1. Introduction
The analysis of performance indicators (PIs) is one of the emerging lines of research in sport sciences [
1]. Using notational analysis, different game indicators have been recorded that make it possible to study the performance of the teams and players during the competition. A series of situational variables has been identified that affect performance indicators such as: playing at home [
2], being ahead on the scoreboard [
3], the partial and final result [
4], the quality of the opposition [
5], and the competition phase [
6], among others.
The events provoked worldwide by COVID-19 have also affected the sport context, making it necessary to study the influence of this variable on PIs. The COVID-19 pandemic declared by the WHO on 11 March 2020 affected everyone’s life both physically and psychologically. This disease, together with the severe measures of confinement in most of the world, brought the suspension of men’s and women’s competitions in all sports. All those who are professionally dedicated to sport, both at the individual and collective level, were affected at the psychological level during and after the pandemic, which caused unprecedented harm [
7].
This exceptional circumstance provided a natural experimental situation in the sport context. The return to sport practice at the top level was accomplished without the presence of spectators as a security measure. Studies on the presence of spectators and their influence on the home advantage have a long history [
8]. However, the available research investigating how the absence of spectators can affect the behavior of teams and players in their sport performance and the results of the competition is scarce. In particular, the research focused on soccer showed a negative impact when returning to play, due to the accumulation of matches, and the accumulation of minutes played in a short space of time for the players, after several months with minimal training; the teams were also very uncertain as to the psychological status of each player [
9]. Then, in order to continue with the competition, it was decided to adopt several measures: to be able to make up to five substitutions in three different blocks during the match, in order to distribute the minutes among all the players on the team, thus obtaining a lower physical load; and it was also decided to have no spectators, so that there would not be large crowds in the same place during a short period of time and so that the numbers of infected people would not soar.
Some studies such as that by Villaseca-Vicuña et al. [
10] stated that the compulsory confinement negatively affected women players in terms of their level of well-being, and the technical staffs had to work much more to monitor individual players and teams so that the absence of normality affected them as little as possible, and did not imply a loss of performance.
In addition, the soccer referees have an essential role on the playing field, and when in a situation of having no spectators, they are witness to any inappropriate comment from the players or technical staff. This can change their behavior or way of working and influence their decision making, which would be different with stadiums full of supporters and the noise that entails. All these decisions, together with the severe restrictions that the referees had to adopt following the competition norms, which caused them to be isolated to a large degree without contact with their relatives or friends as they would normally have had, meant that they were influenced and affected when making decisions during their refereeing [
11]. The decision making of the referees in matches without spectators was less favorable to the home team [
12].
During the 2018–2019 season, there was the inclusion of a regulatory modification that can directly affect the development of the game. This is the VAR (Video Assistant Referee). This tool aims to help the main referee during his intervention to avoid serious and manifest errors during the match. Its application occurs in very specific situations, such as: goals and infractions leading to goals; decisions on penalties and infractions leading to such decisions; direct red cards; mistaken identity. Its application during the 2018 FIFA World Cup has meant a technological revolution in professional soccer, with this technology being implemented in more than one hundred competitions. The influence of the application of this tool is beginning to be studied. Lago-Peñas et al. [
13] demonstrated that the VAR system did not substantially modify the form of play in professional soccer. They identified a decrease in the number of offsides, fouls and yellow cards, as well as an increase in the number of minutes added to playing time in the first half and full match, but not in the second half.
Regarding the home advantage, the percentage of points won by the home teams is over 50% in different leagues worldwide [
14]. Territoriality has been shown in the protective response that the team presents when playing at home to defend its own territory, consequently increasing its probability of winning more points at home than when playing away [
15]. Differences have been found in the different PIs in favor of the home team in the first division of men’s soccer in Spain [
2]. These differences are evident in the German league, whether the teams play with or without their fans, and there is also a tendency to play worse at home and better away when spectators are not present [
16]. In particular, the teams committed more fouls after the pandemic, the home teams scored more goals in the second half of the matches before the pandemic, and the home advantage decreased, thus the return to competition in the absence of spectators had a considerable effect [
17].
Lastly, the different situational variables acquired a great deal of importance during a match and throughout the championship [
18]. Thus, the venue and result are variables that directly affect the players. The support of the spectators and a greater effort on the part of the local team when they have an unfavorable result are important for reaching a tie or getting back into the match. Zajonc’s theory of social facilitation (1965) [
19] and the theory of territoriality [
15] help to justify the importance of the spectators for the home advantage in team sports.
After analyzing the theoretical framework, it is hypothesized that there will be changes in performance indicators depending on the time point analyzed: pre-COVID, COVID and post-COVID. Furthermore, it is hypothesized that, despite the variations that may exist depending on the period of time analyzed, there will be permanent performance indicators. The prior literature that analyzes the evolution of performance indicators in soccer, bearing in mind the influence that COVID may have had, is limited. Thus, the first aim of this research was to analyze the influence of COVID-19 on performance indicators in the Spanish first division of men’s soccer depending on the moment in time: pre-COVID, COVID and post-COVID. The second aim was to analyze and ascertain the persistence of the PIs during the analyzed seasons.
3. Results
Table 2 shows the descriptive (mean and standard deviation) and inferential results of the influence of COVID on the different PIs. Significant differences were found in eight PIs of the pre-COVID and COVID seasons, with increasing attacks and decreasing total and on-goal shots, corners, offsides, saves, dangerous attacks and tackles. From the COVID season to the post-COVID season the different variables became more stabilized and significant differences were only found in five of them, with increasing free kicks and decreasing yellow cards, total shots, attacks, and tackles. The greatest change was shown in the pre-COVID/post-COVID comparison, with significant differences in eleven variables: increases in free kicks and decreases in yellow cards, total, on- and off-goal shots, corners, offsides, saves and fouls committed, dangerous attacks and tackles.
For the covidtime variable, significant differences were found in 11 of the 15 PIs: yellow cards, total, on-goal and off-goal shots, free kicks, corners, offsides, saves, attacks, dangerous attacks, and tackles.
Figure 1 presents a visualization of the differences found in the three COVID moments analyzed.
Table 3 shows the descriptive (mean and standard deviation) and inferential results of the influence of venue on the different PIs. Significant differences were found in 11 of the 15 indicators: possession, total, on-goal and off-goal shots, corners, offsides, saves, attacks, dangerous attacks, total passes, and tackles.
Table 4 shows the descriptive (mean and standard deviation) and inferential results of the influence of the result on the different PIs, as well as the differences among groups.
For the result variable, differences were found in 12 of the 15 PIs: yellow cards, red cards, total, on-goal and off-goal shots, free kicks, corners, offsides, saves, attacks, dangerous attacks, and total passes.
Table 5 shows the inferential results of the interaction of the independent variables with the PIs. No significant differences were found in the interaction among the three independent variables; thus, the table just presents the differences in the interaction between pairs of variables, only showing the PIs that revealed significant differences.
Regarding the covidtime–venue interaction, significant differences were found in 7 out of the 15 PIs: total, on-goal and off-goal shots, corners, saves, attacks and dangerous attacks. The covidtime–result interaction revealed significant differences in 4 out of the 15 PIs: possession, off-goal shots, dangerous attacks and total passes. The venue–result interaction revealed significant differences in 4 out of the 15 PIs: yellow cards, free kicks, offsides and fouls committed.
In the covidtime–venue interaction, on-goal shots and attacks had a high observed power (0.5–0.8), with the rest of the indicators having a very high observed power (>0.8). In the covidtime–result interaction, the variables of possession, off-goal shots, dangerous attacks and total passes had very high observed power (>0.8). In the venue–result interaction, the variables of yellow card and offsides had a high observed power (0.5–0.8), while for free kicks and fouls committed it was very high (>0.8).
Table 6 presents the results of the autocorrelations of the PIs, showing significant differences in 13 of them, but not in 2.
Figure 2 presents the significant differences existing in the performance indicators over the different seasons. Yellow and red cards, total, on-goal and off-goal shots, free kicks, corners, saves, attacks, dangerous attacks, total passes and tackles have a low autocorrelation, while possession has a high autocorrelation [
26]. The means for yellow and red cards, free kicks and tackles increase as the seasons advance, as they have a positive autocorrelation; while in contrast, possession, total, on-goal and off-goal shots, corners, saves, attacks, dangerous attacks, and total passes have a negative autocorrelation, which indicates that their means decrease over the seasons.
5. Conclusions
This study analyzed how the worldwide pandemic and attendant situations have affected different PIs and the result in soccer matches.
Significant differences were found to be evolving in PIs over the different seasons—pre-COVID, COVID and post-COVID—due to the absence of spectators, referee bias, accumulation of matches, change in the rules (five substitutions during the match in three different blocks compared to the three allowed before) players’ physical and psychological state, etc.
Thus, the effect of the COVID-19 pandemic on the Spanish first division of men’s soccer generated an adaptation of play on the part of the teams, showing a decrease in offensive actions and a more passive game. Given these findings, teams should adapt their training to the current game demands, in which there are fewer goals. It is essential to work on stopped-ball plays as given the increase in such actions, they acquire considerable importance for the result.
6. Practical Applications and Limitations
The main practical applications that can be found for this research are as follows: (i) knowledge of the match indicators and how they have affected play before, during and after a worldwide pandemic; (ii) evolution of the match indicators over several seasons; (iii) influence of the PIs in the result with the absence or presence of supporters; (iv) knowledge of how a worldwide pandemic can affect teams’ playing styles.
The main limitation in this study is the difference in the number of matches analyzed before, during and after the pandemic, with a higher number of matches before. There is a lack of existing literature on these PIs in different top-level leagues. Due to the congested calendar after the three months of confinement, and the difficulties that this produced (disease or sport injury), there were teams that could not play normally.
One of the prospects that would allow the continuity of this work could be the study of the influence of VAR on performance indicators, as well as the influence of match status before, during and after the pandemic.