2018
Content may be subject to copyright.
Please
cite
this
article
in
press
as:
Barrett
S,
et
al.
The
within-match
patterns
of
locomotor
efficiency
during
professional
soccer
match
play:
Implications
for
injury
risk?
J
Sci
Med
Sport
(2016),
http://dx.doi.org/10.1016/j.jsams.2015.12.514
ARTICLE IN PRESS
G Model
JSAMS-1266;
No.
of
Pages
6
Journal
of
Science
and
Medicine
in
Sport
xxx
(2016)
xxx–xxx
Contents
lists
available
at
ScienceDirect
Journal
of
Science
and
Medicine
in
Sport
journal
h
om
epage:
www.elsevier.com/locate/jsams
Original
investigation
The
within-match
patterns
of
locomotor
efficiency
during
professional
soccer
match
play:
Implications
for
injury
risk?
Steve
Barretta,b,∗,
Adrian
Midgleyc,
Matt
Reevesd,
Tom
Joeld,
Ed
Frankline,
Rob
Heyworthf,
Andrew
Garretta,
Ric
Lovellg
aDepartment
of
Sport,
Health
and
Exercise
Science,
The
University
of
Hull,
UK
bSport
Medicine
and
Science
Department,
Hull
City
Tigers
FC,
UK
cDepartment
of
Sport
and
Physical
Activity,
Edge
Hill
University,
UK
dMedicine
and
Sport
Science
Department,
Leicester
City
FC,
UK
eMedicine
and
Sport
Science
Department,
Reading
FC,
UK
fMedicine
and
Sport
Science
Department,
Blackburn
Rovers
FC,
UK
gWestern
Sydney
High
Performance
Sports
Group,
University
of
Western
Sydney,
Australia
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
29
June
2015
Received
in
revised
form
9
December
2015
Accepted
19
December
2015
Available
online
xxx
Keywords:
Accelerometry
Football
Injury
risk
Fatigue
a
b
s
t
r
a
c
t
Objectives:
The
principle
aim
of
the
current
study
was
to
examine
within-match
patterns
of
locomotor
effi-
ciency
in
professional
soccer,
determined
as
the
ratio
between
tri-axial
accelerometer
data
(PlayerLoadTM)
and
locomotor
activities.
Between
match
variability
and
determinants
of
PlayerLoadTM during
match
play
were
also
assessed.
Design:
A
single
cohort,
observational
study.
Methods:
Tri-axial
accelerometer
data
(PlayerLoadTM)
was
recorded
during
86
competitive
soccer
matches
in
63
English
championship
players
(574
match
observations).
Accelerometer
data
accumu-
lated
(PlayerLoad
Vector
Magnitude
[PLVM])
from
the
individual-component
planes
of
PlayerLoadTM
(anterior–posterior
PlayerLoadTM [PLAP],
medial–lateral
PlayerLoadTM [PLML]
and
vertical
PlayerLoadTM
[PLV]),
together
with
locomotor
activity
(Total
Distance
Covered
[TDC])
were
determined
in
15-min
seg-
ments.
Locomotor
efficiency
was
calculated
using
the
ratio
of
PLVM and
TDC
(PlayerLoadTM per
metre).
The
proportion
of
variance
explaining
the
within-match
trends
in
PLVM,
PLAP,
APML,
APv,
and
TDC
was
determined
owing
to
matches,
individual
players,
and
positional
role.
Results:
PLVM,
PLAP,
APML,
APvand
TDC
reduced
after
the
initial
15-min
match
period
(p
=
0.001;
2=
0.22–0.43,
large
effects).
PL:TDC
increased
in
the
last
15
min
of
each
half
(p
=
0.001;
2=
0.25,
large
effect).
The
variance
in
PLVM during
soccer
match-play
was
explained
by
individual
players
(63.9%;
p
=
0.001)
and
between-match
variation
(21.6%;
p
=
0.001),
but
not
positional
role
(14.1%;
p
=
0.364).
Conclusions:
Locomotor
efficiency
is
lower
during
the
latter
stages
of
each
half
of
competitive
soccer
match-play,
a
trend
synonymous
with
observations
of
increased
injury
incidence
and
fatigue
in
these
periods.
Locomotor
efficiency
may
be
a
valuable
metric
to
identify
fatigue
and
heightened
injury
risk
during
soccer
training
and
match-play.
©
2015
Sports
Medicine
Australia.
Published
by
Elsevier
Ltd.
All
rights
reserved.
1.
Introduction
Monitoring
load
in
team-sports
players
during
training
and
match
play
is
common
practice
within
industry
settings
in
order
to
reduce
injury
risk
and
optimise
their
readiness
to
perform.1,2
Obtaining
measures
of
internal
load
(e.g.
heart
rate)
during
compe-
tition
can
be
impractical
and
often
prohibited;
hence
practitioners
∗Corresponding
author.
E-mail
address:
s.barrett@2006.hull.ac.uk
(S.
Barrett).
tend
to
rely
on
external
load
measures,
such
as
locomotor
activities,
to
monitor
training
and
competition
loads.
Analyzing
locomotor
activities
during
match
play
has
demonstrated
within-match
pat-
terns,
with
total
distances
covered
(TDC)
and
high-speed
running
distances
(HSR)
decreasing
towards
the
latter
stages
of
each
half.3
These
time
periods
towards
the
end
of
each
half
have
been
asso-
ciated
with
a
high
injury
incidence
rate
in
professional4,5 and
elite
youth6soccer
players,
perhaps
owing
to
fatigue7.
Indeed,
stud-
ies
that
have
simulated
soccer
matches
in
laboratory-controlled
conditions
have
observed
within-match
patterns
in
lower
limb
kinematics,8strength,9,10 and
motor
unit
recruitment11 that
are
http://dx.doi.org/10.1016/j.jsams.2015.12.514
1440-2440/©
2015
Sports
Medicine
Australia.
Published
by
Elsevier
Ltd.
All
rights
reserved.
Please
cite
this
article
in
press
as:
Barrett
S,
et
al.
The
within-match
patterns
of
locomotor
efficiency
during
professional
soccer
match
play:
Implications
for
injury
risk?
J
Sci
Med
Sport
(2016),
http://dx.doi.org/10.1016/j.jsams.2015.12.514
ARTICLE IN PRESS
G Model
JSAMS-1266;
No.
of
Pages
6
2
S.
Barrett
et
al.
/
Journal
of
Science
and
Medicine
in
Sport
xxx
(2016)
xxx–xxx
synonymous
with
injury
incidence
trends.
However,
monitoring
these
specific
injury
risk
markers
during
training
and
competi-
tion
is
not
feasible,
and
real-time
surrogate
measures
are
required
to
enable
practitioners
to
make
informed
load-monitoring
judge-
ments
during
matches
and
training
sessions.
Locomotor
activities
such
as
TDC
and
HSR
can
be
monitored
real-
time
with
advances
in
time-motion
analysis
technology;
however,
these
metrics
neglect
the
energetically
taxing
changes
in
speed.12,13
The
change
in
speed
has
been
determined
as
the
frequency
and/or
distance
covered
in
different
acceleration/deceleration
categories13,14 or
using
more
complex
energetic
models
to
estimate
metabolic
cost.12,16 Whilst
these
contemporary
methods
maybe
valuable
additions
to
monitoring
external
load
for
practitioners,
they
quantify
players’
positional
change
in
a
single
plane
of
motion,
neglecting
soccer’s
three-dimensional
nature
of
movement
and
impacts.
Tri-axial
accelerometers
measure
three-dimensional
move-
ments
and
have
been
used
to
quantify
external
load
in
team
sports,
often
determined
by
a
vector
magnitude
termed
PlayerLoadTM
(PLVM17,18 ).
In
contrast
to
existing
metrics
using
time-motion
analysis,
PLVM has
an
acceptable
signal:
noise
ratio19 owing
to
its
demonstrated
test–retest,20 within-
and
between-device
reliability.19 The
within-match
patterns
of
PLVM were
recently
examined
using
standardised
soccer
simulation
under
laboratory-
controlled
conditions
in
which
the
volume
and
intensity
of
intermittent
and
multi-directional
locomotor
activities
were
fixed
in
15-min
match
segments21.
In
this
study,
PLVM increased
in
the
last
15-min
period
of
each
half,
mirroring
the
within
match
patterns
of
fatigue9,11 and
injury
risk,4,5 indicative
of
a
change
in
movement
strategy17 and/or
a
reduced
locomotor
efficiency.21
Soccer
specific
fatigue
may
manifest
in
the
reduced
stiffness
of
the
musculotendon
unit,15 owing
to
a
reduced
central
motor
output,11,15 which
may
compromise
the
absorption
capacity
and
stability
of
lower-limb
joints,28 increasing
the
injury
risk
to
pas-
sive
joint
structures.
Decreased
stability
and
increased
lower-limb
vibrations
associated
with
the
ground
reaction
force
in
a
fatigued
state30 may
be
detected
with
high-resolution
tri-axial
accelerom-
eter
technology
and
may
explain
the
reduced
locomotor
efficiency
observed
in
the
latter
stages
of
each
half
of
simulated
soccer
match-play.21
However,
during
competitive
games,
within-match
changes
in
locomotor
efficiency
may
not
be
detectable
using
PLVM alone,
given
its
strong
positive
association
with
total
distance
covered.18
Hence,
PLVM is
hypothesised
to
decrease
over
the
course
of
each
half
of
match-play
synonymous
with
the
decline
in
locomotor
activity.3In
this
study,
we
primarily
aimed
to
determine
the
within
match-patterns
of
PlayerLoadTM and
locomotor
variables
in
competitive
fixtures;
however,
we
also
attempted
to
determine
match-related
changes
in
players’
locomotor
efficiency
patterns
by
calculating
a
ratio
of
PLVM:TDC
(or
PlayerLoadTM per
meter).
Whilst
exploratory,
we
hypothesised
that
within-match
declines
in
loco-
motor
activities
(TDC)
would
be
greater
than
PlayerLoadTM metrics,
an
uncoupling
which
may
be
identified
with
the
application
of
PLVM:TDC,
and
indicative
of
a
reduced
locomotor
(movement)
effi-
ciency.
Furthermore,
because
PLVM is
influenced
by
individual
gait
patterns20 and
locomotor
activities18,
and
that
match
running
vari-
ables
are
dictated
by
positional
role22,
our
second
aim
was
to
quantify
the
determinants
of
PLVM together
with
its
between
match
variability.
2.
Method
The
study
gained
ethical
approval
from
a
departmental
ethics
committee
prior
to
the
commencement
of
the
study.
As
these
data
reported
as
part
of
this
retrospective
study
was
collected
as
part
of
the
routine
data
monitoring
of
players
in
industry
practice,
informed
consent
was
not
deemed
necessary.27
Data
was
collected
during
the
2012/2013
and
2013/2014
sea-
sons
from
three
English
Championship
U21
teams
(Age:
20.3
±
1.6
years;
Stature:
1.80
±
0.07
m;
Body
Mass:
81.2
±
6.1
kg).
Official’s
permission
was
gained
to
wear
the
MEMS
devices
(Micromechan-
ical
Electrical
Systems)
prior
to
each
match,
which
were
played
on
natural
turf.
On
match
day,
players
wore
a
customised
tight-
fitting
neoprene
garment
underneath
their
match
day
shirts,
with
the
unit
located
between
the
scapulae.
Prior
to
MEMS
device
(Min-
imaxX
S4,
Catapult
Sports,
Melbourne,
Australia)
placement
in
the
players
garment,
units
were
taken
outside
and
activated
15
min
beforehand
to
attenuate
erroneous
data
owing
to
poor
GPS
sig-
nal
quality.
All
warm-up
data
was
excluded
from
the
study.
Match
play
consisted
of
two
45
min
halves
with
a
15
min
passive
half-time
interval.
Any
additional
time
at
the
end
of
each
half
was
excluded
from
the
analysis
given
the
between-match
variation
in
duration.
Only
players
completing
three
full
90
min
games
were
included
in
the
study,
to
permit
the
assessment
of
between
match-variation
in
our
outcome
measures.
Sixty-four
professional
soccer
players
were
included
in
the
study,
which
provided
574
match
observations
from
86
games
(Team
1,
n
=
221;
Team
2,
n
=
196;
Team
3,
n
=
156).
These
match
recordings
were
then
dissected
into
15
min
periods
to
assess
the
within-match
patterns
of
PLVM and
the
individual
accelerometer
planes.
In
accordance
with
previous
time-motion
analysis
research,14 we
used
the
first
15-min
period
as
a
bench-
mark
from
which
to
identify
within-match
changes
in
our
outcome
measures.
Whilst
the
use
of
this
initial
15-min
period
as
a
refer-
ence
point
from
which
to
draw
conclusions
regarding
fatigue
from
time-motion
analysis
metrics
has
been
questioned,
due
to
the
fran-
tic
nature
of
the
opening
exchanges
in
soccer;23 we
adopted
this
analytical
technique
to
identify
within-match
patterns
of
tri-axial
accelerometer
data
and
to
make
inferences
in
regards
to
locomotor
efficiency,
rather
than
fatigue
per
se.
The
MinimaxX
S4
(Catapult
Innovations,
Scoresby,
Victoria)
contains
a
tri-axial
piezoelectric
linear
accelerometer
(Kionix:
KXP94)
sampling
at
a
frequency
of
100
Hz,
as
part
of
an
iner-
tial
sensor
suite
in
the
micromechanical
system.
The
output
of
the
accelerometer
measures
±13
g,
with
each
device
contain-
ing
its
own
microprocessor
with
a
1GB
flash
memory
and
USB
interface
in
order
to
store
and
download
data.
The
device
is
powered
by
an
internal
lithium
ion
battery
with
5
h
of
life,
weighing
67
g
and
is
88
×
50
×
19
mm
in
dimension.
Vector
Mag-
nitude
PlayerLoadTM (PLVM)
and
individual-component
planes
of
PlayerLoadTM (anterior-posterior
PlayerLoadTM [PLAP],
medial-
lateral
PlayerLoadTM [PLML]
and
vertical
PlayerLoadTM [PLV])
were
recorded.
The
calculation
for
PLVM is
the
square
root
of
the
sum
of
the
squared
instantaneous
rate
of
change
in
acceleration
in
each
of
the
three
vectors
(x,
y
and
z)
and
divided
by
100.19 PLAP,
PLML and
PLVwere
calculated
with
the
same
equation,
using
only
the
relevant
axis
in
the
equation.
Expressed
in
arbitrary
units
(au),
PlayerLoadTM
data
were
recorded
using
the
Catapult
software
(Sprint
5.0.9.2,
Catapultsports,
Melbourne,
Australia).
Prior
to
the
start
of
each
season
units
were
calibrated
using
the
manufacturers
jig
to
com-
ply
with
the
manufacturers
guidelines.
The
device
was
orientated
and
placed
stationary
in
each
plane
of
movement
and
recordings
were
set
at
1
g
for
that
position
to
reduce
any
bias
or
drift.
Every
four
weeks
calibration
values
were
monitored.
All
units
remained
within
the
manufacturer’s
calibration
tolerance
limits
throughout
the
testing
period.
The
MEMS
device
(MinimaxX,
S4,
Firmware
version-6.88)
con-
tains
a
10
Hz
global
positioning
satellite
(GPS)
chip
in
order
to
record
the
time
motion
analysis
data.
Total
Distance
Covered
(TDC)
was
used
as
a
measure
of
the
time
motion
analysis
data
(TMA).
Data
were
included
if
the
number
of
satellites
exceeded
6
and
a
hori-
zontal
displacement
of
positioning
(HDOP)
was
less
than
1.5.
Two
Please
cite
this
article
in
press
as:
Barrett
S,
et
al.
The
within-match
patterns
of
locomotor
efficiency
during
professional
soccer
match
play:
Implications
for
injury
risk?
J
Sci
Med
Sport
(2016),
http://dx.doi.org/10.1016/j.jsams.2015.12.514
ARTICLE IN PRESS
G Model
JSAMS-1266;
No.
of
Pages
6
S.
Barrett
et
al.
/
Journal
of
Science
and
Medicine
in
Sport
xxx
(2016)
xxx–xxx
3
Fig.
1.
The
within-match
changes
in
15
min
match
segments
for
(A)
PLVM;
(B)
PLAP;
(C)
PLML;
(D)
PLVduring
soccer
match
play.
Dashed
line
represents
15
min
half-time
period. adifference
versus
0–15
min; bdifference
versus
15–30
min; cdifference
versus
30–45
min; ddifference
versus
45–60
min;*
denotes
a
significant
difference
of
p
≤
0.01.
PLVM—PlayerLoadTM vector
magnitude;
PLAP—PlayerLoadTM in
the
anterior-posterior
plane;
PLML—PlayerLoadTM in
the
medial-lateral
plane;
PLV—PlayerLoadTM vertical
plane.
match
files
were
excluded
as
a
result.
To
assess
the
within-match
patterns
of
PlayerLoadTM and
its
individual
planes
in
comparison
to
the
locomotor
activities,
PLVM was
made
relative
to
TDC
as
a
measure
of
players
locomotor
efficiency
(PlayerLoadTM per
metre
covered;
PL:TDC).
Prior
to
the
analysis,
both
Q–Q
plots
and
stem
and
leaf
charts
were
monitored
to
check
for
normal
distribution.
Assumptions
of
normality
were
further
assessed
by
plotting
boxplots
of
the
residuals
and
a
scatterplot
of
the
predicted
values.
A
linear
mixed
model
was
then
used
to
assess
the
differences
of
PLVM:
TDC,
PLVM,
PLAP,
PLML,
PLVand
TDC,
between
each
15-min
match
period.
Linear
mixed
models
were
able
to
account
for
the
different
sam-
ples
between
teams.
Post-hoc
pairwise
comparisons,
with
Sidak
adjusted
p
values,
were
conducted
in
the
event
of
a
statistically
sig-
nificant
F-ratio.
A
spline
model
was
then
fitted
to
assess
the
relative
change
of
the
aforementioned
variables
across
the
first
and
second
half.
For
all
players
included
in
the
study,
an
individual
coefficient
of
variation
(CV)
was
calculated
for
each
outcome
variable,
by
dividing
standard
deviation
(SD)
by
the
individual
mean
from
each
game.
To
explain
the
variance
within
the
model
playing
positions,
individual
player
and
each
competitive
fixture
were
included
as
random
fac-
tors
within
the
linear
mixed
model.
The
team
for
which
the
player
represented
was
also
encompassed
in
the
model
to
account
for
any
variation
in
tactical
or
physical
approaches
to
match-play,
however
no
effect
was
observed
(data
not
shown).
Analyses
were
completed
using
IBM
SPSS
Statistics
for
windows
software
(release
20;
SPSS
Inc.,
Chicago,
IL,
USA)
and
all
values
are
reported
as
mean
±
SD.
Two-
tailed
statistical
significance
was
accepted
as
p
<
0.05
and
measures
of
effect
size
were
calculated
using
partial
eta-squared
(2).
Mag-
nitude
of
the
effect
sizes
were
small
(>0.02),
medium
(>0.13)
and
large
(>0.26).25
3.
Results
The
initial
0–15
min
of
match
play
incurred
a
significantly
higher
PLVM (2=
0.36–0.43),
PLAP (2=
0.25–0.38),
PLML (2=
0.22–0.38)
and
PLV(2=
0.29–0.42)
in
comparison
to
all
other
time
periods
(See
Fig.
1).
During
the
second
half,
the
absolute
accelerometer
Please
cite
this
article
in
press
as:
Barrett
S,
et
al.
The
within-match
patterns
of
locomotor
efficiency
during
professional
soccer
match
play:
Implications
for
injury
risk?
J
Sci
Med
Sport
(2016),
http://dx.doi.org/10.1016/j.jsams.2015.12.514
ARTICLE IN PRESS
G Model
JSAMS-1266;
No.
of
Pages
6
4
S.
Barrett
et
al.
/
Journal
of
Science
and
Medicine
in
Sport
xxx
(2016)
xxx–xxx
Table
1
The
individual
coefficient
of
between-match
variation
and
the
contribution
of
the
variance
in
soccer
match
play
for
PLVM,
the
individual
accelerometer
planes,
TDC,
and
locomotor
efficiency.
Coefficient
of
Variation
(95%
CI’s) Game
(%)
Player
(%)
Positions
(%)
PLVM (au)
6.6
±
2.4
(6.0
to
7.2)
21.6*63.9*14.1
PLAP (au)
8.8
±
4.0
(7.4
to
10.4)
27.9*40.5*5.7
PLML (au)
9.0
±
4.1
(6.9
to
11.0)
44.8*36.8*12.9
PLV(au)
7.3
±
2.5
(5.7
to
8.9)
37.1*36.6*13.5
TDC
(m)
6.6
±
2.8
(3.5
to
9.5)
10.3
48.3*8.4
PLVM:
TDC
(au)
6.4
±
2.9
(2.0
to
10.8)
21.0*32.1*3.6
PLVM—PlayerLoad
Vector
Magnitude;
PLAP—PlayerLoad
Anterior-Posterior;
PLML—PlayerLoad
Medial-Lateral;
PLV—PlayerLoad
Vertical.
*Represents
significant
determinant
of
variance
within
the
linear
mixed
model
(p
≤
0.01).
Fig.
2.
The
within-match
patterns
of
PL:TDC
during
soccer
match
play.
Dashed
line
represents
15
min
half-time
period.
a-
is
significantly
greater
than
the
0–15
mins,
d-
is
significantly
greater
than
the
46–60
mins.
*
denotes
a
significant
difference
of
p
≤
0.01.
indices
progressively
decreased
in
successive
15
min
match
periods
(see
Fig.
1),
whereas
there
were
no
within-match
changes
in
the
relative
contributions
(%)
of
each
accelerometer
plane.
TDC
showed
significant
decreases
in
each
15
min
phase
in
com-
parison
to
the
initial
0–15
min
period
(0–15:
1788
±
252;
75–90:
1516
±
224;
2=
0.35,
p
=
0.001).
TDC
showed
similar
significant
changes
across
15
min
time
periods
during
match
play
as
was
observed
for
PLVM.
Significant
increases
were
observed
for
PL:TDC
towards
the
end
of
each
half
(See
Fig.
2;
2=
0.11–0.29).
The
rate
of
increase
for
PL:TDC
was
significantly
greater
in
the
first
half
(0.14
±
0.02
au)
compared
to
the
second
half
(0.06
±
0.3
au;
p
=
0.04).
The
variance
of
PLVM,
the
individual
accelerometer
planes
and
locomotor
efficiency
(PL:TDC)
are
illustrated
in
Table
1.
Significant
findings
were
identified
for
both
games
and
the
individual
player
in
the
current
model,
however,
no
significant
associations
were
found
for
position
per
se.
4.
Discussion
The
aim
of
the
current
study
was
to
determine
the
within
match-patterns
of
PlayerLoadTM and
locomotor
efficiency
(PL:TDC)
in
professional
soccer.
Secondary
aims
of
the
study
were
to
quantify
the
between-match
variation
and
determinants
of
these
external
load
metrics.
The
key
findings
from
the
present
study
included:
(1)
PL:TDC
increased
in
the
last
15
min
of
both
the
first
and
second
halves
in
comparison
to
the
initial
0–15
min
period;
(2)
The
spline
model
showed
that
the
PL:TDC
rate
of
increase
was
significantly
greater
in
the
first
half
compared
to
the
second
half,
indicative
of
an
uncoupling
between
PLVM and
TDC;
(3)
Between-
player
(32.1–63.9%)
and
between-match
(10.3–44.8%)
variability
statistically
explained
the
variance
within
the
current
model
for
PL:TDC,
PLVM,
the
individual
accelerometer
planes
and
TDC.
Increased
injury
occurrence
has
been
shown
to
occur
towards
the
latter
stages
of
each
half
within
elite4and
elite
youth6play-
ers
during
competitive
soccer
match
play.
Soccer
specific
fatigue
has
been
purported
to
have
an
aetiological
role
in
the
increased
injury
incidence
observed
during
these
time
periods.4,5,7 Indeed
simulated
soccer
matches
have
shown
alterations
in
lower
limb
kinematics8,
strength8,9 and
motor
unit
recruitment.11 Monitoring
these
responses
during
competitive
soccer
match
play
is
imprac-
tical
and
contravenes
governing
body
regulations,
hence
external
load
indices
have
traditionally
been
used
to
monitor
fatiguing
trends
(TDC,
HSR11).
However,
methods
utilising
changes
in
two-
dimensional
coordinates
fail
to
quantify
critical
three-dimensional
aspects
of
soccer
match
play,
such
as
tackles,
impacts
and
changes
of
direction.
In
the
current
study,
we
utilised
PlayerLoadTM to
assess
the
within
match
patterns
of
competitive
match
play,
identifying
reductions
in
three
dimensional
loading
in
the
latter
stages
of
each
half,
a
trend
synonymous
with
injury
incidence.
However,
the
locomotor
patterns
have
shown
strong
relation-
ships
to
PlayerLoadTM during
soccer
training,
with
higher
distances
covered
associated
with
greater
loading.18 Therefore,
the
progres-
sive
reductions
in
loading
identified
in
PlayerLoadTM during
each
half
of
match
play
likely
reflects
the
typical
within
match
time-
motion
patterns
of
soccer
match-play,
which
arguably
represents
the
match
context,29 rather
than
fatigue
per
se.
Hence,
in
this
study
we
calculated
a
ratio
of
PlayerLoadVM to
total
distance
covered
as
a
measure
of
locomotor
efficiency,
in
an
attempt
to
identify
any
uncoupling
which
may
be
indicative
of
player
fatigue.
We
observed
a
large
increase
in
the
PL:TDC
during
the
last
15
min
period
of
each
half
when
benchmarked
against
the
initial
0–15
min
period.
During
a
standardised
soccer
simulation
(SAFT90),
Barrett
and
colleagues21
showed
PLVM increased
towards
the
end
of
each
half
when
the
loco-
motor
activities
were
fixed
in
15
min
segments.
Using
the
same
simulation,
Small
et
al.8observed
within
match
alterations
in
hip
extension
and
knee
flexion
during
sprinting,
which
resulted
in
a
decreased
stride
length
in
a
temporal
pattern
that
corroborates
with
the
decreased
locomotor
efficiency
observed
in
this
study,
and
that
of
injury
incidence.4–6
A
fatigue-induced
reduction
in
stride
length
during
running
may
explain
the
locomotor
efficiency
patterns
we
observed
as
its
reciprocal
increase
in
stride
frequency
and
foot
contacts
incurs
loading
detected
by
the
accelerometer.
Furthermore,
increases
in
accelerometer
metrics
in
the
latter
stages
of
each
half
may
reflect
reduced
pre-activation
of
the
musculotendon
unit
associated
with
fatigue,11,15 leading
to
an
impaired
capacity
to
reduce
the
vibra-
tion
amplitudes
in
lower-limb
soft
tissue
(∼20%;30)
that
result
from
ground
reaction
forces.
However,
caution
has
been
advised
when
interpreting
tri-axial
accelerometer
data
collected
at
the
scapulae
to
assess
lower
limb
movement
strategy
changes.21 Whilst
this
Please
cite
this
article
in
press
as:
Barrett
S,
et
al.
The
within-match
patterns
of
locomotor
efficiency
during
professional
soccer
match
play:
Implications
for
injury
risk?
J
Sci
Med
Sport
(2016),
http://dx.doi.org/10.1016/j.jsams.2015.12.514
ARTICLE IN PRESS
G Model
JSAMS-1266;
No.
of
Pages
6
S.
Barrett
et
al.
/
Journal
of
Science
and
Medicine
in
Sport
xxx
(2016)
xxx–xxx
5
unit
positioning
is
necessary
for
MEMS
devices
to
enhance
the
GPS
signal
quality,
laboratory
studies
have
indicated
that
the
posi-
tion
of
the
unit
between
the
scapulae
accrues
different
magnitudes
and
planar
contributions
of
tri-axial
accelerometer
data
versus
its
criterion
positioning
at
the
centre
of
mass
during
both
treadmill
running20 and
a
soccer
match
simulation.21 The
upper
body
move-
ments
of
the
trunk
are
also
non-uniform
during
the
stochastic
and
combative
nature
of
soccer
match-play,
and
may
somewhat
mask
the
lower
limb
changes
in
running
kinematics10 and
lower
limb
stiffness.15,17,30 The
scapulae
unit
positioning
during
competitive
soccer
fixtures
did
not
preclude
us
from
identifying
modulations
in
locomotor
efficiency,
however
future
industry-practice
using
micro-sensor
technology
positioned
at
the
centre
of
mass
may
be
warranted
to
determine
lower
limb
loading,
independent
of
GPS
monitoring.
Whilst
this
study
has
identified
modulations
in
locomotor
effi-
ciency
that
may
be
used
in
industry-practice
to
inform
rotation
policy
by
identifying
players
at
an
exacerbated
risk
of
injury
or
to
denote
the
onset
of
fatigue,
we
recognise
that
further
work
is
nec-
essary
to
confirm
our
speculation.
We
also
acknowledge
the
crudity
of
our
measure
of
locomotor
efficiency,
considering
that
total
dis-
tance
covered
by
players
does
not
represent
the
intermittent
and
intensity
distribution
of
soccer
and
that
accelerometer
loading
rate
is
influenced
by
running
speed.20 Furthermore,
if
observing
locomotor
efficiency
modulations
has
a
role
in
reducing
injury
risk
and
fatigue
management,
real-time
MEMS
data
capture
and
processing
are
necessary,
yet
the
accuracy
of
live
GPS
data
has
been
questioned.26 Accordingly,
further
work
is
required
in
terms
of
both
aetiological
research
and
technological
evolution
to
realise
the
potential
application
of
tri-axial
accelerometer
data
in
professional
sports.
To
our
knowledge,
this
study
is
the
first
to
examine
the
between-
match
variation
in
PlayerLoadTM indices
during
actual
soccer
match
play.
We
observed
low
coefficients
of
variation
for
the
vector
mag-
nitude
(6.4
±
2.4%)
and
its
individual
planes
(7.3–9.0%),
which
in
combination
with
its
sound
test-retest,20 within-
and
between-
device
reliability,19 suggests
that
PlayerLoadTM data
may
be
useful
for
practitioners
to
detect
worthwhile
changes
in
an
athlete’s
exter-
nal
load
or
changes
in
locomotor
efficiency.
Individual
gait
patterns
have
been
speculated
to
cause
the
variation
between-athletes
PLVM
values
during
incremental
treadmill
running20 and
during
a
con-
trolled
fixed
soccer
simulation.21 Consequently,
we
suggested
that
PLVM and
the
individual
planes
should
be
treated
and
measured
within
an
individual-specific
manner
as
a
measure
of
external
load,
findings
which
were
corroborated
in
the
current
study
as
the
indi-
vidual
player
explained
more
variance
in
PLVM (63.9%)
versus
the
match
(21.6%)
and
positional
role
(14.1%)
per
se.
Practitioners
using
accelerometer
data
on
a
routine
basis
are
therefore
recommended
to
limit
their
analyses
to
within-player
contrasts
due
to
the
large
variability
observed
between
individuals.
5.
Conclusions
PL:TDC,
PLVM and
the
individual
component
accelerometer
planes
demonstrated
within-match
patterns
during
elite
profes-
sional
soccer
match
play.
Towards
the
end
of
each
half,
the
locomotor
efficiency
(PL:TDC)
increased,
suggestive
of
an
increase
in
the
loading
required
for
every
given
metre
of
distance
covered
on
the
pitch.
Since
these
within
match
patterns
are
concomitant
with
match-induced
alterations
in
strength,
motor
unit
recruit-
ment
and
lower-limb
kinetics
that
have
been
linked
with
fatigue
and
increased
injury
incidence,
locomotor
efficiency
may
be
a
use-
ful
tool
to
inform
substitutions
or
rotation
policy
in
team
sports.
The
efficacy
of
accelerometer
metrics
are
further
supported
by
their
low
signal
to
noise
ratio,
but
their
large
between-player
variation
limits
comparisons
between
individuals.
Practical
applications
•Locomotor
efficiency,
PLVM and
the
individual
component
accelerometer
planes
detect
within-match
patterns
in
soccer.
•The
latter
stages
of
each
half
show
an
increase
in
locomotor
effi-
ciency,
a
trend
synonymous
with
observations
of
increased
injury
incidence
and
fatigue.
•Locomotor
efficiency
may
be
a
useful
tool
to
inform
substitutions
or
rotation
policy
in
team
sports.
Acknowledgments
No
source
of
funding
was
obtained
for
this
study.
We
would
like
to
thank
all
players
and
clubs
involved
with
the
study.
References
1.
Colby
M,
Dawson
B,
Heasman
J
et
al.
Training
and
game
loads
and
injury
risk
in
elite
Australian
footballers.
J
Strength
Cond
Res
2014.
Apr
7.
Ahead
of
print:
1–8.
2.
Rogalski
B,
Dawson
B,
Heasman
J
et
al.
Training
and
game
loads
and
injury
risk
in
elite
Australian
footballers.
J
Sci
Med
Sport
2013;
16(6):499–503.
3.
Weston
M,
Drust
B,
Gregson
W.
Intensities
of
exercise
during
match-play
in
FA
Premier
League
referees
and
players.
J
Sports
Sci
2011;
29(5):527–532.
4.
Ekstrand
J,
Haggulund
M,
Walden
M.
Epidemiology
of
muscle
injuries
in
profes-
sional
football
(soccer).
Am
J
Sports
Med
2011;
39(6):1226–1232.
5.
Woods
C,
Hawkins
RD,
Maltby
S
et
al.
The
Football
Association
Medical
Research
Programme:
an
audit
of
injuries
in
professional
football—analysis
of
hamstring
injuries.
Br
J
Sports
Med
2004;
38(1):36–41.
6.
Price
RJ,
Hawkins
RD,
Hulse
MA
et
al.
The
Football
Association
medical
research
programme:
an
audit
of
injuries
in
academy
youth
football.
Br
J
Sports
Med
2004;
38(4):466–471.
7.
McCall
A,
Carling
C,
Nedelec
M
et
al.
Risk
factors,
testing
and
preventative
strategies
for
non-contact
injuries
in
professional
football:
current
perceptions
and
practices
of
44
teams
from
various
premier
leagues.
Br
J
Sports
Med
2014;
48(18):1352–1357.
8.
Small
K,
McNaughton
L,
Greig
M
et
al.
The
effects
of
multi-directional
soccer-
specific
fatigue
on
markers
of
hamstring
injury
risk.
J
Sci
Med
Sport
2010;
13(1):120–125.
9.
Lovell
R,
Midgley
AW,
Barrett
S
et
al.
Effects
of
different
half-time
strategies
on
second
half
soccer-specific
speed,
power
and
dynamic
strength.
Scan
J
Med
Sci
Sports
2013;
23(1):105–113.
10.
Small
K,
McNaughton
LR,
Greig
M
et
al.
Soccer
fatigue,
sprinting
and
hamstring
injury
risk.
Int
J
Sports
Med
2009;
30(8):573–578.
11.
Marshall
PW,
Lovell
R,
Jeppesen
GK
et
al.
Hamstring
muscle
fatigue
and
central
motor
output
during
a
simulated
soccer
match.
PLoS
ONE
2014:1–11.
Apr
7.
(ahead
of
print).
12.
Gaudino
P,
Iaia
FM,
Alberti
G
et
al.
Monitoring
training
in
elite
soccer
players:
systematic
bias
between
running
speed
and
metabolic
power
data.
Int
J
Sports
Med
2013;
34(11):963–968.
13.
Varley
V,
Aughey
R.
Acceleration
profiles
in
elite
Australian
soccer.
Int
J
Sports
Med
2013;
34(1):34–39.
14.
Akenhead
R,
Hayes
PR,
French
D
et
al.
Diminutions
of
acceleration
and
decel-
eration
output
during
professional
football
match
play.
J
Sci
Med
Sport
2013;
16(6):556–561.
15.
Oliver
JL,
De
Ste
Croix
MBA,
Lloyd
RS
et
al.
Altered
neuromuscular
control
of
leg
stiffness
following
soccer-specific
exercise.
Eur
J
Appl
Physiol
2014;
114(1):2241–2249.
16.
Osgnach
C,
Posner
S,
Bernadini
R
et
al.
Energy
cost
and
metabolic
power
in
elite
soccer:
a
new
match
analysis
approach.
Med
Sci
Sports
Exerc
2010;
42(1):170–178.
17.
Cormack
S,
Mooney
MM,
Morgan
W
et
al.
Influence
of
neuromuscular
fatigue
on
accelerometer
load
in
elite
Australian
football
players.
Int
J
Sports
Physiol
Perform
2013;
8(4):373–380.
18.
Scott
TJ,
Black
CR,
Quinn
J
et
al.
Validity
and
reliability
of
the
session-RPE
method
for
quantifying
training
in
Australian
football:
a
comparison
of
the
CR10
and
CR100
scales.
J
Strength
Cond
2013;
27(1):270–276.
19.
Boyd
LJ,
Ball
K,
Aughey
R.
The
reliability
of
MinimaxX
accelerometers
for
mea-
suring
physical
activity
in
Australian
football.
Int
J
Sports
Physiol
Perform
2011;
6(3):311–321.
20.
Barrett
S,
Midgley
AW,
Lovell
R.
PlayerLoadTM:
reliability,
convergent
valid-
ity
and
influence
of
unit
position
during
treadmill
running.
Int
J
Sports
Physiol
Perform
2014;
9(6):945–952.
21.
Barrett
S,
Midgley
AW,
Towlson
C
et
al.
Within-match
PlayerLoadTM patterns
during
simulated
soccer
match
(SAFT90):
Potential
implications
for
unit
posi-
tioning
and
fatigue
management.
Int
J
Sports
Physiol
Perform
2016;
11(1):
135–140.
22.
Gregson
W,
Drust
B,
Atkinson
G
et
al.
Match-match
variability
of
high-speed
activities
in
premier
league
soccer.
Int
J
Sports
Med
2010;
31(4):237–242.