This article was downloaded by: [Universidad Pablo de Olavide]

On: 22 November 2011, At: 08:55

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

An ergonomics model of the soccer training process

Thomas Reilly
a
Research Institute for Sport and Exercise Sciences, Liverpool John Moores University,
Liverpool, UK
b
Research Institute for Sport and Exercise Sciences, Liverpool John Moores University,
The Henry Cotton Campus, 15 – 21 Webster Street, Liverpool, L3 2ET, UK E-mail:
t.p.reilly@livjm.ac.uk

Available online: 18 Feb 2007

To cite this article: Thomas Reilly (2005): An ergonomics model of the soccer training process, Journal of Sports Sciences,
23:6, 561-572

To link to this article: http://dx.doi.org/10.1080/02640410400021245

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.
 Journal of Sports Sciences, June 2005; 23(6): 561 – 572

An ergonomics model of the soccer training process

THOMAS REILLY

Research Institute for Sport and Exercise Sciences, Liverpool John Moores University, Liverpool, UK

(Accepted 24 July 2004)

Abstract
Downloaded by [Universidad Pablo de Olavide] at 08:55 22 November 2011

An ergonomics model of training is described in which the demands of the game and the fitness profiles of soccer players are
placed in perspective. The demands of the game may be gauged by monitoring the work rate of players during matches and
the concomitant physiological responses. These indices suggest an increased tempo in contemporary professional soccer
compared with previous decades, a trend replicated in the fitness levels of players. The simulation of the exercise intensity
corresponding to match-play has enabled sport scientists to study discrete aspects of play under laboratory conditions.
Observations highlight the value of exercising with the ball where possible, notably using activity drills in small groups.
Small-sided games have particular advantages for young players, both in providing a physiological training stimulus and a
suitable medium for skills work. While complementary training may be necessary in specific cases, integrating fitness training
into a holistic process is generally advisable.

Keywords: Aerobic power, body composition, field tests, small-sided games, specificity

are incorporated in the ergonomics model illustrated

Introduction
in Figure 1.
Training for soccer may be placed in an ergonomics Fitness tests can provide an indication of players’
context. The game can be viewed as imposing a distinctive strengths and deficiencies. They can also
range of demands on its players who must possess be employed to determine the effectiveness of any
the necessary fitness to cope with these demands. systematic change in training regimen. A profile of
Fitness for soccer calls for a combination of different fitness test data juxtaposed alongside physiological
physiological characteristics. Training these factors responses to match-play can highlight the extent to
enables the player to add training effects to his or her which players can impose demands on themselves
endowed characteristics so that ultimate potential and provide pointers as to when they are under-
can be realized. Even though the training effects for performing in matches by not meeting the
some characteristics may be moderate in magnitude,
the impact on performance characteristics can still be
substantial.
As soccer is a team game, a priority in preparing
players for match-play must be to harness their
individual capabilities so that the group becomes an
effective competitive unit. The dilemma for the
coach or manager is often which team selection is
the most appropriate for the forthcoming engage-
ment. One difficulty the trainer faces is the
identification of individual weaknesses that can be
remedied in training, over and above the overall
conditioning work for the team. There is also the
need to take positional role into account to ensure
specificity of the training programme. These factors Figure 1. An ergonomics model for the analysis of football.

Correspondence: T. Reilly, Research Institute for Sport and Exercise Sciences, Liverpool John Moores University, The Henry Cotton Campus, 15 – 21
Webster Street, Liverpool L3 2ET, UK. E-mail: t.p.reilly@livjm.ac.uk
ISSN 0264-0414 print/ISSN 1466-447X online ª 2005 Taylor & Francis Group Ltd
DOI: 10.1080/02640410400021245
 562 T. Reilly

requirements of the game. In such instances, this employs aerobic metabolism. Bangsbo (1994) calcu-
individual may be omitted from the team selection lated that the major substrate used by the active
until fitness is remedied or may be assigned an muscles is glycogen, the stores of which that are
alternative tactical role. located in active muscles contribute more than do
The self-imposed demands chosen by players glycogen sources mobilized from hepatic depots.
reflect both their commitment to the team’s efforts There is also a substantial engagement of lipid
and their own abilities to pace themselves throughout metabolism compatible with endurance exercise.
the game. The choice of exercise intensity represents The contribution of protein is relatively minor at
a subjective judgement with regard to what is thought around 2 – 4%, although Wagenmakers, Brookes,
to be tolerable. Since the majority of activity is spent Coakley, Reilly and Edwards (1989) demonstrated
in running ‘‘off-the-ball’’ (Reilly & Thomas, 1976), that protein metabolism is employed in exercise
the individual who has been over-committed may not sustained for 90 min well before muscle glycogen
recover in time to engage in the next bout of critical concentrations are depleted.
activity ‘‘on-the-ball’’. The fine timing of self-chosen It has been estimated that the energy expenditure
strenuous efforts is an important product of the during match-play averages 70 – 75% of maximal
Downloaded by [Universidad Pablo de Olavide] at 08:55 22 November 2011

training process. oxygen consumption (Bangsbo, 1994, Reilly, 1997).

This review represents an attempt to place the It has been calculated that the total energy expended
training process in soccer within the ergonomics in a game approximates 6.317 MJ (Reilly & Thomas,
model. Some consideration is first given to the 1979). These estimates were based on extrapolations
physiological demands of the game. Some insight from heart rate and motion analysis data during
into these demands is added by reviewing fitness training games, utilizing individually calibrated heart
data on contemporary professional players. The rate – oxygen uptake relationships determined under
attention is directed to training programmes, both laboratory conditions. Although such an approach to
soccer-specific and general conditioning. The use estimating energy expenditure during intermittent
of small-sided games for young players is consid- exercise may be criticized, errors appear to be
ered before the review culminates in an overview averaged out over the period of measurement
that places training in the context of talent (Ainslie, Reilly, & Westerterp, 2003; Bangsbo,
development. 1994). Saltin (1973) demonstrated that glycogen in
the vastus lateralis approached depletion towards the
end of a game. Players who undertook strenuous
Physiological demands of the game
training the day before the match had severely
Playing soccer constitutes intermittent exercise in lowered concentrations of glycogen even at half-time
which the timing of the high-intensity efforts is and played at a reduced work rate during the second
acyclical and therefore unpredictable. A team coa- half.
ched and trained to perform at a high tempo has a The contemporary game at the professional level
competitive edge over an equally skilled but less ‘‘fit’’ seems to be more demanding than suggested in
opposition. If recovery periods in between bouts of much of the early literature (e.g. Reilly & Thomas,
strenuous efforts are inadequate in duration, tran- 1976) and therefore calls for a more systematic
sient fatigue will ensue. Mohr, Krustrup and approach to training. Strudwick and Reilly (2001)
Bangsbo (2003) showed that activity was reduced compared the work rates of English Premier League
in the 5 min immediately after a 5-min period of players over two seasons (1998 – 1999 and 1999 –
sustained exercise at high intensity. 2000) with previous observations of top English
In a review of the prolonged run-up, which the League players before 1992. The current players
Korean team adopted in preparing for the 2002 covered approximately 1.5 km more in a game than
World Cup Finals, Verheijen (2003) described how their earlier counterparts, the differences being
initially the team could not keep up the desired pace manifest among all the playing positions (see Figure
of the game for the full 90 min. Players made high- 2). Williams, Lee and Reilly (1999) have provided
intensity runs less frequently and fewer explosive evidence of a faster tempo of the game in the 1997 –
actions as the second half progressed. After a 1998 season compared with the 1991 – 1992 season,
systematic training programme, they were able to including more movement of the ball and shorter
maintain a high tempo for the entire match. By the intermissions in play. Changes in the rules, such as
end of the next phase of training, the team was the rule prohibiting the goalkeeper from picking up a
prepared to raise the pace of the game even higher, back-pass, the penalizing of time-wasting and per-
and recover more quickly between actions. mission to use three substitutes, have contributed to
Notwithstanding the repeated sprints and the the rise in tempo. The end result is an increased
contests for possession of the ball that entail reliance on optimum training programmes to meet
anaerobic exercise, most activity during a game these elevated demands.
 An ergonomics model of soccer training 563
Downloaded by [Universidad Pablo de Olavide] at 08:55 22 November 2011

Figure 2. Distance covered per game in the Premier League compared with former First Division (pre-1992). Data taken from Strudwick
and Reilly (2001).

Drust, Reilly, Carter, & Martin, 2000) and levels of

Assessment of game demands
play (Florida-James & Reilly, 1995). The observa-
Various methods have been used to determine the tions can also be broken down into distinct
demands of competitive soccer match-play. The categories, including the distance covered in (and
most common have been indirect methods, such as frequency of) sprints and other high-intensity activ-
assessment of distance covered, while physiological ities. They may also include discrete actions, such as
responses have also been employed either in compe- jumping, tackling and kicking, which are related
titive or friendly matches. A further alternative has directly to involvement in play (Bangsbo, 1993).
been to simulate the exercise intensity corresponding Summaries of the main methods adopted for
to match-play on the basis of work rate profiles and motion analysis are reported in the literature (Reilly,
to study such simulations in laboratory conditions. 1994, 2003). These include applications of trigono-
The classical method of motion analysis has been metry, global positioning systems and the use of
used by different research teams (e.g. Bangsbo, synchronized multiple cameras. Global positioning
Norregaard, & Thorso, 1991; Reilly & Thomas, systems have been used to a limited extent (Kirken-
1976: Withers, Maricic, Wasilewski, & Kelly, 1982) dall, Leonard, & Garrett, 2003), and their precision
to determine the work rate of players during a game. remains in question. Customized computer-aided
The aggregated data can be expressed as overall analysis has been popular with coaches, and compu-
distance covered in a game; this index is sufficiently terized and digital video-based systems have been
sensitive to differentiate between positional roles commercialized for use in notation analysis. The
(Reilly & Thomas, 1976), environmental conditions software allows for recording of each action, asso-
(Ekblom, 1986), styles of play (Drust, 1999, Rienzi, ciated player involvement and position on the pitch,
 564 T. Reilly

so that all actions can be aggregated to quantify state activity (Bangsbo, 1994; Bot & Hollander,
patterns of play. The observations are most useful as 2000).
feedback on performance and the majority of Drills associated with training have been examined
computer programs do not have the capability to using field studies (Kawakami, Nozaki, Matsuo, &
determine distances covered due to limitations of Fukunaga, 1992). These authors used a portable
camera work. These drawbacks are overcome with respiratory gas analyser to measure metabolic re-
the use of multiple cameras placed strategically at sponses to small-group activities in a training context.
elevated positions on the stands, allowing both the The study provided a model for relevant field studies
monitoring of movements of each individual player of training activities. An alternative approach was used
and notation analysis of patterns of play (see Carling, by Reilly and Ball (1984), who isolated the practice of
2002). Although such systems have been used by dribbling a ball for study in a laboratory setting.
many of the top professional clubs in Europe, their Energy expenditure, blood lactate and perceived
reliability has not been formally established and small exertion were elevated when dribbling was compared
errors in data collection could significantly influence to running on a treadmill, the increase being similar at
their interpretation. The high cost of the service all four running speeds investigated. For a given speed
Downloaded by [Universidad Pablo de Olavide] at 08:55 22 November 2011

restricts its accessibility to the exclusive few rich of locomotion, the training stimulus is higher when
clubs that can afford it. working with the ball compared with running nor-
Rahnama, Reilly and Lees (2002) adapted mally, suggesting benefits of soccer-specific work
computer-aided notation analysis to study injury wherever possible. Additionally, Reilly and Bowen
risk in soccer. Their approach employed a critical (1984) compared orthodox running, moving back-
incident technique in which all actions in a game wards and moving sideways; there was a progressive
were classified according to their potential for increase in energy expenditure at each speed studied
causing injury. Tackling constituted the major for the unorthodox movements. Since the ability to
source of injury risk. The first 15-min and the last move backwards and sideways quickly is an important
15-min periods in the game were when critical skill for defenders, and is an unorthodox mode of
incidents were most in evidence. The method was motion, special attention should be given to such
useful in highlighting behavioural precursors of activities in training.
injury occurrence. Another alternative open to applied physiologists
Physiological responses to match-play have in- is to simulate the work rates associated with soccer
cluded monitoring metabolic variables (such as for experimental investigations. This approach is an
oxygen uptake, blood lactate accumulation, glucose advance of the 90-min continuous exercise or the
and fatty acid utilization), thermoregulatory re- more regimental intermittent exercise protocols
sponses, circulating metabolic and stress hormone previously adopted by researchers. Nicholas, Nut-
concentrations. Studies of the adenonucleotide and tall and Williams (2000) employed a repeated
adenosine monophosphate (AMP) systems (Apor, shuttle run, average intensity of which corre-
1988; Ekblom, 2002) have also been conducted in sponded roughly to the work rate of football
the context of soccer play. Energy expenditure in match-play; the activity was continued for 75 min
training has been placed in the context of habitual and subsequently altered to provide a test to
activity and analysed by means of combined diary voluntary exhaustion. The protocol was designed
card and physiological methods (Reilly & Thomas, with a view to study nutritional interventions that
1979), including doubly labelled water (Ebine et al., might offset fatigue towards the end of a normal
2002). The average daily energy expenditure has game or benefit performance in the event of extra
been estimated as 14.8 + 1.7 MJ. The availability of time. The protocol was later modified by Edwards,
short-range radio telemetry systems for monitoring Clark and MacFayden (2003a) and, in a restricted
heart rate during exercise has led to its widespread form, used for purposes of fitness testing. Drust,
use in a soccer training context. The so-called ‘Team Reilly and Cable (2000b) designed a different
System’ (Polar Electro, Finland) enables physiolo- protocol, for use on either a motorized or non-
gists to monitor heart rates of all players motorized treadmill, in which the exercise inten-
simultaneously during sessions: responses may also sities were changed frequently to reflect the
be used to regulate rather than monitor training dynamics of soccer play and correspond to its
intensity, for example when heart rates in the zone intensity. It has been used to study effects of pre-
120 – 130 beats × min71 are desired on recovery days. cooling before a match (Drust, Cable, & Reilly,
While heart rate responses have been criticized for 2000a). On a non-motorized treadmill, the power
use in intermittent exercise due to the influence of output can be calculated by tethering the partici-
non-metabolic factors, such as emotional and ther- pant to a force transducer via a running harness
moregulatory stimuli, heart rate has proved to be a and monitoring the revolutions of the treadmill belt
robust index of physiological strain in non-steady- (Lakomy, 1987). Each intensity can be regulated or
 An ergonomics model of soccer training 565

the high-intensity efforts can be freely chosen by The average value of the top Hungarian team
the individual to yield a measure of performance. studied by Apor (1988) (66.6 ml × kg71 × min71) was
This paradigm can be used to study experimental unusually high for football players at the time (see
interventions, such as the relevance of creatine or review by Douge, 1988). Such values are regularly
carbohydrate loading. attained by contemporary teams. Casajus (2001)
In summary, a number of approaches to the reported a mean value of 65.5 ml × kg71 × min71 for a
complex assessment of game demands have been Spanish League team early in the season, rising to
employed by researchers. It is important that the 66.4 ml × kg71 × min71 by mid-season. Even higher
method chosen should fit the purpose of the values (67.6 ml × kg71 × min71) were reported by
investigation. The limitations of particular ap- Wisløff, Helgerud and Hoff (1998) for Rosenburg
proaches should be appreciated if the researcher is FC, a top Norwegian professional side. This figure
to interpret the observations correctly. provides a reasonable guideline for current teams to
adopt while recognizing that figures may vary
according to positional role and individual factors.
The fitness of contemporary players
The increasing use of field tests for soccer players
Downloaded by [Universidad Pablo de Olavide] at 08:55 22 November 2011

Information about the fitness of current elite has meant that few teams are assessed entirely in
players provides an insight into the requirements laboratory contexts. The 20-m shuttle run test
for competing at the top level. The fitness (Ramsbottom, Brewer, & Williams, 1988) and the
requirements of soccer are related to the physiolo- Yo-Yo tests (Bangsbo & Michalsik, 2000) have been
gical, physical and technical aspects of the game. A adopted for administrative convenience and lack a
combination of demands is imposed on the player physiological criterion that maximal effort was
during competition, as a result of which tests with achieved. Where maximal oxygen uptake values are
a high degree of specificity to the game should be available, they indicate that there are fewer low
employed for the assessment of performance values among the current squads playing at a high
capability. This principle has led to the promotion level. Repeated measures taken throughout the
of field tests conducted in applied settings, season suggest that values do not change much
although more useful physiological insights have during the competitive campaign; the Spanish team
been gained from laboratory tests. studied by Casajus (2001) improved from 65.5 + 8.0
Maximal oxygen uptake has proved to be a good to 66.4 + 7.0 ml × kg71 × min71 between early season
indicator of aerobic prowess (Table I). Reilly and and mid-season. It would also appear that due to
Thomas (1976) reported a high correlation between personalized fitness training in the off-season,
maximal aerobic power and distance covered in a detraining effects during this period are minimized
game. Apor (1988) found a direct correlation (Edwards, Clark, & MacFayden, 2003b).
between the average maximal oxygen uptake of a While the change in maximal oxygen consumption
team and its finishing position in the Hungarian among the Spanish players (Casajus, 2001)
League. Bangsbo and Michalsik (2000) showed that amounted only to 1.7%, greater improvements were
midfield players had on average higher values than observed in the lactate threshold. The relative
full-backs, who were in turn superior to forwards, intensity at which the lactate threshold (the inflation
with central defenders having the lowest values point in the blood lactate responses to graded
among outfield players. These observations were exercise) was observed increased from 76.6 to
broadly similar to the results of a field test (i.e. the 79.4% V_ O2max. The velocity at which the lactate
‘‘Yo-Yo’’ test) applied to 115 Danish players, threshold occurred increased from 12.4 + 1.5 to
although in this case the forwards performed as 13.1 + 1.4 km × h71, a relative improvement of
poorly as the defenders. 5.7%. It would appear that blood lactate responses

Table I. Maximal oxygen uptake (VO2max) as measured in professional football teams (average values and sources are cited)

Team V_ O2max (ml × kg71 × min71) Reference

Elite players 61.0 Douge (1988)

Austrian national team 58.3 Bachl & Prokop (1977)

German national team 62.0 Hollmann et al. (1981)
Bundesliga team 58.4 Dickhut, Simon, Bachl, Lehman, & Keul (1981)
Hungarian champions 66.6 Apor (1988)
Portugese First League team 59.7 Puga et al. (1993)
Spanish League 65.5 Casajus (2001)
Norwegian champions 67.6 Wisløff et al. (1998)
 566 T. Reilly

to incremental exercise may be more useful to factor. This index has been termed the ‘‘dynamic
applied sports scientists than are maximal values. control ratio’’ (Aagaard, Simonsen, Magnusson,
Grant and McMillan (2001) showed how lactate Larson, & Dyhre-Poulson, 1998).
profiles were sensitive to detraining effects due to The muscularity of professional soccer players is
injury and to subsequent improvements in fitness as reflected in the physique or somatotype. Interna-
rehabilitation goals were being reached. tional players competing in the Copa America
Physiological variables strongly related to endur- championships displayed a mean somatotype with
ance running performance do appear therefore to be high mesomorphy values but low endomorphy values
relevant to soccer. These sports differ in that the (Rienzi et al., 2000). The gains in muscular strength
latter is also heavily reliant on muscle strength and should not compromise speed of movement. Strud-
power. There is a wealth of data supporting the wick, Reilly, and Doran (2002) showed that players
superiority of soccer players to age-matched counter- in an English Premier club had similar fitness profiles
parts in measures of muscle strength determined (including aerobic power) to those of a top Gaelic
using isometric (Reilly & Thomas, 1980) and Football side, except for their faster running speeds
isokinetic dynamometry (Fowler & Reilly, 1993). over 10 and 30 m. Besides the ability to accelerate
Downloaded by [Universidad Pablo de Olavide] at 08:55 22 November 2011

Professional players perform significantly better at all over short distances, another distinguishing feature
angular velocities and in eccentric as well as in soccer is agility or the ability to change direction of
concentric modes than do players of a lesser standard movement quickly (Reilly, Williams, Nevill, &
(Rahnama, Reilly, Lees, & Graham-Smith, 2003). Franks, 2000).
The profiling of individuals can help to identify Accompanying the muscular make-up of contem-
asymmetric imbalances in peak torque between left porary professional players is a low level of adiposity
and right limbs and between leg flexors and (see Table II). As is to be expected, the percent
extensors. Once weaknesses are identified in indivi- adiposity tends to be decreased as the season
dual profiles, these should be rectified using progresses (Casajus, 2001) and in the off-season does
appropriate physical conditioning regimens. not regress to the high figures reported in previous
The strength of the lower limbs of soccer players decades (White, Emery, Kane, Groves, & Risman,
has been assessed using isokinetic dynamometry. An 1988). There is a problem in comparing values
asymmetrical difference between limbs or a dispro- expressed as a percentage of body fat when the figures
portionate balance between the knee flexors and knee are derived from different methodologies and com-
extensors may predispose a player towards muscu- puted using different formulae. The majority of the
loskeletal injury. Rahnama et al. (2003) reported that data reported were acquired from estimations based
Premier League players were superior in strength to on skinfold thicknesses, but occasionally the reference
size-matched sub-elite players at a range of angular method of hydrodensitometry has been used. The use
velocities, for dominant and non-dominant limbs of skinfold thicknesses has been criticized on the basis
and for concentric as well as eccentric modes of of a weak relationship with internal fat depots and the
action. Oberg, Moller, Gillquist and Ekstrand (1986) concentration of upper body sites in the case of
found relatively greater fast speed/slow speed ratios Durnin and Womersley’s (1974) equations (Clarys,
in national players compared with other players, Martin, & Drinkwater, 2002). Nevertheless, there is a
highlighting the relevance of strength assessments to good relationship between values recorded using dual
actions in the game. For injury prevention purposes, energy X-ray absorptiometry (DEXA) and percent
it would appear that the relative eccentric strength of body fat predicted from four upper body skinfold
the knee flexors compared with the concentric measures (Egan, Reilly, Chantler, & Lawlor, 2004), as
strength of the knee extensors is a determining shown for observations on 16 Premier League players

Table II. Adiposity in soccer players (mean percentages for players, including goalkeepers, are cited)

Team % Body fat References

English First Division 19.3 White et al. (1998)

Portugese First Division 10.9 Puga et al. (1993)
English Premier (early season) 12.6 Dunbar & Power (1997)
Spanish professionals 7.9 Mujika, Padilla, Ibañez, Izquierdo, & Gorostiaga (2000)
English Premier team 12.3 Strudwick et al. (2002)
Copa America 10.6 Rienzi et al. (2000)
La Liga
Early season 8.6 Casajus (2001)
Mid-season 8.2
 An ergonomics model of soccer training 567

pre season (see Figure 3) The DEXA technique is rate responses that the more intense parts of the
likely to be the reference method of the future and, session are when the ball is used (Reilly, 1979).
while the technology is expensive, it has the added Reilly and White (2003) compared the effects of a
benefit of indicating bone mineral density. The extent classical interval training programme in a group of
to which the skeletal system responds to football young players in an English Premier League club.
training or protects against injury is yet to be The training consisted of additions to the normal
established. programme, the players doing interval running
In summary, it would appear that elite contem- consisting of six 4-min runs with a 3-min recovery
porary footballers have distinct features. These in between efforts, three times a week for 6 weeks.
include few low values of maximal oxygen uptake The experimental group played five-a-side games for
among the squad, combined with good endurance an equivalent duration and recovery, with vocal
capabilities in mid-season. They have a muscular encouragement from the trainer. Both groups main-
make-up and lower adiposity than players of previous tained their aerobic fitness and lactate profiles
decades. They are further characterized by speed equally, the training being conducted in mid-season.
over short distances and by their agility. It is clear from the typical heart rate responses in
Downloaded by [Universidad Pablo de Olavide] at 08:55 22 November 2011

Figure 5 that the soccer-specific work provided a

comparable training stimulus to classical interval
Soccer-specific training
training.
In view of the extra energetic costs of exercising with In a related study, Sassi, Reilly and Impellizzeri
the ball at a given speed of locomotion (Reilly & Ball, (2003) compared small-sided games under different
1984), working with the ball has been advocated as a conditions with formal interval training. In this
means of facilitating training stimuli. This practice study, the interval training consisted of 4 6 1000-
would be economical in use of players’ time, when m runs, while the number of players, dimensions of
physical training is integrated with technical aspects the playing area, technical factors and so on were
of preparation. This strategy does not exclude the varied systematically. The results, summarized in
use of additional general conditioning, which may Table III, show how the inclusion of a goalkeeper
include running, weight-training, flexibility exercises and an emphasis on pressing the player in possession
and agility drills, and other regimens such as influence the responses. They show that appropri-
plyometrics (stretch – shortening exercises) and ately designed exercises can raise the intensity to a
deep-water running. level that exceeds interval running. It was concluded
The traditional training programme may be that small-group work with the ball can present
divided into different components, as shown in physiological training stimuli comparable with and
Figure 4. The first four elements (warm-up, sometimes exceeding interval running without the
calisthenics, running, circuit training) constitute ball. In contrast, technical/tactical training presents a
physical training, whereas the remainder represent moderate challenge to the circulatory system, more
work with the ball. The recovery component is the compatible with maintenance programmes or recov-
sum of intermissions between the discrete sections of ery on days following competitive matches.
the training session. It is evident from mean heart The observations on elite players have been
corroborated in participants at a recreational level
in indoor four-a-side leagues. MacLaren, Davids,
Isokawa, Mellor and Reilly (1988) showed that
metabolic and hormonal responses of players ex-
ceeded those found in typical outdoor games at this
level. Noradrenaline, adrenaline and blood lactate
concentrations were greatly increased, values during
the league final reaching 414%, 228% and 130%,
respectively, of the values noted after recreational
games. It was concluded that the four-a-side version
of indoor soccer has important implications for its
use as a training method for the outdoor competitive
format or as a recreational activity in its own right.
Similar conclusions were drawn by Miles, MacLa-
ren, Reilly and Yamanaka (1993) in a study of female
Figure 3. Percent body fat of Premier League soccer players participants. It would appear that small-sided games
(n = 16) pre season determined by dual-energy absorptiometry
(DEXA) and estimated from skinfolds according to Durnin and
can provide an acceptable means of training the
Womersley (1974). Mean values for DEXA were 13.26 + 1.84 oxygen transport system in both sexes at elite and
and for skinfolds 12.99 + 1.97 (r = 0.88). recreational levels.
 568 T. Reilly
Downloaded by [Universidad Pablo de Olavide] at 08:55 22 November 2011

Figure 4. The traditional allocation of training time to its different components. Sessions started with a warm-up and callisthenics (flexibility
and agility) and ended with drills and games (from Reilly, 1979).

A variety of drills for execution with the ball have more involvement with play, dribbled more with
been designed to train the metabolic systems for the ball and had more opportunities for executing
application to soccer. These have incorporated the skills such as passing and shooting (see Figure 6).
training of the aerobic system (Reilly & Bangsbo, Furthermore, they had more relative success in
1998), anaerobic endurance (see Bangsbo, 1993) these skills than their counterparts in the five-a-side
and running speed (Wilson, 2001b). They also game. The results support the promotion of the
accommodate the period between matches, includ- three-a-side game for young players, although
ing recovery days and more intense training, suitable pitch dimensions should be established to
regulated by blood lactate responses, heart rate and fit the fitness levels of the group.
perceived exertion (Bangsbo, 1994; Drust, 1999). In In contrast, Castagna, Belardinelli and Abt (2003)
particular, attention has been drawn to the opportu- found that moderately trained and skilled Italian
nities offered by small-sided games as a means of high-school students (aged 16 – 17 years) reached an
simultaneously acquiring physiological and skills average only of 53% of maximal oxygen uptake
training effects. playing five-a-side on a court measuring 40 6 29 m.
They concluded that this intensity would have little
impact on the development of aerobic power. While
Small-Sided games and young players
acknowledging that dimensions of the playing area,
Young players need to develop their physiological number of players and rules employed affect the
determinants of fitness in conjunction with games exercise intensity, the duration of play may also be an
skills and decision-making in a soccer context. These influential factor, as is the motivational climate in
factors depend not only on the intensity of exercise which the practice is conducted. Castagna et al.
but also on the direct involvement in activities with advised that the physiological load of small-sided
the ball. Youth soccer matches can be played by a games should be quantified before they are used
reduced number of players on smaller pitches, within a training programme for the development of
depending on the players’ age. aerobic power.
Platt, Maxwell, Horm, Williams and Reilly The dribbling elements of the game could
(2001) examined the influence of small-sided enhance important game skills. Reilly et al.
games on players’ actions, combining physiological (2000) reported that dribbling capability distin-
investigation with a notation analysis of players guished elite under-16 players from age-matched
aged under 12. Three-a-side was found to be players at a lower level more than did shooting,
superior to five-a-side on all of the criteria passing or ball control tests. Since dribbling
employed. The mean heart rate was higher in the typically entails changes in direction to deceive an
three-a-side version throughout the 15-min period opponent, it also offers a convenient means of
of observation. The smaller group of players had training agility.
 An ergonomics model of soccer training 569
Downloaded by [Universidad Pablo de Olavide] at 08:55 22 November 2011

Figure 5. Typical heart rate response to interval training (4 6 4 min with 3 min recovery) and small-sided games over the same time periods
(from Reilly & White, 2003).

ing may require individual attention. The main

Complementary training
reason for this requirement is that competitive play
Although recognizing the value of integrating the may not provide the necessary training stimulus at a
physiological preparation of players with game- sufficient intensity, for long enough or often enough
related activities, some aspects of physical condition- to secure a physiological adaptation. A further reason
 570 T. Reilly

Table III. Heart rate (beats × min71) and blood lactate concentration (mmol × l71) during two different training drills (mean + s) (data from
Sassi et al., 2003)

4 vs 4 8 vs 8

Without With Free Free touch Technical – tactical

4 6 1000 m goalkeeper goalkeeper touch (pressing) drills

Heart rate 167 + 4 178 + 7 174 + 7 160 + 3 175 + 4 140 + 5

Blood lactate 7.9 + 3.4 6.4 + 2.7 6.2 + 1.4 3.3 + 1.2 — 2.9 + 0.8
Downloaded by [Universidad Pablo de Olavide] at 08:55 22 November 2011

Figure 6. A comparison of activity during three-a-side and five-a-side games in boys aged under 12 years (from Platt et al., 2001).

is that conditioning embraces protecting against lifted. Alternative modes of exercise, such as various
injury as well as enhancing performance. In a forms of running in water, offer an acceptable means
classical study, Ekstrand (1982) demonstrated that for promoting recovery after matches (Reilly, Dow-
a systematic programme of flexibility training re- zer, & Cable, 2003). Running in deep water avoids
duced the incidence of injuries over a season of impact loading on muscles likely to be sore following
Swedish League soccer. His programme of exercises games: it can be used as part of a warm-down when
targeted muscles susceptible to injury in the game, facilities are available, or as recovery training the
notably the hamstring group and the thigh adduc- following day. During the off-season, any formal
tors. training may be for general conditioning purposes
There are also positional differences in fitness (Edwards et al., 2003b).
requirements whereby certain players need to focus
on conditioning work that enhances their roles. For
Overview
example, plyometric and complex training is likely to
be more beneficial to central defenders than midfield An ergonomics model allows training to be con-
players (Wilson, 2001a,b), whereas whole-body sidered as interfacing with the demands of the game
agility is important for goalkeepers (Whall, 2001). on the one hand and with the capabilities of players
Non-specific activities can also help to add variety on the other. Preparation for competition is opti-
to the habitual activity of players. Running and mized when technical, tactical and physiological
weight-training can have a positive influence on requirements are integrated in the planning cycles.
players because of the psychophysical calibration Such a holistic approach favours the use of exercises
inherent in these activities. Wisløff et al. (2004) with the ball as far as it is possible. This principle is
demonstrated how the benefits of weight-training underlined in the case of young players whose talent
transferred positively to performance in a game is to be developed concomitantly with natural growth
context. They also provided a quantitative measure and development processes. In this instance, priority
of performance in terms of time, distance or load is given to deliberate practice and acquisition of
 An ergonomics model of soccer training 571

game skills and awareness (Williams & Reilly, 2000), Drust, B., Reilly, T., & Cable, N. T. (2000b). Physiological
while recognizing the value of physiological criteria responses to laboratory-based soccer-specific intermittent and
continuous exercise. Journal of Sports Sciences, 18, 885 – 892.
for training appropriate to age (Stratton, Reilly, Dunbar, G. M. J., & Power, L. (1997). Fitness profiles of English
Williams, & Richardson, 2004). professional and semi-professional soccer players using a
battery of field tests. In T. Reilly, J. Bangsbo, & M. Hughes
(Eds.), Science and football III (pp. 27 – 31). London: E & FN
Spon.
References Durnin, J. V. G. A., & Womersley, J. (1974). Body fat assessed
from total body density and its estimation from skinfold
Aagaard, P., Simonsen, E. B., Magnusson, S. P., Larson, B., &
thickness: Measurements on 481 men aged 16 to 72 years.
Dyhre-Poulsen, P. (1998). A new concept for isometric British Journal of Nutrition, 32, 77 – 97.
hamstring:quadriceps muscle ratio. American Journal of Sports Ebine, N., Rafamantavantosa, H. H., Nayuki, Y., Yamanaka, K.,
Medicine, 26, 231 – 237. Ono, T., Saitoh, S., & Jones, P. H. (2002). Measurement of
Ainslie, P. N., Reilly, T., & Westerterp, K. (2003). Estimating
total energy expenditure by the doubly-labelled water method
human energy expenditure: A review of techniques with in professional soccer. Journal of Sports Sciences, 20, 391 – 397.
particular reference to doubly-labelled water. Sports Medicine, Edwards, A. M., Clark, N., & MacFadyen, A. M. (2003a). Test
33, 683 – 698.
performance indicators from a single soccer specific fitness test
Apor, P. (1988). Successful formulae for fitness training. In T.
Downloaded by [Universidad Pablo de Olavide] at 08:55 22 November 2011

differentiates between highly trained and recreationally active

Reilly, A. Lees, K. Davids, & W. J. Murphy (Eds.), Science and soccer players. Journal of Sports Medicine and Physical Fitness, 43,
football (pp. 95 – 107). London: E & FN Spon. 14 – 20.
Bachl, N., & Prokop, L. (1977). Wie gut trainiert sina
Edwards, A. M., Clark, N., & MacFadyen, A. M. (2003b). Lactate
Osterriechische Fussballer? Leistungsdiagnostische Standort- and ventilatory thresholds reflect the training status of profes-
bestimmung aus der Sicht des Sportphysiologen. Osterreiche sional soccer players where maximum aerobic power is
Zeitschrift Sportsmedizen, 7, 3 – 10. unchanged. Journal of Sports Science and Medicine, 2, 23 – 29.
Bangsbo, J. (1993). Fitness training in football. Copenhagen:
Egan, E., Reilly, T., Chantler, P., & Lawlor, J. (2004). Body
HO + Storm. composition before and after six weeks preseason training in
Bangsbo, J. (1994). The physiology of soccer with special professional football players. Communication to the Pre-
reference to intense intermittent exercise. Acta Physiologica
Olympic Scientific Conference, Thessaloniki, Greece.
Scandinavica, 151 (suppl. 159), 1 – 155. Ekblom, B. (1986). Applied physiology of soccer. Sports Medicine,
Bangsbo, J., & Michalsik, L. (2000). Assessment of the physio- 3, 50 – 60.
logical capacity of elite soccer players. In W. Spinks, T. Reilly, Ekblom, B. (2002). Assessment of fitness and player profiles.
& A. Murphy (Eds.), Science and football IV (pp. 53 – 62).
Communication to the International Football and Sports Medicine
London: Routledge. Conference, Beverly Hills, CA, 22 – 24 March.
Bangsbo, J., Norregaard, L., & Thorso, F. (1991). Activity profile Ekstrand, J. (1982). Soccer injuries and their prevention. Medical
of professional soccer. Canadian Journal of Sports Science, 16, dissertation #130, Linköping University, Linköping, Sweden.
110 – 116.
Florida-James, G., & Reilly, T. (1995). The physiological
Bot, S. D. M., & Hollander, A. P. (2000). The relationship demands of Gaelic football. British Journal of Sports Medicine,
between heart rate and oxygen uptake during non-steady state 29, 41 – 45.
exercise. Ergonomics, 43, 1578 – 1592.
Fowler, N., & Reilly, T. (1993). Assessment of muscle strength
Carling, C. (2002). Football: A game of chance or does match
asymmetry in soccer players. In E. J. Lovesey (Ed.), Con-
analysis have the answers? Insight: The FA Coaches Association temporary ergonomics (pp. 327 – 332). London: Taylor &
Journal, 5(2), 41 – 42. Francis.
Casajus, J. A. (2001). Seasonal variation in fitness variables in
Grant, S., & McMillan, K. (2001). The role of blood lactate
professional soccer players. Journal of Sports Medicine and response to sub-maximal exercise in the monitoring of aerobic
Physical Fitness, 41, 463 – 467. fitness in footballers. Insight: The FA Coaches Association
Castagna, C., Belardinelli, R., & Abt, G. (2003). The VO2 and Journal, 4(2), 34 – 35.
heart rate response to training with a ball in youth soccer
Hollmann, W., Liesen, H., Mader, A., Heck, H., Rost, R., Dufaux,
players. Communication to the Fifth World Congress on Science B., Schurch, P., Lagerstrom, D., & Föhrenbach, R. (1981). Zur
and Football, Lisbon, Portugal, 13 – 17 April. Höchst – und Dauerleistungsfahigkeit der Deutschen Fussball –
Clarys, J., Martin, A., & Drinkwater, D. (2002). Body composi- Spitzenspieler. Deutsch Zeitschift Sportsmedizin, 32, 113 – 120.
tion, Part II. ‘‘Whole-body adiposity’’ prediction: Males versus
Kawakami, Y., Nozaki, D., Matsuo, A., & Fukunaga, T. (1992).
females. In T. Reilly (Ed.), Musculoskeletal disorders in health- Reliability of measurement of oxygen uptake by a portable
related occupations (pp.151 – 161). Amsterdam: IOS Press. telemetric system. European Journal of Applied Physiology, 65,
Dickhut, W. H., Simon, G., Bachl, N., Lehman, M., & Keul, J.
409 – 414.
(1981). Zur Höchst – und Dauerleistungsfähigkeit von Bun- Kirkendall, D. T., Leonard, K., & Garrett, W.E., Jr. (2003). On
diesligafussballspielern. Leistungssport, 11, 148 – 152. the relationship of fitness to running volume and intensity in
Douge, B. (1988). Football: The common threads between the female soccer players. Communication to the Fifth World
games. In T. Reilly, A. Lees, K. Davids, & W. J. Murphy
Congress on Science and Football, Lisbon, 11 – 15 April.
(Eds.), Science and football (pp. 3 – 19). London: E & FN Spon. Lakomy, H. (1987). The use of a non-motorised treadmill for
Drust, B. (1999). Analysis of work-rate in soccer. Sports Exercise analyzing sprint performance. Ergonomics, 3, 627 – 637.
and Injury, 4, 151 – 155. MacLaren, D., Davids, K., Isokawa, M., Mellor, S., & Reilly, T.
Drust, B., Cable, N. T., & Reilly, T. (2000a). Investigation of the
(1988). Physiological strain in 4-a-side soccer. In T. Reilly, A.
effect of pre-cooling on the physiological responses to soccer- Lees, K. Davids, & W. J. Murphy (Eds.), Science and football
specific intermittent exercise. European Journal of Applied (pp. 76 – 80). London: E & FN Spon.
Physiology, 81, 11 – 17.
 572 T. Reilly

Miles, A., MacLaren, D., Reilly, T., & Yamanaka, K. (1993). An Reilly, T., & Thomas, V. (1980). The stability of fitness factors
analysis of physiological strain in four-a-side women’s soccer. over a season of professional soccer as indicated by serial factor
In T. Reilly, J. Clarys, & A. Stibbe (Eds.), Science and football II analysis. In M. Ostyn, G. Beunen, & J. Simons (Eds.),
(pp. 140 – 145). London: E & FN Spon. Kinanthropometry II (pp. 247 – 257). Baltimore, MD: University
Mohr, M., Krustrup, P., & Bangsbo, J. (2003). Match perfor- Park Press.
mance of top-level soccer players with special reference to Reilly, T., & White, C. (2003). Small-sided games as an alternative
development of fatigue. Journal of Sports Sciences, 21, 519 – 528. to interval-training for soccer players. Communication to the
Mujika, I., Padilla, S., Ibañez, J., Izquierdo, M., & Gorostiaga, E. Fifth World Congress on Science and Football, Lisbon, 11 – 15
(2000). Creatine supplementation and sprint performance in April.
soccer players. Medicine and Science in Sports and Exercise, 32, Reilly, T, Williams, A. M., Nevill, A., & Franks, A. (2000). A
518 – 525. multidisciplinary approach to talent identification in soccer.
Nicholas, C. W., Nuttall, F. E., & Williams, C. (2000). The Journal of Sports Sciences, 18, 695 – 702.
Loughborough Intermittent Shuttle Test: A field test that Rienzi, E., Drust, B., Reilly, T., Carter, J. E. L., & Martin, A.
simulates the activity pattern of soccer. Journal of Sports Sciences, (2000). Investigation of anthropometric and work-rate profiles
18, 97 – 104. of elite South American international soccer players. Journal of
Oberg, B., Moller, M., Gillquist, J., & Ekstrand, J. (1986). Sports Medicine and Physical Fitness, 40, 162 – 169.
Isokinetic torque levels for knee extensors and knee flexors in Saltin, B. (1973). Metabolic fundamentals in exercise. Medicine
soccer players. International Journal of Sports Medicine, 17, 50 – and Science in Sports, 5, 137 – 146.
Downloaded by [Universidad Pablo de Olavide] at 08:55 22 November 2011

53. Sassi, R., Reilly, T., & Impellizzeri, F. (2003). A comparison of

Platt, D., Maxwell, A., Horn, R., Williams, M., & Reilly, T. small-sided games and interval training in elite professional
(2001). Physiological and technical analysis of 3 v 3 and 5 v 5 soccer players. Communication to the Fifth World Congress of
youth football matches. Insight: The FA Coaches Association Science and Football, Lisbon, 11 – 15 April.
Journal, 4(4), 23 – 24. Stratton, G., Reilly, T., Williams, A. M., & Richardson, D.
Puga, N., Ramos, J., Agostinho, J. Lomba, I., Costa, O., & de (2004). Science and youth soccer. London: Routledge.
Freitas, F. (1993). Physical profile of a First Division Portugese Strudwick, T., & Reilly, T. (2001). Work-rate profiles of elite
professional football team. In T. Reilly, J. Clarys, & A. Stibbe Premier League football players. Insight: The FA Coaches
(Eds.), Science and football II (pp.40 – 42). London: E & FN Association Journal, 4(2), 28 – 29.
Spon. Strudwick, A., Reilly, T., & Doran, D. (2002). Anthropometric
Rahnama, N., Reilly, T., & Lees, A. (2002). Injury risk associated and fitness profiles of elite players in two football codes. Journal
with playing actions during competitive soccer. British Journal of of Sports Medicine and Physical Fitness, 42, 239 – 242.
Sports Medicine, 36, 354 – 359. Verheijen, R. (2003). Periodisation in football: Preparing the
Rahnama, N., Reilly, T., Lees, A., & Graham-Smith, P. (2003). A Korean national team for the 2002 World Cup. Insight: The FA
comparison of musculoskeletal function in elite and sub-elite Coaches Association Journal, 6(2), 30 – 33.
English soccer players. In T. Reilly & M. Masrfell-Jones (Eds.), Wagenmakers , A. J. M., Brookes, J. H., Coakley, J. H., Reilly, T.,
Kinanthropometry VIII: Proceedings of the 8th International & Edwards, R. H. T. (1989). Exercise-induced activation of the
Conference of the International Society for the Advancement of branched chain 2-oxo acid dehydrogenase in human muscle.
Kinanthropometry (pp. 151 – 164). London: Routledge. European Journal of Applied Physiology, 59, 159 – 167.
Ramsbottom, R., Brewer, J., & Williams, C. (1988). A progressive Whall, R. (2001). Conditioning the goalkeeper: A scientific
shuttle run test to estimate maximal oxygen uptake. British approach. Insight: The FA Coaching Association Journal, 4(3),
Journal of Sports Medicine, 22, 141 – 144. 23 – 25.
Reilly, T. (1979). What research tells the coach about soccer. Reston, White, J. E., Emery, T. M., Kane, J. E., Groves, R., & Risman, A.
VA: AAHPERD. B. (1988). Pre-season fitness profiles of professional soccer
Reilly, T. (1994) Motion characteristics. In B. Ekblom (Ed.), players. In T. Reilly, A. Lees, K. Davids, & W. J. Murphy
Football (soccer) (pp. 31 – 42). Oxford: Blackwell Scientific. (Eds.), Science and football (pp. 164 – 171). London: E & FN
Reilly, T. (1997). Energetics of high intensity exercise (soccer) Spon.
with particular reference to fatigue. Journal of Sports Sciences, 15, Williams A. M., Lee, D., & Reilly, T. (1999). A quantitative
257 – 263. analysis of matches played in the 1991 – 92 and 1997 – 98 seasons.
Reilly, T. (2003). Motion analysis and physiological demands. In London: The Football Association.
T. Reilly & A. M. Williams (Eds.), Science and soccer (pp. 59 – Williams, A. M., & Reilly, T. (2000). Talent identification and
72). London: Routledge. development in soccer. Journal of Sports Sciences, 18, 657 – 667.
Reilly, T., & Ball, D. (1984). The net energetic cost of dribbling a Wilson, D. (2001a). Complex training for performance enhance-
soccer ball. Research Quarterly for Exercise and Sport, 55, 267 – ment and injury prevention. Insight: The FA Coaches Association
271. Journal, 4(2), 56.
Reilly, T., & Bangsbo, J. (1998). Anaerobic and aerobic training. Wilson, D. (2001b). The physiological basis of speed endurance.
In B. Elliott (Ed.), Training in sport: Applying sport science Insight: The FA Coaches Association Journal, 4(3), 36 – 37.
(pp.351 – 409). Chichester: Wiley. Wisløff, U., Castagna, C., Helgerud, J., Jones, R., & Hoff, J.
Reilly, T., & Bowen, T (1984). Exertional costs of changes in (2004). Strong correlation of maximal squat strength with
directional modes of running. Perceptual and Motor Skills, 58, sprint performance and vertical jump height in elite soccer
149 – 150. players. British Journal of Sports Medicine, 38, 285 – 288.
Reilly, T., Dowzer, C. N., & Cable, N. T. (2003). The physiology Wisloff, U., Helgerud, J., & Hoff, J. (1998). Strength and
of deep-water running. Journal of Sports Sciences, 21, 959 – 972. endurance of elite soccer players. Medicine and Science in Sports
Reilly, T., & Thomas, V. (1976). A motion analysis of work-rate in and Exercise, 30, 462 – 467.
different positional roles in professional football match-play. Withers, R., Maricic, Z., Wasilewski, S., & Kelly, L. (1982).
Journal of Human Movement Studies, 2, 87 – 97. Match analysis of Australian professional soccer players. Journal
Reilly, T., & Thomas, V. (1979). Estimated daily energy of Human Movement Studies, 8, 159 – 176.
expenditures of professional association footballers. Ergonomics,
22, 541 – 548.