Validity and reliability of a 6-a-side small-sided game as an indicator of match-related physical performance in elite youth Brazilian soccer players

Abstract

The aims of this study were: (i) to compare the external and internal load during a 6-aside small-sided game (6v6-SSG) according to age-group (i.e., sensitivity); (ii) to relate these parameters between the 6v6-SSG and official matches (i.e., construct validity); and (iii) to test the reliability of the 6v6-SSG. A total of 51 Brazilian youth soccer players participated in this study (U11 [n=16]; U13 [n=10]; U15 [n=9]; U17 [n=8]; U20 [n=8]). Three experiments were conducted. Experiment A: fifty-one U11 to U20 players were submitted to 6v6-SSGs (n=10 games; two for each age-group). Experiment B: thirty-two players were randomized to also play official matches (n=6 matches). Experiment C: thirty-five youth players played the 6v6-SSG twice for test and retest reliability analysis. External load was obtained using Global Positioning Systems and the internal load parameter was calculated through mean heart rate. Statistical approaches showed progressive increases in all parameters according to categories (U11<U13<U15<U17<U20; p<0.05; ES=0.42˗23.68). Even controlling for chronological age, all parameters showed likely to almost certain correlations between 6v6-SSG and official matches (r=0.25˗0.92). Collectively, the proposed protocol indicates good reliability (CV%=2.0˗12.6; TE%=2.3˗2.7%; ICC=0.78˗0.90). This research suggests that the 6v6-SSG is an alternative tool to indicate match-related physical performance in youth soccer players.

Full text

Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=rjsp20
Journal of Sports Sciences
ISSN: 0264-0414 (Print) 1466-447X (Online) Journal homepage: https://www.tandfonline.com/loi/rjsp20
Validity and reliability of a 6-a-side small-sided
game as an indicator of match-related physical
performance in elite youth Brazilian soccer players
Rodrigo Aquino, Bruno Melli-Neto, João Victor S. Ferrari, Bruno L. S.
Bedo, Luiz H. Palucci Vieira, Paulo Roberto P. Santiago, Luiz Guilherme C.
Gonçalves, Lucas P. Oliveira & Enrico F. Puggina
To cite this article: Rodrigo Aquino, Bruno Melli-Neto, João Victor S. Ferrari, Bruno L. S. Bedo,
Luiz H. Palucci Vieira, Paulo Roberto P. Santiago, Luiz Guilherme C. Gonçalves, Lucas P. Oliveira
& Enrico F. Puggina (2019) Validity and reliability of a 6-a-side small-sided game as an indicator
of match-related physical performance in elite youth Brazilian soccer players, Journal of Sports
Sciences, 37:23, 2639-2644, DOI: 10.1080/02640414.2019.1608895
To link to this article: https://doi.org/10.1080/02640414.2019.1608895
Published online: 07 May 2019.
Submit your article to this journal
Article views: 148
View related articles
View Crossmark data

SPORTS PERFORMANCE
Validity and reliability of a 6-a-side small-sided game as an indicator of
match-related physical performance in elite youth Brazilian soccer players
Rodrigo Aquino
a,b
, Bruno Melli-Neto
c
, João Victor S. Ferrari
c
, Bruno L. S. Bedo
a
, Luiz H. Palucci Vieira
d
,
Paulo Roberto P. Santiago
a,c
, Luiz Guilherme C. Gonçalves
e
, Lucas P. Oliveira
e
and Enrico F. Puggina
a,c
a
Post-Graduate Program in Rehabilitation and Functional Performance, Ribeirão Preto Medical School, University of São Paulo, Ribeirão Preto,
Brazil;
b
Faculty of Physical Education, University of Estácio, Ribeirão Preto, São Paulo, Brazil;
c
School of Physical Education and Sport of Ribeirão
Preto, University of São Paulo, Ribeirão Preto, Brazil;
d
Human Movement Research Laboratory, Post-graduate Program in Movement Sciences, São
Paulo State University, Bauru, Brazil;
e
Department of Performance Analysis, Botafogo Football (Soccer) Club, Ribeirão Preto, Brazil
ABSTRACT
The aims of this study were: (i) to compare the external and internal load during a 6-a-side small-
sided game (6v6-SSG) according to age-group; (ii) to relate these parameters between the 6v6-SSG
and official matches; and (iii) to test the reliability of the 6v6-SSG. A total of 51 Brazilian youth
soccer players participated in this study (U11 [n = 16]; U13 [n = 10]; U15 [n = 9]; U17 [n = 8]; U20
[n = 8]). Three experiments were conducted. Experiment A: fifty-one U11 to U20 players were
submitted to 6v6-SSGs (n = 10 games; two for each age-group). Experiment B: thirty-two players
were randomized to also play official matches (n = 6 matches). Experiment C: thirty-five youth
players played the 6v6-SSG twice for test and retest reliability analysis. External load was obtained
using Global Positioning Systems and the internal load parameter was calculated through mean
heart rate. Statistical approaches showed progressive increases in all parameters according to
categories (U11< U13< U15< U17< U20; p < 0.05; ES = 0.42˗23.68). Even controlling for chronolo-
gical age, all parameters showed likely to almost certain correlations between 6v6-SSG and official
matches (r = 0.25˗0.92). Collectively, the proposed protocol indicates good reliability (CV
%=2.0˗12.6; TE% = 2.3˗2.7%; ICC = 0.78˗0.90).Thisresearchsuggeststhatthe6v6-SSGisan
alternative tool to indicate match-related physical performance in youth soccer players.
ARTICLE HISTORY
Accepted 12 April 2019
KEYWORDS
Association football; time-
motion analysis; construct
validity; GPS; sports sciences
Introduction
A variety of methods exist to assess soccer player physical
performance, both outside the match context (Stølen,
Chamari, Castagna, & Wisløff,2005), as well as during actual
competitions (Carling, Bloomfield, Nelsen, & Reilly, 2008).
Traditionally, assessments of the former have mainly been
conducted in laboratory settings, e.g., incremental treadmill
tests (Cerda-Kohler et al., 2016), the Wingate Anaerobic Test
(Meckel, Machnai, & Eliakim, 2009), and the maximal anaerobic
oxygen deficit test (Andrade et al., 2015). However, to improve
ecological estimations of physical performance, there has
been an emergent trend of using field tests to assess physical
fitness, e.g., Yo-Yo Intermittent Recovery Test (Yo-Yo IR)
(Krustrup et al., 2003), repeated sprint ability test (Impellizzeri
et al., 2008), HoffTest (Chamari et al., 2005), and Footeval
(Manouvrier, Cassirame, & Ahmaidi, 2016). In general, these
physical fitness tests have various purposes, notably to iden-
tify player individual strengths and weaknesses, investigate
the effects of training interventions, aid in rehabilitation pro-
grams following injury, and profile and monitor youth player
development (Svensson & Drust, 2005).
In order to ensure that the protocol and subsequent infor-
mation gathered from a given physical test have practical
meaning, large relationships between the player’s physical
fitness (derived from laboratory and field tests) and match
running performance assessments using time-motion analyses
are required, to gain insights into construct validity (Aquino,
Palucci Vieira, de Paula Oliveira, Cruz Gonçalves, & Pereira
Santiago, 2018). In the 1990s, Bangsbo and Lindquist (1992)
were the first to address this issue in professional soccer
players. Subsequently, there was a rapid increase in similar
research (Buchheit, Mendez-Villanueva, Simpson, & Bourdon,
2010; Castagna, Manzi, Impellizzeri, Weston, & Barbero Alvarez,
2010), mainly related to the growth and availability of a range
of portable match-play tracking equipment, e.g., acceler-
ometers, local position measurement, global positioning sys-
tems (GPS) (Carling et al., 2008). However, Drust, Atkinson, and
Reilly (2007) reported that the match-play running profile is
not-continuous; therefore its patterns defy precise modeling
and are difficult to replicate or predict, hindering the creation
of specific and valid tests outside the match context to eval-
uate soccer players.
Small-sided games (SSGs) have been widely used as
a training methodology in team sports (e.g. soccer)(Hill-Haas,
Dawson, Impellizzeri, & Coutts, 2011). Therefore, since SSGs are
highly specific and frequently used during the season, an
additional practical application could be their use as a fitness
indicator, mainly since: (i) more players can be evaluated at
the same time; (ii) SSGs involve technical and tactical
CONTACT Rodrigo Aquino rodrigo.thomaz@live.estacio.br Faculty of Physical Education, University of Estácio: Abrahão Issa Halach, 980, Ribeirânia, Ribeirão
Preto, São Paulo, Brazil
JOURNAL OF SPORTS SCIENCES
2019, VOL. 37, NO. 23, 2639–2644
https://doi.org/10.1080/02640414.2019.1608895
© 2019 Informa UK Limited, trading as Taylor & Francis Group

demands; and (iii) players are performing soccer training dur-
ing the test. Stevens, De Ruiter, Beek, and Savelsbergh (2016)
tested this hypothesis and found that SSGs alone cannot be
used as a valid and reliable fitness indicator. However, the
authors used a field test as the criterion measure (Yo-Yo IR)
to relate the external (e.g., GPS measures) and internal (e.g.,
heart rate) load obtained during the SSGs. With substantial
methodological alterations, the aims of the present study,
therefore, were: (i) to test the sensitivity of the 6v6-SSG to
discriminate external and internal load according to age (U11
to U20); (ii) to evaluate construct validity by relating these
parameters between the 6v6-SSG and official matches; and
(iii) to test the reliability of the 6v6-SSG.
Methods
Participants
A total of 51 Brazilian youth soccer players participated in this
study (U11 [n = 16]; U13 [n = 10]; U15 [n = 9]; U17 [n = 8]; U20
[n = 8]). The sample size was calculated using the correlation
coefficients between the Yo-Yo IR2 and running demands
reported in a previous study (Stevens et al., 2016) (statistical
power = 0.85; alpha = 5%; software G*Power, Dusseldorf,
Germany). All measurements were performed in-season as
part of the regular testing program of the teams analyzed.
The players were members of a professional soccer club that
plays in the 1
st
division of the São Paulo State Championship,
Brazil –the leading state-level tournament in the country
(Aquino et al., 2016). This study was approved by the
Research Ethics Committee of the School of Physical
Education and Sport of Ribeirão Preto/USP (CAAE:
61884716.9.0000.5659) and was conducted in accordance
with the Declaration of Helsinki. All participants and their
legal guardians signed an Assent and Consent Term,
respectively.
Experimental design
Three experiments were conducted. First, all 51 players were
submitted to 6-a-side small-sided games (6v6-SSG) to investi-
gate the sensitivity of the protocol (n = 10 games; two for
each age-group). In this step, we compared external and
internal load parameters between U11 to U20 age-groups
(Experiment A). Second, thirty-two players were randomized
to play official matches according to the official rules of their
category (U11: n = 16 players, n = 2 matches; U15: n = 8
players, n = 2 matches; U17: n = 8 players, n = 2 matches). This
experiment was performed to test the construct validity of the
6v6-SSG as an indicator of match-related physical performance
(Experiment B). Finally, thirty-five youth players (U11: n = 16;
U13: n = 10; U15: n = 9) played the 6v6-SSG twice to test and
retest (i.e., after seven days) the reliability analysis (Experiment
C). During all study phases, each test (6vs6-SSG or official
matches) was followed by at least 48-hours of recovery. In
addition, all tests were performed on artificial grass (~105
x 68 m) between 09:00 AM and 12:00 AM. Prior to the tests,
the players performed a habitual warm-up, consisting of
5-minutes of low-intensity running and 10-minutes of
coordinative running (i.e., skipping, dribbling, anfersen, hop-
serlauf, and kick-out).
Measures
6v6-SSG: the protocol was played including goalkeepers, in
6 × 6 min with 90 s of active rest between periods (playing
time: 36 minutes; pitch size = 49 x 25 m). Field size and
playing time were adapted from previous studies (Rebelo,
Silva, Rago, Barreira, & Krustrup, 2016; Stevens et al., 2016).
Consistent coach encouragement was given at all times dur-
ing the 6v6-SSGs. We recommended the players use the 1–2–
2–1 team formation (i.e., goalkeeper –central + external
defender –central + external midfielder –forward). All official
match rules were applied with the exception of “offside”and
yellow/red cards. The coaches organized the players in evenly
balanced teams to maintain the competitiveness of the play-
ing teams. The players were familiar with the 6v6-SSG struc-
ture from their training routines. Goalkeepers could be
involved with the play, but restricted to a maximum of 3 ball
possessions (Stevens et al., 2016). For the reliability analysis,
the teams were composed of the same players (e.g., test: team
A vs. team B; re-test: team A vs. team B).
Official matches: all matches were performed during the in-
competitive season. Two matches (i.e., one at home –against
top-ranking opponent team [draw]; and one away –against
worst-ranking opponent team [won]) were monitored for each
age-group (U11: 20ʹx20ʹwith 10 minutes of passive rest; U15:
30ʹx30ʹwith 15 minutes of passive rest; U17: 40ʹx40ʹwith
15 minutes of passive rest), to prevent the possible confound-
ing factor of situational variables (Aquino, Carling, et al., 2018).
External and internal measures were obtained (see below). The
team formation in all categories was 1–4–4–2 with occasional
minor variations. In the U11 category, the corner kick was
performed at the intersection between the lateral line and
penalty area (Palucci Vieira, Carling, Barbieiri, Aquino, &
Santiago, 2019). The players were allowed to drink water and
isotonic beverage (6% tangerine-flavored carbohydrate-
electrolyte) freely before, during, and after the matches.
External and internal load: external load was measured using
GPStechnology(Sports®,QSTARZ,BT-Q1300ST,5Hz,Taipei,
Taiwan). Previous studies have reported good coefficients of varia-
tion and error rate, i.e., <5% in all running outputs (Aquino et al.,
2018,2017). In this study, a complementary quality-control assess-
ment was conducted. The players wearing the GPS devices cov-
ered a known distance (calculated by tape measure) at different
intensities (6, 13, and 15km·h
−1
). The error rate was <5% for all
running categories. The GPS units were attached to the players’
shorts. All players used the same unit throughout the study. The 2D
reconstruction of the geographical coordinates (latitude and long-
itude) of each player at each time point were exported to a CSV
format file using QSports software (Taipei, Taiwan) for later analysis
in Matlab environment (The MathWorks Inc., Natick, USA). The
geographic coordinates were converted to Cartesian coordinates
(xy) and smoothed by a third order Butterworth digital filter
(cutofffrequency = 0.4 Hz) for further calculation of the total
distance covered (m), high-intensity running (speed
thresholds individualized according to distance covered > 60% of
peak game speed; U11 [mean≥14 km.h
−1
], U13 [mean≥15km.h
−1
],
2640 R. AQUINO ET AL.

U15 [mean≥16 km.h
−1
], U17 [mean≥17 km.h
−1
], U20
[mean≥18 km.h
−1
]), acceleration (≥2m·s
2
), and deceleration (≥
−2m·s
2
)(Castagna, Varley, Póvoas, & D’Ottavio, 2017). The mean
heart rate (HR
MEAN
) was calculated after beat-to-beat monitoring
usingthePolarTeamSystem(PolarElectroOY,Kempele, Finland)
(Hill-Haas,Rowsell,Coutts,&Dawson,2008).
Statistical analysis
Data normality and homogeneity of variance were checked
using the Shapiro-Wilk and Levene tests, respectively. The
p-value threshold was pre-fixed at 5% (p < 0.05).
Comparisons between internal and external load according
to age-groups were performed by ANOVA one-way
(Experiment A). When necessary, we used the Bonferroni post-
hoc and log-transformed data. Partial correlations (to control
for the possible effect of chronological age) of Pearson (para-
metric) and Spearman (nonparametric) were used to assess
the relationships between 6v6-SSGs and official matches
(Experiment B) according to each age-group. This was per-
formed while controlling for the possible casual relationships
between 6v6- SSGs and official matches. The test and retest
reliability of the internal and external load parameters was
tested using the t-test for dependent samples, coefficients of
variation (CV%), absolute and typical percentage error (TE),
and intraclass correlation coefficient (ICC). The typical percen-
tage error was obtained by dividing the resulting estimate of
the typical error by the mean for the participants in all trials,
then multiplying by 100 (Hopkins, 2000). In addition, the
smallest detectable difference (SDD) was calculated using the
following equation: SDD = [1.96 · √2 · (SD of the test and retest
differences · √1–ICC)](Haley & Fragala-Pinkham, 2006)
(Experiment C). The analyses were performed using the soft-
ware IBM SPSS Statistics for Windows, version 22.0 (Armonk,
NY: IBM Corporation®). A magnitude-based inferential (MBI)
statistical approach was also used (confidence level = 90%,
number of independent inferences = 1; maximum risk of
harm = 0.5%; minimum chance of benefit = 25%; benefit/
harm odds ratio = 66). Raw data outcomes in standardized
Cohen units were used (Effect Size [ES]). Quantitative chances
of higher or lower differences were assessed qualitatively as
follows: < 1%, almost certainly not; 1 −5 %, very unlikely; 5 −
25%, unlikely; 25 −75%, possibly; 75 −95%, likely; 95 −99%,
very likely; > 99%, almost certain. If the chance of higher or
lower differences was > 5%, the true difference was assumed
as unclear. Otherwise the effect was deemed clear (Hopkins,
Marshall, Batterham, & Hanin, 2009). For greater impact of the
results, only likely chances (> 75%) of differences were
considered true (Lacome, Simpson, Cholley, Lambert, &
Buchheit, 2018).
Results
Experiment A
Table 1 presents the values of the external and internal load
obtained during the 6v6-SSGs protocol in the analyzed age-
groups (U11 to U20). We observed progressive increases in all
parameters according to categories (U11 < U13 < U15 < U17
< U20; p < 0.05; ES = 0.42˗23.68; likely˗almost certain).
Experiment B
Figure 1 show the correlation coefficients (r) observed in the
relationships between 6v6-SSGs and official matches con-
trolled by chronological age. All parameters showed likely to
almost certain correlations (r = 0.25˗0.92). The distance cov-
ered in HIR was the best predictor in the relationship between
6v6- SSGs and official matches (r = 0.92; almost certain). Within
age-groups (Table 2), U11 and U17 players demonstrated likely
to almost certain relations in all variables (r = 0.46˗0.98), with
the exception of HR
MEAN
(r = 0.29˗0.54; unclear). The U15
category showed likely to very likely relations for TD, DAcHI,
and HIR (r = 0.42˗0.90), and unclear for DDecHI and HR
MEAN
(r = 0.004˗0.05).
Experiment C
In general, Table 3 indicates good reliability of the 6v6-SSG. In
relation to CV%, the DAcHI, DDecHI, and HIR demonstrated
greater values (CV% = 8.1˗12.6); however, TD and HR
MEAN
presented good reproducibility (CV < 5%). All parameters of
external and internal load demonstrated low values of typical
percentage error (TE = 2.3˗2.7%) and high ICC (0.78˗0.89). The
test and retest measures did not differ substantially according
to statistical analysis (p > 0.05; ES = −0.19˗0.25; possibly).
Discussion
The present investigation was conducted in youth soccer
players with the aim of verifying the sensitivity, validity, and
reliability of a 6v6-SSG as an indicator of match-related physi-
cal performance. The main findings were: (i) the protocol
presented sensitivity in discriminating age-related perfor-
mance (U11 to U20, see Table 1); (ii) even controlling for
chronological age, all external and internal measures showed
likely to almost certain correlations (see Figure 1); iii) the
Table 1. Mean (standard deviation) of external and internal load obtained during the 6-a-side small-sided games (n = 10 games; playing time: 36 minutes per game)
according to age-groups analyzed (U11 [n = 16 players], U13 [n = 10 players], U15 [n = 9 players], U17 [n = 8 players], U20 [n = 8 players].
Variables U11 U13 U15 U17 U20
TD (m) 2786.3 (304.6) 3421.6 (382.2) 5699.1 (770.2) 6086.0 (214.9) 6416.3 (149.4)
DAcHI (m) 42.1 (15.1) 75.9 (14.6) 191.4 (51.8) 243.9 (40.9) 298.8 (12.4)
DDecHI (m) 24.7 (10.6) 79.3 (17.7) 160.9 (47.4) 258.4 (34) 298.6 (22)
HIR (m) 179 (57.7) 413.8 (137.7) 447 (40.5) 629.3 (68.7) 757.8 (55.5)
HR
MEAN
(bpm) 160.8 (5.1) 174.5 (4.1) 188.2 (2.4) 189.4 (2.1) 188.4 (3.3)
Note: TD = Total Distance covered; DAcHI = Distance covered at High-Intensity Accelerations (≥2 m.s
2
); DDecHI = Distance covered at High-Intensity Deceleration (≥
−2 m.s
2
); HIR = Distance covered at High-Intensity Running (60.1 to 100% of maximum running speed); HR
MEAN
= Mean Heart Rate.
JOURNAL OF SPORTS SCIENCES 2641

distance covered in HIR during the 6v6-SSG explained 84% of
the variance in the distance covered in HIR during the official
matches; and (iv) the proposed protocol demonstrated good
reliability (see Table 3).
Sports scientists continuously address physical tests that
are closely linked to match physical demands. Krustrup et al.
(2003) reported that high-intensity running (> 18.0 km·h
−1
)
covered by the players during matches was correlated to Yo-
Yo test performance (r = 0.71) but not to
_
VO2
MAX
or an
incremental treadmill test in professional soccer players. In
addition, two previous recent systematic reviews indicated
trends of associations between several physical fitness tests
(e.g., Yo-Yo IR-1, IR-2; multistage fitness test, Carminatti test,
20-m shuttle run test, Zig-Zag test, Hofftest) and match run-
ning performance (e.g., total distance covered, sprinting, high-
intensity running) in youth soccer players (Palucci Vieira et al.,
2019; Paul & Nassis, 2015). However, one can argue that
laboratory and field tests per se are not sensitive enough to
profile in-match performance (Stølen et al., 2005). This state-
ment is supported by the apparent lack of laboratory/field
tests which reproduce construct validity with respect to the
motion types, directions, and intensities corresponding to
match running demands. Therefore, the representativeness
of these tests to predict official match physical demands
remains questionable (Drust et al., 2007). Thus, soccer practi-
tioners should use laboratory and field tests to prescribe
training sessions (e.g., maximal aerobic speed during incre-
mental tests) and not to predict/associate match-related phy-
sical performance.
In contrast, the integrated approach of the SSGs (i.e.,
including physical and tactical/technical aspects compared to
traditional strength and endurance exercises) has been
deemed high enough to promote soccer-specific adaptations
(Hill-Haas, Coutts, Rowsell, & Dawson, 2009; Hill-Haas et al.,
2011), supporting recent criticisms. Recently, Stevens et al.
(2016) observed that running demands during small-sided
games cannot serve as a valid and reliable fitness indicator
for professional and amateur soccer players. Nonetheless, in
this study, the 6v6-SSG presented good sensitivity, validity,
and reliability to indicate match-related physical performance
in youth soccer players. First, as expected, we observed pro-
gressive increments in running demands during the 6v6-SSGs
across the age-groups analyzed (U11 to U20). Second, we
verified positive likely to almost certain correlations between
Figure 1. Correlation coefficients (± 90% confidence interval and quantitative chances) of external and internal loads between 6-a-side small-sided games (n = 6
games) and official matches (n = 6 matches) controlled by chronological age. Note: TD = Total Distance covered; DAcHI = Distance covered at High-Intensity
Accelerations (≥2 m.s
2
); DDecHI = Distance covered at High-Intensity Deceleration (≥−2 m.s
2
); HIR = Distance covered at High-Intensity Running (60.1 to 100% of
maximum running speed); HR
MEAN
= Mean Heart Rate.
Table 2. Correlation coefficients (± 90% confidence interval and quantitative chances) of external and internal load measures of the 6-a-side small-sided games
(n = 6 games) with official matches (n = 6 matches) per age-group.
U11 U15 U17
Variables r ±90% CI QC R ±90% CI QC r ±90% CI QC
TD (m) 0.94 0.06 almost certain 0.62 0.45 likely 0.71 0.39 very likely
DAcHI (m) 0.97 0.03 almost certain 0.42 0.21 likely 0.92 0.14 almost certain
DDecHI (m) 0.98 0.02 almost certain 0.05 0.63 unclear 0.98 0.04 almost certain
HIR (m) 0.96 0.04 almost certain 0.71 0.39 very likely 0.97 0.06 almost certain
HR
MEAN
(bpm) 0.29 0.40 unclear 0.004 0.63 unclear 0.54 0.50 unclear
Note: TD = Total Distance covered; DAcHI = Distance covered at High-Intensity Accelerations (≥2m.s
2
); DDecHI = Distance covered at High-Intensity Deceleration (≥−2 m.s
2
);
HIR = Distance covered at High-Intensity Running (60.1 to 100% of maximum running speed); HR
MEAN
=MeanHeartRate;90%CI=Confidence Interval; QC = Quantitative
Chance.
2642 R. AQUINO ET AL.

running outputs in the 6v6-SSG and official matches (even
when controlling for chronological age), especially for distance
covered in HIR (explained 84% of the common variance
between 6v6-SSGs and official matches). A previous study
demonstrated that HIR is one of the best variables to discri-
minate won vs. lost matches (Aquino et al., 2018); therefore,
the current results demonstrated the construct validity of the
6v6-SSG. Third, collectively the protocol demonstrated good
reliability. Only the external load related to high-intensity
efforts (i.e., DAcHI, DDecHI, and HIR) presented greater values
of CV% (CV% = 8.1˗12.6), in agreement with previous studies
(Hill-Haas et al., 2008; Stevens et al., 2016). These results
promote novel insights for coaches and practitioners, primarily
by providing scientific evidence that the 6v6-SSG can be use-
ful to indicate match-related physical performance in youth
soccer players.
Furthermore, SSGs are frequently used in training routines
during the soccer season. Thus, soccer coaches in youth acad-
emy should implement the 6v6-SSG utilized in this study during
the training session to indicate match-related physical perfor-
mance, without wasting time with extended periods of physical
evaluation. Stevens et al. (2016) complement that SSGs can be
used to signal possible limitations in physical endurance fitness
of individual players and in case of doubt, additional maximum
endurance tests (e.g., Yo-Yo IR) can be performed for the
selected player, without having to test all players. However,
the external load parameters of the 6v6-SSG are lower
(~30–40% depending on the age-group) compared to official
matches. Therefore, coaches and practitioners should not use
the 6v6-SSG structure to simulate an official game.
We consider two main limitations of this study. First, we
used a cross-sectional design. Further studies should adopt
repeated assessment during the season and analyze the
consistency of the relationships reported in this
study. Second, in this study it was not possible to separate
the players according to their positional role. In contrast, this
study also has some novel aspects: (i) to the best of our
knowledge, this is the first study to demonstrate the possi-
ble use of SSGs to indicate match-related physical perfor-
mance in youth soccer players; (ii) external and internal
parameters were considered to define physical performance;
and (iii) the results proved to be efficient in supporting
conditioning coaches during evaluation periods. In addition,
we highlighted five crucial points for when clubs opt to use
SSGs as a match-fitness indicator: (i) consistent coach
encouragement should be given at all times during the 6v6-
SSG, as this can improve the reliability analysis; (ii) although
the SSGs are better standardized than official matches (e.g.,
less position-dependent; Stevens et al., 2016), we recom-
mend players use the 1–2-2–1 team formation, and the
same teams in the comparisons between two or more
moments (e.g., test: team A vs. team B; re-test: team A vs.
team B); (iii) all official rules should be applied with the
exception of “offside”andyellow/redcards–increasing
the representativeness of SSGs with official matches; (iv) in
practice, the players often like competitive playing teams,
therefore, the coaches and sports scientists should use
evenly balanced teams (Hill-Haas et al., 2011); and (v)
youth players should be familiar with the 6v6-SSG structure.
Practical application
Physical fitness evaluated outside the match context may well
guide training and research, but does not seem to have the best
potential to predict match running demands in youth soccer
players (Aquino, Palucci Vieira, et al., 2018). Our research suggests
that the 6v6-SSG could be an alternative tool to indicate match-
related physical performance in youth soccer players, with greater
specificity and representativeness (i.e., technical-tactical aspects)
than traditional approaches; therefore, coaches and practitioners
should include the 6v6-SSGs in physical assessment routines.
Conclusion
The overall aim of our study was to determine whether the
use of the 6v6-SSG internal and external load can serve as an
indicator of match-related physical performance for youth
soccer players. We performed three experiments to test this
hypothesis: (i) we compared running demands and HR
MEAN
according to age-groups (U11 to U20); (ii) we checked the
relationships between the 6v6-SSG and official matches; and
(iii) we verified the reliability analysis of the protocol used.
Collectively, we demonstrated that the 6v6-SSG presented
sensitivity, validity, and reliability to indicate match-related
physical performance in elite youth Brazilian soccer players.
Acknowledgments
This study was financed in part by the Coordenação de Aperfeiçoamento
de Pessoal de Nível Superior - Brasil (CAPES - Finance Code 001), FAPESP
(grant number - 2017/11698-0 and 2014/16164-5), and a Productivity and
Research Exchange with the University of Estácio (Ribeirão Preto; grant
number –2018/19; 2019/20).
Table 3. Reliability analysis for external and internal load obtained during the 6-a-side small-sided games (n = 6 games; playing time: 36 minutes per game).
Variables Test Retest CV% TE ICC SDD ES QC
TD (m) 3716.8 (1300.9) 3522.6 (1354.4) 4.4 (2.8) 86.9 (2.4%) 0.89 339.9 −.019 possibly
DAcHI (m) 90.2 (68.2) 98.7 (62.4) 10.0 (8.0) 2.2 (2.3%) 0.79 7.7 0.19 possibly
DDecHI (m) 75.3 (61.7) 84.1 (66.7) 12.6 (13.2) 1.8 (2.3%) 0.80 6.4 0.22 possibly
HIR (m) 315.0 (151.9) 345.2 (163.2) 8.1 (4.7) 8.8 (2.7%) 0.78 33.5 0.21 possibly
HR
MEAN
(bpm) 171.7 (12.2) 175.0 (14.8) 2.0 (1.7) 4.1 (2.4%) 0.87 15.4 0.25 possibly
Note: TD = Total Distance covered; DAcHI = Distance covered at High-Intensity Accelerations (≥2m.s
2
); DDecHI = Distance covered at High-Intensity Deceleration (≥−2 m.s
2
);
HIR = Distance covered at High-Intensity Running (60.1 to 100% of maximum running speed); HR
MEAN
= Mean Heart Rate; CV% = Coefficient of Variation; TE = Typical Error
Absolute (relative);ICC=IntraclassCoefficient Correlation (all values were almost certain); SDD = Smallest Detectable Difference; ES = Effect Size; QC = Quantitative Chance.
JOURNAL OF SPORTS SCIENCES 2643

Disclosure statement
No potential conflict of interest was reported by the authors.
ORCID
Rodrigo Aquino http://orcid.org/0000-0002-4885-7316
Bruno L. S. Bedo http://orcid.org/0000-0003-3821-2327
Luiz H. Palucci Vieira http://orcid.org/0000-0001-6981-756X
Paulo Roberto P. Santiago http://orcid.org/0000-0002-9460-8847
Enrico F. Puggina http://orcid.org/0000-0002-8379-2247
References
Andrade, V. L., Zagatto, A. M., Kalva-Filho, C. A., Mendes, O. C.,
Gobatto, C. A., Campos, E. Z., & Papoti, M. (2015). Running-based
anaerobic sprint test as a procedure to evaluate anaerobic power.
International Journal of Sports Medicine,36(14), 1156–1162.
Aquino, R., Carling, C., Palucci Vieira, L. H., Martins, G., Jabor, G.,
Machado, J., . . . Puggina, E. (2018). Influence of situational variables,
team formation, and playing position on match running performance
and social network analysis in brazilian professional soccer players.
Journal of Strength and Conditioning Research, epud ahead of print.
doi: 10.1519/JSC.0000000000002725.
Aquino,R.,CruzGonçalves,L.G.,PalucciVieira,L.H.,Oliveira,L.P.,Alves,G.F.,
PereiraSantiago,P.R.,&Puggina,E.F.(2016). Periodization training focused
on technical-tactical ability in young soccer players positively affects bio-
chemical markers and game performance. Journal of Strength and
Conditioning Research,30(10), 2723–2732.
Aquino,R.,PalucciVieira,L.H.,dePaulaOliveira,L.,CruzGonçalves,L.G.,&
Pereira Santiago, P. R. (2018). Relationship between field tests and match
running performance in high-level young Brazilian soccer players. The
Journal of Sports Medicine and Physical Fitness,58(3), 256–262.
Aquino, R., Vieira, L. H. P., Carling, C., Martins, G. H. M., Alves, I. S., &
Puggina, E. F. (2017). Effects of competitive standard, team formation
and playing position on match running performance of Brazilian pro-
fessional soccer players. International Journal of Performance Analysis in
Sport,17(5), 695–705.
Bangsbo, J., & Lindquist, F. (1992). Comparison of various exercise tests
with endurance performance during soccer in professional players.
International Journal of Sports Medicine,13(2), 125–132.
Buchheit, M., Mendez-Villanueva, A., Simpson, B. M., & Bourdon, P. C.
(2010). Match running performance and fitness in youth soccer.
International Journal of Sports Medicine,31(11), 818–825.
Carling, C., Bloomfield, J., Nelsen, L., & Reilly, T. (2008). The role of motion
analysis in elite soccer: Contemporary performance measurement tech-
niques and work rate data. Sports Medicine,38(10), 839–862.
Castagna, C., Manzi, V., Impellizzeri, F., Weston, M., & Barbero Alvarez, J. C.
(2010). Relationship between endurance field tests and match perfor-
mance in young soccer players. Journal of Strength and Conditioning
Research,24(12), 3227–3233.
Castagna,C.,Varley,M.,Póvoas,S.C.A.,&D’Ottavio, S. (2017). Evaluation of the
MatchExternalLoadinSoccer:MethodsComparison.International Journal
of Sports Physiology and Performance,12(4), 490–495.
Cerda-Kohler, H., Burgos-Jara, C., Ramírez-Campillo, R., Valdés-Cerda, M.,
Báez, E., Zapata-Gómez, D., . . . Izquierdo, M. (2016). Analysis of agree-
ment between 4 lactate threshold measurements methods in profes-
sional soccer players. Journal of Strength and Conditioning Research,30
(10), 2864–2870.
Chamari, K., Hachana, Y., Kaouech, F., Jeddi, R., Moussa-Chamari, I., &
Wisløff,U.(2005). Endurance training and testing with the ball in
young elite soccer players. British Journal of Sports Medicine,39(1),
24–28.
Drust, B., Atkinson, G., & Reilly, T. (2007). Future perspectives in the
evaluation of the physiological demands of soccer. Sports Medicine,37
(9), 783–805.
Haley, S. M., & Fragala-Pinkham, M. A. (2006). Interpreting change scores of
tests and measures used in physical therapy. Physical Therapy,86(5),
735–743.
Hill-Haas, S., Rowsell, G., Coutts, A., & Dawson, B. (2008). The reproduci-
bility of physiological responses and performance profiles of youth
soccer players in small-sided games. International Journal of Sports
Physiology and Performance,3(3), 393–396.
Hill-Haas, S. V., Coutts, A. J., Rowsell, G. J., & Dawson, B. T. (2009). Generic
versus small-sided game training in soccer. International Journal of
Sports Medicine,30(9), 636–642.
Hill-Haas, S. V., Dawson, B., Impellizzeri, F. M., & Coutts, A. J. (2011).
Physiology of Small-Sided Games Training in Football. Sports Medicine,
41(3), 199–220.
Hopkins, W. G. (2000). Measures of reliability in sports medicine and
science. Sports Medicine,30(1), 1–15.
Hopkins, W. G., Marshall, S. W., Batterham, A. M., & Hanin, J. (2009).
Progressive statistics for studies in sports medicine and exercise
science. Medicine & Science in Sports & Exercise,41(1), 3–13.
Impellizzeri, F., Rampinini, E., Castagna, C., Bishop, D., Ferrari Bravo, D.,
Tibaudi, A., & Wisloff,U.(2008). Validity of a repeated-sprint test for
football. International Journal of Sports Medicine,29(11), 899–905.
Krustrup, P., Mohr, M., Amstrup, T., Rysgaard, T., Johansen, J.,
Steensberg, A., . . . Bangsbo, J. (2003). The yo-yo intermittent recovery
test: Physiological response, reliability, and validity. Medicine and
Science in Sports and Exercise,35(4), 697–705.
Lacome, M., Simpson, B. M., Cholley, Y., Lambert, P., & Buchheit, M. (2018).
Small-Sided Games In Elite Soccer: Does one size fit all? International
Journal of Sports Physiology and Performance,13(5), 568–576.
Manouvrier, C., Cassirame, J., & Ahmaidi, S. (2016). Proposal for a specific
aerobic test for football players: The “Footeval.”.Journal of Sports
Science & Medicine,15(4), 670–677.
Meckel, Y., Machnai, O., & Eliakim, A. (2009). Relationship among repeated
sprint tests, aerobic fitness, and anaerobic fitness in elite adolescent
soccer players. Journal of Strength and Conditioning Research,23(1),
163–169.
Palucci Vieira, L. H., Carling, C., Barbieiri, F., Aquino, R., & Santiago, P. R. P.
(2019). Match running performance in young soccer players:
A systematic review. Sports Medicine,49(2), 289–318.
Paul, D. J., & Nassis, G. P. (2015). Physical fitness testing in youth soccer:
issues and considerations regarding reliability, validity and sensitivity.
Pediatric Exercise Science,27(3), 301–313.
Rebelo, A. N. C., Silva, P., Rago, V., Barreira, D., & Krustrup, P. (2016).
Differences in strength and speed demands between 4v4 and 8v8
small-sided football games. Journal of Sports Sciences,34(24),
2246–2254.
Stevens, T. G. A., De Ruiter, C. J., Beek, P. J., & Savelsbergh, G. J. P. (2016).
Validity and reliability of 6-a-side small-sided game locomotor perfor-
mance in assessing physical fitness in football players. Journal of Sports
Sciences,34(6), 527–534.
Stølen, T., Chamari, K., Castagna, C., & Wisløff,U.(2005). Physiology of
soccer: An update. Sports Medicine,35(6), 501–536.
Svensson, M., & Drust, B. (2005). Testing soccer players. Journal of Sports
Sciences,23(6), 601–618.
2644 R. AQUINO ET AL.