Journal of Science and Medicine in Sport (2008) 11, 487490

SHORT REPORT

Variability of acute physiological responses and

performance proles of youth soccer players
in small-sided games
Stephen Hill-Haas a,c,, Aaron Coutts b, Greg Rowsell c, Brian Dawson a
a

University of Western Australia, Australia

University of Technology, Sydney, Australia
c South Australian Sports Institute, Australia
b

Received 30 April 2007 ; received in revised form 19 July 2007; accepted 20 July 2007
KEYWORDS
Reliability;
Reproducibility;
Heart rate;
Lactate;
Training load

Summary The aim of this study was to examine the variability in physiological and
perceptual responses and timemotion proles of various small-sided soccer game
(SSG) formats (2 versus 2, 4 versus 4 and 6 versus 6 players) and regimes (interval and
continuous). Typical error (TE) was calculated for mean heart rate as a percentage
of maximum heart rate (%HRmax ), global ratings of perceived exertion (RPE), blood
lactate [La ] and various timemotion characteristics for 16 male soccer players
(mean 16.2 years, range 15.617.9). The TE for HR responses were <5% for all SSGs.
RPE also demonstrated small variability across all SSGs, with TE ranging between 1
and 2 units. In contrast, the TE% for [La ] was higher, ranging from 16% (2 versus
2-interval) to 34% (4 versus 4-interval). The TE% for total distance (TD) and distance
covered at 06.9 km/h was <5% for all SSGs, with 2 versus 2 interval and continuous games recording the lowest TE (2.2% and 2.9%, respectively). An increase in
game format size does not appear to inuence the variability of the acute physiological responses to SSGs, although continuous formats display less variability than
interval formats. The TD, distance covered and percentage of total time moving at
06.9 km/h demonstrated small variability across all formats and regimes. However,
higher movement speeds zones (>8 km/h) reected increased variability, irrespective of game format or regime. Collectively, these results suggest that SSG training
can provide a reliable aerobic training stimulus.
2007 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

Introduction
The use of small-sided games (SSGs) for technical and tactical development of soccer players is

Corresponding author.
E-mail address: hillhs01@student.uwa.edu.au (S. Hill-Haas).

well established.1,2 However, SSGs are increasingly

being used as an alternative to interval training as
an aerobic training stimulus. Although some studies
support this approach,1,3,4 a better understanding
of the reliability of the acute physiological and perceptual responses (heart rate, blood lactate and
ratings of perceived exertion) and performance

1440-2440/$ see front matter 2007 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

doi:10.1016/j.jsams.2007.07.006

488
proles (quantied using global positioning system
units) during SSGs is needed. Consequently, the aim
of this study was to examine the reliability of various small-sided game (SSG) formats (2 versus 2, 4
versus 4 and 6 versus 6 players) and regimes (interval, SSGI and continuous, SSGC ).

Methods
Sixteen male soccer players (mean 16.2 years,
range 15.617.9) participated in the study. All
participants were notied of the research procedures, requirements, benets and risks before
giving informed consent.
The games were played (in duplicate) over a
9-week in-season period in random order, with
participants selected on the same teams against
the same opponents as often as possible. All SSGs
were played at the start of each training session
(with consistent coach encouragement), held twice
weekly, with at least 48 h separating each training session. The participants were well familiarised
with all SSG formats and regimes. The duration
of SSGC was 24 min. The SSGI regime involved
4 6 min bouts with 1.5 min passive rest. The pitch
sizes (length width) used for 2 versus 2, 4 versus 4
and 6 versus 6 games were 28 m 21 m, 40 m 30 m
and 49 m 37 m, respectively.
Heart rate (HR) was measured (5 s recording
intervals) via short-range radiotelemetry (Polar
Team Sport System, Polar Electro, Finland). Maximum heart rate was determined from a 20 m shuttle
test. Global ratings of perceived exertion (RPE)
were recorded immediately after the SSGs using the
Borg scale (620 scale).5 Capillary blood samples
were drawn from an earlobe at rest and immediately following each SSG. Blood lactate [La ]
was determined using a blood-gas analyser (ABL
625 Radiometer, Copenhagen). Timemotion characteristics were measured using portable global
positioning system (GPS) units (SPI 10, GPSports,
Canberra, Australia). The distance and time spent
in six speed zones were selected for analysis:
zone 1 (06.9 km/h); zone 2 (79.9 km/h); zone
3 (1012.9 km/h); zone 4 (1315.9 km/h); zone 5
(1617.9 km/h) and zone 6 (>18 km/h). The GPS
sampling rate was 1 Hz.
The within group reliability of each dependent
measure was calculated using the typical standard error of measurements (TE).6 The TE was
also expressed as a percentage of the mean (TE%).
Scores were calculated for mean heart rate as a
percentage of maximum heart rate (%HRmax ), RPE,
([La ]) and various timemotion characteristics.

S. Hill-Haas et al.

Results
The TE% scores (Table 1) for %HRmax and HRpeak
demonstrated good reliability (<5%) across all formats and regimes. Similarly, RPE also demonstrated
good reliability, with absolute TE scores ranging
between 1 and 2 units. However, [La ] was more
variable across all formats and regimes.
The timemotion characteristics of total distance (TD), distance covered and percentage of
time moving at 06.9 km/h displayed small variability (TE% < 11%) across all formats and regimes,
although absolute TE scores increased as game format size increased (Table 1). Variability of TE%
increased in the higher running speed categories,
irrespective of format or regime.
Additionally, absolute TE scores for the physiological responses to SSGI were higher than SSGC
with some exceptions (Table 1). In contrast, sprint
duration and distance appear more variable in SSGC
compared with SSGI (Table 1).

Discussion
In contrast to previous research,7 the results
demonstrated that with the exception of [La ],
alterations in format do not appear to affect
the reliability of acute physiological or perceptual responses. Greater variability of [La ] may
be attributed to measurement variability and the
stochastic nature of SSGs,3 which may limit its
usefulness as an indicator of internal training
load.8 However, the consistent HR response 2 and
magnitude of TE scores (lowest for heart rate,
followed by RPE and [La ]) agree with previous
research.2,3
This is the rst study to report on the testretest
variation in GPS data as a measure of external
training load during SSG soccer training. In this
study, TD, distance covered and percentage of
total time at 06.9 km/h exhibited small variability across all formats and regimes. However, the
testretest variability increased for higher velocity
movement speeds zones, irrespective of SSG format
or regime. A possible explanation for the greater
variation in the higher movement velocity zones is
that the GPS units only sample distance at 1 Hz.
Given that the duration of high intensity efforts
during soccer are usually very brief (often <2 s),9
it is possible that a higher sampling rate may have
demonstrated lower variation between the training
sessions.
The results of this study also showed that physiological, perceptual and timemotion measures

Reliability of physiological, perceptual and timemotion variables

Variable
%HRmax
HRpeak
RPE
Blood [La ]
Timemotion characteristics
Total distance
Distance (06.9 km/h)
Distance (1315.9 km/h)
Distance (>18.0 km/h)
% Time (06.9 km/h)
% Time (1315.9 km/h)
% Time (>18.0 km/h)
Mean sprint duration
Mean sprint distance

Format regime

2 versus 2C

2 versus 2I

TE (%)
TE%
TE (bpm)
TE%
TE (AU)
TE%
TE (mmol L1 )
TE%

1.7
1.9
2.2
1.1
1.7
12.6
2.0
25.8

2.9
3.3
6.0
3.0
0.8
5.9
0.8
16.5

3.8
4.4
4.3
2.3
1.2
9.6
1.4
28.3

2.1
2.5
4.1
2.1
1.4
12.2
1.8
34.4

2.2
2.6
5.4
2.8
1.6
15.3
0.6
16.6

2.8
3.4
6.9
3.6
0.8
7.4
1.4
32.2

TE (m)
TE%
TE (m)
TE%
TE (m)
TE%
TE (m)
TE%
TE (%)
TE%
TE (%)
TE%
TE (%)
TE%
TE (s)
TE%
TE (m)
TE%

97.9
3.8
33.8
2.9
39.1
14.2
23.4
51.1
2.5
4.0
0.1
16.3
0.4
51.7
0.2
16.4
1.38
21.2

57.7
2.2
39.8
3.4
38.3
14.4
12.5
25.7
3.8
6.1
3.5
54.2
1.4
141.7
0.1
8.5
1.27
20.4

108
4.1
50.1
4.5
64.5
23.5
16.3
31.2
6.4
11.0
1.2
20.9
0.2
29.1
0.3
14.4
1.44
16.8

124.5
4.6
38.5
3.4
42.1
14.4
25.5
32.5
5.6
9.0
1.5
22.5
0.4
41.0
0.4
22.1
1.95
23.3

134
5.2
45.3
3.9
53.9
21.3
38.4
56.0
6.4
10.1
1.0
19.2
0.5
54.1
0.5
26.1
3.04
30.5

167.9
6.4
40.2
3.5
63
21.0
25.9
31.3
6.7
10.7
2.8
44.7
1.4
105.9
0.2
13.0
1.26
14.5

TE, typical error; TE%, typical error as % of mean; C, continuous; I, intermittent; AU, absolute units.

4 versus 4C

4 versus 4I

6 versus 6C

6 versus 6I

Variability of acute physiological responses and performance proles of youth soccer players in small-sided games489

Table 1

490

S. Hill-Haas et al.

from interval SSGs are generally more variable than

continuous formats. A possible explanation may
be that SSGI superimpose an additional intermittent element in the form of planned rest between
each bout. The rest period between each interval may result in a more lagged HR response in
subsequent bouts and therefore facilitate greater
[La ] clearance during planned rest. Moreover,
the between-bout rest may promote physiological recovery between bouts and therefore allow
some players to work at higher intensity during subsequent bouts. Combined, these factors
may reduce the reliability of the HR and [La ]
responses to SSGI regimes compared to the SSGC
regimes.
The results of this study demonstrated that various small-sided soccer formats played in either
a continuous or interval regime, provide reliable
internal responses and external loads and therefore represent a viable alternative to traditional
running interval training for developing and maintaining aerobic tness. Additionally, due to the low
variation in most of the internal training load measures (i.e. RPE and HR) and greater variation in
the external training load measures (i.e. GPS measures), we recommend that coaching staff place a

higher priority on players HR and RPE responses

when monitoring soccer training.

References
1. Rampinini E, Coutts A, Castagna C, Sassi R, Impellizzeri F.
Variation in top level soccer match performance. Int J Sport
Med 2007;28:17.
2. Little T, Williams AG. Suitability of soccer training drills for
endurance training. J Strength Cond Res 2006;20(2):3169.
3. Rampinini E, Impellizzeri F, Castagna C, et al. Factors inuencing physiological responses to small-sided soccer games.
J Sports Sci 2007;25(6):65966.
4. Impellizzeri FM, Marcora SM, Castagna C, et al. Physiological and performance effects of generic versus specic
aerobic training in soccer players. Int J Sport Med
2006;27(6):48392.
5. Borg G. Borgs perceived exertion and pain scales, human
kinetics. Illinois: Champaign; 1998.
6. Hopkins WG. Measures of reliability in sports medicine and
science. Sports Med 2000;30(1):115.
7. Little T, Williams AG. Measures of exercise intensity during soccer training drills with professional soccer players.
J Strength Cond Res 2007;21(2):36771.
8. Impellizzeri FM, Rampinini E, Marcora SM. Physiological
assessment of aerobic training in soccer. J Sports Sci
2005;23(6):58392.
9. Bangsbo J. Aerobic and anaerobic training in soccer.
Stormtryk, Bagsvaerd: Copenhagen; 2007.

Available online at www.sciencedirect.com