Physiology of Small-Sided Games Training in Football · Physiology of Small-Sided Games Training in Football ... sures are lower in smaller game formats ... logical responses associated

Physiology of Small-Sided GamesTraining in FootballA Systematic Review

Stephen V. Hill-Haas,1 Brian Dawson,1 Franco M. Impellizzeri2,3 and Aaron J. Coutts4

1 School of Sports Science, Exercise & Health, University of Western Australia, Perth, Western Australia,

Australia

2 Neuromuscular Research Laboratory, Schulthess Clinic, FIFA Centre of Excellence, Zurich, Switzerland

3 Research Centre for Sport, Mountain and Health (CSMS) of Rovereto, University of Verona, Verona Italy

4 School of Leisure, Sport & Tourism, University of Technology, Lindfield, New South Wales, Australia

Contents

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1991. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2002. Small-Sided Games (SSGs) in Football . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201

2.1 Quantifying Exercise Intensity During SSGs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2012.2 Time-Motion Measurement in SSGs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2022.3 Variables Affecting SSG Intensity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202

2.3.1 Pitch Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2032.3.2 Player Number. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2032.3.3 Concurrent Manipulation of Pitch Area and Player Number . . . . . . . . . . . . . . . . . . . . . . . . 2062.3.4 Rule Modifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2072.3.5 Goalkeepers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2082.3.6 Training Regimen (Including Game Duration and Work : Rest Ratios) . . . . . . . . . . . . . . . . . 2092.3.7 Coach Encouragement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2112.3.8 Logistics and Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2122.3.9 Comparisons of SSG Training Intensity with Competitive Match Play . . . . . . . . . . . . . . . . . 213

3. Studies Comparing SSGs Training with Interval Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2133.1 Acute Physiological Comparisons of SSGs Training with Interval Training . . . . . . . . . . . . . . . . . . . 2133.2 Training Studies Comparing SSGs Training with Interval Training . . . . . . . . . . . . . . . . . . . . . . . . . . . 214

4. Limitations of SSGs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2165. Future Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2176. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218

Abstract Small-sided games (SSGs) are played on reduced pitch areas, often usingmodified rules and involving a smaller number of players than traditionalfootball. These games are less structured than traditional fitness trainingmethods but are very popular training drills for players of all ages and levels.At present, there is relatively little information regarding how SSGs can bestbe used to improve physical capacities and technical or tactical skills infootballers. However, many prescriptive variables controlled by the coachcan influence the exercise intensity during SSGs. Coaches usually attempt tochange the training stimulus in SSGs through altering the pitch area, player

REVIEW ARTICLESports Med 2011; 41 (3): 199-220

0112-1642/11/0003-0199/$49.95/0

ª 2011 Adis Data Information BV. All rights reserved.

number, coach encouragement, training regimen (continuous vs intervaltraining), rules and the use of goalkeepers. In general, it appears that SSGexercise intensity is increased with the concurrent reduction in player numberand increase in relative pitch area per player. However, the inverse relation-ship between the number of players in each SSG and exercise intensity doesnot apply to the time-motion characteristics. Consistent coach encourage-ment can also increase training intensity, but most rule changes do not appearto strongly affect exercise intensity. The variation of exercise intensity mea-sures are lower in smaller game formats (e.g. three vs three) and have ac-ceptable reproducibility when the same game is repeated between differenttraining sessions or within the same session. The variation in exercise intensityduring SSGs can also be improved with consistent coach encouragement butit is still more variable than traditional generic training methods. Other stud-ies have also shown that SSGs containing fewer players can exceed matchintensity and elicit similar intensities to both long- and short-duration high-intensity interval running. It also appears that fitness and football-specificperformance can be improved equally with SSG and generic training drills.Future research is required to examine the optimal periodization strategies ofSSGs training for the long-term development of physiological capacity,technical skill and tactical proficiency.

1. Introduction

The main purpose of this review is to provide asummary of the research that has examined thephysiological and performance benefits of small-sided games (SSGs) in football. The review ispresented in six sections. The first section brieflydescribes the origins, definition and advantagesof SSGs. The second section reviews the use ofSSGs in football. The three aspects addressedin the second section include findings from stud-ies that have examined (i) the validity and relia-bility of quantifiable exercise intensity measuresin SSGs; (ii) time-motion analysis of SSGs usingGlobal Positioning System (GPS) technology; and(iii) variables affecting SSGs training intensity.

The third section contains two parts. Part Aexamines studies that describe the acute physio-logical responses associated with various SSGsand part B examines training studies that com-pare the effectiveness of using both interval andSSGs training for conditioning. The fourth sec-tion describes the limitations of SSGs as a fitnesstraining mode. The final two sections providesuggestions for future research and conclude thereview.

The articles reviewed here were acquired bysearching the electronic databases of AUSPORT,ProQuest 5000, PubMed, SPORTDiscus� andGoogle Scholar. The following keywords wereused in various combinations: ‘small-sided soccergames’, ‘small-sided football games’, ‘metabolicconditioning’, ‘soccer-specific conditioning’,‘football-specific conditioning’, ‘skill-based train-ing’, ‘skill-based conditioning’, ‘soccer training’,‘football training’ and ‘game-based training’.Due to the focus on football, this reduced thenumber of articles retrieved and, consequently,no limit to the search period was applied. Elec-tronic database searching was supplementedby examining the bibliographies of relevantarticles.

This review is justified, given the increasingamount of research conducted into SSGs infootball. It represents a useful synthesis of all re-search into SSGs in football, and helps to identifyareas for future research, including the in-vestigation of the technical load and tacticaltransfer of SSGs to match performance. Finally,it serves to further establish SSGs training as analternative conditioning method for footballplayers.

200 Hill-Haas et al.

ª 2011 Adis Data Information BV. All rights reserved. Sports Med 2011; 41 (3)

2. Small-Sided Games (SSGs) in Football

SSGs, also referred to as skill-based condition-ing games[1] or game-based training,[2] aremodifiedgames played on reduced pitch areas, often usingadapted rules and involving a smaller number ofplayers than traditional football games. Formal-ized SSGs, such as those implemented in footballclubs throughout the world and which underpinmany junior football development programmes(e.g. Royal Dutch Football Association, FootballFederation Australia), appear to have evolvedfrom informal unstructured games of street foot-ball. Indeed, many of the world’s top players wereintroduced to football informally, via street, parkor beach football.[3] Although it is still commonto observe informal football SSGs being playedin the street, park or beach, a structured ap-proach to SSGs training has been adopted in theclub setting.[3]

SSGs in football are widely considered to offermany practical advantages that have lead to itspopularity as a training modality in football at allages and levels. The primary benefits of SSGs arethat they appear to replicate the movement de-mands, physiological intensity and technical re-quirements of competitive match play,[4-7] whilstalso requiring players to make decisions underpressure and fatigue.[8] SSGs have also beensuggested to facilitate the development of tech-nical skills and tactical awareness within the ap-propriate context of a game.[7,9] Compared withtraditional fitness training sessions, SSGs arethought to increase player compliance and moti-vation, since it is perceived to be sport specific.[6,7]

Finally, SSGs are considered to be more time ef-ficient, as physical performance, technical skillsand tactical awareness, can be developed con-currently.[6,7] However, the realization of theseadvantages is dependent on game design.

2.1 Quantifying Exercise Intensity During SSGs

Exercise intensity in SSGs has typically beenassessed via heart rate (HR), blood lactate con-centration and rating of perceived exertion(RPE). Indeed, HR is the most common measureused for objectively monitoring training intensity

in many sports,[10] and several studies have shownHR to be a valid indicator of exercise intensity infootball.[11,12] For example, the mean HR andoxygen consumption (

.VO2) relationship have

been reported to be similar during treadmill-based intermittent exercise that reproduced thedemands of a football game.[11] Similarly, severalstudies have shown that the HR/

.VO2 relationship

established in the laboratory is similar to theHR/

.VO2 relationship measured at different in-

tensities during football-specific exercises (five vsfive SSGs).[12-14] Collectively, the findings in-dicate that HR is a valid measure of exercise in-tensity during football.

There are, however, some limitations to usingHR to assess exercise intensity during football-specific activities. For example, it has been sug-gested that factors inherent in football training,including emotion and the intermittent nature ofthe activity, may result in HR values that over-estimate actual energetic cost of exercise.[15] Incontrast, there is also evidence showing that HRmonitoring may underestimate the intensity offootball drills that have a high anaerobic com-ponent, including short-duration SSGs involvingfew players (e.g. 2-minute bouts; two vs two).[16]

Therefore, it seems that other measures of ex-ercise intensity may provide a more appropriatemeasure of exercise intensity during SSGs.

Blood lactate, a by-product of anaerobic gly-colysis, has been extensively used as an indicatorof exercise intensity in football. The blood lactateconcentration has been suggested to represent anoverall accumulation of lactate productionduring football-specific exercise.[17] However, gi-ven the intermittent nature of football, bloodlactate concentration is a poor indicator ofmuscle lactate concentration during footballmatch play[17] and, consequently, may be mis-representative of individual exercise intensities.

In contrast to blood lactate concentration,RPE is a simple, non-invasive and inexpensivemethod of monitoring exercise intensity.[18] Sev-eral studies have shown that RPE can be validlyused to assess exercise intensity at a specific timeduring exercise[19] and as a global indicator ofoverall session intensity (session RPE).[20,21] Forexample, to validate RPE as a measure of exercise

Small-Sided Games Training Physiology in Football 201


intensity during football SSGs, Coutts et al.[19]

examined the relationship between RPE withboth HR and blood lactate concentration mea-sures. The findings of this study demonstratedthat the combination of HR and blood lactateconcentration predicted RPE better than HR orblood lactate concentration measures alone.Therefore, it was suggested that RPE may be amore valid marker of global exercise intensitythan any physiological measures independently.

Similarly, other studies have assessed the va-lidity of the session RPE for assessing exerciseintensity in football-specific exercise.[20,21] Thesession RPE method requires that players pro-vide a single RPE relating to the exercise intensityof the entire session, usually 30 minutes followingexercise.[22] However, although several studieshave reported that session RPE is a valid in-dicator of overall perception of effort for inter-mittent aerobic football-specific exercises (includingSSGs) training, it may not be a valid substitutefor HR-based methods.[20,21] Nonetheless, due itspsychobiological foundations,[18] session RPEmeasures may be a more valid global measure ofexercise intensity during high-intensity inter-mittent exercise such as SSGs.

However, all the methods currently availableto assess exercise intensity during SSGs do havelimitations. There is no clear evidence to suggestthat one particular method is superior to theothers. The methodology chosen may depend onwhat the variable of interest is. Therefore, on thebasis of studies examining the validity of HR,RPE and blood lactate concentration duringfootball-specific training, it has been suggestedthat SSGs training is best monitored via a com-bination of each of these measures of internalexercise intensity.[19]

2.2 Time-Motion Measurement in SSGs

In addition to physiological measures of ex-ercise intensity during SSGs, recent technologicaladvances now allow for movement characteristicsof football players to be collected.[23] This in-formation may be used to design game-relatedconditioning activities.[23] Specifically, GPS micro-technology is now used by various professional

football codes to quantify themovement demandson players during training and games.[23]

The validity and reliability of the measuresprovided by these commercially available (non-differential) GPS receivers has recently beendescribed.[24-28] In general, the error for total dis-tance travelled (metres/minute) has been reportedto be between 3% and 5%.[24] Moreover, the cor-relations between speed measured by electronictiming gates and values obtained fromGPS units,have also been reported to be very high.[25,29]

However, there are several limitations associatedwith this technology, including reduced reliabilitywith increased movement speeds.[24,28] For ex-ample, the coefficient of variation (CV) for high-intensity running (>14.4 km/h) is reported as11.2–32.4%, and 11.5–30.4% for very high-intensity running (>20.0 km/h).[24] Moreover,lower sampling rates (i.e. 1 vs 5Hz) may also be alimitation, as this may reduce the devices abilityto detect changes in direction at high speed.[24,28]

Other limitations, including the number of satellitesavailable fromwhich to collect data, as well as theinability to sample data indoors, should also beconsidered. However, despite these limitations,the information obtained from these devices,specifically measures of exercise intensity such astotal distance, distance covered in wide speedzones (i.e. speed zones that include a wide rangeof velocities) and peak velocity, may still provideuseful data regarding variations in movementdemands in the various SSGs.

2.3 Variables Affecting SSG Intensity

The exercise intensity of SSGs can be demon-strated through a player’s movement and/orphysiological/perceptual responses. Many pre-scriptive variables that can be controlled by thecoach may influence the exercise intensity duringSSGs.[30] These factors include pitch area, playernumber, coach encouragement, training regimen(continuous or interval, including work : restmanipulations) rule modifications, and the use ofgoals and/or goalkeepers.[7,31] The following sec-tion will review how each of these factors havebeen manipulated to alter the exercise intensityduring football SSGs.



Despite the recent increases in the number ofresearch studies that have investigated the influ-ence of adjusting each of these variables upon theexercise intensity in SSGs, caution should be ap-plied when interpreting the practical suitabilityof a specific SSG on the basis of a statisticalobservation. It has been suggested that the smallbut significant changes in isolated physiologicalvariables between the various SSG designs mayhave a relatively minor influence on trainingadaptation.[32] Nonetheless, it is possible thatwhen an alteration in SSG design elicits changesin a combination of physiological variables to-gether (e.g. blood lactate concentration and HR),that a vastly different training response may beelicited. Accordingly, because of these complexinteractions, it is important that coaches andscientists carefully interpret changes in the phy-siological responses to various SSG designs in thecontext of the global response, rather than simplyon the basis of a statistical assessment of singlephysiological variables.

2.3.1 Pitch Area

The total pitch area, both in absolute and re-lative terms, can be varied, and this may influencethe intensity of SSGs. The relative pitch area perplayer is defined as the total pitch area divided bythe total number of players. Table I is a summaryof all the studies that examined the effect of ma-nipulating absolute and relative pitch area (whilekeeping the number of players constant) on SSGintensity. The majority of studies report an in-creased HR, RPE and blood lactate concentrationresponse with increased pitch area. For example,Rampinini et al.[32] increased the pitch area by20% across a variety of SSG formats (three vsthree to six vs six, inclusive). Both the percentageof maximum HR (%HRmax) and blood lactateconcentration were higher during SSGs playedon a large pitch than on a medium-sized or smallpitch. RPE was also higher on medium and largepitch sizes compared with small pitches.[32]

2.3.2 Player Number

The number of players on each team in a SSGcan also be altered to regulate the intensity of thistraining mode. Studies that have investigated the

effect of altering player number on SSGs trainingintensity have altered player numbers while, atthe same time, held many other factors constant,including the pitch area. A summary of all thestudies that examined the effect of altering playernumbers on SSG intensity is presented in table II.

In summary, despite some methodologicalconcerns (very short game duration; differingwork : rest ratios), most studies have shown thatSSGs containing smaller numbers of players elicitgreater HR, blood lactate and perceptual re-sponses.[34,35,37,39] On closer analysis, the resultssuggest the possible existence of a threshold pitcharea. For example, the most pronounced reduc-tions in HR occurred when two versus two wasincreased to three versus three, and three versusthree was increased to four versus four, on a25 · 20 metre pitch area. In contrast, less pro-nounced reductions in HR occurred when twoversus two was increased to three versus three,and three versus three was increased to four ver-sus four on 20 · 15 metre and 30 · 25 metre pitchareas, respectively.[34]

As illustrated in table II, these previous studiesonly examined the influence of altering the playernumbers on teams containing equal numbers ofplayers (e.g. two vs two or three vs three). Intraining situations, SSGs are often implementedthat contain teams of unequal numbers (e.g. fourvs three players or six vs five). Reasons forcreating an imbalance between opposing teamsmay include technical development and unavail-ability of players due to injury. A further varia-tion in player number involves creating tempor-ary ‘overload’ and ‘underload’ situations betweenopposing teams, via the use of a ‘floater’ player.This neutral player transitions to the team inpossession of the ball, to create temporary‘overload’ and ‘underload’ situations. This SSGgame design is typically used to develop defensiveor attacking proficiency or to increase the physi-cal load on the ‘floating’ player.

The impact of creating fixed and temporary‘overload’ and ‘underload’ situations (includingthe use of a ‘floater’) on the physiological, per-ceptual and time-motion responses in SSGs in-volving elite youth football players have recentlybeen investigated.[38] The findings from this study



were that there were no significant differencesbetween the fixed (four vs three or six vs five) andvariable (three vs three +one floater or five vsfive +one floater) SSGs in terms of physiologicaland perceptual responses (see table II). Despitethis, either may provide a useful SSGs trainingvariation, or as a technical/tactical trainingmethod for defensive, transition and attackingplays. The possibility of fixed and variable SSGs

providing a greater technical load needs to beexamined by further research. Finally, the use ofa floater appears to be more effective in SSGscontaining fewer players (e.g. three vs three + onefloater), and may be appropriate for eithermaintaining or developing aerobic fitness.[38] Forexample, the floater travelled a significantlygreater total distance and recorded a greater RPEcompared with four-player teams in fourversus

Table I. Summary of studies examining the effects of pitch dimensions on small-sided game intensity in football players

Study Sample

size

Game

design

Training

prescription

Pitch

dimensions

(m)

Area per

playera

(m2)

%HRmax

[mean – SD]b[BLa-]

(mmol/L)

[mean – SD]

RPE (6–20 AU)c

[mean – SD]

Aroso

et al.[33]

14 4 vs 4 3 · 6 min/90 s

rest

30 · 20

50 · 30

75

188

70.0 – 9.0

-2.6 – 1.7

› (no value)

13.3 – 0.9

› (no value)

Owen

et al.[34]

13 1 vs 1

2 vs 2

3 vs 3

4 vs 4

5 vs 5

1 · 3 min/12 min

rest

10 · 5

15 · 10

20 · 15

15 · 10

20 · 15

25 · 20

20 · 15

25 · 20

30 · 25

25 · 20

30 · 25

30 · 25

35 · 30

40 · 35

25

75

150

38

75

125

50

83

125

63

94

75

105

140

86.0

88.0

89.0

84.2

87.4

88.1

81.7

81.8

84.8

72.0

78.5

75.7

79.5

80.2

--------------

--------------

Williams

and

Owen[35]

9 3 vs 3 - 20 · 15

25 · 20

30 · 25

50

83

125

164 – 12d (mean HR)

166 – 9d (mean HR)

171 – 11d (mean HR)

–

–

–

–

–

–

Rampinini

et al.[32]

20 3 vs 3 (CE)

4 vs 4 (CE)

5 vs 5 (CE)

6 vs 6 (CE)

3 · 4 min/3 min

rest

20 · 12

25 · 15

30 · 18

24 · 16

30 · 20

36 · 24

28 · 20

35 · 25

42 · 30

32 · 24

40 · 30

48 · 36

40

63

90

48

75

108

56

88

126

64

100

144

89.5 – 2.9

90.5 – 2.3

90.9 – 2.0

88.7 – 2.0

89.4 – 1.8

89.7 – 1.8

87.8 – 3.6

88.8 – 3.1

88.8 – 2.3

86.4 – 2.0

87.0 – 2.4

86.9 – 2.4

6.0 – 1.8

6.3 – 1.5

6.5 – 1.5

5.3 – 1.9

5.5 – 1.8

6.0 – 1.6

5.2 – 1.4

5.0 – 1.7

5.8 – 1.6

4.5 – 1.5

5.0 – 1.6

4.8 – 1.5

8.1 – 0.6 (CR10)

8.4 – 0.4 (CR10)

8.5 – 0.4 (CR10)

7.6 – 0.5 (CR10)

7.9 – 0.5 (CR10)

8.1 – 0.5 (CR10)

7.2 – 0.9 (CR10)

7.6 – 0.6 (CR10)

7.5 – 0.6 (CR10)

6.8 – 0.6 (CR10)

7.3 – 0.7 (CR10)

7.2 – 0.8 (CR10)

Kelly and

Drust[36]

8 5 vs 5 (CE) 4 · 4 min/2 min

rest

30 · 20

40 · 30

50 · 40

60

120

200

91.0 – 4.0

90.0 – 4.0

89.0 – 2.0

---

---

a Total pitch area divided by total number of players.

b Data for Owen et al.[34] are presented as mean values.

c RPE is 6–20 AU unless otherwise stated.

d Age predicted HR values, mean – SD.

AU = arbitrary units; [BLa-] = blood lactate concentration; CE = with coach encouragement; CR10 = category ratio 10 scale; HR = heart rate;

%HRmax = percentage of maximum HR; RPE = rating of perceived exertion; ›› indicates increase; - indicates no data.



Tab

leII

.S

um

mary

ofstu

die

sexam

inin

gth

eeff

ects

ofpla

yer

num

ber

on

sm

all-

sid

ed

gam

ein

tensity

info

otb

all

pla

yers

Stu

dy

Sam

ple

siz

eG

am

edesig

nT

rain

ing

pre

scription

Pitch

dim

ensio

ns

(m)

Are

aper

pla

yera

(m2)

%H

Rm

ax

[mean

–S

D]

[BLa

- ](m

mol/L

)

[mean

–S

D]

RP

E(6–20

AU

)b

[mean

–S

D]

Aro

so

etal.

[33]

14

2vs

2

3vs

3

4vs

4

3·

1.5

min

/90

sre

st

3·

4m

in/9

0s

rest

3·

6m

in/9

0s

rest

30

·20

30

·20

30

·20

150

100

75

84.0

–5.0

87.0

–3.0

70.0

–9.0

8.1

–2.7

4.9

–2.0

2.6

–1.7

16.2

–1.1

14.5

–1.7

13.3

–0.9

Ow

en

etal.

[34]

13

1vs

1

2vs

2

1vs

1

2vs

2

3vs

3

2vs

2

3vs

3

4vs

4

4vs

4

5vs

5

1·

3m

in/1

2m

inre

st

15

·10

15

·10

20

·15

20

·15

20

·15

25

·20

25

·20

25

·20

30

·25

30

·25

75

38

150

75

50

125

83

63

94

75

88.0

c

84.2

c

89.0

c

87.4

c

81.7

d

88.1

c

81.8

d

72.0

e

78.5

e

75.7

e

- - - - - - - - - -

- - - - - - - - - -

Sam

paio

etal.

[37]

82

vs

2

3vs

3

2·

1.5

min

/90

sre

st

2·

3m

in/9

0s

rest

30

·20

30

·20

150

100

83.7

–1.4

80.8

–1.7

- -15.5

–0.6

15.8

–0.2

Will

iam

sand

Ow

en

[35]

91

vs

1

2vs

2

3vs

3

2vs

2

3vs

3

4vs

4

3vs

3

4vs

4

5vs

5

- - - - - - - - -

20

·15

20

·15

20

·15

25

·20

25

·20

25

·20

30

·25

30

·25

30

·25

150

75

50

125

83

63

125

94

75

183

–7

(mean

HR

)

179

–7

(mean

HR

)

164

–12

(mean

HR

)

180

–5

(mean

HR

)

166

–9

(mean

HR

)

152

–14

(mean

HR

)

171

–11

(mean

HR

)

165

–5

(mean

HR

)

152

–6

(mean

HR

)

- - - - - - - - -

- - - - - - - - -

Hill

-Haas

etal.

[38]

12

16 8

20

24 4

3pla

yers

4pla

yers

Flo

ate

r

5pla

yers

6pla

yers

Flo

ate

r

24

f

24

f

24

f

24

f

24

f

24

f

37

·28

37

·28

37

·28

47

·35

47

·35

47

·35

148

148

148

149

149

149

82.3

–3.5

2543

–187

(TD

m)

83.1

–4.0

2408

–231

(TD

m)

82.7

–3.0

2668

–220

(TD

m)

82.5

–5.0

2526

–302

(TD

m)

81.4

–5.1

2524

–247

(TD

m)

82.5

–5.6

2610

–201

(TD

m)

2.5

–0.7

553

–187

(Dm

)

2.5

–0.9

482

–178

(Dm

)

2.3

–0.8

628

–132

(Dm

)

2.5

–1.0

649

–190

(Dm

)

2.6

–1.1

589

–177

(Dm

)

2.8

–0.2

673

–194

(Dm

)

16.3

–1.6

10

–6

(SP

)

14.6

–1.9

8–

4(S

P)

16.3

–1.5

9–

6(S

P)

15.2

–1.0

9–

5(S

P)

14.9

–0.9

8–

4(S

P)

16.3

–1.7

15

–3

(SP

)

Continued

nextpage



three-player games[38] (figure 1). The floater alsocompleted a significantly greater amount of sprints(>18km/h) comparedwith five- and six-player teamsin sixversus five-player games[38] (see table II).

2.3.3 Concurrent Manipulation of Pitch Areaand Player Number

Few studies have systematically examined theinfluence of the concurrent manipulation of pitcharea and player number on exercise intensity inSSGs.[32,40-42] In addition, there are several dif-ferences in the design and prescription of theSSGs in the studies that inadvertently manipu-lated both player number and pitch area, makingcomparisons between these studies very difficult.Indeed, tables III and IV show that there aresubtle differences in the training prescriptions,age and ability of players, intensity measures andsizes in pitch area amongst the studies, all of whichmay affect the exercise intensity in these SSGs.

In general, it appears that a concurrent in-crease in player number and relative pitch areaper player in SSGs elicits lower exercise intensity.For example, Rampinini et al.[32] investigated theeffects of concurrently increasing the player num-ber and pitch area on %HRmax, blood lactateconcentration and RPE in 20 amateur footballplayers. The main finding of this study was thatthe exercise intensity during all game formats wasdecreased when there was an increase in the numberof players and more pitch area per player[32] (seetable III). Similarly, Jones and Drust[41] also re-ported a reduction in %HRmax when both playernumber and pitch areawere increased (see table III).

One important aspect that has not been con-sidered by studies where both pitch size andplayer number were altered concurrently was theinfluence of the relative pitch area per play-er.[16,32,41-44] In all of these studies, an increase inabsolute pitch area and player number also re-sulted in a greater relative pitch area per player.Therefore, the observed reduction in SSG intensityby several of these studies[32,41,42] may have beendue to either the independent effects of increasingthe number of players or the inability of the addi-tional players to cover more of the available pitcharea. Clearly, more research is required to deter-mine the effect of an increase in player number onT

ab

leII

.C

ontd

Stu

dy

Sam

ple

siz

eG

am

edesig

nT

rain

ing

pre

scription

Pitch

dim

ensio

ns

(m)

Are

aper

pla

yera

(m2)

%H

Rm

ax

[mean

–S

D]

[BLa

- ](m

mol/L

)

[mean

–S

D]

RP

E(6–20

AU

)b

[mean

–S

D]

Matc

hed

PN

g

Overload

PN

Underload

PN

3vs

3and

5vs

5

6pla

yer

and

4pla

yer

team

s

5pla

yer

and

3pla

yer

team

s

82.5

–4.6

2585

–204

(TD

m)

82.3

–4.5

2458

–243

(TD

m)

82.3

–4.0

2535

–247

(TD

m)

2.6

–1.1

582

–190

(Dm

)

2.6

–1.0

528

–184

(Dm

)

2.6

–1.0

598

–192

(Dm

)

15.2

–1.4

- 14.7

–1.5

- 15.8

–1.5

-a

Tota

lpitch

are

adiv

ided

by

tota

lnum

ber

ofpla

yers

.

bR

PE

is6–20

AU

unle

ss

oth

erw

ise

sta

ted.

cin

tensity

>11

vs

11

com

petitive

matc

h.

din

tensity

=11

vs

11

com

petitive

matc

h.

ein

tensity

<11

vs

11

com

petitive

matc

h.

fG

am

edura

tion

(min

).

gM

atc

hed

team

exclu

din

gth

efloate

r.

AU

=arb

itra

ryunits;[

BL

a- ]

=blo

od

lacta

teconcentr

ation;D

=dis

tance

(m):

>13.0

km

/h;H

R=

heart

rate

;%H

Rm

ax

=perc

enta

ge

ofm

axim

um

HR

;PN

=pla

yernum

bers

;RP

E=

rating

of

perc

eiv

ed

exert

ion;S

P=

num

ber

ofsprints

>18.0

km

/h;T

D=

tota

ldis

tance

(m);

-in

dic

ate

sno

data

.



SSG intensity (or vice versa). However, it is im-portant that future studies control for the influenceof relative pitch area per player so that an im-proved understanding of increasing pitch area andplayer number in SSGs can be obtained.

More recently, a study involving youth foot-ball players examined the acute physiological andperceptual responses and time-motion char-acteristics during three variations of SSGs (two vstwo, four vs four and six vs six) with a constantratio of player number to pitch area applied toeach SSG variation.[45] The main findings were,as the number of players in the SSG teams de-creased, when the relative pitch area per playerremained constant, the overall physiological andperceptual responses increased. Notably, the in-verse relationship between the number of playersin each SSG and exercise intensity did not extendto the time-motion characteristics. In general, thelargest game format (six vs six) was associatedwith a greater range of distances travelled atspeeds >18 km/h. In contrast, the four versus fourformat, compared with the two versus two, was

characterized by significantly longer (average andmaximal) effort durations and distances forspeeds >18 km/h.[45] However, since it is theinternal response to training (e.g. HR and RPE)and not the external training load (e.g. distancestravelled in speed zones) that determines eachplayers adaptation to a training stimulus,[46] it isrecommended that each player’s internal load bemonitored to assess how players are coping withdifferent SSGs design (see table IV).

2.3.4 Rule Modifications

In practice, football coaches quite often mod-ify playing rules in SSGs to achieve greater ex-ercise intensity, or develop specific technical andtactical skills. However, there have only been a fewstudies that have examined how the modificationof rules can influence these variables. Table Vprovides a summary of studies that have investigatedthe effects of rule changes on exercise intensityduring football SSGs. Two studies[47,48] reportedan increase in %HRmax and another reported anincrease in blood lactate concentration due torule changes[33] (table V). Simple rule changeshave also been reported to increase the percep-tion of effort[37] (table V), which may be due tothe increased cognitive load required of players asa consequence of new rules. To date, the onlystudy to have reported on the influence of rulechanges on movement characteristics is by Malloand Navarro.[48] Compared with normal footballrules, these specific rule changes resulted in anincrease in total distance travelled (table V) andtime spent performing high-intensity running,with less spent time spent stationary.[33,48]

Although these simple rule modifications re-late to technical aspects of the game, other studieshave investigated the influence of providing ‘ar-tificial’ changes.[38] An example of an artificialrule change is the requirement for a player tocomplete a series of sprints of planned dura-tion during a SSG. Hill-Haas et al.[38] recentlyexamined the acute physiological responses andtime-motion characteristics associated with fourdifferent rule changes, including the addition of‘artificial’ rules. The main finding was thatchanges in SSG playing rules can influence thephysiological and time-motion responses, but not

3000a

b

TD

(m

)R

PE

(6−

20 A

U)

2750

2500

2250

2000

20

18

16

14

12

3 Players 4 Players Floater

Fig. 1. Comparison of (a) total distance (TD; m) and (b) rating ofperceived exertion (RPE) [6–20 arbitrary units; AU] with ‘floating’players and other players in various smaller game formats.[38]



perceptual responses, in young elite footballplayers (table V).[38] The artificial rule changethat required players to complete extra sprint ef-forts around the pitch during each SSG at pre-settimes, imposed a greater external training load onthe players, but did not affect HR, blood lactateconcentration or RPE. In contrast, changes intechnical rules that were related to a team’s chancesof scoring, may have improved player motivationand thereby increased the exercise intensityduring the SSGs.[38] Although there have beenrelatively few studies that have examined the in-fluence of rule modifications on exercise intensityduring SSGs, the rule changes that have been in-vestigated are by no means exhaustive. To date,the rule changes that have been investigated havealtered either the physiological and/or perceptualresponses, as well as the time-motion character-istics of various SSGs. However, this may not bethe case for all types of rule changes that couldpossibly be implemented. Future studies shouldaim to more systematically classify the types ofrules changes that appear to have differential ef-fects on physiological, perceptual and time-motion

responses during SSGs. Future studies shouldexamine the effect of common rules modificationson the technical and tactical skills of football play-ers. Factors such as decision making and cognitiveload of players should also be assessed (table V).

2.3.5 Goalkeepers

One common rule modification in SSGs is theremoval of goalkeepers from the game in an at-tempt to increase the number of goals scored.Goalkeepers are an integral part of football;however, surprisingly few studies have investigatedthe use of goalkeepers and their possible effecton SSGs training intensity. Table VI provides asummary of the SSGs studies that investigatedthe effects of goalkeepers on SSG intensity. Malloand Navarro[48] reported a significant decrease in%HRmax, total distance and time spent in high-intensity running, in three versus three SSGs withgoalkeepers. It was suggested that the reducedphysiological and time-motion responses weredue to increased defensive organization near thegoal area, which reduced the tempo of play andsubsequently the physiological and time-motion

Table III. Summary of studies examining the effects of concurrent changes in player number and pitch dimensions on small-sided game

intensity in football players

Study Sample

size;

age (y)

Game

design

Training

prescription

Pitch

dimensions

(m)

Area per

playera

(m2)

%HRmax

[mean – SD]b[BLa-] (mmol/L)

[mean – SD]

RPE (6–20 AU)c

[mean – SD]

Platt

et al.[43]

2; 10–12 3 vs 3

5 vs 5

1 · 15 min continuous


27 · 18

37 · 27

81

100

88.0d

82.0d

--

--

Little and

Williams[16]

28; NR 2 vs 2

3 vs 3

4 vs 4

5 vs 5

6 vs 6

8 vs 8

4 · 2 min/2 min rest

4 · 3.5 min/90 s rest


4 · 6 min/90 s rest



27 · 18

32 · 23

37 · 27

41 · 27

46 · 27

73 · 41

122

123

125

111

104

187

88.9 – 1.2

91.0 – 1.2

90.1 – 1.5

89.3 – 2.5

87.5 – 2.0

87.9 – 1.9

9.6 – 1.0

8.5 – 0.8

9.5 – 1.1

7.9 – 1.7

5.6 – 1.9

5.8 – 2.1

16.3 – 0.9

15.7 – 1.1

15.3 – 0.7

14.3 – 1.5

13.6 – 1.0

14.1 – 1.8

Jones and

Drust[41]

8; 7 4 vs 4

8 vs 8



30 · 25

60 · 40

94

150

83.0

79.0

--

--

Rampinini

et al.[32]

20; NR 3 vs 3 (CE)

4 vs 4 (CE)

5 vs 5 (CE)

6 vs 6 (CE)

3 · 4 min/3 min rest 30 · 18

36 · 24

42 · 30

48 · 36

90

108

126

144

90.9 – 2.0

89.7 – 1.8

88.8 – 2.3

86.9 – 2.4

6.5 – 1.5

6.0 – 1.6

5.8 – 1.6

4.8 – 1.5

8.5 – 0.4 (CR10)

8.1 – 0.5 (CR10)

7.5 – 0.6 (CR10)

7.2 – 0.8 (CR10)


b Data for Platt et al.[43] and Jones and Drust[41] are presented as mean values.

c RPE is 6–20 AU unless otherwise stated.

d Age predicted heart rate values.

AU = arbitrary units; [BLa-] = blood lactate concentration; CE = coach encouragement; CR10 = category ratio 10 scale; %HRmax = percentage

of maximum heart rate; NR = not reported; RPE = rating of perceived exertion; - no data.



responses.[48] In contrast, Dellal et al.[44] reporteda 12% increase in heart rate response in eightversus eight SSGs with goalkeepers. The presenceof goalkeepers may have increased the player’smotivation to both attack and defend, therebyincreasing the physiological load.[44] At present,the influence of goalkeepers on exercise intensityin football SSGs is not clear. They may have animportant role in keeping team structures andformations intact, as well as increasing commu-nication, all of which may influence movement,skill and physiological demands. Future studiesare required to determine the influence of goalkeepers on the physiological and technical/tac-tical demands in SSGs.

2.3.6 Training Regimen (Including Game Durationand Work : Rest Ratios)

Similar to interval running, many prescriptivevariables can be used in SSGs to alter exercise

intensity. The majority of the studies have useda traditional ‘interval’ training format, wherebyseveral consecutive bouts of SSGs play are inter-spersed with active or passive rest periods(table VII). The duration of each SSG bout in-terval, alternating with planned rest periods, isused to determine work : rest ratios. Althoughmost studies examining SSGs have prescribed theSSG bouts using intervals with short rests, somerecent studies have used continuous SSG formatsof differing duration (e.g. 10–30 minutes).

Unfortunately, previous studies have not usedconsistent work : rest ratios and there is a largevariation in the length, duration, and number ofwork bouts and rest intervals amongst studies(table VII), which makes comparison difficult.For example, a SSG ‘interval’ training prescrip-tion consisting of a 1 · 3-minute work bout with a12-minute rest represents a very low work : restratio (1 : 4) and a very short total game duration

Table IV. Summary of studies examining the effects of concurrent changes in player number and pitch dimensions on small-sided game

intensity in football players

Study Sample

size; age

(y)

Game

design

Training prescription Pitch

dimensions

(m)

Area per

playera

(m2)

%HRR

[mean – SD]

[BLa-]

(mmol/L)

[mean – SD]

RPE (6–20 AU)b

[mean – SD]

Dellal et al.[44] 10;

24–27c

1 vs 1

2 vs 2

4 vs 4 + GK

8 vs 8 + GK

8 vs 8

10 vs 10+ GK

4 · 1.5 min/90 s rest

6 · 2.5 min/2.5 min rest





10 · 10

20 · 20

30 · 25

60 · 45

60 · 45

90 · 45

50

100

94

169

169

203

77.6 – 8.6

80.1 – 8.7

77.1 – 10.7

80.3 – 12.5

71.7 – 6.3

75.7 – 7.9

------

------

Hill-Haas

et al.[45]

16;

16–18c

2 vs 2

4 vs 4

6 vs 6

24 min continuous 28 · 21

40 · 30

49 · 37

150

150

150

89.0 – 4.0

(%HRmax)d

2574 – 16 TD (m)

85.0 – 4.0

(%HRmax)d

2650 – 18 TD (m)

83.0 – 4.0

(%HRmax)d

2590 – 33 TD (m)

6.7 – 2.6

1176 – 8 (D m)

4.7 – 1.6

1128– 10 (D m)

4.1– 2.0

1142– 16 (D m)

13.1 – 1.5

44 – 24 (SP m)

12.2 – 1.8

65 – 36 (SP m)

10.5 – 1.5

71 – 36 (SP m)

Katis and

Kellis[42]

34;

13 – 0.9e

3 vs 3

6 vs 6

10 · 4 min/3 min rest 25 · 15

40 · 30

63

100

87.6 – 4.8

82.8 – 3.2

--

--


b RPE is 6–20 AU unless otherwise stated.

c Age range.

d Age predicted heart rate values.

e Age presented as mean – SD.

AU = arbitrary units; [BLa-] = blood lactate concentration; D = distance: 13.0–15.9 km/h; GK = including goalkeepers; %HRmax = percentage of

maximum heart rate; %HRR = percentage of heart rate reserve; RPE = rating of perceived exertion; SP = number of sprints >18.0 km/h;

TD = total distance; - indicates no data.



Tab

leV

.S

um

mary

ofstu

die

sexam

inin

gth

eeff

ects

ofru

lem

odific

ations

on

sm

all-

sid

ed

gam

ein

tensity

info

otb

all

pla

yers

Stu

dy

Sam

ple

siz

e

Gam

e

desig

n

Tra

inin

gpre

scription

Pitch

dim

ensio

ns

(m)

Rule

s%

HR

max

[mean

–S

D]a

[BLa

- ](m

mol/L

)

[mean

–S

D]

RP

E(6–20

AU

)

[mean

–S

D]

TD

(m)

[mean

–S

D]

Aro

so

etal.

[33]

14

2vs

2

3vs

3

3·

1.5

min

/90

sre

st

3·

4m

in/9

0s

rest

30

·20

Pla

yer-

to-p

layer

mark

ing

Maxim

um

of3

consecutive

touches

- -›

8.1

–2.7

4.9

–2.0

- -- -

Sassi

etal.

[47]

98

vs

8

+G

K

8vs

8

+G

K

4·

4m

in/2

.5m

inre

st

50

·30

Fre

eto

uch

Fre

eto

uch

with

pre

ssure

82.0

›91.0

3.3

–1.2

-

- -

- -

Sam

paio

etal.

[37]

82

vs

2

3vs

3

2·

1.5

min

/90

sre

st

2·

3m

in/9

0s

rest

30

·20

Pla

yer-

to-p

layer

mark

ing

Maxim

um

of2

consecutive

touches

Pla

yer-

to-p

layer

mark

ing

Maxim

um

of2

consecutive

touches

2 2 2 2

- - - -

›17.1

–0.5

›16.8

–0.5

›16.5

–0.5

›16.5

–0.5

- - - -

Little

and

Will

iam

s[4

0]

23

5vs

5

6vs

6

5·

2m

in/2

min

rest

5·

2m

in/2

min

rest

55

·32

59

·27

Pre

ssure

half

sw

itch

Pre

ssure

half

sw

itch

89.9

90.5

- -- -

- -

Mallo

and

Navarr

o[4

8]

10

3vs

31

·5

min

/10

min

rest

33

·20

Possessio

n

Possessio

nw

ith

2outs

ide

neutr

alpla

yers

Norm

alr

ule

s+

GK

91.0

2

91.0

2

88.0

fl

- - -

- - -

747

–24

749

–29

638

–34

Hill

-Haas

etal.

[38]

24

23

23

26

3vs

4and

3vs

3+

1

floate

r

24

min

continuous

37

·28

Conditio

na

b+

bc

Conditio

na

+b

+c

d

Conditio

na

+b

+c

+d

e

Conditio

na

+b

+c

+d

+e

f

83.3

–3.8

84.8

–3.8

80.3

–4.8

83.7

–4.0

2.8

–1.0

2.4

–0.8

2.3

–1.1

2.8

–1.1

15.8

–1.6

15.6

–2.3

14.8

–1.2

15.1

–1.6

2439

–166

2405

–201

2450

–223

2677

–192

21

22

20

21

5v

6and

5v

5+

1

floate

r

24

min

continuous

47

·35

Conditio

na

b+

bc

Conditio

na

+b

+c

d

Conditio

na

+b

+c

+d

e

Conditio

na

+b+c

+d

+e

f

81

–4

83

–5

83

–5

80

–3

2.2

–1.0

3.2

–1.2

2.3

–1.1

2.4

–0.9

15.3

–1.1

14.9

–1.4

14.6

–0.9

14.9

–1.1

2471

–355

2583

–147

2614

–178

2639

–189

aD

ata

for

Sassie

tal.,[4

7]Little

and

Will

iam

s[4

0]and

Mallo

and

Navarr

o[4

8]are

pre

sente

das

mean

valu

es.

bC

onditio

na:off

sid

eru

lein

eff

ect(f

rontone-t

hird

zone

ofth

epitch).

cC

onditio

nb:kic

k-in

only

(ball

cannotbe

thro

wn

inif

itle

aves

the

pitch).

dC

onditio

nc:all

att

ackin

gte

am

pla

yers

mustbe

infr

onttw

ozones

for

agoalt

ocount.

eC

onditi

on

d:o

uts

ide,b

uta

long

the

two

length

sofe

ach

pitc

h,t

wo

neutr

alp

laye

rsca

nm

ove

up

and

dow

nth

epitc

h,b

utn

ote

nte

rth

egrid.B

efo

rea

shoto

ngoali

sperm

itted,t

he

attack

ing

team

must

pass

the

ball

toeith

erofth

ese

pla

yers

.T

he

ball

can

als

obe

pass

ed

toeith

erpla

yer

inth

edefe

nsi

ve

half.

Each

pla

yer

isonly

allo

wed

am

axi

mum

ofone

touch

on

the

ball.

fC

onditi

on

e:o

ne

pla

yerfr

om

each

team

(apair)co

mple

tefo

urre

petit

ions

of‘

sprintt

he

wid

ths/

jog

the

length

s’on

a90

sin

terv

al(

3vs

4and

3vs

3+

1gam

es)

orth

ree

repetit

ions

on

a80

sin

terv

al(

5vs

6and

5vs

5+

1gam

es)

.T

Dtr

ave

lled

per

pla

yer,

regard

less

ofgam

efo

rmat,

would

be

appro

xim

ate

ly440

m.

AU

=arb

itrary

units

;[B

La

- ]=

blo

od

lact

ate

conce

ntr

atio

n;

GK

=in

cludin

ggoalk

eepers

;%

HR

max

=perc

enta

ge

of

maxi

mum

heart

rate

;R

PE

=ra

ting

of

perc

eiv

ed

exert

ion;

TD

=to

tal

dis

tance;››

indic

ate

sin

crease

;fl

indic

ate

sdecr

ease

;2

indic

ate

sno

change;

-in

dic

ate

sno

data

.



(3 minutes). Other studies have used different work:rest ratios across various SSGs (table VII).[16]

Together, these may confound the physiologicaland perceptual responses, as well as the time-motion characteristics of the games. A recent studyinvolving youth football players examined theacute physiological and perceptual responses andtime-motion characteristics of two different train-ing regimens (continuous and intermittent). Theseintermittent (4 · 6-minute bouts with 1.5 minutespassive rest) and continuous (24 minutes) regi-mens were applied to various SSGs including twoversus two, four versus four and six versus six.[49]

The main finding of this study was that inter-mittent regimens were characterized by increaseddistances covered at speeds of >13 km/h. How-ever, paradoxically, the global RPE and %HRmax

was significantly higher in continuous regimens.The results of this study demonstrated that bothSSG training regimens could be used during aseason for match-specific aerobic conditioning,but were unlikely to provide a sufficient stimulusoverload for fully developing maximal oxygenconsumption (

.VO2max).

[49] Another study re-cently investigated the effect of SSG duration,using a 2-, 4- and 6-minute interval format, onboth exercise intensity and technical performanceduring three versus three SSGs.[50] The mainfindings were that although there was a signif-icant decrease in HR between the 4- and 6-minute

game durations and an increase in RPE, the4-minute bouts appear to provide the optimalphysical training stimulus for interval formatSSGs.[50] However, the various interval durationsdid not affect technical performance and, giventhat the magnitude of changes between each ofthe different interval bouts was small, footballcoaches can be confident in using various SSGinterval durations to provide an adequate physicaland technical training stimulus.[50] In summary,research shows that neither training regimen ap-pears to offer any major advantage over theother, and that both regimens could be used forin-season aerobic fitness maintenance training.

2.3.7 Coach Encouragement

Direct supervision and coaching of exercise ses-sions have been shown to improve adherence to anexercise programme, increase training intensity andincrease performance measures in a variety of train-ing modes.[51,52] In football, active, consistent coachencouragement has also been suggested to have aninfluence on training intensity.[30,32,37] For ex-ample, Rampinini et al.[32] demonstrated that HR,blood lactate concentration and RPE were higherwhen coaches provided consistent encouragementduring SSGs with 20 amateur football players ina variety of SSG formats (three vs three, four vsfour, five vs five and six vs six players and onsmall, medium and large-sized pitches). Similarly,

Table VI. Summary of studies examining the effects of goalkeepers on small-sided game intensity in football players

Study Sample

size

Game

design

Training

prescription

Pitch

dimensions (m)

Rules %HRmaxa

[mean – SD]b[BLa-] (mmol/L)

[mean – SD]

Time motion

Sassi et al.[47] 9 4 vs 4

4 vs 4

+ GK

4 · 4 min/2.5 min

rest

30 · 30

33 · 33

Possession 91.0

fl 88.8

6.4 – 2.7

6.2 – 1.4

-

-

Mallo and

Navarro[48]

10 3 vs 3

+ GK

1 · 5 min/10 min

rest

33 · 20 Normal

rules88.0 fl - flTD; flHIR;

›S + W

Dellal et al.[44] 10 8 vs 8 4 · 4 min/3 min

rest

60 · 45 - 71.7 – 6.3

(%HRR)

- -

8 vs 8

+ GK

2 · 10 min/5 min

rest

60 · 45 - › 80.3 – 12.5

(%HRR)

- -

a %HRmax unless otherwise stated.

b Data for Sassi et al.[47] and Mallo and Navarro[48] are presented as mean values.

[BLa-] = blood lactate concentration; GK = including goalkeepers; HIR = high-intensity running; %HRmax = percentage of maximum heart rate;

%HRR = percentage of heart rate reserve; S + W: standing and walking; TD = total distance; › indicates increase; fl indicates decrease;

- indicates no data.



Sampaio et al.[37] reported a significant increase inRPE (for two vs two and three vs three SSGs) withverbal encouragement, but no significant changein %HRmax. Collectively, these studies support therole of the coach in providing consistent encour-agement during SSGs, especially when it is plan-ned that high intensities be achieved.

2.3.8 Logistics and Planning

The logistical considerations associated withorganizing SSGs training are also importantconsiderations for coaches, as these have thepotential to influence player motivation andexercise intensity. For example, the total numberof players available (including goalkeepers) toparticipate in any session will determine the num-ber of SSG teams that can be formed, as wellas the type of games implemented, particularly ifthe objective is to use evenly balanced teams.[7] In

practice, coaches often like to create ‘competitiveplaying structures’, which typically require allSSG teams in one session to play against eachother for an equal number of times. This type ofplaying structure is thought to increase motiva-tion levels by increasing competition and placingan emphasis on results; however, this has not yetbeen empirically tested. It is possible that overuseof a competitive playing structure may result inthe selection of an inappropriate training regimenand therefore a suboptimal training stimulus. Ifthis occurs frequently, it may compromise longerterm training adaptations. Therefore, it is sug-gested that coaches should select SSGs judi-ciously. They should also be aware that not allSSG formats will provide sufficient internal stressto provide the desired physiological adaptation.

Careful planning and organization of trainingsessions for SSGs is also important if the appro-

Table VII. Summary of different training regimens implemented in small-sided game studies with football players

Study Sample size Game design Training prescription Work : rest ratio Regimen

Balsom et al.[30] 6 3 vs 3 6 · 3 min/2 min rest

15 · 70 s/20 s rest

36 · 30 s/15 s rest

36 · 30 s/30 s rest

1 · 30 min

1.5 : 1

3.5 : 1

2 : 1

1 : 1

-

Interval

Interval

Interval

Interval

Continuous

Owen et al.[34] 13 1 vs 1 - 5 vs 5 1 · 3 min/12 min rest 1 : 4 Interval

Aroso et al.[33] 14 2 vs 2

3 vs 3

4 vs 4

3 · 1.5 min/90 s rest



1 : 1

2.6 : 1

4 : 1

Interval

Interval

Interval

Jones and Drust[41] - 4 vs 4 and 8 vs 8 1 · 10 min - Continuous

Rampinini et al.[32] 20 3 vs 3 - 5 vs 5 3 · 4 min/3 min rest 1.3 : 1 Interval

Kelly and Drust[36] 8 5 vs 5 4 · 4 min/2 min rest 2 : 1 Interval

Little and Williams[16] 28 2 vs 2

3 vs 3

4 vs 4

5 vs 5

6 vs 6

8 vs 8


4 · 3.5 min/90 s rest





1 : 1

2.3 : 1

2 : 1

4 : 1

5.3 : 1

5.3 : 1

Interval

Interval

Interval

Interval

Interval

Interval

Dellal et al.[44] 10 1 vs 1

2 vs 2

4 vs 4 + GK

8 vs 8 + GK

8vs 8

10 vs 10 + GK

4 · 1.5 min/90 s rest

6 · 2.5 min/2.5 min rest





1 : 1

1 : 1

1.3 : 1

2 : 1

1.3 : 1

4 : 1

Interval

Interval

Interval

Interval

Interval

Interval

Hill-Haas et al.[49] 16 2 vs 2; 4 v 4; 6 vs 6

2 vs 2; 4 vs 4; 6 vs 6

4 · 6 min/90 s passive rest

1 · 24 min

4 : 1

-Interval

Continuous

Fanchini et al.[50] 19 3 vs 3 3 · 2 min; 3 · 4 min;


1 : 2; 1 : 1; 1.5 : 1 Interval

GK = including goalkeepers; - indicates 1 vs 1, 2 vs 2, 3 vs 3, 4 vs 4 and 5 vs 5 small-sided games were used; - indicates no data.



priate training stimulus is to be achieved. Forexample, factors such as planning SSGs accord-ing to a prospective training plan designed to meetthe physical, technical and tactical requirementsof the team, along with the appropriate use ofcoach encouragement, pitch area, player number,goalkeepers, rule modification and selection ofwork and rest periods, will help achieve optimalexercise intensity. The variation in individual re-sponses to the various SSG structures within asession and between training sessions should alsobe considered.[32,53,54] Finally, it is advisable toavoid skill and fitness mismatches between op-posing teams in order to avoid compromisingtraining intensity.

2.3.9 Comparisons of SSG Training Intensitywith Competitive Match Play

Several studies have examined how the exerciseintensity of various SSGs compares with the ex-ercise intensity of competitive match play.[8,44,55,56]

The findings of these studies can also be used todetermine if the most intense periods of matchescompare with the intensity of various SSGs. Forexample, Gabbett andMulvey[8] recruited 13 elitefemale football players and compared three ver-sus three and five versus five SSGs with (i) do-mestic football matches against male youthteams; (ii) Australian National Women’s Leaguefootball matches; and (iii) international women’sfootball matches. The main finding was that al-though SSGs simulate the overall movementpatterns of domestic, national and internationalcompetition, they do not simulate the high-in-tensity repeated-sprint demands of internationalcompetition.[8] In contrast, Allen et al.[55] re-ported that although total distance was similar,the ratio of high- to low/moderate-intensity workin five versus five SSGs was higher comparedwith 11 versus 11 games. Similarly, the intensityof two versus two was found to exceed the in-tensity of State Premier League under 19 matches,while four versus four were similar to, and sixversus six were below match intensity (figure 2).Capranica et al.[56] reported that the physiologi-cal intensity and movement demands of sevenversus seven and 11 versus 11 in prepubescentfootball players were similar, with HRs exceeding

170 beats per minute. In summary, it appears thatselected SSG formats containing fewer playerscan exceed mean match intensity in youth foot-ball players. Coaches can use this information forchoosing SSGs that are either more intense thanmatch demands to overload the players, or lowerthan 11 versus 11 match intensity when eithertechnical/tactical requirements or recovery andregeneration is the goal of training.

3. Studies Comparing SSGs Trainingwith Interval Training

Despite the widespread use of SSGs in foot-ball, there are surprisingly few studies comparingtheir effectiveness in comparison to traditionalforms of fitness training. The previous studies thathave been completed can be divided into the fol-lowing two categories: (i) studies that investigatedacute physiological responses of SSGs and com-pared these with generic (interval) training re-sponses;[30,44,47] and (ii) studies involving thecomparison of each training mode on either phy-siological performance measures and/or directmatch performance.[57-59]

3.1 Acute Physiological Comparisons ofSSGs Training with Interval Training

Several studies have compared the physiolo-gical responses between generic interval trainingwith football-specific SSG training drills. Indeed,

95

90

85

80

Inte

nsity

(%

HR

max

)

75

70

2 vs 2 4 vs 4

Playing format

6 vs 6 Match

Fig. 2. Box and whisker plot of exercise intensity (percentage ofmaximum heart rate [%HRmax]) in various small-sided games andmatches.[45]



many studies have shown that the exercise intensityachieved during SSGs are similar to generic fit-ness training drills of similar duration.[30,44,47]

For example, Sassi et al.[47] compared the acutephysiological responses of two formats of fourversus four and eight versus eight SSGs with in-terval running (4 · 1000 metre repeats, separatedby 150 seconds of recovery), using 11 elite pro-fessional players from a Spanish first divisionfootball club. Although there was no systematicmanipulation of pitch area, game format (playernumber) or rule modifications in this study, theSSG formats elicited a greater %HRmax responsecompared with the interval running (91% vs 85%HRmax).

[47] More recently, Dellal et al.[44] com-pared the HR response of short-duration (5- to30-second efforts) high-intensity interval runningwith a variety of SSG formats, using ten elitefootballers from a French first division footballclub. In contrast to the previous studies, only thetwo versus two (no goalkeepers) and eight versuseight (including goalkeepers) SSG formats gen-erated similar HR responses compared with theshort-duration interval running protocols. Theone versus one (no goalkeepers) and four versusfour (including goalkeepers) formats generatedthe lowest HR responses of both the SSGs andinterval running.[44] In general, the results of thesestudies demonstrated that many smaller-formatSSGs played on a relatively large pitch area perplayer, can elicit similar intensities to both long-duration interval running[47] and short-durationhigh-intensity interval protocols.[44] However, itappears that the variability in exercise stimulus isgreater in SSGs compared with generic intervaltraining (figure 3), which may be due to the un-structured and stochastic nature of the movementdemands in SSGs.

3.2 Training Studies Comparing SSGs Trainingwith Interval Training

There have been few studies that have exam-ined the efficacy of using SSGs as a conditioningstimulus compared with traditional forms of fit-ness training. In the first controlled training studyto compare both SSGs and generic training,Reilly and White[57] recruited 18 professional youth

footballers from an English Premier League foot-ball club. Using a parallel matched-group design,players were allocated to a SSGs group or anaerobic interval training group (ITG). Playerscompleted the training twice per week, as part oftheir normal training, over a 6-week period duringthe competitive season. The SSGs involved fiveversus five games, played in intervals of 6 · 4minutes, interspersed with 3-minute active recoveryat 50–60% HRmax. The interval running durationwas matched with the SSGs, with a target intensityof 85–90% HRmax (active recovery of 3 minutes at50–60% HRmax). All physiological performancemeasures, including counter movement jumps,10–30 metre sprints, 6 · 30 second anaerobicshuttle test, the agility T-test and the multi-stagefitness test, demonstrated similar changes duringthe study.[57] Based on these results, the authorsconcluded that both SSGs and interval trainingare equally effective for maintaining in-seasonaerobic and anaerobic fitness in elite youth foot-ballers.[57] Unfortunately, the HR responses toeach type of training were not reported, making itdifficult to determine if both groups received asimilar internal training load during the studyperiod. A further limitation of this study was thatthere was little detail of the periodization andprescription of the SSGs training. For example,the game format was restricted to five versus fivefor all sessions, and no detail relating to pitcharea, rules or coach encouragement was provided.

In a comprehensive training study comparingSSGs with generic interval training, Impellizzeri

100

Inte

nsity

(%

HR

max

)

90

Tacticaltraining SSGs Circuit Interval

80

70

60

50

0

Fig. 3. Mean (–90% CI) exercise intensity (percentage of maximumheart rate [%HRmax]) in various football training activities. SSGs =small-sided games.



et al.[58] used a parallel matched-group researchdesign, where 29 youth football players from twojunior teams of Italian professional football clubswere randomly allocated to either a SSG or ITG.The 12-week training intervention spanned over4 weeks of the pre-season and 8 weeks of thecompetitive season in which the players com-pleted two sessions per week designed to improveaerobic fitness. The interval training compriseda fixed prescription of 4 · 4-minute efforts at atarget intensity of 90–95% of HRmax, interspersedby 3 minutes of active recovery at 60–70% ofHRmax. The SSGs training involved a mix ofSSGs, including three versus three, four versusfour and five versus five players. Both the dura-tion and training intensity were matched betweenthe groups. The results demonstrated no differ-ence in mean exercise intensity (%HRmax) orweekly training load (session RPE) between thegroups, with the exception of time spent at >95%HRmax, where the SSGs group spent ~30 secondsper session longer in this zone.[58] Fitness test re-sults revealed similar improvements for the ITGand SSG groups for peak oxygen consump-tion (

.VO2peak) [8% and 7%, respectively], lactate

threshold (13% and 11%, respectively) and run-ning economy (3% for both groups) over the12 weeks of training. Notably, the improvementsin

.VO2peak for ITG and SSGs for the in-season

phase of the study were also very similar to theearlier study of Reilly and White[57] (0.8% and0.7%, and 0.3% and 0.2%, respectively).

Impellizzeri et al.[58] also examined the influ-ence of generic and specific training strategies onphysical performance during matches. The resultsrevealed non-significant increases (pre-seasontraining phase only) in low-intensity activity(forwards, backwards and sideways jogging),high-intensity activity (higher speed running andsprinting) and total distance travelled for boththe ITG and SSG groups following the 12-weektraining period.However, whenmatch performancemeasures for the in-season phase of training wereanalysed, the magnitude of the increases (for bothgroups) in low- and high-intensity activity areconsiderably smaller.[58]

Previous training studies comparing SSGstraining with interval running have demonstrated

good research design and high internal validity.However, in the field, there are certain aspects ofthese studies that rarely occur. For example, it ispractically difficult to apply a rigid prescriptionof interval training that does not have progressiveoverload when training elite football players.Moreover, in practice, the systematic manipula-tion of SSGs for the purpose of physical devel-opment is problematic, as the technical/tacticaltraining goals of the coach do not always relateto physiological development needs or priorities.Therefore, to examine these issues, Hill-Haaset al.[59] assessed the efficacy of a coach-led SSGsprogramme and a progressive mixed-methodsgeneric fitness training programme in 25 eliteyouth football players. Using a parallel matched-group research study design, the players wererandomly allocated to either SSG ormixed-generictraining groups over a 7-week pre-season trainingperiod. In contrast to previous research,[58] thisstudy implemented a mixed-generic training pro-gramme (consisting mainly of aerobic powertraining and prolonged intermittent high-intensityinterval training), and a SSGs training pro-gramme, incorporating a broad range of gameformats (i.e. two vs two to seven vs seven).[59]

Although the manipulation of the SSGs trainingvariables (such as pitch area and rules) was lesssystematic than previous studies, a key differencewas the planning and implementation of the SSGstraining programme by an experienced coach,which increased the external validity of the study.The main finding of this study was that bothcoach-selected SSGs training and mixed-generictraining (comprising short duration, high-intensityintervals of <90 seconds) were effective at sig-nificantly improving yo-yo intermittent recoverytest (level 1) performance, but not

.VO2max.

[59]

Notably, there were no between-group or training-induced changes in any other performancemeasures.

In general, the results of these training studiesshow that SSGs provide similar changes in aerobicfitness and match performance measures, withthe majority of changes in fitness/performanceobserved during pre-season training. The studiesalso suggest that more effective use of this train-ing mode is still possible. This may be achieved



through systematic manipulation of the train-ing variables. However, it is clear that carefulselection of SSG formats and training regimens isrequired to optimize fitness and performancegains. Combined, the evidence suggests that bothSSG and interval training drills are suitable forimproving fitness and performance in footballplayers. It is most likely that a mixed-methodsapproach is appropriate for football training;however, the selection of these should be basedon the technical, tactical and performance needsof the players.

4. Limitations of SSGs

The first limitation relates to the current re-search knowledge into the prescriptive variablesthat affect SSGs intensity. The current volume ofsystematic research in this area is small and,consequently, definitive conclusions are difficultto form. Despite offering several advantages,there are also a number of suggested limitationsthat relate to the implementation of SSGs, in-cluding (i) the ceiling effect in achieving high-exercise intensities for highly fit or skilled players;(ii) the ability to replicate the demands of themost intense periods of match play; (iii) the re-quirement of a high level of technical and tac-tical proficiency to achieve appropriate exerciseintensity; (iv) the risk of contact injuries duringtraining; and (v) and the availability of enoughcoaches to control and monitor this type oftraining.

It has been reported that players with thehighest

.VO2peak elicited the lowest percentage of.

VO2peak during SSGs,[60] suggesting that eitherthe technical/tactical constraints of the game orthe intermittent nature of the exercise can preventsome players from reaching appropriate trainingintensities.[61] Therefore, it was suggested thatplayers with a high fitness level and a good skilllevel will not exercise at sufficient intensity toelicit aerobic fitness adaptations under thesetraining conditions. However, in contrast, wehave observed a weak but significant positivecorrelation between fitness level and exercise in-tensity during various SSGs (figure 4). These re-sults suggest that players with a high fitness level

exercise at a higher intensity during SSGs.Therefore, future research is required to elucidatethe possible relationships between fitness, skilland exercise intensity during SSGs.

Additionally, the intermittent nature of SSGshas been suggested to limit the ability of playersto achieve sufficient cardiac load for aerobic fit-ness adaptations. Indeed, Hoff and Helgerud[61]

argue that optimal aerobic adaptations are onlypossible if cardiac output remains elevated forsustained periods during football training, andthat exercise intensities of >90% HRmax are re-quired for improvements in aerobic fitness. SinceSSGs are more intermittent than interval run-ning, it has been suggested that the continualre-setting of the muscular venous pump will com-promise cardiac output and consequently preventa sustained high stroke volume being achieved.[61]

It has also been reported that SSGs training maynot always simulate the high-intensity, repeated-sprint demands of high level competition,[2] and itis not known if they can be used to replicate themost intense periods of the game. However, thesepotential physiological limitations to SSGs train-ing may be countered by appropriate manipula-tion of SSGs training variables.

Moreover, since SSGs involve a combinationof technical/tactical ability, decision making andphysical exertion, it seems that concurrent abil-ities may be required to achieve appropriate ex-ercise intensities. Consequently, it is possible that

100

95

90

Inte

nsity

(%

HR

max

)

85

80

75

1750 2250

MSFT distance (m)

2750 3250

Fig. 4. Relationship between player fitness (multi-stage fitness test[MSFT] distance [m]) and exercise intensity (percentage of maximumheart rate [%HRmax]) during various small-sided games, [r = 0.26,p = 0.04].[45] Full line represents the line of best fit and the dashed linerepresents 95% confidence intervals.



less-skilled players may not be able to con-sistently sustain the technical skill or tacticalproficiency to achieve and maintain the requiredmetabolic strain; as such, training may be coun-terproductive in terms of playing performance.[14]

However, this has not been empirically tested andfuture studies should examine if low technicalskill ability limits the exercise intensity of in-dividual players during SSGs.

Due to the competitive nature of SSGs infootball, there may be an increased risk of contactinjuries during training,[7] although rule mod-ifications may help minimize this potential pro-blem. The incidence of injuries in skill-basedconditioning games in rugby league have beenreported to be lower than that of traditional fit-ness training.[1] However, to date, there have beenno studies that have examined the incidence ofinjuries during SSGs training in comparison togeneric training in football.

Other logistic factors involved in the planningof SSGs (e.g. pitch area available, number ofstaff, number of players available) can also affectthe effectiveness of this training mode. These in-clude the ability to control and monitor the in-tensity of multiple, concurrent SSGs being playedon various pitches at any one time. Therefore, ahigh level of organization and consistent coachencouragement is also needed to maintain playermotivation. The use of technology, includingreal-time HR monitoring of individual playersduring SSGs, may also promote more effectiveimplementation of SSGs training.

In summary, there are several potential lim-itations to SSGs training in football. Coachesshould be aware of these factors, which may re-duce the effectiveness of this mode of training fordeveloping both physical attributes and footballproficiency. Therefore, for optimal use of SSGstraining to improve aerobic fitness, it is suggestedthat a systematic approach to manipulating SSGprescriptive variables is adopted, with an emphasison careful control and real-time monitoring

5. Future Research

Future research is required to further developour understanding of the training stimulus pro-

vided by football-specific SSGs. One importantarea that requires further investigation is the in-fluence of modifying SSG design variables on theexercise intensity of SSGs training. This systematicreview has demonstrated that, with the possibleexception of player number, the majority of pre-scriptive variables have not been investigatedthoroughly. Therefore, future research shouldexamine the influences of manipulating selectedvariables such as pitch area, technical involve-ments and rule changes. Further research is stillrequired before a complete understanding of howeach of the SSG prescriptive variables may in-fluence exercise intensity is gained.

Another important area for future research isthe influence of different periodization strategiesof SSGs training for the development of physio-logical, technical skill and tactical proficiency.A number of interesting research questions couldbe posed. For example, are larger SSG formats(e.g. six vs six) more effectively used in early pre-season training, while smaller game formats (e.g.two vs two) be used just prior to the competitiveseason? Is the overall effectiveness of SSGstraining improved when implemented as part of atraditional linear periodization approach, or is itbetter to implement these games using a ‘blockperiodization model’[62] approach? To date, thetraining studies comparing the effectiveness ofSSGs and interval running suggest that both areequally effective. Consequently, future studiesshould examine optimal periodization strategiesfor using both types of training methods for de-veloping football-specific physical qualities.

Additionally, although many studies haveinvestigated the technical requirements ofSSGs,[8,34,36,39,41-43,48,50,55,56,63,64] research con-ducted to date has not been very systematic. Fu-ture studies should include detailed notationalanalysis to provide an improved understandingof the technical skill requirements of variousSSGs. This may assist coaches to better under-stand the link between the technical load andexercise intensity of SSGs training.

One of the major advantages of SSGs trainingis thought to be the development of tacticalawareness and decision-making capabilities, andthe transfer of these to match performance. Future



research is also needed to understand the natureof the tactical awareness and decision makingdevelopment provided by different SSG formats.Once established, further research should establisha link between SSGs and the transfer of theseskills to match performance.

6. Conclusions

Despite the extensive use of SSGs in football,our understanding of their effectiveness as atraining tool for developing physical, technicaland tactical skills in football players is not com-plete. Nevertheless, recent research has improvedour understanding of some of the variables af-fecting SSGs intensity. Future studies are requiredto increase the understanding of the interactionbetween the technical, tactical and physical de-mands of SSGs, and how these can be manipu-lated to improve the training process for footballplayers. However, at present, it seems that ex-ercise intensity in SSGs can be manipulated byaltering factors such as player number, numericalbalance between teams, rules of play, the use ofgoalkeepers, pitch area and coach encourage-ment. It also appears that similar fitness andperformance gains can be made with SSGs as isachieved with traditional interval trainingmethods.

Acknowledgements

Inmemory ofMartyn Crook, the former head coach of theAustralian National under 17 and South Australian SportsInstitute (SASI) men’s football squads. The authors thankMrCrook for his coaching expertise and commitment to thisproject. To all the players, thank you for your time and effortduring the SSGs. To Dr Greg Rowsell, thank you for pro-viding valuable feedback on earlier versions of this manu-script. No sources of funding were used to assist in the pre-paration of this article. The authors have no conflicts ofinterest that are directly relevant to the content of this article.

References1. Gabbett T. Skill-based conditioning games as an alternative

to traditional conditioning for rugby league players. JStrength Cond Res 2006; 20 (2): 309-15

2. Gabbett T, Jenkins D, Abernethy B. Game-based trainingfor improving skill and physical fitness in team sport ath-letes. Int J Sports Sci Coach 2009; 4 (2): 273-83

3. Football Federation Australia. Small-sided games hand-book [online]. Available from URL: http://www.penrithfc.com/SSG%20Handbook%20for%20clubs%2019Dec07.pdf[Accessed 2010 Dec 10]

4. Gamble P. A skill-based conditioning games approach tometabolic conditioning for elite rugby football players.J Strength Cond Res 2004; 18 (3): 491-7

5. Owen A. Physiological and technical analysis of small-sidedconditioned training games within professional football.Wrexham: SAGE Publications, 2003

6. Gregson W, Drust B. The physiology of football drills. In-sight 2000; 3 (4): 1-2

7. Little T. Optimizing the use of soccer drills for physiologicaldevelopment. Strength Cond J 2009; 31 (3): 1-8

8. Gabbett T, Mulvey M. Time-motion analysis of small-sidedtraining games and competition in elite women soccerplayers. J Strength Cond Res 2008; 22 (2): 543-52

9. Allison S, Thorpe R. A comparison of the effectiveness oftwo approaches to teaching games within physical educa-tion: a skills approach versus a games for understandingapproach. Br J Phys Ed 1997; 28 (3): 9-13

10. Achten J, Jeukendrup A. Heart rate monitoring-applica-tions and limitations. Sports Med 2003; 33 (7): 517-38

11. Drust B, Reilly T, Cable N. Physiological responses tolaboratory-based soccer-specific intermittent and continuousexercise. J Sports Sci 2000; 18 (11): 885-92

12. Esposito F, Impellizzeri FM,Margonato V, et al. Validity ofheart rate as an indicator of aerobic demand during socceractivities in amateur soccer players. Eur J Appl Physiol2004; 93: 167-72

13. Hoff J, Wisløff U, Engen L, et al. Soccer specific aerobicendurance training. Br J Sports Med 2002; 36: 218-21

14. Castagna C, Belardinelli R, Abt G. The oxygen uptake andheart rate response to training with a ball in youth soccerplayers. J Sports Sci 2004; 22: 532-3

15. Bangsbo J. The physiology of soccer: with special referenceto intense intermittent exercise. Acta Physiol Scand 1994;619: 1-155

16. Little T, Williams A. Measures of exercise intensity duringsoccer training drills with professional soccer players.J Strength Cond Res 2007; 21 (2): 367-71

17. Krustrup P, Mohr M, Steensberg A, et al. Muscle and bloodmetabolites during a soccer game: implications for sprintperformance. Med Sci Sports Exerc 2006; 38 (6): 1165-7

18. Borg G. Psychophysical basis of perceived exertion. Med SciSports Exerc 1982; 14 (5) 377-81

19. Coutts AJ, Rampinini E, Marcora S, et al. Heart rate andblood lactate correlates of perceived exertion during small-sided soccer games. J Sci Med Sport 2009; 12 (1): 79-84

20. Impellizzeri FM, Rampinini E, Coutts AJ, et al. Use of RPE-based training load in soccer. Med Sci Sports Exerc 2004;36 (6): 1042-7

21. Alexiou H, Coutts AJ. A comparison of methods used forquantifying internal training load in women soccer players.Int J Sports Physiol Perform 2008; 3: 1-12

22. Foster C, Florhaug J, Franklin J, et al. A new approach tomonitoring exercise training. J Strength Cond Res 2001;15 (1): 109-15



23. Carling C, Bloomfield J, Nelsen L, et al. The role of motionanalysis in elite soccer. Sports Med 2008; 38 (10): 839-62

24. Coutts AJ, Duffield R. Validity and reliability of GPS unitsfor measuring movement demands of team sports. J SciMed Sport 2010; 13 (1): 133-5

25. Macleod H, Morris J, Nevill A, et al. The validity of a non-differential global positioning system for assessing playermovement patterns in field hockey. J Sports Sci 2009; 27(2): 121-8

26. Townsend A, Worringham C, Stewart I. Assessment ofspeed and position during human locomotion using non-differential GPS. Med Sci Sports Exerc 2008; 40: 124-32

27. Edgecomb S, Norton K. Comparison of global positioningand computer-based tracking systems for measuring playermovement distance during Australian Football. J Sci MedSport 2006; 9: 25-32

28. Petersen C, Pyne D, Portus M, et al. Validity and reliabilityof GPS units to monitor cricket-specific movement pat-terns. Int J Sports Physiol Perform 2009; 4: 381-93

29. Barbero Alvarez JC, Coutts AJ, Granda J, et al. The validityand reliability of a Global Positioning Satellite system de-vice to assess speed and repeated sprint ability (RSA) inathletes. J Sci Med Sport 2010; 13 (2): 232-5

30. Balsom P, Lindholm T, Nilsson J, et al. Precision football.Kempele: Polar Electro Oy, 1999

31. Jeffreys I. The use of small-sided games in the metabolictraining of high school soccer players. Strength Cond J2004; 26 (5): 77-8

32. Rampinini E, Impellizzeri FM, Castagna C, et al. Factorsinfluencing physiological responses to small-sided soccergames. J Sports Sci 2007; 25 (6): 659-66

33. Aroso J, Rebelo A, Gomes-Pereira J. Physiological impactof selected game-related exercises [abstract]. J Sports Sci2004; 22 (6): 522

34. Owen A, Twist C, Ford P. Small-sided games: the physio-logical and technical effect of altering pitch size and playernumbers. Insight FACA J 2004; 7 (2): 50-3

35. Williams K, Owen A. The impact of player numbers onthe physiological responses to small sided games [abstract].J Sports Sci Med 2007; 6 Suppl. 10: 100

36. Kelly D, Drust B. The effect of pitch dimensions on heartrate responses and technical demands of small-sided soccergames in elite players. J Sci Med Sport 2009; 12 (4): 475-9

37. Sampaio J, Garcia G, Macas V, et al. Heart rate and per-ceptual responses to 2 · 2 and 3· 3 small-sided youth soc-cer games. J Sports Sci Med 2007; 6 Suppl. 10: 121-2

38. Hill-Haas S, Coutts AJ, Dawson B, et al. Time-motioncharacteristics and physiological responses of small-sidedgames in elite youth players: the influence of player numberand rule changes. J Strength Cond Res 2010; 24 (8):2149-56

39. Duarte R, Batalha N, Folgado H, et al. Effects of exerciseduration and number of players in heart rate responses andtechnical skills during Futsal small-sided games. OpenSports Sci J 2009; 2: 37-41

40. Little T, Williams A. Suitability of soccer training drills forendurance training. J Strength Cond Res 2006; 20 (2):316-9

41. Jones S, Drust B. Physiological and technical demands of 4 v4 and 8 v 8 games in elite youth soccer players. Kinesiology2007; 39 (2): 150-6

42. Katis A, Kellis E. Effects of small-sided games on physicalconditioning and performance in young soccer players. JSports Med 2009; 8: 374-80

43. Platt D, Maxwell A, Horn R, et al. Physiological and tech-nical analysis of 3 v 3 and 5 v 5 youth football matches.Insight FACA J 2001; 4 (4): 23-5

44. Dellal A, Chamari K, Pintus A, et al. Heart rate responsesduring small-sided games and short intermittent runningtraining in elite soccer players: a comparative study. J StrengthCond Res 2008; 22 (5): 1449-57

45. Hill-Haas S, Dawson B, Coutts AJ, et al. Physiological re-sponses and time-motion characteristics of various small-sided soccer games in youth players. J Sports Sci 2009; 27(1): 1-8

46. Impellizzeri FM, Rampinini E, Marcora SM. Physiologicalassessment of aerobic training in soccer. J Sports Sci 2005;23 (6): 583-92

47. Sassi R, Reilly T, Impellizzeri FM. A comparison of small-sided games and interval training in elite professional soc-cer players [abstract]. J Sports Sci 2004; 22: 562

48. Mallo J, Navarro E. Physical load imposed on soccer playersduring small-sided training games. J Sports Med Phys Fit2008; 48 (2): 166-72

49. Hill-Haas S, Rowsell G, Coutts AJ, et al. Acute physiologi-cal responses and time-motion characteristics of two small-sided training regimes in youth soccer players. J StrengthCond Res 2008; 22 (6): 1-5

50. Fanchini M, Azzalin A, Castagna C, et al. Effect of boutduration on exercise intensity and technical performance ofsmall-sided games in soccer. J Strength Cond Res. Epub2010 May 28

51. Coutts AJ, Murphy A, Dascombe B. Effect of direct super-vision of a strength coach onmeasures of muscular strengthand power in young rugby league players. J Strength CondRes 2004; 18 (2): 316-23

52. Mazzetti S, Kraemer W, Volek J, et al. The influence ofdirect supervision on strength performance. Med SciSports Exerc 2000; 32: 1175-84

53. Hill-Haas S, Coutts AJ, Rowsell G, et al. Variability of acutephysiological responses and performance profiles of youthsoccer players in small-sided games. J Sci Med Sport 2008;11: 487-90

54. Hill-Haas S, Rowsell G, Coutts AJ, et al. The reproducibilityof physiological responses and performance profiles ofyouth soccer players in small-sided games. Int J SportsPhysiol Perform 2008; 3 (3): 393-6

55. Allen J, Butterly R, Welsch M, et al. The physical and phy-siological value of 5-a-side soccer training to 11-a-sidematch play. J Hum Movement Stud 1998; 34: 1-11

56. Capranica L, Tessitore A, Guidetti L, et al. Heart rate andmatch analysis in pre-pubescent soccer players. J Sports Sci2001; 19: 379-84

57. Reilly T, White C. Small-sided games as an alternative tointerval-training for soccer players [abstract]. J Sports Sci2004; 22 (6): 559

58. Impellizzeri FM,Marcora S, Castagna C, et al. Physiologicaland performance effects of generic versus specific aerobic



training in soccer players. Int J Sports Med 2006; 27 (6):

483-92

59. Hill-Haas S, Coutts AJ, Rowsell G, et al. Generic versus

small-sided game training in soccer. Int J Sports Med 2009;

30 (9): 636-42

60. BuchheitM, Laursen P,Kuhnle J, et al. Game-based training in

young elite handball players. Int J Sports Med 2009; 30: 251-8

61. Hoff J, Helgerud J. Endurance and strength training for

soccer players. Sports Med 2004; 34 (3): 165-80

62. Issurin VB. New horizons for the methodology and physiology

of training periodization. Sports Med 2010; 40 (3): 189-206

63. Grant A, Williams M, Johnson S. Technical demands of 7 v7 and 11 v 11 youth football matches. Insight FACA J1999; 2 (4): 1-2

64. Grant A, Williams M, Dodd R, et al. Physiological andtechnical analysis of 11 v 11 and 8 v 8 youth football mat-ches. Insight FACA J 1999; 2 (3): 3-4

Correspondence: Dr Aaron J. Coutts, School of Leisure, Sport& Tourism, University of Technology, Sydney, Kuring-gaiCampus, P O Box 222, Lindfield, NSW 2070, Australia.E-mail: [email protected]



Documents

Physiology of Small-Sided Games Training in Football · Physiology of Small-Sided Games Training in Football ... sures are lower in smaller game formats ... logical responses associated