Rev Bras Fisiol Exerc 2021;20(3):325-34
doi: 10.33233/rbfex.v20i3.4178
ORIGINAL ARTICLE
Influence of match location, competitive level and
match outcome on match running performance in professional soccer players
Influências
do mando de jogo, nível competitivo e resultado da partida sobre o desempenho
físico em jogadores profissionais de futebol
Gabriel
Colatto Sagaz1, Lourenço Zini Moreira
Aresi1, Bruno Luiz Souza Bedo2, Filipe Mesquita3,
Paulo Roberto Pereira Santiago2,3, Angelo Melim Azevedo1, Helder Souza1, Eder
Gonçalves1, Rodrigo Aquino1
1LabSport, Departamento de Desportos,
Centro de Educação Física e Desportos, Universidade Federal do Espírito Santo,
Vitória, ES, Brasil
2Programa de Pós-Graduação em
Reabilitação e Desempenho Funcional, Faculdade de Medicina de Ribeirão Preto,
Universidade de São Paulo, Ribeirão Preto, SP, Brasil
3Escola de Educação Física e Esporte de
Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto, SP, Brasil
Received: 2th Jun 2020;
Accepted: 30th Apr 2021
Correspondence: Rodrigo Aquino, Universidade Federal do Espírito Santo,
Av. Fernando Ferrari, 514 Goiabeiras 29075-910 Vitória ES
Gabriel Colatto Sagaz:
gabrielcsagaz@hotmail.com
Lourenço Zini Moreira Aresi:
lourenco20@gmail.com
Bruno Luiz Souza Bedo:
bruno.bedo@usp.br
Filipe Mesquita: fmesquita@usp.br
Paulo Roberto Pereira Santiago: paulosantiago@usp.br
Angelo Melim
Azevedo: angelo.melim.azevedo@gmail.com
Helder Souza: helder_ldesouza@hotmail.com
Eder Gonçalves: egoncalves.ef@hotmail.com
Rodrigo Aquino: aquino.rlq@gmail.com
Abstract
Introduction: The context of
soccer matches (e.g., match location, competitive level, match outcome) can
impact the players’ physical performance. Aim: To investigate the
effects of the match location, competitive level and match outcome on match
running performance in professional Brazilian players. Methods: The
performance data were monitored in 16 male soccer players of the same team
during the following championships: 1st Division of the Capixaba
Championship 2020 (n = 8 matches; n = 64 individual observations) and 1st and
2nd Phases of the Brazil Cup 2020 (n = 2 matches; n = 16 individual
observations). The match running performance was calculated using Global Positioning
System: 1) total distance covered; 2) total distance covered in high-intensity;
3) total distance covered in high-acceleration. The t-test for independent
measurements was used to compare the home vs away, national vs. state, and win
vs. loss matches. The magnitude of the effects (ES) was calculated using
Cohen's "d". Multiple linear regression was used to estimate the
relative contribution of independent variables to the variation of dependent
variables. Results: The matches played at “home” resulted in greater
physical demands compared to “away” (p < 0,001-0,01, ES = moderate-very
large). Matches at the national level required greater running demands than
matches at the state level (p < 0,001, ES = very large). Winning matches
were more physically demanding compared to losing matches (p < 0,001-0,002,
ES = very large). Conclusion: Coaches can use this information to
prescribe training sessions that are more representative of the match context.
Keywords: task
performance and analysis; geographic information systems; exercise; sports nutrition sciences.
Resumo
Introdução: O contexto das partidas de futebol
(e.g., mando de jogo, nível competitivo, resultado da partida) pode impactar no
desempenho físico dos jogadores. Objetivo: Investigar os efeitos do mando de
jogo, nível competitivo e resultado final sobre o
desempenho físico em jogadores profissionais. Métodos: Foram monitorados
16 jogadores da mesma equipe durante a participação em duas competições: 1ª
Divisão do Campeonato Capixaba de 2020 (n = 8 partidas; n = 64 observações
individuais), 1ª e 2ª Fase da Copa do Brasil de 2020 (n = 2 partidas; n = 16
observações individuais). O desempenho físico foi obtido usando o Sistema de
Posicionamento Global: distância total percorrida; distância total percorrida
em alta intensidade; distância total percorrida em alta aceleração. Foi
utilizado o teste-t para medidas independentes. A
magnitude do efeito (ES) foi calculada usando o "d" de Cohen. A
regressão linear múltipla foi usada para estimar a contribuição relativa das
variáveis independentes para a variação das variáveis dependentes. Resultados:
Os jogos realizados “em casa” resultaram em maior demanda física em comparação
aos jogos “fora de casa” (p < 0,001-0,01, ES = moderado-muito grande). Os
jogos a nível nacional exigiram mais fisicamente dos jogadores em comparação
aos jogos a nível estadual (p < 0,001, ES = muito grande). Partidas com
vitória apresentaram maior exigência física quando comparado as partidas com
derrota (p < 0,001-0,002, ES=muito grande). O local da partida e o nível
competitivo têm maiores contribuições relativas para as variações na distância
total (R2 = 29%) e na distância em alta aceleração (R2 = 46%). Conclusão:
Os treinadores podem usar essas informações para prescrever sessões de treinamento
mais representativas ao contexto de jogo.
Palavras-chave:
análise de desempenho e tarefas;
sistemas de informação geográfica;
exercício físico; ciências da
nutrição e do
esporte.
The match
analysis is often used in team sports (e.g., soccer) to obtain objective
information about the players’ and teams’ performance and can assist coaches in
planning training sessions more in line with match demands. In addition,
information of match running performance (e.g., total distance covered at
high-speed running) is important physical performance indicator [1,2,3]. A
classic study about the evolution of match physical demands showed that the
total distance covered at high-speed running and the number of passes increased
50% across seven seasons of the England Premier League (season 2006-07 to
season 2012-13) [4]. This growth can affect the behavior of ball possession
during the matches, as a greater number of successful passes can result in
higher possession values [5].
In Brazil
context, there are few studies that characterized match running performance in
professional soccer players [6]. In 2007, Barros & Misuta
[7] verified that the players cover ~10000 ± 1024 meters during the games,
being 1,128 ± 361 meters in high intensity (≥ 19.0 km/h) in games of the
1st Division of the Brazilian Championship. Twelve years later, Vieira et
al. [8] observed similar values for total distance covered (10147 ± 971
meters); however, the authors demonstrated an increase of ~71% in high-speed
distances during 1st Division of the São Paulo State Championship. This
increase in match intensity can impact the match outcome. Faude
et al. [9] showed that 83% of goals during the German National League
were preceded by some high-intensity action. In Brazil, two studies observed
that matches that resulted in a victory during the 3rd and 4th Divisions of the
Brazilian National Championship had greater actions in high-intensity
compared to games with defeat [6,10]. Furthermore, previous studies
demonstrated that match location (i.e., home vs. away) and competitive level
(i.e., different divisions) can influence players' performance [11,12].
Regarding the
match location, studies have shown that home matches resulted in greater
physical and technical-tactical performance compared to away matches [13,14].
For example, higher values of total distance covered in high-speed running and
teams’ ball possession were verified in home vs. away matches [3,13,15]. These results
can be explained by the familiar environment of playing at home and a more
solid playing style [16,17]. Regarding the competitive level, the results in
the scientific literature are divergent. In Europe (i.e., Union of European
Football Associations (UEFA) and Italian League), high-level players covered
greater distances in high-speed running compared to low-level players [18,19].
In contrast, Aquino et al. [11] showed higher values of high-speed
actions in lower divisions (e.g., 4th Division of the Brazilian National
Championship) vs. Upper division (i.e., 1st Division of the São Paulo State
Championship).
Despite the
existence of previous studies in Brazilian soccer demonstrating the effects of
contextual variables on the physical and technical-tactical performance of
professional players [6,10,11,20], the data are restricted to teams from São
Paulo, which does not guarantee “nomothetic” observational designs [21].
Therefore, more studies are needed at different competitive levels (state and national
championships). This information can be crucial for sports coaches and
scientists in understanding the real physical and technical-tactical demands
according to the match context in Brazilian soccer. Thus, the aim of this study
was to investigate the effects of match location (home vs. away), competitive
level (state vs. national) and match outcome (won vs. loss) on match running
performance during the 1st Division of the Capixaba
Championship (edition 2020) and the Continental Cup in Brazil (edition 2020).
Participants and match sample
Match running
performance data (80 individual observations) were monitored in 16 professional
outfield male soccer players (mean ± standard deviation; age: 25 ± 10 years;
height: 177.1 ± 9.2 cm; body mass: 82.3 ± 7.5 kg) of the same team during two
competitions: 1st Division of the 2020 Capixaba
Championship (n = 8 matches; n = 64 individual observations), 1st and 2nd
Phases of the 2020 Continental Cup in Brazil (n = 2 matches; n = 16
observations). As an inclusion criterion, only data from players who
participated in ≥ 80 minutes were analyzed. In the present study, a
previous sample calculation was not performed, as it is a follow-up of the
entire season of the analyzed team. However, based on the study of Aquino et
al. [6] and based on the results found in the present study for the comparisons
of the total distance covered between the win (10019.6 ± 832.1 m) vs. loss
matches (8322.4 ± 1238.6 m), there is a real power = 0.82, with an effect size
of 1.60 and an α = 0.05 for the match sample size. This study complies
with the Code of Ethics of the World Medical Association (approved by the
Swansea University Ethics Advisory Board), approved by the Research Ethics
Committee of the Universidade de São Paulo at School of Physical Education and
Sport of Ribeirão Preto (protocol 108137/2015) and
was carried out in accordance with the Declaration of Helsinki. Written
informed consent was obtained from all players prior to the proceedings.
Measures
Dependent
variables: Global Positioning System (GPS) devices were used
(QSTARZ - 5 Hz, Taipei, Taiwan) [6]. GPS technology has been widely used to
measure running performance in team sports and its accuracy and reliability
have been previously determined [22,23]. Although the low acquisition frequency
of the devices used in this study (5 Hz) could potentially underestimate the
total distance covered at high-speed running compared to computerized optical
tracking [24], the quality control analyzes that we performed showed good
reliability (coefficient of variation ± 5%). All players used the same unit
throughout the competition season [25].
The GPS devices
were attached between the upper scapulae approximately at the T3-4 junction and
were activated 15 minutes before the matches. After the matches, data were
downloaded using the software (QStarz International
Co., GPS View, version 1.2.24) and exported to CSV format for further analysis
in the Matlab® environment (The Math Works Inc
Natick, USA). Using specific scripts [10], the geographic coordinates were
converted into Cartesian coordinates (x, y) and smoothed by a Butterworth
digital filter (third order; cutoff frequency = 0.4 Hz) for later calculation
of the total distance covered (TD; meters), total distance covered in high speed running (HSR; ≥ 18 km/h; meters) and total
distance covered in high acceleration (> 2m/s2; meters). The third-order
Butterworth filter with a cut-off frequency of 0.4 Hz was used to smooth the
positional data according to a quality control assessment of the GPS units in a
pilot study and in previous research.
Independent
variables: three independent variables were considered [10,11]:
1) match location (“home” vs. “away”); 2) competitive level (state vs. national
championship) and; 3) match outcome (loss vs. win).
Regarding the match location, during the 1st Division of the 2020 Capixaba Championship, 1st and 2nd
Phases of the 2020 Continental Cup in Brazil, 10 matches were monitored,
including seven home matches (n = 56 observations) and three away matches (n =
24 observations). Regarding the competitive level, data were collected from
eight matches at the state level (n = 64 observations) and two games at the
national level (n = 16 observations). Over the two competitions analyzed, the
reference team in this study had nine wins (n = 72 observations) and one loss
(n = 8 observations).
Statistical analysis
The normality
and homogeneity of variance of the data were confirmed by the
Kolmogorov-Smirnov and Levene tests, respectively.
Therefore, the results were analyzed and presented as mean and standard
deviation (SD). To compare the dependent variables (total distance, high-speed
running, and high-acceleration) according to match location, competitive level
and match outcome, the t-test for independent measures was used. The magnitude
of the effect (Effect Size [ES]) was calculated using Cohen's "d"
[26]. The values of "d" were considered as: d < 0.1 (trivial), 0.1
< d < 0.2 (small), 0.2 < d < 0.5 (moderate), 0.5 < d < 0.8
(large), d > 0.8 (very large). Multiple linear regression (stepwise method)
was used to estimate the relative contribution of the independent variables
(match location, competitive level, and match outcome) to the variation of the
dependent variables. Data for regression analysis assumed homoscedasticity,
independence, normal distribution, and no multicollinearity between the
independent variables. The level of significance was set at 5% (p < 0.05).
Analyzes were performed using the software IBM SPSS Statistics, for Windows,
version 22.0 (IBM Corporation).
Table I shows
the effects of contextual variables on the match running performance over the
season. Home matches resulted in greater physical demands compared to away
matches (p < 0.001-0.01, ES = moderate-very large). Win matches presented
higher physical demands than loss matches (p < 0.001-0.002, ES = very
large). Furthermore, national level matches presented greater physical demands
compared to state level matches (p < 0.001, ES = very large).
Multiple linear
regression analysis showed that all regression coefficients were significant (p
< 0.05). Regarding the relative contribution of each independent variable,
we verified that 29%, 8% and 6% of the total variance of the total distance
covered was explained by the match location, competitive level, and match
outcome, respectively. In relation to the total distance covered in high-speed
running, 22% and 8% of its total variance was explained by the competitive
level and the match outcome, respectively. Finally, 46% and 6% of the total
variance of the high acceleration was explained by the competitive level and
match location, respectively.
Table I - Effects of
contextual variables on physical performance in professional soccer players
(mean ± standard deviation)
ES = Effect Size;
High-speed running = total distance covered above 18 km/h; High acceleration =
total distance covered above 2 m/s2
The aim of the
present study was to examine the effects of match location (home vs. away
home), competitive level (state vs. national) and match (loss vs. win) on the
match running performance in a professional Brazilian soccer team. In addition,
the relative contributions of each independent were analyzed. The results
showed that home, winning, and national level matches presented greater running
demands compared to their counterparts. Furthermore, we verified that the match
location and the competition level have greater relative contributions to the
variations in the total distance (R2 = 29%) and in the total distance covered
in high acceleration (R2 = 46%), respectively.
Few studies have
analyzed the independent and interactive effects of match contextual variables
on running performance in professional players in Brazilian soccer [15,17,27].
Our data for the total distance covered (i.e., TD = 9849.9 ± 1010.3 m) are like
the values found in professional soccer players from Serie A of the Brazilian
National Championship in the 2007 season (TD = 10012 ± 1924 m) [7] and the 1st
Division of the São Paulo Championship in the 2016/2017 season (TD= 10147 ±
971m) [8]. However, we verified lower values for the distance covered at high-speed
running (HSR = 849.1 ± 113 m) compared to matches of the 1st Division of the
São Paulo Championship in the 2015/2016 season (HSR = 1924.9 m) [8].
Research about
the influence of match location on players’ performance has been received
extensive coverage in the last two decades [13,15,28]. Previous studies showed
that teams change their playing style and match strategies according to the
match location [3,5,28,29]. Therefore, the match location was considered an
important factor that influence offensive and defensive performance in soccer
[30,31]. For example, Thomas et al. [32] showed that the home advantage
occurred in 60.7% of the 4426 matches in the English Football Premiership. Lago
and Martín [5] showed that teams playing at home have more possession than
visiting teams, using data from 170 matches from the Spanish Football League
(season 2003-4). The same behavior was found by several other studies
[27,33,34], in line with our findings. In this study, home matches presented greater
running outputs than away matches. Previous research in sports psychology has
listed some factors that may explain this behavior, such as spectators’ effects
[35,36], familiarity with the location and absence of travel [37]. In addition,
the tactical strategy adopted by the team at home (i.e., controlling the match
with ball possession strategy) may explain this advantage [5]. In fact, players
must adapt physiologically and psychologically to different match scenarios.
For example, Pollard [16] showed that players are more familiar with the
installation and environment when playing at home, which can result in more
positive physiological and psychological states [38].
Another
important variable to explain the results refers to the competitive level. In this
study, when the analyzed team played the national championship, we observed
greater TD and match intensity (higher values of high-speed running and high
acceleration). Similar findings were reported by Rampinini
& Coutts [19], in a study of players from a semi-finalist UEFA Champions
League club. This behavior seems to be more physically demanding due to the
need to "surprise" and "uncontrol" the opposing team's
playing style, making it difficult for opponents to retrieve the ball, which
indicates that players must be physically prepared to play against strong
opponents. Although there are studies showing that matches against weak
opponents have a higher percentage of possession than matches against strong
opponents. This demonstrates that stronger teams dominate possession against
their weaker opponents and, therefore, print less intensity of play [3,15,34].
This can be explained by the system and style of play adopted by the team,
preferring to “control” the game by maintaining possession of the ball. This
style is known as positional attack, in which the team has a slower game,
usually using short passes and seeking to create spaces between the opposing
defensive lines [3].
Regarding the
match outcome, there are two possible reasons to explain the lower match
intensity when the analyzed team lost. First, Lago [3], in a case study of the
Espanyol Fútbol Club (season 2005-2006), showed that ball possession was
greater when the team was losing vs. winning. This can be explained by the
change in the team formation and playing style adopted by the team, preferring
to “control” the game by positional attack. Under these conditions, it is
suggested that players present low running intensity, as demonstrated in the
present study, in which the variables that determine the intensity of the game
(high-speed and high acceleration) were significantly lower when the team lost.
In addition, Moura et al. [39] showed reduced values of team surface
area without vs. with ball possession in professional Brazilian soccer.
Therefore, as reduced tactical performance is expected to occur simultaneously
with reduced physical performance [40], an excessively defensive strategy, such
as when the team mainly adopts a more compact style of play and/or without
possession of the ball for a long time, can reduce the running demands.
Overall, this
study supports the criticisms of Mackenzie & Cushion [30], that highlight
the importance of considering the contextual factors of the match in the
analysis of soccer performance. Coaches can use this information to prepare
their teams for the specific competitive situation in which they will play
[15]. Therefore, coaches must be aware that the physical demands of the match
are influenced by match location, competitive level, and match outcome. This
information allows the creation of strategies to maximize the athletes'
physical performance during the matches.
It is important
to highlight that this study has some limitations. First, the fact that the
influence of the contextual variables of the match was followed by only one
team limits the generalizability of the data. Second, there is a lack of
information in this study about possible influences of other variables related
to the match context, such as team formation (1-4-4-2, 1-4-3-3), and
technical-tactical performance indicators. However, several studies have
evaluated the influence of contextual variables on technical-tactical
performance [14,27,41,42]. Third, the match sample size for each independent
variable were different and this can be considered a limitation of this study.
However, previous studies with the same observational design often show this
sampling difference, largely due to real observation throughout the season,
which makes it difficult to balance the number of matches for each context
analyzed [3,6,10,11,12,13].
The findings of
this study are new and provide pertinent information about physical
requirements during soccer season. The analyzed players covered greater total
distance, distance in high-speed running (> 18 km/h) and distance in high
acceleration (> 2 m/s2) in home location, national level and when
win the matches. Coaches can use this information to prescribe more
representative training sessions, and to adapt post-match recovery strategies
in relation to the physical requirements during the matches. Further studies in
Brazilian soccer can consider tactical-technical indicators and include other
contextual variables (e.g., coach change, travel).
Conflict of interest
No conflict of interest
with relevant potential.
Financing source
Fundação de Amparo à Pesquisa do Estado de São Paulo
(FAPESP; processos: 2019/17729-0; 2014/1616 4-5) e Coordenação de
Aperfeiçoamento de Pessoal de Nível Superior (CAPES) – Código de financiamento
001.
Author’s contributions
Conception and design
of the research: Sagaz CS, Aresi
LZ, Azevedo AM, Souza H, Aquino R. Data collection, statistical analysis and
writing of the manuscript: Sagaz CS, Aresi LZ, Bedo B, Mesquita F,
Santiago PRP, Gonçalves E, Azevedo AM, Souza H, Aquino R. Critical review of
the manuscript: Sagaz CS, Aresi
LZ, Gonçalves E, Aquino R. Publication of the document: Sagaz
CS, Aresi LZ, Aquino R.
Acknowledgement
Our thanks to the
Vitória Football Club (Espírito Santo State). In
addition, our thanks to the Prof. Dr. Danilo Sales Bocalini
and Prof. Dr. Ubirajara de Oliveira for the
suggestions.