Content uploaded by Paul S Bradley
Author content
All content in this area was uploaded by Paul S Bradley on Jul 20, 2015
Content may be subject to copyright.
This article was downloaded by: [University of Sunderland]
On: 30 April 2012, At: 12:54
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Journal of Sports Sciences
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/rjsp20
High-intensity running in English FA Premier League
soccer matches
Paul S. Bradley a , William Sheldon a , Blake Wooster b , Peter Olsen c , Paul Boanas b &
Peter Krustrup d
a Department of Sport and Exercise Sciences, University of Sunderland, Sunderland, UK
b Department of Research and Performance Analysis, ProZone Sports Ltd., Leeds, UK
c Department of Sport and Exercise, University of Teesside, Middlesbrough, UK
d Department of Exercise and Sport Sciences, University of Copenhagen, Copenhagen,
Denmark
Available online: 19 Jan 2009
To cite this article: Paul S. Bradley, William Sheldon, Blake Wooster, Peter Olsen, Paul Boanas & Peter Krustrup (2009): High-
intensity running in English FA Premier League soccer matches, Journal of Sports Sciences, 27:2, 159-168
To link to this article: http://dx.doi.org/10.1080/02640410802512775
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to
anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should
be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,
proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in
connection with or arising out of the use of this material.
High-intensity running in English FA Premier League soccer matches
PAUL S. BRADLEY
1
, WILLIAM SHELDON
1
, BLAKE WOOSTER
2
, PETER OLSEN
3
,
PAUL BOANAS
2
, & PETER KRUSTRUP
4
1
Department of Sport and Exercise Sciences, University of Sunderland, Sunderland, UK,
2
Department of Research and
Performance Analysis, ProZone Sports Ltd., Leeds, UK,
3
Department of Sport and Exercise, University of Teesside,
Middlesbrough, UK, and
4
Department of Exercise and Sport Sciences, University of Copenhagen, Copenhagen, Denmark
(Accepted 29 September 2008)
Abstract
The aims of this study were to (1) determine the activity profiles of a large sample of English FA Premier League soccer
players and (2) examine high-intensity running during elite-standard soccer matches for players in various playing positions.
Twenty-eight English FA Premier League games were analysed during the 2005–2006 competitive season (n¼370), using a
multi-camera computerised tracking system. During a typical match, wide midfielders (3138 m, s¼565) covered a greater
distance in high-intensity running than central midfielders (2825 m, s¼473, P¼0.04), full-backs (2605 m, s¼387,
P50.01), attackers (2341 m, s¼575, P50.01), and central defenders (1834 m, s¼256, P50.01). In the last 15 min of a
game, high-intensity running distance was *20% less than in the first 15-min period for wide midfielders (467 m, s¼104 vs.
589 m, s¼134, P50.01), central midfielders (429 m, s¼106 vs. 534 m, s¼99, P50.01), full-backs (389 m, s¼95
vs. 481 m, s¼114, P50.01), attackers (348 m, s¼105 vs. 438 m, s¼129, P50.01), and central defenders (276 m, s¼93
vs. 344 m, s¼80, P50.01). There was a similar distance deficit for high-intensity running with (148 m, s¼78 vs. 193 m,
s¼96, P50.01) and without ball possession (229 m, s¼85 vs. 278 m, s¼97, P50.01) between the last 15-min and first
15-min period of the game. Mean recovery time between very high-intensity running bouts was 72 s (s¼28), with a 28%
longer recovery time during the last 15 min than the first 15 min of the game (83 s, s¼26 vs. 65 s, s¼20, P50.01). The
decline in high-intensity running immediately after the most intense 5-min period was more evident in attackers (216 m,
s¼50 vs. 113 m, s¼47, P50.01) and central defenders (182 m, s¼26 vs. 96 m, s¼39, P50.01). The results suggest that
high-intensity running with and without ball possession is reduced during various phases of elite-standard soccer matches
and the activity profiles and fatigue patterns vary among playing positions. The current findings provide valuable information
about the high-intensity running patterns of a large sample of elite-standard soccer players, which could be useful in the
development and prescription of specific training regimes.
Keywords: ProZone, high-intensity running, recovery time, fatigue, playing position
Introduction
Findings from time–motion studies are useful for
quantifying the physiological demands of soccer and
can provide the conceptual framework for the
development of specific performance tests and
training regimes (Bangsbo, Nørregaard, & Thorsøe,
1991; Drust, Reilly, & Cable, 2000; Drust, Reilly, &
Rienzi, 1998). Researchers have examined the
activity patterns of Australian (Withers, Maricic,
Wasilewski, & Kelly, 1982), Belgian (Van Gool, Van
Gerven, & Boutmans, 1988), Canadian (Mayhew
and Wenger, 1985), Danish (Bangsbo et al., 1991;
Mohr, Krustrup, & Bangsbo, 2003), Swedish
(Andersson, Ekblom, & Krustrup, 2007), Japanese
(Ohashi, Togari, Isokawa, & Suzuki, 1988), Italian
(Mohr et al., 2003), and Spanish League soccer
players (Di Salvo et al., 2007). However, there is
limited information on movement patterns of con-
temporary elite-standard English League soccer
players. Studies conducted to date have used small
sample sizes (Rienzi, Drust, Reilly, Carter, & Martin,
2000; Strudwick and Reilly, 2001) or were per-
formed more than 30 years ago (Reilly & Thomas,
1976). Therefore, a study using a large sample of
players could provide a more accurate depiction of
the current demands of the elite-standard English
League, since there is variability in movement
patterns both among and within players according
to match and playing position (Bangsbo, 1994; Mohr
et al., 2003).
Some researchers have suggested that distances
covered during high-intensity running in matches are
valid measures of physical performance in soccer
Correspondence: P. S. Bradley, Department of Sport and Exercise Sciences, University of Sunderland, Darwin Building, Chester Road, Sunderland SR1 3SD,
UK. E-mail: paul.s.bradley@sunderland.ac.uk
Journal of Sports Sciences, January 15th 2009; 27(2): 159–168
ISSN 0264-0414 print/ISSN 1466-447X online Ó2009 Taylor & Francis
DOI: 10.1080/02640410802512775
Downloaded by [University of Sunderland] at 12:54 30 April 2012
because of their strong relationship with training
status (Krustrup et al., 2003; Krustrup, Mohr,
Ellingsgaard, & Bangsbo, 2005) and are a distin-
guishing characteristic between different standards of
player (Bangsbo et al., 1991; Mohr et al., 2003).
Recent findings using computerised time–motion
analysis of elite-standard Italian League soccer
players demonstrated that high-intensity running
was reduced towards the end of the game and
temporarily after intense periods of the game (Mohr,
Krustrup, & Bangsbo, 2003, 2005). Mohr et al.
(2003) did not investigate the extent to which
temporary and end-game fatigue occurred for players
in various playing positions. In addition, it is unclear
whether decrements in high-intensity running to-
wards the end of the game are with or without ball
possession. Detailed examination of high-intensity
running by position in 5-min periods of multiple
matches would provide valuable information regard-
ing patterns of within-game fatigue. Such detailed
analyses could also be used to provide information
regarding differences in high-intensity running in the
most intense periods of play and recovery times after
very high-intensity running bouts. Finally, quantify-
ing players’ maximal running speeds could provide
insight into any fatigue-related changes in high-speed
performance during matches and positional varia-
tions in speed profiles.
Therefore, the aims of the present study were to
(1) determine the activity profiles of a large sample
of English FA Premier League soccer players and
(2) examine high-intensity running during elite-
standard soccer matches for players in various
playing positions.
Methods
Match analysis
With institutional ethics approval, 28 English FA
Premier League games were analysed during the
2005–2006 competitive season, using a multi-camera
computerised tracking system (ProZone Version 3.0,
ProZone Sports Ltd
1
, Leeds, UK). All outfield
players’ movements were captured during each game
by eight colour cameras (Vicon surveyor dome
SVFT-W23, Oxford, UK) positioned in each of the
stadiums at roof height (Figure 1). During the
installation process, each camera’s position, zoom,
and field of vision were fixed to guarantee the long-
term stability of the capture system. The position of
the cameras allowed the complete playing area to be
covered from all corners of the stadium to address
issues of accuracy, occlusion, resilience, and resolu-
tion. Furthermore, to facilitate the tracking of each
player, every section of the playing area was covered
by at least two cameras. The data captured were
analysed using match-analysis software (Stadium
Manager, ProZone Sports Ltd
1
, Leeds, UK) to
produce a single data set on each player’s activity
pattern during a match. Signals received from the
cameras were digitised at a sampling frequency of
10 Hz. This system has recently been independently
validated to verify the capture process and subse-
quent accuracy of the data (Di Salvo, Collins,
McNeill, & Cardinale, 2006).
The games selected for analysis were played
between teams of a similar placing in the league
table and at a similar time of day to minimise the
influence of circadian variations on performance
(Reilly & Brooks, 1986). To avoid indirect compar-
isons with playing time, only data from players
completing an entire match were used. This allowed
profiling of 370 players in various playing positions
(92 central defenders, 84 full-backs, 80 central
midfielders, 52 wide midfielders, and 62 attackers).
Furthermore, to ensure team and player confidenti-
ality, all data were desensitised before analysis.
Movement categories and speed thresholds
Players’ activities were coded into the following
categories and speed thresholds: standing
(0–0.6 km h
71
), walking (0.7–7.1 km h
71
),
jogging (7.2–14.3 km h
71
), running (14.4–
19.7 km h
71
), high-speed running (19.8–
25.1 km h
71
), and sprinting (425.1 km h
71
).
High-intensity running consisted of running, high-
speed running, and sprinting (running speed
414.4 km h
71
). Very high-intensity running
consisted of high-speed running and sprinting
(running speed 419.8 km h
71
). The speed thresh-
olds for each category are similar to those reported
previously (Bangsbo et al., 1991; Mohr et al., 2003;
Rampinini, Coutts, Castagna, Sassi, & Impellizzeri,
2007). High-intensity running with ball possession
was defined as the high-intensity running
distance covered when the players’ own team was
in possession of the ball. High-intensity running
without ball possession was defined as the
subtraction of high-intensity running distance with
possession of the ball from total high-intensity
running distance. Peak high-intensity running
distance represented the 5 min that contained most
high-intensity running in a match and was
specific for each player profiled (Mohr et al.,
2003). Distance covered, frequency of occurrence,
and time spent in each category were obtained at 5-,
15-, and 45-min time periods. Mean recovery time
was defined as the time that elapsed between very
high-intensity running bouts (running speed
419.8 km h
71
). Maximal running speed was
defined as the top speed a player reached in 5-min
periods of a match.
160 P. S. Bradley et al.
Downloaded by [University of Sunderland] at 12:54 30 April 2012
Inter- and intra-observer reliability
An intra- and inter-observer reliability study was
conducted using five players (two central defenders,
two central midfielders, and one attacker) from a
randomly selected English FA Premier League game.
Two trained observers tracked each player on two
separate occasions, interspersed by 7 days, and
coefficients of variation were determined to assess
reliability (Atkinson & Nevill, 1998). The intra-
observer coefficients of variation for total distance
covered, walking, running, high-speed running,
high-intensity running, very high-intensity running,
and sprinting were 1.0, 0.9, 1.2, 1.6, 1.1, 1.8, and
2.4%, respectively. The inter-observer coefficients
of variation for total distance covered, walking,
running, high-speed running, high-intensity running,
and very high-intensity running were 52% with
Figure 1. The ProZone
1
capture system uses eight colour cameras that cover the entire pitch from all four corners of the stadium: (A)
cameras 1–3 positioned in corner 1; (B) cameras 4–6 positioned in corner 2; (C) cameras 7 and 8 positioned opposite one another.
High-intensity running in soccer matches 161
Downloaded by [University of Sunderland] at 12:54 30 April 2012
the exception of sprinting, which was 3.5%. These
values compare favourably with those reported by
Impellizzeri and colleagues (Impellizzeri, Sassi, &
Rampinini, 2006) for similar computerised tracking
systems used in soccer.
Statistical analyses
All statistical analyses were conducted using SPSS
for Windows Version 14.0 (SPSS Inc., Chicago, IL,
USA). Descriptive statistics were calculated on each
variable and z-scores were used to verify normality.
Differences between the first and second half were
determined using paired sample t-tests. Differences
between 5-, 15-, and 45-min periods within a match
were determined using one-way analysis of variance
(ANOVA) with repeated measures. In the event of a
significant difference, Tukey’s post-hoc tests were
used to identify any localised effects. Statistical
significance was set at P50.05. Data are presented
as means and standard deviations unless otherwise
stated.
Results
Activity profile
During matches, players stood for 5.6% (s¼2.4) of
the total time. Low-intensity activity represented
85.4% of total time, which consisted of 59.3%
(s¼4.8) walking and 26.1% (s¼3.8) jogging.
High-intensity runs represented 9.0% of total time,
which consisted of 6.4% (s¼1.6) running, 2.0%
(s¼0.6) high-speed running, and 0.6% (s¼0.2)
sprinting. Players spent longer standing and walk-
ing in the second than the first half (standing:
181 s, s¼81 vs. 148 s, s¼70, P50.01; walking:
1739 s, s¼143 vs. 1674 s, s¼150, P50.01).
More time was spent jogging and running in the
first than the second half (jogging: 769 s, s¼109
vs. 723 s, s¼126, P50.01; running: 192 s, s¼51
vs. 179 s, s¼56, P50.01). There was no differ-
ence in time spent in high-speed running or
sprinting between the two halves (high-speed
running: 58 s, s¼19 vs. 56 s, s¼18, P¼0.06;
sprinting: 16 s, s¼8 vs. 17 s, s¼9, P¼0.07).
Players performed more bouts of high-intensity
running in the first than the second half (279,
s¼66 vs. 267, s¼67, P50.01). No differences
were observed for sprint frequency between the
two halves (17, s¼8 vs. 18, s¼9, P¼0.56).
Distances covered
Mean total distance covered during a match was
10,714 m (s¼991). The distance covered during the
first half was greater than during the second half of a
match (5422 m, s¼561 vs. 5292 m, s¼508,
P50.01). Greater distances were covered in the
first half than the second half when jogging (2172 m,
s¼338 vs. 2052 m, s¼315, P50.01) and running
(879 m, s¼237 vs. 827 m, s¼221, P50.01),
whereas the distance covered when walking was
greater in the second than in the first (1929 m,
s¼145 vs. 1889 m, s¼183, P50.01). There were
no differences between first and second half distance
for high-speed running (326 m, s¼111 vs. 336 m,
s¼114, P¼0.98) and sprinting (123 m, s¼59 vs.
132 m, s¼68, P¼0.06).
High-intensity profile
Mean distances covered in high-intensity running
and very high-intensity running were 2492 m
(s¼625) and 905 m (s¼285). Distance covered in
high-intensity running was 17% less in the last
15 min of the first half and 21% less in the last
15 min of the second half than in the first 15-min
period of the game (391 m, s¼131 and 374 m,
s¼119 vs. 466 m, s¼137, P50.01) (Figure 2). The
distance covered by players when sprinting was lower
in the last 15-min versus the first 15-min period in
both the first (34 m, s¼23 vs. 43 m, s¼17,
P¼0.02) and second half (36 m, s¼20 vs. 44 m,
s¼19, P¼0.02). The distance covered in high-
intensity running with possession of the ball in the
last 15-min period of the game was 23% less than in
the first 15-min period (148 m, s¼78 vs. 193 m,
s¼96 m, P50.01) (Figure 3).
Peak distance covered in high-intensity running in
a 5-min period was 231 m (s¼53). In the subse-
quent 5-min period, the amount of high-intensity
running was 126 m (s¼52), which was 6% less than
the mean distance covered during all 5-min periods
minus the peak 5-min period (134 m, s¼35,
Figure 2. Distance covered in high-intensity running by players
during the first and second half within 15-min periods of match-
play. Significant difference between first and second half:
*P50.01;
#
P50.05.
{
Different from last 15 min (P50.05).
Values are means and standard deviations.
162 P. S. Bradley et al.
Downloaded by [University of Sunderland] at 12:54 30 April 2012
P¼0.03) (Figure 4). When peak distance covered in
high-intensity running was analysed in each half,
high-intensity running in a 5-min period was 213 m
(s¼54) and 210 m (s¼52) for first and second half
respectively. In the subsequent 5-min period, the
amount of high-intensity running was 126 m (s¼48)
and 122 m (s¼49) for the first and second half
respectively, which was 7% and 6% less than the
mean high-intensity distance covered during all 5-
min periods in each half minus the peak 5-min
period (136 m, s¼37 and 130 m, s¼34 , P¼0.01).
Recovery time and maximal running speed
Mean recovery time between very high-intensity
running bouts was 72 s (s¼28). Recovery times
during the second half were 15% longer than in the
first half (77 s, s¼25 vs. 67 s, s¼23, P¼0.01) and
28% longer during the last 15 min than the first
15 min of the game (83 s, s¼26 vs. 65 s, s¼20,
P50.01). Moreover, players’ recovery time in-
creased over the duration of each half, with longer
(P¼0.01) recovery times found for the second 5-min
period and in particular the last five 5-min periods of
the first half and the last three 5-min periods of the
second half compared with the first 5-min period
(Figure 5).
Maximal running speeds during the second half
were not different from the first half (7.64 m s
71
,
s¼0.40 vs. 7.62 m s
71
,s¼0.44, P¼0.58). The
maximum running speeds were lower during the last
15 min of the first half (7.51 m s
71
,s¼0.61,
P¼0.01) but not the second half (7.59 m s
71
,
s¼0.61, P¼0.29) versus the first 15-min period of
the game (7.65 m s
71
,s¼0.56).
Positional differences
Wide and central midfielders covered a greater total
distance (P50.01) than full-backs, attackers, and
central defenders (Table I). Wide midfielders also
covered a greater distance in high-intensity running
(P50.01) than central defenders, full-backs, central
midfielders, and attackers. Central defenders under-
took less (P50.01) high-intensity running than all
other positions. Wide midfielders and full-backs
covered a greater (P50.01) distance when sprinting
than central midfielders, attackers, and central
defenders. The attackers demonstrated no significant
decline in high-intensity running distance (P¼0.10),
whereas central defenders and full-backs showed a
greater decline (P50.05) from the first to the
second half than central and wide midfielders
(Figure 6). Distances covered in high-intensity
running in the last 15 min of the game were 21,
20, 19, 21, and 20% less than in the first 15-min
period for wide midfielders (467 m, s¼104 vs.
589 m, s¼134, P50.01), central midfielders
(429 m, s¼106 vs. 534 m, s¼99, P50.01), full-
backs (389 m, s¼95 vs. 481 m, s¼114, P50.01),
attackers (348 m, s¼105 vs. 438 m, s¼129,
Figure 3. Contribution of high-intensity running with and without the ball to the high-intensity distance deficit during match-play. High-
intensity running distances with and without the ball were different from the first 15-min values at all time periods: *P50.01;
#
P50.05.
Figure 4. Peak high-intensity running distance covered by players
in 5-min periods, next 5-min, final 5-min, and mean values of the
remaining 5-min periods minus the peak value. *Different from
mean 5-min period (P50.05). Values are means and standard
deviations.
High-intensity running in soccer matches 163
Downloaded by [University of Sunderland] at 12:54 30 April 2012
P50.01), and central defenders (276 m, s¼93 vs.
344 m, s¼80, P50.01).
The decline in high-intensity running immediately
after the most intense 5-min period was more evident
in central defenders (182 m, s¼26 vs. 96 m, s¼39 or
47%, P50.01) and attackers (216 m, s¼50 vs.
113 m, s¼47 or 48%, P50.01) than full-backs
(243 m, s¼39 vs. 137 m, s¼53 or 44%, P50.01),
wide midfielders (282 m, s¼49 vs. 162 m, s¼56 or
43%, P50.01), and central midfielders (251 m,
s¼46 vs. 137 m, s¼50 or 45%, P50.01). However,
there was no difference in the relative decline in high-
intensity running between positions. Wide and
central midfielders demonstrated less recovery time
(P50.01) between very high-intensity running bouts
than full-backs, attackers, and central defenders.
Wide midfielders, attackers, and full-backs also
demonstrated a higher maximal running speed in
the first 15-min period of the game than central
midfielders (wide midfielders: 7.98 m s
71
,s¼
0.38, P50.01; attackers: 7.81 m s
71
,s¼0.58,
P50.01; full-backs: 7.70 m s
71
,s¼0.49, P¼
0.02; central defenders: 7.61 m s
71
,s¼0.57). This
was also the case in the final 15-min period of the first
half (wide midfielders: 7.80 m s
71
,s¼0.50,
P50.01; attackers: 7.57 m s
71
,s¼0.51, P¼0.03;
full-backs: 7.81 m s
71
,s¼0.54, P50.01; central
midfielders: 7.33 m s
71
,s¼0.60) and second half
(wide midfielders: 7.93 m s
71
,s¼0.61, P50.01;
attackers: 7.77 m s
71
,s¼0.44, P¼0.01; full-
backs: 7.62 m s
71
,s¼0.65, P¼0.04; central
midfielders: 7.32 m s
71
,s¼0.59) (Figure 7).
Maximal running speed over the course of the game
was higher (P50.05) for wide midfielders, attackers,
and full-backs than central midfielders and central
defenders.
Figure 5. Mean recovery time between very high-intensity bouts (running speed 419.8 km h
71
) attained in 5-min periods of match-play.
*Different from first 5 min of each half (P50.01).
#
Different from the first 5 min of each half (P50.05). Broken line denotes mean values
over a 45-min period. Values are means and standard deviations.
Table I. Match performance variables in relation to playing position.
Match performance variables
Central
defenders
(n¼92)
Full-backs
(n¼84)
Central
midfielders
(n¼80)
Wide
midfielders
(n¼52)
Attackers
(n¼62)
Distances covered
Total (m) 9885 +555 10710 +589 11450 +608
a
11535 +933
a
10314 +1175
High-intensity running (m) 1834 +256 2605 +387 2825 +473 3138 +565
b
2341 +575
Very high-intensity running (m) 603 +132 984 +195 927 +245 1214 +251
b
955 +239
Sprinting (m) 152 +50 287 +98
c
204 +89 346 +115
c
264 +87
Other variables
Maximal running speed (m s
71
) 7.31 +0.30 7.74 +0.24
d
7.52 +0.32 7.93 +0.31
d
7.76 +0.28
d
Recovery time (s) 101 +15 74 +23 62 +19
a
51 +16
a
73 +22
Notes:
a
Different from central defenders, full-backs, and attackers (P50.05).
b
Different from all other playing positions (P50.05).
c
Different from central defenders, central midfielders, and attackers (P50.01).
d
Different from central defenders and central midfielders
(P50.05).
164 P. S. Bradley et al.
Downloaded by [University of Sunderland] at 12:54 30 April 2012
Discussion
The results of this study provide evidence of the
high physical demands of the modern elite-stan-
dard English League and that the amounts of high-
intensity running of central and wide midfielders,
full-backs, and central defenders, but not attackers,
are comparable to recent results obtained from
elite-standard Italian and Spanish Leagues. Despite
large positional differences in high-intensity run-
ning, the pattern of high-intensity running changed
after the most intense periods and towards the end
of the game for players in all playing positions. The
mean recovery time between very high-intensity
running bouts increased markedly over the dura-
tion of the game, resulting in an end-game
‘‘distance deficit’’ of *20% for high-intensity
running with and without ball possession. After
the most intense 5-min period, high-intensity
running was reduced by *50% to levels below
the game mean, due to fewer and shorter high-
intensity running bouts.
Figure 6. High-intensity running in each half and during the entire game for various playing positions. *Difference between the first and
second half (P50.05).
#
Different from full-backs, central midfielders, wide midfielders, and attackers.
{
Different from central midfielders
and wide midfielders (P50.05).
D
Different from central defenders, central midfielders, full-backs, and attackers (P50.05). Values are
means and standard deviations.
Figure 7. Mean maximal running speeds for players in various playing positions within 15-min periods and during the entire match.
Different from central defenders: *P50.01;
#
P50.05. Values are means and standard deviations.
High-intensity running in soccer matches 165
Downloaded by [University of Sunderland] at 12:54 30 April 2012
Using a large sample of English FA Premier
League players representing all playing positions,
the results show that high-intensity running and total
distance covered were higher for wide (3138
and 11,535 m) and central midfielders (2825
and 11,450 m) than for full-backs (2605 and
10,710 m), attackers (2341 and 10,314 m), and
central defenders (1834 and 9885 m). Furthermore,
it was observed that wide midfielders, full-backs, and
attackers cover a greater distance when sprinting
(346, 287, and 264 m, respectively) than central
midfielders (204 m) and central defenders (152 m).
These data show that the total distances covered in
the modern elite-standard English League are much
higher than 30 years ago (Reilly & Thomas, 1976),
but also reveal that the amount of high-intensity
running is similar to the Italian Serie A and the
Spanish Primera Division (Di Salvo et al., 2007;
Mohr et al., 2003). The category for high-intensity
running used in the present study (414.4 km h
71
)
can be used for a direct comparison with recent
studies using similar categories (415 km h
71
,
Mohr et al., 2003; 414.1 km h
71
, Di Salvo
et al., 2007). Strikingly similar values for high-
intensity running were obtained for elite-standard
English and Spanish League central defenders (1834
vs. 1869 m), full-backs (2605 vs. 2784 m), wide
midfielders (3138 vs. 3171 m), and central mid-
fielders (2825 vs. 2991 m), whereas the English
League attackers performed less high-intensity run-
ning than their Spanish counterparts (2341 vs.
2708 m). Mean values for high-intensity running of
2430 m in the Italian Serie A (Mohr et al., 2003) are
also close to the mean values obtained in the present
study. Furthermore, the present data suggest that the
amount of high-intensity running is 10–15% higher
in the English FA Premier League than in the Danish
(Mohr et al., 2003) and Swedish Premier Leagues
(Andersson et al., 2007). It should be noted that
different methods of observation and variations in
the classification of movements may account for
differences between studies for high-intensity run-
ning. However, tactical and physical reasons may
explain the increased intensity in the modern English
League and the differences in high-intensity running
performed by the English League attackers com-
pared with attackers in other elite-standard European
leagues. Evidence for increased intensity during
modern English soccer is further supported by a
quantitative comparative analysis of English League
matches played in the 1991–1992 and 1997–1998
seasons (Williams, Lee, & Reilly, 1999). Matches
played in the 1997–1998 season included more
dribbling, passing, running with possession of the
ball, and crosses. The differences in high-intensity
running between the present study and the Danish
(Mohr et al., 2003) and Swedish Premier Leagues
(Andersson et al., 2007) could be attributed to
differences in playing style. In the English game,
players are required to maintain a high level of
activity when not directly involved in play to create
space to receive passes or to pressurise opponents
into making mistakes in order to regain possession.
Furthermore, it has been speculated that the fitness
level of attackers are not sufficient to meet the
demands of elite-standard European leagues. In
support of this notion, attackers performed more
poorly in the game-specific Yo-Yo intermittent
recovery test (IR) 1 and 2 (Krustrup et al., 2003,
2006). In the combined aerobic and anaerobic Yo-
Yo IR2 test, they performed poorer than players in all
other playing positions, including central defenders.
Further studies are required to investigate the
physical fitness of English FA Premier League
attackers and its influence on game performance.
An interesting finding of the present study is that
the maximal running speeds reached during games
were 6–8% higher for wide midfielders and attackers
than for central defenders. Although the reliability of
maximal running speeds were not determined in the
present study, it is reasonable to conclude that large
differences in maximal running speeds are present
between playing positions. For wide midfielders, the
maximal running speed may be related to the fact
that their high-intensity runs are the longest, giving
them more time to reach full acceleration. However,
it is also likely that the attackers and wide midfielders
have the highest running speeds, as shown by 30-
and 40-m sprint tests (Bangsbo, Mohr, & Krustrup,
2006).
Mohr et al. (2003) showed that the amount of
high-intensity running decreased after the most
intense periods (suggesting a temporary form of
fatigue) and decreased markedly towards the end
of the game (suggesting a more permanent form of
fatigue). The number of players examined in this
study was not sufficiently high to allow for differ-
ences in playing positions to be elucidated. The
present data demonstrate that temporary and per-
manent decrements in high-intensity running occur
for players in all playing positions. The amount of
high-intensity running in the most intense period
also varied between playing positions but relatively
less than the full game, with values of 282, 251, 243,
216, and 182 m, respectively, for wide midfielders,
central midfielders, full-backs, attackers, and central
defenders. Although the amount of high-intensity
running in the most intense 5-min period varied
between playing positions, the amount of high-
intensity running decreased by around 50% in the
subsequent 5 min, reaching values that were 6–8%
lower than the game mean. The temporary drop in
high-intensity running may have been underesti-
mated, as our results are based on pre-defined 5-min
166 P. S. Bradley et al.
Downloaded by [University of Sunderland] at 12:54 30 April 2012
periods, meaning that the true temporary drop in
high-intensity running could have been even greater.
These data show the need for a high anaerobic
capacity when a large number of high-intensity runs
have to be performed within a 5-min period. The
amount of high-intensity running in the most intense
period of the game has been suggested to be related
to the physical capacity of the player, as evaluated
in the Yo-Yo IR2 test (Randers, Rostgaard, &
Krustrup, 2007). Thus, the present study provides
clear evidence of a need for recovery after the most
intense periods and physical training aimed at coping
with multiple intense actions. Throughout the game,
the amount of high-intensity running decreased
gradually but the distance deficit for high-intensity
running was most pronounced in the last 15-min
period of the game, being 18–21% lower for the five
playing positions investigated. The decrement in
high-intensity running from the first to the last 15-
min period was somewhat lower than observed by
Mohr et al. (2003). Caution is needed when
interpreting the reason for the different findings
between studies. However, this difference may be
due to the number of matches and players analysed
in each study. Mohr et al. (2003) focused on the
performance of 18 elite-standard players in both
domestic and European games, compared with the
370 elite-standard English League players profiled in
the present study. Therefore, as large inter- and
intra-player variability in movement patterns are
evident from one game to another and within each
position, the present study may provide a better
picture of the average high-intensity running deficit
observed over the duration of the game.
The present study is the first to report mean
recovery times between very high-intensity bouts
within games and across 5-min periods. From the
first to the last 15-min period, recovery times were
28% longer with roughly similar increases among
playing positions. The observation that not only the
total amount of high-intensity running but also the
frequency of high-intensity running bouts decreased
may indicate that fatigue manifests over the course of
the game and therefore affects the style of play.
Players’ increased need for recovery and less
frequent bouts of high-intensity actions must have
impacts on the tactical possibilities of a team and
may indicate that players cannot maintain high-
intensity running in support of team-mates or
covering for them. The high-intensity distance deficit
was roughly similar with and without possession of
the ball, indicating that all parts of the play are
affected by fatigue. It should not be discounted that
these observations could also indicate a reduced pace
of the game, whereby players reduce the amount of
high-intensity running they perform. Despite
the present study controlling for various factors
(standard of opposition, circadian variation, fixture
congestion), caution is necessary when interpreting
the results, as the movement patterns of players
could be influenced by factors not controlled or
measured for (environmental factors, tactical system
employed, home advantage, importance of the game,
and previous results).
In summary, the present results show that the
demands placed on players are high in the modern
elite-standard English League and that temporary
and permanent decrements occur in high-intensity
running. Furthermore, the frequency of high-inten-
sity bouts both with and without ball possession is
affected by fatigue and the activity patterns vary
between playing positions.
Acknowledgements
We would like to thank Phil Tordoff and the
performance analysts at ProZone Sports Ltd
1
for
assistance with the reliability study and Chris Barnes
at Middlesbrough Football Club for suggestions.
The study was financially supported by the Uni-
versity of Sunderland.
References
Andersson, H., Ekblom, B., & Krustrup, P. (2007). Elite football
on artificial turf versus natural grass: Movement pattern,
technical standard and player opinion. Journal of Sports Sciences,
8, 1–10.
Atkinson, G., & Nevill, A. M. (1998). Statistical methods for
assessing measurement error (reliability) in variables relevant to
sports medicine. Sports Medicine,26, 217–238.
Bangsbo, J. (1994). The physiology of soccer – with special
reference to intense intermittent exercise. Acta Physiologica
Scandinavica,151(suppl. 619).
Bangsbo, J., Mohr, M., & Krustrup, P. (2006). Physical and
metabolic demands of training and match-play in the elite
football player. Journal of Sports Sciences,24, 665–674.
Bangsbo, J., Nørregaard, L., & Thorsøe, F. (1991). Activity profile
of competition soccer. Canadian Journal of Sports Sciences,16,
110–116.
Di Salvo, V., Baron, R., Tschan, H., Calderon Montero, F. J.,
Bachl, N., & Pigozzi, F. (2007). Performance characteristics
according to playing position in elite soccer. International
Journal of Sports Medicine,28, 222–227.
Di Salvo, V., Collins, A., McNeill, B., & Cardinale, M. (2006).
Validation of ProZone
1
: A new video-based performance
analysis system. International Journal of Performance Analysis in
Sport,6,108–119.
Drust, B., Reilly, T., & Cable, N. T. (2000). Physiological
responses to laboratory-based soccer-specific intermittent
and continuous exercise. Journal of Sports Sciences,18, 885–
892.
Drust, B., Reilly, T., & Rienzi, E. (1998). Analysis of the work rate
in soccer. Sport, Exercise and Injury,4, 151–155.
Impellizzeri, F. M., Sassi, A., & Rampinini, E. (2006). Accuracy
and reliability of a commercial video-computerised, semi-
automatic soccer-match analysis system: Preliminary results.
In Proceedings of the Eleventh Annual Conference of the European
College of Sport Science, Lausanne, Switzerland (p. 319).
High-intensity running in soccer matches 167
Downloaded by [University of Sunderland] at 12:54 30 April 2012
Krustrup, P., Mohr, M., Amstrup, T., Rysgaard, T., Johansen, J.,
& Steensberg, A. et al. (2003). The yo-yo intermittent recovery
test: Physiological response, reliability, and validity. Medicine
and Science in Sports and Exercise,35, 697–705.
Krustrup, P., Mohr, M., Ellingsgaard, H., & Bangsbo, J. (2005).
Physical demands of elite female soccer games: Importance of
training status. Medicine and Science in Sports and Exercise,37,
1242–1248.
Krustrup, P., Mohr, M., Nybo, L., Jensen, J. M., Nielsen, J. J., &
Bangsbo, J. (2006). The Yo-Yo IR2 test: Physiological
response, reliability and application to elite soccer. Medicine
and Science in Sports and Exercise,38, 1666–1673.
Mayhew, S. R., & Wenger, H. A. (1985). Time–motion analysis of
professional soccer. Journal of Human Movement Studies,11,49–
52.
Mohr, M., Krustrup, P., & Bangsbo, J. (2003). Match performance
of high-standard soccer players with special reference to devel-
opment of fatigue. Journal of Sports Sciences,21, 519–528.
Mohr, M., Krustrup, P., & Bangsbo, J. (2005). Fatigue in soccer:
A brief review. Journal of Sports Sciences,23, 593–599.
Ohashi, J., Togari, H., Isokawa, M., & Suzuki, S. (1988).
Measuring movement speeds and distances covered during
soccer match play. In T. Reilly, A. Lees, K. Davids, & W. J.
Murphy (Eds.), Science and football II (pp. 329–333). London:
E & FN Spon.
Rampinini, E., Coutts, A. J., Castagna, C., Sassi, R., & Impellizzeri,
F. M. (2007). Variations in top level soccer match performance.
International Journal of Sports Medicine,28,1018–1024.
Randers, M. B., Rostgaard, T., & Krustrup, P. (2007). Physical
match performance and Yo-Yo IR2 test results of successful
and unsuccessful football teams in the Danish Premier League.
Journal of Sports Science and Medicine,6, 70–71.
Reilly, T., & Brooks, G. A. (1986). Exercise and the circadian
variation in body temperature measures. International Journal of
Sports Medicine,7, 358–368.
Reilly, T., & Thomas, V. (1976). A motion analysis of work-rate in
different positional roles in in professional football match-play.
Journal of Human Movement Studies,2, 87–89.
Rienzi, E., Drust, B., Reilly, T., Carter, E. L., & Martin, A.
(2000). Investigation of anthropometric and work-rate profiles
of elite South American international soccer players. Journal of
Sports Medicine and Physical Fitness,40, 162–169.
Strudwick, T., & Reilly, T. (2001). Work-rate profiles of
elite Premier League football players. FA Insight Journal,2,
28–29.
Van Gool, D., Van Gerven, D., & Boutmans, J. (1988). The
physiological load imposed on soccer players during real
match-play. In T. Reilly, A. Lees, K. Davids, & W. J. Murphy
(Eds.), Science and football II (pp. 51–59). London: E & FN
Spon.
Williams, A. M., Lee, D., & Reilly, T. (1999). A quantitative
analysis of matches played in the 1991–92 and 1997–98 seasons.
London: The Football Association.
Withers, R. T., Maricic, Z., Wasilewski, S., & Kelly, L. O. (1982).
Match analysis of Australian professional soccer players. Journal
of Human Movement Studies,8, 159–176.
168 P. S. Bradley et al.
Downloaded by [University of Sunderland] at 12:54 30 April 2012
