A new approach to quantifying physical demand in rugby union

Abstract

Abstract The objective of the study was to describe an original approach to assessing individual workload during international rugby union competitions. The difference between positional groups and between the two halves was explored. Sixty-seven files from 30 French international rugby union players were assessed on a computerised player-tracking system (Amisco Pro (®) , Sport Universal Process, Nice, France) during five international games. Each player's action was split up into exercise and recovery periods according to his individual velocity threshold. Exercise-to-recovery (E:R) period ratios and acceleration were calculated. Results indicated that about 65% of exercise periods lasted less than 4 s; half of the E:Rs were less than 1:4, and about one-third ranged between 1 and 1:4 and about 40% of exercise periods were classified as medium intensity. Most acceleration values were less than 3 m·s(-2) and started from standing or walking activity. Back row players showed the highest mean acceleration values over the game (P < 0.05). No significant decrease in physical performance was seen between the first and second halves of the games except for back rows, who showed a significant decrease in mean acceleration (P < 0.05). The analysis of results emphasised the specific activity of back rows and tended to suggest that the players' combinations of action and recovery times were optimal for preventing large decrease in the physical performance.

Full text

This article was downloaded by: [88.175.61.234]
On: 11 October 2013, At: 06:52
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
A new approach to quantifying physical demand in
rugby union
Mathieu Lacome
a
b
, Julien Piscione
a
, Jean-Philippe Hager
a
& Muriel Bourdin
b
a
Research Department , French Rugby Union Federation (FFR) , Marcoussis , France
b
LBMC, IFSTTAR , University of Lyon 1 , Oullins , France
Published online: 09 Sep 2013.
To cite this article: Mathieu Lacome , Julien Piscione , Jean-Philippe Hager & Muriel Bourdin , Journal of Sports
Sciences (2013): A new approach to quantifying physical demand in rugby union, Journal of Sports Sciences, DOI:
10.1080/02640414.2013.823225
To link to this article: http://dx.doi.org/10.1080/02640414.2013.823225
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
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. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

A new approach to quantifying physical demand in rugby union
MATHIEU LACOME
1,2
, JULIEN PISCIONE
1
, JEAN-PHILIPPE HAGER
1
,
& MURIEL BOURDIN
2
1
Research Department, French Rugby Union Federation (FFR), Marcoussis, France, and
2
LBMC, IFSTTAR,
University of Lyon 1, Oullins, France
(Accepted 3 July 2013)
Abstract
The objective of the study was to describe an original approach to assessing individual workload during international rugby
union competitions. The difference between positional groups and between the two halves was explored. Sixty-seven files
from 30 French international rugby union players were assessed on a computerised player-tracking system (Amisco Pro
®
,
Sport Universal Process, Nice, France) during five international games. Each player’s action was split up into exercise and
recovery periods according to his individual velocity threshold. Exercise-to-recovery (E:R) period ratios and acceleration
were calculated. Results indicated that about 65% of exercise periods lasted less than 4 s; half of the E:Rs were less than 1:4,
and about one-third ranged between 1 and 1:4 and about 40% of exercise periods were classified as medium intensity. Most
acceleration values were less than 3 m·s
−2
and started from standing or walking activity. Back row players showed the highest
mean acceleration values over the game (P < 0.05). No significant decrease in physical performance was seen between the
first and second halves of the games except for back rows, who showed a significant decrease in mean acceleration
(P < 0.05). The analysis of results emphasised the specific activity of back rows and tended to suggest that the players’
combinations of action and recovery times were optimal for preventing large decrease in the physical performance.
Keywords: Time–motion analysis, Rugby, Acceleration, Ratio
Introduction
In recent years, video graphic motion analysis has
become a common method of estimating the physi-
cal demands of rugby union, as in soccer. The
player’s displacement can be analysed using two
methods, depending on the video system used. (1)
The notational method defines subjective intensity
zones assessed from the observation of each player’s
movement characteristics (Austin, Gabbett, &
Jenkins, 2011a; Austin, Jenkins, & Gabbett, 2011b;
Deutsch, Kearney, & Rehrer, 2007; Duthie, Pyne, &
Hooper, 2005). (2) Alternative time–motion techni-
ques are now available that use automatic and semi-
automatic (Eaton & George, 2006; Quarrie,
Hopkins, Anthony, & Gill, 2013) or manual
(Roberts, Trewartha, Higgitt, El-Abd, & Stokes,
2008) player-tracking techniques to provide velocity
data over the course of a match; intensity zones can
thus be defined using absolute velocity values. Both
approaches have limitations. First, the choice of a
single absolute velocity threshold for all players
does not take into ac count the player’s individual
physiological characteristic despite the noticeable
heterogeneity of physiological profiles between posi-
tional groups (e.g. Duthie, Pyne, & Hooper, 2003).
Abt and Lovell (2009) highlighted the importance of
individualising the high-intensity threshold: the dis-
tance covered at high intensity by a professional
soccer player was substantially underestimated if
the same absolute running velocity value was used
for all players. Second, the usual method of data
analysis consists in determining the frequency and
duration of activity in different intensity zones.
However, as illustrated by Figure 1 and pointed out
by Cunniffe, Proctor, Baker, and Davies (2009), this
analysis provides frequency of entry per velocity zone
but not the number of real exercise bouts. Since
velocity almost never reached a steady state, exercise
bouts could be counted twice due to acceleration
and deceleration phases. Likewise, each individual
activity was analysed in terms of duration per velo-
city zone, but never analysed globally as such.
Moreover, neither approach accounts for the fact
that the metabolic cost of running at a consta nt
velocity is significantly lower than the cost of accel-
eration (di Prampero et al., 2005; Osgnach, Poser,
Bernardini, Rinaldo, & di Prampero, 2010).
Correspondence: Muriel Bourdin, University of Lyon 1, IFSTTAR, LBMC, Oullins, France. E-mail: muriel.bourdin@univ-lyon1.fr
Journal of Sports Sciences, 2013
http://dx.doi.org/10.1080/02640414.2013.823225
© 2013 Taylor & Francis
Downloaded by [88.175.61.234] at 06:52 11 October 2013

Osgnach et al. (2010) also demonstrated that the
energy cost of acceleration increased as a function
of running velocity. To the best of our knowledge,
only one previous study has considered acceleration
in the analysis of a rugby game, but only 2 players
were analysed (Cunniffe et al., 2009). The analysis
of acceleration data will arguably improve the under-
standing of physical demand during rugby match
play.
The intermittent nature of rugby play was
demonstrated by McLean (1992), who first calcu-
lated the ratio between exerc ise a nd recovery period
durations (E:R) t o quantify exercise density during
the Five Nations matches in 1991. Subsequently,
mean E:R duration was analysed in each positional
group to theori se physiological demand (D eutsch
et al ., 20 07) and training pre scrip tion (Eaton &
George, 2006). However, for a given E:R , physio-
logic al workl oad depe nds on exercise duration
(Åstrand, Åstrand, Christensen, & Hedman, 1960;
Balsom, Seger, Sjodin, & Ekblom, 1992) and, for a
given ratio and exercise duration, on the intensity
of exercise and recovery periods (Åstrand et al.,
1960; Bogdanis, Nevill, Lakomy, Graham, &
Louis , 199 6; Spencer, Bishop, & Dawson, 2006;
Spencer, Dawson, Goodma n, Dascombe, &
Bisho p, 2008).
The primary aim of the present study was to
propose a complementary approach to classic time–
motion analysis during international rugby competi-
tions by analysing the mean intensity of each player’s
activity, acceleration and E:R distribution according
to exercise duration and intensity in different posi-
tional groups.
A secondary aim was to investigate decrease in
physical performance during the course of a match.
Methods
Participants
The expe riment was carried out on 30 French inter-
national rugby pla yers. To make inter-position com-
parisons, players were categ orised as forwards or
backs, and subdivided into 4 positional groups fol-
lowing Deutsch, Maw, Jenkins, and Reaburn (1998):
Front row forwards (props and second rows, n = 9);
back row forwards (hookers and back rows, n = 8);
inside backs (outside halves and centre, n = 6) and
outside backs (wingers and full-backs, n = 7).
Scrum-halves (n = 2) were excluded from the analy-
sis due to the limited sample and unique physical
demands of that position (Deutsch et al., 1998;
Duthie et al., 2005; Roberts et al., 2008 ). The results
of annual physiological testing were used in agree-
ment with the Fédération Française de Rugby (FFR,
French Rugby Federation). Match recordings were
bought by the FFR, which consented to the use of
the data. The study was conducted in accordance
with the guid elines of the ethics committee of the
University of Lyon (France). Anthropometric char-
acteristics of the players are listed in Table I.
Procedures
Physiological tests were conducted in November, in
a single training session.
Five home matches were then analysed a poster-
iori. As can be seen in Table II, games 1 and 2 were
international competitions played in November
while games 3–5 were video-recorded during the
2010 Six Nations Tournament in February and
March. Players participating in at least 1 complete
half were included in the analysis which represented
Figure 1. Typical evolution of the velocity (km·h
−1
) observed during a game. Each horizontal line indicates a different intensity threshold
(MAV: maximal aerobic velocity; MI run.: medium-intensity run).
2 M. Lacome et al.
Downloaded by [88.175.61.234] at 06:52 11 October 2013

67 files. The results, real playing time and number of
files analysed are listed in Table II.
Analysis and calculations
Physiological assessment. Velocity corresponding to a
blood lactate concentration ([La]b) of 4 mmol·l
−1
(VLa4) and maximal aerobic velocity were deter-
mined using an intermittent progressive running test
(adapted from the test described by Leger and
Boucher [1980]) consisting of bouts of 3 min inter-
spersed with 1 min passive rest. Velocity was
increased by 2 km·h
−1
from 8 to 12 km·h
−1
and then
by 1 km·h
−1
until voluntary exhaustion. The test was
performed on a Tartan outdoor track. Players were all
fully familiarised with the test procedure. During the
test, heart rate was continuously recorded using a
heart-rate monitor (Polar Team2, Kempele,
Finland). Blood was sampled from the fingertip, dur-
ing each rest period and 3 min after end of test
.
[La]b
was measured using a lactate analyser (YSI2300, YSI
Inc., Yellow Springs, OH, USA). VLa4 was deter-
mined by straight-line interpolation between the two
closest measured lactate values. Body fat was assessed
on whole-body dual-energy X-ray absorptiometry
(D-XA) (Discovery W, Hologic Inc., Marlborough,
MA, US A).
Video data collection and analysis. A computerised
player-tracking system (Amisco Pro
®
, Sport
Universal Process, Nice, France) was used to record
the activity of each player during the game . Six to
eight cameras were positioned under the roof of the
Stade de France stadium (Saint Denis, France) and
subsequently calibrated and synchronised. The sig-
nals and angles obtained by the encoders were
sequentially converted into digital data at 25 Hz
and recorded on 6 computers for post-game analysis.
Processing of the raw data gave the distance covered
every 0.1 s (10 Hz).
In line with Vigne, Gaudino, Rogowski, Alloatti,
and Hautier (2010), displacement at velocities
higher than VLa4 and static activity were considered
as exercise, and activity lower than VLa4 as recovery.
For each period of exercise corresponding to run-
ning, mean exercise velocity was calculated and sub-
divided into medium-intensity running velocity
(medium-intensity run, ranging between VLa4 and
maximal aerobic velocity) and supramaximal run-
ning velocity (higher than maximal aero bic velocity).
Table I. Anthropometric and physiological characteristics of the 4 positional groups.
Positional group (n)
Forwards Backs
Front row (9) Back row (8) All forwards (17) Inside backs (6) Outside backs (7) All backs (13)
Age (years) 29.9 ± 3.3 27.4 ± 4.1 28.7 ± 3.8 25.9 ± 3.5 27.7 ± 3.4 27.1 ± 3.4
Height (m) 1.88 ± 0.09 1.88 ± 0.05 1.88 ± 0.07 1.82 ± 0.05 1.83 ± 0.06 1.83 ± 0.05
Mass (kg) 112.4 ± 6.7 103.7 ± 7.5
a
108.3 ± 8.2 93.0 ± 14.6
a
90.9 ± 6.4
abc
94.0 ± 8.2*
Body fat (%) 17.5 ± 5.8 13.3 ± 4.7 15.5 ± 5.5 15.3 ± 5.8 11.4 ± 1.8 13.5 ± 4.8
MAV (km·h
−1
) 14.9 ± 0.5 16.1 ± 1.3 15.5 ± 0.8 15.4 ± 0.8 15.7 ± 0.9 15.4 ± 1.1
VLa4 (km·h
−1
) 12.2 ± 0.9 12.5 ± 1.4 12.3 ± 1.1 12.3 ± 0.8 13.2 ± 1.1 12.7 ± 0.9
VLa4 (% MAV) 82.0 ± 5.6 77.5 ± 4.9
d
79.9 ± 5.7 80.0 ± 3.6 83.5 ± 5.1 81.9 ± 4.7
Notes:
a
Significantly different to front row forwards, P < 0.05;
b
Significantly different to back row forwards, P < 0.05;
c
Significantly different
to inside backs, P < 0.05;
d
Significantly different to outside backs, P < 0.05. *Significant between group (forwards and backs) differences,
P < 0.05. MAV: maximal aerobic velocity; VLa4: velocity corresponding to blood lactate concentration of 4 mmol·l
−1
.
Data are presented as mean ± standard deviation; (n) number of players.
Table II. Result, real playing time and number of files analysed for each of the 5 matches.
Game Result Real playing time (min)
Number of files analysed
Forwards Backs
Front row Back row Inside backs Outside backs
Total files
1 Won 20–13 38.7 4 4 2 3 13
2 Lost 12–39 36.8 4 4 3 3 14
3 Won 33–10 44.5 4 4 3 3 14
4 Won 46–20 37.7 4 4 3 2 13
5 Won 12–10 44.8 4 4 3 2 13
Total files 20 20 14 13 67
Physical demand in international rugby union 3
Downloaded by [88.175.61.234] at 06:52 11 October 2013

Total distance, duration and mean intensity in the
various previously defined intensity zones were
determined automatically on computerised match
analysis software (Amisco Viewer
®
, Sport Universal
Process, Nice, France). The study of Zu billaga
(2006) compared the distances and velocities
obtained with the Amisco Pro
®
system to that calcu-
lated from time performances over known distances.
It was concluded by the author that the Amisco Pro
®
system demonstrated high level of accuracy and
reliability.
Scrums, rucks and mauls were classified as “static
activity”. Since the system used was unable to iden-
tify and quantify static activities accu rately and auto-
matically, active participation in a ruck or maul was
monitored manually from the moment a player’s
shoulder entered into contact with the ruck or maul
to the moment it broke contact. Tackles whereby the
ball carrier was stopped on their feet were also clas-
sified as “static activity”. Scrums, rucks and mauls
were identified from match videos and listed by an
experienced opera tor. The reliability of a similar
time–motion analysis method has previously been
documented (Deutsch et al., 1998, 2007). The
time spent and distance travelled during static activ-
ities was excluded from the analysis done automati-
cally by Amisco Viewer
®
.
In order to establish a representative profile for
each positional group, all 5 game s were pooled for
analysis (Hughes, Evans, & Wells, 2001).
Calculations. For each player data file , E:R was
calculated by dividing exercise period duration by
the following recovery duration. The results were
related to the total number of E:R ratios (i.e.
exercise + recovery events) and expressed in per-
centages (%). Ratios were divided in to 3 groups:
lower than 1:4 (E:R < 1 :4), ranging betwee n 1:4
and 1 (1:4 < E:R < 1) and higher than 1 (E:R > 1)
and expressed as percentages (%) of the total
number of ratios. For each group, E:Rs were clas-
sified according to exercise duration and e xercise
intensity.
Acceleration values were calculated from distance
data sampled at 10 Hz and derived twice. A
Butterworth low-pass second-order (cut-off fre-
quency: 1 Hz) double phase-lag filter was used
after each derivation. Accord ing to the manufac-
turer, the mean error for a pixel was 75 mm (ranging
between 50 and 100 mm according to the position of
the player relative to the cameras). To test the accu-
racy and the reliability of our calculation, we gener-
ate 30 s noisy displacement data sets (sampled at
10 Hz) using the Monte Carlo method and consid-
ering a standard deviation of 75 mm. We calculated
the standard deviation of the noisy data sets to assess
the accuracy and the reliability of the calculation of
both velocity and acceleration values and to verify
the convergence of the standard deviati on computed
from these noisy sets regarding the number of sets
(between 10 and 500 30 s noisy sets). Standard
deviation was assessed 0.17 m·s
−1
and 0.34 m·s
−2
for velocity and acceleration, respectively, whatever
the number of noisy sets.
Only acceleration values reaching at least 1 m·s
−2
for a duration greater than 0.5 s were taken into
account in the analysis. In accordance with Osgnach
et al. (2010), data were pooled in 3 groups: 1–2, 2–3
and greater than 3 m·s
−2
. Acceleration values were
then classified into the following activity categories
according to mean exercise-period velocity: standing–
walking (0–7 km·h
−1
), jogging (7 km·h
−1
– VLa4),
medium-intensity run and higher than maximal aero-
bic velocity.
To analys e changes in activity patterns over the
whole match, only players participating in a com-
plete match were analysed. This represented 37
files (front rows, 9 files; back rows, 10 files; inside
backs, 9 files and outside backs, 9 files). The study
parameters were averaged per match and per half.
Statistical analysis
All data are presented as mean ± standard deviation.
The study parameters studied were not normally
distributed on the Shapiro–Wilk test, and non-
parametric tests were therefore used. Differences
between groups and between the first and second
halves of the matches were tested using the
Wilcoxon test. Statistical significance was set at
P < 0.05. All statistics were run on JMP Pro 9
(SAS, Cary, NC, USA).
Results
Anthropometric and physiological characteristics of the 4
positional groups
The results demonstrated that players in the 4 posi-
tional groups had similar heights. On the other hand,
body mass was greater in forwards than in backs.
Front rows were the heaviest and outside backs the
lightest. There was no significant difference between
positional groups relative to body fat, VLa4, max-
imal aerobic velocity or age. Vla4 related to the
maximal aerobic velocity (VLa4%) was significantly
lower in back rows than in outside backs (P < 0.05).
Total exercise time and distance covered
Front rows and back rows showed similar mean total
exercise times (11.9 ± 1.3 and 11.2 ± 2.6 min,
respectively). Inside backs showed a mean total exer-
cise time of 10 ± 1.4 min, similar to back rows,
4 M. Lacome et al.
Downloaded by [88.175.61.234] at 06:52 11 October 2013

shorter than front rows (P < 0.05) and longer than
outside backs (7.1 ± 1.3 min, P < 0.05).
Backs covered more distance than forwards
(7,944 ± 659 vs 7,006 ± 356 m, respectively;
P < 0.05). Inside backs and outside backs covered
similar distances. Front rows covered the shortest
distance (6,935 ± 334). Back rows and outside
backs covered similar distances (7,215 ± 266 and
7,764 ± 789, respectively), back rows covered
shorter distance than inside backs (8,079 ± 539;
P < 0.05);
Exercise recovery ratio
The total number of ratios (exercise + recovery
events) per match was significantly lower in backs
than in forwards (137.1 ± 28.8 vs 178.7 ± 25.7,
respectively; P < 0.05). The total number of ratios
in front rows was similar to that in back rows
(185.9 ± 20.3 and 172.2 ± 29.3, respectively) and
significantly higher in inside backs than in outside
backs (155.4 ± 13.9 and 118 ± 16.1 respectively,
P < 0.05). Inside backs showed a lower total number
of ratios than front rows (P < 0.05); outside backs
also showed a lower total number of ratios than back
rows (P < 0.05).
To facilitate the reading of Tabl e III , data hig her
than 5% are colour ed in grey. Overall, as illustrated
in Table III and synthesised in Ta ble IV , forwards
were more involve d in static activity and backs were
more involved in supramaximal intensity bouts. No
significant differe nce was found for medium-
intensity runs. All positional groups had si milar
percentages of E:Rs in per ex ercise duration
category.
Back rows had significantly more E:R >1 ratios
than front rows or outside backs; outside backs had
the lowest percentage. Outside backs had signifi-
cantly more E:R <1:4 ratios than front rows, back
rows or inside backs.
It can be seen that, for exercise durat ions less
than 4 s, whatever the type of exercise, a large
part of the action showed a E:R <1:4. For exercise
durations >4 s and in static activity, forwards
showed a prepon derance of E:R <1. For exercise
intensities greater than maximal aerobic velocity,
backs showed a large percentage of exercise dura-
tions >4 s, r egardless of t he ratio. In the medium-
intensity run category, forwards showed a lower
percentage of E:R <1:4 than backs for exercise
durations 2–4s.
Acceleration
Mean acceleration duration was significantly higher
in backs than in forwards (0.85 ± 0.06 vs
0.76 ± 0.03 s, respectively; P < 0.05), as was mean
maximal acceleration duration (3.67 ± 0.35 vs
3.24 ± 0.39 s, respectively; P < 0.05).
Mean acceleration value was significantly higher in
forwards than in backs (2.46 ± 0.92 vs
2.36 ± 0.93 m.s
−2
, respectively). Back rows had the
highest mean acceleration value, compared to front
rows, inside backs and outside backs (2.50 ± 0.95,
2.41 ± 0.89, 2.38 ± 0.90 and 2.34 ± 0.98 m·s
−2
,
respectively; P < 0.001). Inside backs and outside
backs both had significantly lower mean acceleration
values than front rows (P < 0.05), while outside
backs had significantly lower mean values than inside
backs (P < 0.05).
Overall analysis of results, illustrated in Figure 2,
demonstrated that the majority of match accelera-
tion values (41.2 ± 7.6%) r anged between 1 and
2m·s
−2
.Thepercentageofaccelerationvalues
of the category “acceleration ranging between 2
and 3 m·s
−2
” and of the category “accelera-
tion >3 m·s
−2
” represented 37.7 ± 3.9 and
21.1 ± 6% of the total number of accelerations,
respectively. A cceleration started mainly from jog-
ging (31.8 ± 5%) or standing–walking activity
categories (53.4 ± 5.5%); the percentage was sig-
nificantly higher in front rows and outside backs
than in inside backs (86.5 ± 3.0% and 85.7 ± 4.7%
vs 82.7 ± 3.1%, respectively; P < 0.05) but no
significant difference emerged between forwa rds
and backs as a whole.
As can be seen in Figure 2, back rows had a lower
percentage of acceleration values ranging between 1
and 2 m.s
−2
than the other subgroups, with a higher
percentage of acceleration values >3 m.s
−2
than
inside backs and outside backs in the medium-inten-
sity run, jogging and walking–standing zones. Front
rows had the lowest percentage of acceleration values
from running velocities higher than maximal aerobic
velocity.
First Half vs Second Half
As seen in Table V, whatever the positional group,
no significant differences in the percentage of exer-
cise-period type or intensity, exercise duration or E:
R were observed between the two halves, except that
the percentage of ratios ranging between 1:4 and 1
decreased significantly between the two halves of the
match for backs.
Mean acceleration, on the other hand, was sig-
nificantly greater during the first half than the sec-
ond half (2.45 ± 0.95 vs 2.38 ± 0.90 m·s
−2
,
respectively; P <0.001),inbothforwardsand
backs (2.49 ± 0.93 vs 2.42 ± 0.91 m·s
−2
,
and 2.39 ± 0.97 vs 2.33 ± 0.89 m·s
−2
,respectively).
Analysis by subgroup found that mean a cceleration
Physical demand in international rugby union 5
Downloaded by [88.175.61.234] at 06:52 11 October 2013

Table III. Distribution of the ratio of exercise duration to recovery duration (E:R; % of the total number of ratios) classified by exercise duration for each group and subgroup and for each exercise
intensity.
Exercise duration <2 s 2–4 s >4 s
Ratio E:R < 1:4 1:4 < E:R < 1 E:R > 1 E:R <1:4 1:4 < E:R < 1 E:R > 1 E:R <1:4 1:4 < E:R < 1 E:R > 1
Static activity
Front row 6.2 ± 3.2 1.5 ± 1.1 0.1 ± 0.3 8.7 ± 2.7 3.0 ± 1.4 1.0 ± 1.1 5.5 ± 1.5 4.0 ± 1.8 3.3 ± 2.0
Back row 5.9 ± 2.9 1.5 ± 1.2 0.4 ± 0.5 7.6 ± 3.0 2.8 ± 1.3 1.4 ± 1.0 5.0 ± 2.1 2.9 ± 1.7 4.5 ± 3.1
All forwards 6.0 ± 3.0 1.5 ± 1.2 0.3 ± 0.4 8.2 ± 2.9 2.9 ± 1.3 1.2 ± 1.1 5.2 ± 1.8 3.4 ± 1.8 3.9 ± 2.7
Inside backs 3.4 ± 1.6
ab
0.7 ± 0.6
a
0.3 ± 0.5 2.3 ± 1.7
ab
1.3 ± 1.0
ab
0.4 ± 0.6
b
0.2 ± 0.5
ab
0.4 ± 0.4
ab
0.1 ± 0.3
ab
Outside backs 3.4 ± 2.4
ab
0.5 ± 0.5
ab
0.3 ± 0.5 1.9 ± 1.1
ab
0.2 ± 0.5
abc
0.5 ± 0.6
b
0.7 ± 0.7
ab
0.2 ± 0.5
ab
0.1 ± 0.2
ab
All backs 3.4 ± 2.0* 0.6 ± 0.6* 0.3 ± 0.5 2.1 ± 1.4* 0.8 ± 0.9* 0.4 ± 0.6* 0.4 ± 0.6* 0.3 ± 0.5* 0.1 ± 0.3*
>MAV Front row 3.9 ± 2.3 1.7 ± 1.4 1.3 ± 1.2 3.0 ± 2.2
cd
1.7 ± 1.0
cd
1.9 ± 1.3
c
4.0 ± 2.6
cd
4.6 ± 1.6
cd
5.2 ± 2.1
cd
Back row 2.6 ± 2.5
ac
1.1 ± 0.9
c
0.7 ± 0.9 2.6 ± 1.4
cd
1.4 ± 1.0
cd
2.2 ± 1.2
c
4.1 ± 2.8
cd
4.0 ± 2.0
cd
5.4 ± 2.6
cd
All forwards 3.2 ± 2.4* 1.4 ± 1.2* 1.0 ± 1.1 2.8 ± 1.8* 1.6 ± 1.0* 2.1 ± 1.2* 4.1 ± 2.7* 4.3 ± 1.8* 5.3 ± 2.4*
Inside backs 4.8 ± 1.9 2.3 ± 1.2 0.7 ± 0.7 6.4 ± 2.6
d
3.3 ± 1.6 3.6 ± 1.1 9.7 ± 3.2
d
8.2 ± 3.2 10.2 ± 4.7
Outside backs 5.7 ± 2.8 2.5 ± 2.1 1.4 ± 2.3 10.7 ± 5.1 4.3 ± 2.6 2.9 ± 1.4 12.6 ± 3.8 7.4 ± 3.2 8.0 ± 2.7
All backs 5.2 ± 2.4 2.4 ± 1.7 1.0 ± 1.7 8.5 ± 4.5 3.8 ± 2.2 3.3 ± 1.3 11.1 ± 3.8 7.8 ± 3.2 9.1 ± 4.0
Medium-intensity run Front row 8.8 ± 3.7
c
3.8 ± 2.2 1.7 ± 1.1 7.3 ± 3.7 3.7 ± 2.0 2.5 ± 1.5 4.1 ± 1.8 4.0 ± 2.4 3.6 ± 2.3
Back row 10.4 ± 3.7 4.5 ± 2.5 1.7 ± 1.5 6.3 ± 1.7
cd
4.3 ± 1.8 3.3 ± 2.2 4.6 ± 2.7 4.3 ± 2.1 4.6 ± 2.8
All forwards 9.6 ± 3.8 4.1 ± 2.3 1.7 ± 1.3 6.8 ± 2.9* 4.0 ± 1.9 2.9 ± 1.9 4.4 ± 2.3 4.1 ± 2.2 4.1 ± 2.6
Inside backs 10.9 ± 2.2 3.6 ± 2.5 1.3 ± 1.1 8.3 ± 2.1 4.1 ± 2.1 2.5 ± 1.6 3.1 ± 1.3
b
4.9 ± 2.1 3.2 ± 2.7
Outside backs 10.7 ± 3.6 2.7 ± 1.9
b
0.7 ± 1.0
ab
9.5 ± 3.4 3.4 ± 2.3 1.6 ± 2.1
b
3.9 ± 3.1 2.7 ± 2.3
bc
1.7 ± 1.9
ab
All backs 10.8 ± 2.9 3.2 ± 2.3 1.0 ± 1.1* 8.8 ± 2.8 3.8 ± 2.2 2.1 ± 1.8 3.5 ± 2.3 3.9 ± 2.4 2.5 ± 2.4*
Notes:
a
Significantly different to front row forwards, P < 0.05;
b
Significantly different to back row forwards, P < 0.05;
c
Significantly different to inside backs, P < 0.05;
d
Significantly different to
outside backs, P < 0.05. *Significant between group (forward and back) differences, P <0.05.Greycells,percentageofE:Rhigherthan5%.MAV:maximalaerobicvelocity.
6 M. Lacome et al.
Downloaded by [88.175.61.234] at 06:52 11 October 2013

Table IV. Percentage of exercise periods according to exercise intensity, exercise duration and ratio of exercise duration to recovery duration (E:R) for each group and subgroup.
Exercise intensity Exercise duration Ratio
Medium-intensity run >MAV Static activity 0–2s 2–4 s >4s E:R < 1:4 1:4 < E:R < 1 E:R > 1
Front row (20) 39.5 ± 12.2 27.2 ± 9.2
cd
33.3 ± 5.6 29.0 ± 4.0 32.8 ± 4.4 38.2 ± 4.2 51.4 ± 6.2
d
27.9 ± 4.8 20.6 ± 4.4
b
Back row (20) 44.0 ± 14.4 24.1 ± 9.0
cd
31.8 ± 7.8 28.7 ± 4.8 32.0 ± 2.8 39.3 ± 4.4 49.1 ± 5.4
d
26.7 ± 5.6 24.2 ± 4.4
Forwards (40) 41.8 ± 13.4 25.7 ± 9.2* 32.6 ± 6.8 28.8 ± 4.4 32.4 ± 3.7 38.8 ± 4.3 50.3 ± 5.9 27.3 ± 5.1 22.4 ± 4.7
Inside back (14) 41.9 ± 7.2 49.0 ± 7.8 9.1 ± 3.1
ab
27.9 ± 4.1 32.1 ± 4.1 40.0 ± 6.4 48.9 ± 5.0
d
28.8 ± 4.0 22.3 ± 6.3
Outside back (13) 37.0 ± 12.2 55.4 ± 13.0 7.6 ± 3.4
ab
27.8 ± 4.6 35.0 ± 6.0 37.2 ± 4.9 59.0 ± 6.6 23.9 ± 7.2 17.1 ± 4.3
abc
Backs (27) 39.6 ± 10.1 52.1 ± 10.9 8.4 ± 3.2* 27.8 ± 4.3 33.5 ± 5.2 38.7 ± 5.8 53.8 ± 7.7 26.4 ± 6.1 19.8 ± 5.9
Notes:
a
Significantly different to front row forwards, P < 0.05;
b
Significantly different to back row forwards, P < 0.05;
c
Significantly different to inside backs, P < 0.05;
d
Significantly different to
outside backs, P < 0.05. *Significant between group (forwards and backs) differences, P < 0.05. MAV: maximal aerobic velocity. In parenthesis, the number of files analysed.
Physical demand in international rugby union 7
Downloaded by [88.175.61.234] at 06:52 11 October 2013

values decreased significantly only for back rows
(2.55 ± 0.96 vs 2.45 ± 0.94; P < 0.001), whereas
only a trend was observed for the o ther subgroups.
Discussion
The purpose of the current study was to propose a
complementary approach to classic time–motion ana-
lysis to improve the understanding of the phys ical
demand of international rugby union match play.
Results indicated that about 65% of exercise periods
lasted less than 4 s; half of the E:Rs were less than 1:4
and about one-third ranged between 1 and 1:4; and
about 40% of exercise periods were classified as med-
ium intensity. Most acceleration values were less than
3 m·s
−2
and started from standing or walking activity.
Back rows showed the highest mean acceleration
values over the game. Analysis in subgroups demon-
strated that only back rows showed a significant
decrease in the mean acceleration between the two
halves. The E:Rs and their repartition in duration and
intensity categories did not vary over the game.
The anthropometric characteristics of the studied
group were in line with those presented in previous
studies concerning elite professional rugby union
(Fuller, Taylor, Brooks, & Kemp, 2013; Higham,
Pyne, & Mitchell, 2013; Holway & Garavaglia,
2009; Sedeaud et al., 2012). To our knowledge,
the current study is the first to present mean values
of maximal aerobic velocity and of VLa4% in a
group of international rugby union players.
The methodological originality of t he present
study was the choice of individual v elocity thresh-
old. In a recent study by C ahill, Lamb, Worsfold,
Headey, and Murray (2013), i ndividual thresholds
corresponding to maximal running velocity were
used to quantify the movement characteristics of
English premiership players. It is generally
accepted that fatigue occurs quickly during supra-
maximal e xercise intensities (i.e . greater than max-
imal aerobic velocity). Accordingly, the choice of
maximal aerobic velocity as thre shold to de fine
high intensity seems justified, espec ially for asses-
sing physic al demand. I n t he present s tud y,
Front row forwards; Back row forwards; Inside backs; Outside backs
Figure 2. Distribution of acceleration (percentage of total number of acceleration values) according to intensity category (67 files).
a Significantly different to front row forwards, P < 0.05; b Significantly different to back row forwards, P < 0.05; c Significantly different to
inside backs, P < 0.05; d Significantly different to outside backs, P < 0.05. MAV: maximal aerobic velocity. MI run: medium-intensity
running.
Table V. Evolution of the percentage of exercise periods according to exercise intensity, exercise duration and the ratio of exercise duration
to recovery duration (E:R) between the two halves (H1 and H2).
Exercise intensity Exercise duration Ratio
Medium-
intensity run >MAV
Static
activity 0–2s 2–4s >4s E:R < 1:4
1:4 < E:
R < 1 E:R > 1
Forwards (19) H1 36.1 ± 11.6 28.0 ± 8.3 35.9 ± 6.3 28.8 ± 6.1 32.6 ± 4.3 38.6 ± 5.2 51.1 ± 7.6 26.4 ± 5.8 22.4 ± 5.0
H2 36.7 ± 10.2 27.6 ± 10.2 35.8 ± 3.5 29.1 ± 4.8 31.8 ± 3.4 39.1 ± 6.0 51.3 ± 5.8 24.6 ± 4.5 24.0 ± 6.3
Backs (18)
H1 42.0 ± 10.3 49.4 ± 10.7 8.6 ± 3.4 27.0 ± 4.8 36.0 ± 5.1 37.0 ± 6.2 51.7 ± 8.1 28.5 ± 6.4 19.8 ± 6.6
H2 43.1 ± 11.4 48.3 ± 11.3 8.5 ± 5.0 28.7 ± 6.9 34.8 ± 7.0 36.5 ± 8.0 54.4 ± 7.9 23.8 ± 6.0* 21.9 ± 5.9
Notes: *Significantly different to H1, P < 0.05. MAV: maximal aerobic velocity. In parenthesis, the number of files analysed.
8 M. Lacome et al.
Downloaded by [88.175.61.234] at 06:52 11 October 2013

maximal aerobic ve locity tended to be lower in
front rows. This confirms that by using a single
velocity threshold to define high intensity tends to
systematically underestimate physical demand in
players with low er maximal aerobic ve locity values
(Abt & Lovell, 2009). It can be assumed that pre-
vious studies, which used a fixed-int e n si t y th r e sh -
old, to define high-intens ity category, did not
dispose of individual reference values such as max-
imal aerobic velocity.
In most studies, active participation in scrums,
rucks and mauls is timed from when the player has
any contact with another player, whereas in the cur-
rent study, participation was only considered to be
when specifically the player’s shoulder was in con-
tact. Consequently, the mean duration of static
activity in the present study (4.15 and 2.21 s for
forwards and backs, respectively) was lower than in
other reports (5 to 7.1 and 3.6 to 4.2 s, respectively)
(Deutsch et al., 2007; Duthie et al., 2005; Roberts
et al., 2008). The mean number of static activity
bouts for forwards (about 53, including 22 scrums)
was significantly less than in previous studies (ran-
ging between 80 and 105) (Deutsch et al., 2007;
Duthie et al., 2005; Roberts et al., 2008). It is not
possible to determine whether these differences are
due to the methodology and/or to the playing level
and style of the teams studied. Nevertheless, the
methods used in different studies cannot influence
the total number of scrums. The current results
demonstrated a mean value of 22 scrums per
match. This is consistent with the value of around
25 scrums per match reported by Quarrie et al.
(2013) during south hemisphere international game.
In agreement with the studies of Austin et al.
(2011b), total exercise time was similar for front
rows (11.9 min), back rows (11.2 min) and inside
backs (10.4 min), but significantly shorter for out-
side backs (7.1 min). Consistent with this, the total
number of ratios (exercise + recovery events) was the
lowest in outside backs (118). Table IV shows that
forwards were more involved in static activity than
backs, while backs were more involved in supramax-
imal runs than forwards. Forwards (7,006 m) were
found to cover a shorter total distance than backs
(7,944 m). This is related to the fact that front rows
covered the shortest total distance (6,935 m). This
finding is in line with those of Austin et al. (2011b)
and Roberts et al. (2008). It is noteworthy that, in
agreement with Austin et al. (2011b), back rows and
outside backs covered similar total distances whereas
inside backs tended to cover the longest. The shorter
distance covered by front rows is consistent with
their lower maximal aerobic velocity. Further studies
are needed to investigate whether maximal aerobic
velocity value could represent the limitation to total
distance covered.
The players’ competitive level and/or the metho-
dological approach of the present study differs from
previous studies (Austin et al., 2011b; Deutsch et al.,
2007; Duthie et al., 2005; Quarrie et al., 2013;
Roberts et al., 2008), making comparison difficult.
However, it can be concluded that the overall char-
acteristics and between-group differences observed
in the present study were similar to those observed
in elite southern hemisphere and English rugby
union teams.
Nevertheless, total distance covered gives no
information on intensity of displacement, which is
what determines metabolic workload. During inter-
mittent activity, the exercise duration in different
velocity zones or analysis of mean velocity gives par-
tial information on metabolic workload: the accel-
eration should be taken into account in order to
better estimate metabolic workload.
Figure 2 shows that most acceleration values were
less than 3 m·s
−2
. In line with the present results,
Cunniffe et al. (2009) reported that most accelera-
tion values attained in 1 s ranged between 1.5 and
2.5 m·s
−2
. The present results also demonstrated
that the mean duration of accele ration phases
exceeding 1 m.s
−2
was less than 1 s, with a mean
maximal duration ranging between 3 and 4 s. Mean
acceleration value was 4.2% greater in forwards than
in backs, and back rows showed the greatest mean
acceleration over the match as a whole. The studies
of Crewther, Lowe, Weatherby, Gill, and Keogh
(2009) reported that the ability to accele rate over
short distance (10 and 20 m) is greater in backs
than in forwards. It could be supposed that the
higher acceleration value observed in forwards does
not reflect a better ability to accelerate but rather
reflects the game demands of this position.
Consistent with the finding of Cahill et al. (2013)
and the normal game demands, the present results
also underline the high-intensity workload in back
rows compared to front rows, inside backs or outside
backs.
According to Osgnach et al. (2010), metabolic
power is influenced by both initial running velocity
and acceleration: for a given acceleration, the higher
the initial running velocity the higher the metabolic
power. As illustrated by Figure 2, most acceleration
values started at intensities corresponding to recov-
ery. It is interesting to note that forwards, and parti-
cularly back rows, were more involved in acceleration
values exceeding 3 m·s
−2
, while inside backs were
more involved in acceleration values corresponding
to a medium- or high-intensity run. Consistent with
the finding of Duthie, Pyne, Marsh, and Hooper
(2006), in forwards, most acceleration values have
started when they were standing or walking. This
may due to forwards having to produce higher accel-
eration to attain the same running velocity than backs
Physical demand in international rugby union 9
Downloaded by [88.175.61.234] at 06:52 11 October 2013

involved in acceleration values corresponding to a
medium- or high-intensity run.
As discussed earlier, intermittent activity could be
characterised by the analysis of the distribution of E:
Rs according to exercise durations and intensity cate-
gories. The mean ratio (calculated by dividing mean
exercise duration by mean recovery duration) was
1:6.5 and 1:8.5 in forwards and backs respectively.
This is consistent with the range of 1:4 and 1:25
observed in previous studies (Austin et al., 2011b;
Deutsch et al., 2007; Duthie et al., 2005; Eaton &
George, 2006). An overall analysis of the data shown
in Table IV demonstrates that: (1) in agreement with
Duthie et al. (2005), about 65% of exerc ise periods
lasted less than 4 s; (2) half the E:Rs were less than 1:4
and about one-third ranged between 1 and 1:4; and
(3) about 40% of exercise periods were classified as of
medium intensity. If one considers ratio distribution
and the fact that short-duration high-intensity (near
maximal) phases of static activity or sprints occurred
evenly and were interspersed by exercise and recovery
periods varying in intensity and duration, a decreas e
in the performance due to fatigue is not obvious. In
line with previous studies (Cunniffe et al., 2009 ;
Duthie et al., 2005; Roberts et al., 2008), analysis of
the evolution of ind ividual activity between halves
failed to demonstrate a significant decrease in the
performance when the characteristics of individual
exercise periods were considered. Nevertheless, a
slight but significant decrease of 2.9% in the mean
acceleration was found during the second half, which
might be attributed to fatigue and/or to a strategic
decrease in activity level linked to scores (Lago,
Casais, & Dominguez, 2010). Mean acceleration
decreased (by 2.8% and 2.5%, respectively) in both
forwards and backs during the second half, but the
decrease was significant only in back rows, which may
be related to the significantly higher percentage of E:
Rs >1 and higher percentage of acceleration values
exceeding 3 m·s
−2
in back rows than in the other
subgroups. This decrease in acceleration could also
be related to significantly lower values of Vla4% in
back rows, a parameter generally associated with
endurance quality.
This result also underlines the interest of analysing
acceleration in intermittent activity. Nevertheless, the
accuracy and reliability of acceleration calculated from
a double derivative of displacement sampled at 10 Hz
and the small number of players within each subgroup
call for caution in generalising the results. Further
studies using new technologies (GPS and accelerome-
try) are needed to confirm the present findings.
Conclusion
To the best of our knowledge, this is the first study to
analyse the physical demands of elite rugby union
during five complete international matches using
semi-automated video analysis, and to consider
mean intensity during each individual exercise per-
iod. Despite methodological limitations, the analysis
of results demonstrates the interest of analysing accel-
eration and E:R ratios to estimate physical demand.
This approach enabled us to show that back rows
were the players with the highest acceleration level
but tended to decrease it during the second half.
These current results tended to show that the combi-
nation of action and recovery times was optimal to
prevent any serious decrease in performance.
Consistent with the earlier studies, the results demon-
strated that international rugby matches place specific
physical demands on players occupying the various
positional groups. This has practical implications.
First, conditioning programmes could differ from a
qualitative point of view between positional sub-
groups: for example, forwards and particularly back
rows should improve their ability to accelerate and to
repeat acceleration. The distribution of ratios pre-
sented in the current study could also be helpful in
designing specific intermittent exercises. Second, the
current results suggested that substitutions of players
during a match should mainly concerned forwards
and particularly back rows.
Acknowledgements
We are very grateful to Paola Sibi, Julien Robineau,
Julien Deloire, Nicolas Barizien and Hugo
Maciejewski for their precious help in field data col-
lection. We also gratefully thank Professor J.R.
Lacour and the reviewers for their constructive com-
ments that have significantly improve the manuscript.
References
Abt, G., & Lovell, R. (2009). The use of individualized speed and
intensity thresholds for determining the distance run at high-inten-
sity in professional soccer. Journal of Sports Sciences, 27(9), 893–898.
Åstrand, I., Åstrand, P. O., Christensen, E. H., & Hedman, R.
(1960). Intermittent muscular work. Acta Physiologica
Scandinavica, 48, 448–453.
Austin, D., Gabbett, T., & Jenkins, D. (2011a). Repeated high-
intensity exercise in professional rugby union. Journal of Sports
Sciences, 29(10), 1105–1112.
Austin, D., Jenkins, D., & Gabbett, T. (2011b). The physical
demands of Super 14 rugby union. Journal of Science and
Medicine in Sport, 14(3), 259–263.
Balsom, P. D., Seger, J. Y., Sjodin, B., & Ekblom, B. (1992).
Physiological responses to maximal intensity intermittent exer-
cise. European Journal of Applied Physiology, 65(2), 144–149.
Bogdanis, G. C., Nevill, M. E., Lakomy, H. K., Graham, C. M.,
& Louis, G. (1996). Effects of active recovery on power output
during repeated maximal sprint cycling. European Journal of
Applied Physiology, 74(5), 461–469.
Cahill, N., Lamb, K., Worsfold, P., Headey, R., & Murray, S.
(2013). The movement characteristics of English Premiership
rugby union players. Journal of Sports Sciences, 31(3), 229–237.
10 M. Lacome et al.
Downloaded by [88.175.61.234] at 06:52 11 October 2013

Crewther, B. T., Lowe, T., Weatherby, R. P., Gill, N., & Keogh,
J. (2009). Neuromuscular performance of elite rugby union
players and relationships with salivary hormones. Journal of
Strength and Condiotioning Research, 23, 2046–2053.
Cunniffe, B., Proctor, W., Baker, J. S., & Davies, B. (2009). An
evaluation of the physiological demands of elite rugby union
using Global Positioning System tracking software. Journal of
Strength and Conditioning Research, 23(4), 1195–1203.
Deutsch, M. U., Kearney, G. A., & Rehrer, N. J. (2007). Time-
motion analysis of professional rugby union players during
match-play. Journal of Sports Sciences, 25(4), 461–472.
Deutsch, M. U., Maw, G. J., Jenkins, D., & Reaburn, P. (1998).
Heart rate, blood lactate and kinematic data of elite colts
(under-19) rugby union players during competition. Journal of
Sports Sciences, 16(6), 561–570.
di Prampero, P. E., Fusi, S., Sepulcri, L., Morin, J. B., Belli, A., &
Antonutto, G. (2005). Sprint running: A new energetic approach.
Journal of Experimental Biology, 208(Pt 14), 2809–2816.
Duthie, G., Pyne, D., & Hooper, S. (2003). Applied physiology
and game analysis of rugby union. Sports Medicine, 33(13),
973–991.
Duthie, G., Pyne, D., & Hooper, S. (2005). Time motion analysis
of 2001 and 2002 Super 12 rugby. Journal of Sports Sciences, 23
(5), 523–530.
Duthie, G. M., Pyne, D. B., Marsh, D. J., & Hooper, S. L. (2006).
Sprint patterns in rugby union players during competition.
Journal of Strength and Conditioning Research, 20(1), 208–214.
Eaton, C., & George, K. (2006). Position specific rehabilitation
for rugby union players. Part I: Empirical movement analysis
data. Physical Therapy in Sport, 7(1), 22–29.
Fuller, C. W., Taylor, A. E., Brooks, J. H. M., & Kemp, S. P. T.
(2013). Changes in the stature, body mass and age of English
professional rugby players: A 10-year review. Journal of Sports
Sciences, 31(7), 795–802.
Higham, D. G., Pyne, D. B., & Mitchell, J. A. (2013). Rugby
union players. In R. K. Tanner & C. J. Gore (Eds.),
Physiological tests in elite athletes (2nd ed.) (pp. 379–396).
Canberra: Australian Institute of Sport.
Holway, F. E., & Garavaglia, R. (2009). Kinanthropometry of
Group I rugby players in Buenos Aires, Argentina. Journal of
Sports Sciences, 27(11), 1211–1220.
Hughes, M., Evans, E., & Wells, J. (2001). Establishing normative
profiles in performance analysis. International Journal of
Performance Analysis in Sport, 1,1–26.
Lago, C., Casais, L., & Dominguez, E. (2010). The eff ects of
situational variables on distance covered at various speeds
in elite soccer. European Journal of Sport Science, 10(2),
103–10 9.
Leger, L., & Boucher, R. (1980). An indirect continuous running
multistage field test: The Université de Montreal track test.
Canadian Journal of Applied Sport Sciences, 5(2), 77–84.
McLean, D. A. (1992). Analysis of the physical demands of interna-
tional rugby union. Journal of Sports Sciences, 10(3), 285–296.
Osgnach, C., Poser, S., Bernardini, R., Rinaldo, R., & di
Prampero, P. E. (2010). Energy cost and metabolic power in
elite soccer: A new match analysis approach. Medicine and
Science in Sports and Exercise, 42(1), 170–178.
Quarrie, K. L., Hopkins, W. G., Anthony, M. J., & Gill, N. D.
(2013). Positional demands of international rugby union:
Evaluation of player actions and movements. Journal of
Sciences and Medicine in Sport, 16(4), 353–359.
Roberts, S. P., Trewartha, G., Higgitt, R. J., El-Abd, J., & Stokes,
K. A. (2008). The physical demands of elite English rugby
union. Journal of Sports Sciences, 26(8), 825–833.
Sedeaud, A., Marc, A., Schipman, J., Tafflet, M., Hager, J.-P., &
Toussaint, J.-F. (2012). How they won Rugby World Cup
through height, mass and collective experience. British Journal
of Sports Medicine, 46(8), 580–584.
Spencer, M., Bishop, D., & Dawson, B. (2006). Metabolism and
performance in repeated cycle sprints: Active versus passive
recovery. Medicine and Science in Sports and Exercise, 38(8),
1492–1499.
Spencer, M., Dawson, B., Goodman, C., Dascombe, B., &
Bishop, D. (2008). Performance and metabolism in repeated
sprint exercise: Effect of recovery intensity. European Journal of
Applied Physiology, 103(5), 545–552.
Vigne, G., Gaudino, C., Rogowski, I., Alloatti, G., & Hautier, C.
(2010). Activity profile in elite Italian soccer team. International
Journal of Sports Medicine, 31
(5), 304–310.
Zubillaga, A. (2006). La actividad del jugador de fútbol en alta
competición: Análisis de variabilidad (Doctoral thesis).
University of Málaga, Spain.
Physical demand in international rugby union 11
Downloaded by [88.175.61.234] at 06:52 11 October 2013