Influence of Game Indicators on the Ranking of Teams in the Spanish Soccer League
by
, , and
José Fernández-Cortés
1,2
,
Juan M. García-Ceberino
3,4,*
,
Javier García-Rubio
1,2 and
Sergio J. Ibáñez
1,2
1
Optimization of Training and Sports Performance Research Group (GOERD), Faculty of Sports Science, University of Extremadura, 10003 Caceres, Spain
2
Faculty of Sports Science, University of Extremadura, 10003 Caceres, Spain
3
Universidad de Extremadura, Facultad de Educación y Psicología, Avenida de Elvas s/n, 06006 Badajoz, Spain
4
Faculty of Education, Psychology and Sports Science, University of Huelva, 21007 Huelva, Spain
*
Author to whom correspondence should be addressed.
Appl. Sci. 2023, 13(14), 8097; https://doi.org/10.3390/app13148097
Submission received: 28 May 2023
/
Revised: 2 July 2023
/
Accepted: 10 July 2023
/
Published: 11 July 2023
(This article belongs to the Special Issue Advances in Sports Performance Analysis and Applied Technologies)
Abstract
:Soccer is dominated by game indicators that can influence the performance of teams and their players. Therefore, this study analyzed the influence of game indicators on the partial and final ranking of Spanish LaLiga teams, independently for the pre-COVID-19, COVID-19 and post-COVID-19 periods. In addition, the statistical differences between the pandemic periods were identified. A total of 2660 Spanish LaLiga matches played over seven seasons (from the 2014–2015 to the 2020–2021 season) were analyzed. The game indicators analyzed were the yellow cards, red cards, ball possession, total shots, shots on goal, shots off goal, free kicks, corners, offsides, goalkeeper saves, fouls committed, attacks, dangerous attacks, total passes, and tackles. Data were collected from the official Spanish LaLiga website, and recorded on a post hoc observation sheet. The intra-observer concordance was almost perfect (Cohen’s kappa values > 0.83). In each pandemic period, the findings indicated that the statistically significant game indicators had a greater influence on the final ranking, with an intermediate and large effect (η2 ≥ 0.060), than on the partial ranking (little or no effect). In this regard, the LaLiga teams ranked in a European competition position (final ranking) reported a higher ball possession (p < 0.001) and total passes (p < 0.001). A higher ball possession allowed them to take more shots (offensive actions), and therefore to have a better chance of winning. Similarly, these game indicators were higher post-COVID-19, compared to pre-COVID-19 and during COVID-19. This is interesting information for the preparation and management of matches.
1. Introduction
Soccer is one of the most popular sports in the world. In elite soccer, to achieve success in league rankings, a team must win the highest number of matches, influenced by game indicators/game actions (e.g., free kicks, cards, ball possession, shots, corner kicks, goalkeeper saves, fouls committed, or offsides) [1]. For this reason, there are assistants who collect statistical data from both their own, and the opposing, players, to identify potential strengths and weaknesses [2]. The authors of this paper highlight the influence of the aforementioned game indicators on the final result of a match, and how they thus become performance indicators. In addition, the opponent’s quality and the match location are situational variables that can influence the performance of players and soccer teams [3].
Some of these game indicators and situational variables have been studied in the Spanish league, LaLiga [2]. The study [2] shows that the match location influences the final result in favor of the home team. A similar study [1] indicates that the match location, the effect of scoring first, and the quality of the opposing team influence the final result. Home teams are motivated by their fans, creating a feeling of territoriality, and visiting teams are pressured [4] during matches. This means that home teams tend to take more shots and, therefore, the opposing team’s goalkeeper makes more saves [5]. The differences in the movement profiles (intensities) in matches depending on the quality of the opponent have also been studied in the elite Spanish soccer leagues, in the First and Second Division [6], and the Second B Division [3]. Furthermore, it has been shown that teams participating in international competitions (e.g., the Champions League or the Europa League) are composed of coaches, assistants, and players who are more likely to withstand pressures, such as playing in crowded stadia, playing outside of the country, long-distance travel, etc. [7,8].
The COVID-19 pandemic interrupted the major European soccer leagues in mid-March 2020. The first league to resume competition was the Bundesliga on May 16. Subsequently, other European leagues resumed competition (e.g., LaLiga, the Premier League, Serie A, etc.). Except for the Russian League (at 10% capacity in each stadium), these leagues resumed competition without the attendance of fans [9]. Specifically in LaLiga, the competition was interrupted in March (the 27th round), and resumed on June 8 [7]. The COVID-19 pandemic brought major changes to the elite soccer leagues after their resumption. As a result, the Fédération Internationale de Football Association introduced new rules: (1) the banning of fans, in so-called ghost matches, or matches with a very limited fan attendance [9], and (2) an increase in player substitutions from 3 to 5 players [10]. For these reasons, the COVID-19 pandemic presents a unique situation in history for investigating the influence of game indicators that occurred in elite soccer matches.
Regarding performance in elite soccer matches, the current results on the effects of the COVID-19 pandemic report that the home advantage decreased in ghost matches [11,12]. In this regard, the number of fouls committed, and cards shown to visiting teams decreased, as the result of a lesser perceived social pressure on referees from the stands [13]. Moreover, a study on the four major leagues (i.e., Bundesliga, LaLiga, the Premier League, and Serie A), comparing statistics from before and after the COVID-19 pandemic, indicates a decrease in the expected points and goals scored by the home team. The decrease in the home team’s advantage is due to the absence of fans in the stadium during the COVID-19 pandemic [14]. In other words, the performance of the home team is halved when the stadium is empty [7]. Nowadays, with stadia again crowded with fans, the video assistant referee helps referees in their decision-making during matches [15]. It can only be used in the following four basic situations: (1) goal, (2) penalty, (3) red card, and (4) player mix-up (in awarding a yellow, yellow–red, or red card) [9].
The results of soccer matches can be influenced by different game indicators [1,3,16]. This study analyzed, independently for the pre-COVID-19, COVID-19, and post-COVID-19 periods, the influence of game indicators on the partial (Study 1-S1) and final (Study 2-S2) ranking of Spanish LaLiga teams over seven seasons. The statistical differences between the pandemic periods were also identified in the two studies. For each pandemic period, we hypothesized that: (1) in the partial ranking, there would be differences in the game indicators between balanced, unbalanced, and very unbalanced teams (S1); (2) in the final ranking, there would be differences in the game indicators between the Europe, mid-table, and relegation teams (S2); and (3) in both studies, there will be differences in these game indicators between the pre-COVID-19, COVID-19, and post-COVID-19 periods, according to the team’s ranking.
2. Materials and Methods
2.1. Study Design
For three different pandemic periods (pre-COVID-19, COVID-19, and post-COVID-19), a notational study was performed, using an arbitrary observation code [17]. In addition, the study was carried out in a natural context (LaLiga), and the researchers did not intervene in any way. It was divided into two stages: (S1) the analysis of different game indicators based on the mid-season (partial) ranking; and (S2) the analysis of the same game indicators based on the season’s end (final) ranking.
2.2. Sample and Procedure
A total of 2660 matches in LaLiga (the First Division of men’s soccer), played over seven seasons, were analyzed (pre-COVID-19: the 2014–2015, 2015–2016, 2016–2017, 2017–2018, and 2018–2019 seasons; COVID-19: the 2019–2020 season; and post-COVID-19: the 2020–2021 season). All seasons developed normally, except for the 2019–2020 season, when there was a three-month hiatus due to the COVID-19 pandemic, and the adoption of new rules for its resumption.
LaLiga is composed of 20 teams (38 matches per season), and is characterized by the aesthetic aspect of the game, and by the purpose of the teams to have a greater control of the game [3]. The scoring system of the Fédération Internationale de Football Association is used; i.e., three points for a win, one for a draw, and zero for a loss. Therefore, the sum of the points obtained in each match determines the partial and final ranking, and the team with the most points at the end of each season is the winner.
The following data (game indicators) were collected for each team and match played: (1) yellow cards, (2) red cards, (3) ball possession, (4) total shots, (5) shots on goal, (6) shots off goal, (7) free kicks, (8) corners, (9) offsides, (10) goalkeeper saves, (11) fouls committed, (12) attacks, (13) dangerous attacks, (14) total passes, and (15) tackles. These were the dependent variables. The quality of the opponents, the independent variable, was divided into three different groups according to the partial ranking (S1: equaled = 0–5 places; unbalanced = 6–10 places; very unbalanced = 11–19 places), and the final ranking (S2: Europe position = 1–7 places; middle position = 8–17 places; relegation = 18–20 places). The partial ranking referred to the position of each team at the mid-season point, while the final ranking referred to the position of each team at the end of the season.
All the data were collected from the official LaLiga website (http://www.laliga.es; accessed on 10 April 2022), and recorded on a post hoc observation sheet. They were also contrasted with two different official websites (https://www.flashscore.es and https://optaplayerstats.statsperform.com/en_GB/soccer; accessed both on 10 May 2022), to minimize errors. In addition, an intra-observer concordance analysis was performed, to ensure correct data entry. The agreement found in the game indicators was almost perfect [18], with Cohen’s kappa values above 0.83. Finally, the data obtained were entered into the statistical software, for descriptive and inferential analysis.
2.3. Statistical Analysis
First, the normality of the data (p > 0.05) was assumed, after statistical analysis with the Kolmogorov–Smirnov test, using the Lilliefors significance correction [19]. Subsequently, for each study (S1 and S2), and the three pandemic periods (pre-COVID-19, COVID-19, and post-COVID-19), a descriptive analysis was performed using the mean and standard deviation of the game indicators.
In each pandemic period (pre-COVID-19, COVID-19, and post-COVID-19), an analysis of variance was performed, using an one-factor ANOVA test and a Bonferroni post hoc test to identify statistical differences in the game indicators, according to the partial (S1) and final (S2) ranking. The effect size was calculated using eta squared and Cohen’s d (point estimate) [19,20,21]. The ranges used for eta squared were 0.000–0.009 (no effect), 0.010–0.059 (small effect), 0.060–0.139 (intermediate effect), and 0.140–0.200 (large effect). The ranges used for Cohen’s d were <0.000 (adverse effect), 0.000–0.199 (no effect), 0.200–0.499 (small effect), 0.500–0.799 (intermediate effect), and 0.800–≥1.000 (large effect) [20,21]. Thus, the reported effect size complemented the significance value, to provide an estimate of the extent of our results, interpreting results with no effect or a small effect as unimportant.
A linear mixed model and Bonferroni post hoc [22] were also calculated, to identify the statistical differences between the pandemic periods in the two studies. In this model, the individual response of each team was controlled (R2 conditional random factor = team), to analyze whether the repetition of the matches affected the results. An interclass correlation coefficient close to 0 indicated that the responses were stable, and the LRT test indicated whether the random factor had any effect or not.
Statistical analysis was performed using SPSS software, version 27 (IBM Corp. Released 2020. IBM SPSS Statistics for Windows, Version 27, IBM Corp., Armonk, NY, USA) and Jamovi software [23], version 2.3.24. A significance level of p ≤ 0.05 was considered.
3. Results
Table 1 and first table in Section 3.2. show the statistically significant and non-significant results. However, due to space limitations, only statistically significant results are shown in other tables.
3.1. Partial Ranking of the LaLiga Teams
The differences in game indicators according to the partial ranking are shown in Table 1. Thus, in the pre-COVID-19 period, there were significant statistical differences in six game indicators: yellow cards, red cards, total shots, shots on goals, free kicks, and fouls committed. In the COVID-19 period, there were only significant statistical differences in two game indicators: shots off goal, and total passes. In the post-COVID-19 period, there were significant statistical differences in five game indicators: shots on goals, free kicks, offside, fouls committed, and total passes. Despite the significant differences noted, there was no effect (0.000–0.009) or a small effect (0.010–0.059) in these game indicators (Table 1).
Table 1.
The differences in game indicators according to the partial ranking, independent of the pandemic period.
Table 1.
The differences in game indicators according to the partial ranking, independent of the pandemic period.
| Game Indicator | Ranking | Pre-COVID-19 | COVID-19 | Post-COVID-19 | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M | SD | F | p (η2) | M | SD | F | p (η2) | M | SD | F | p (η2) | ||
| Yellow cards | Equaled | 2.83 | 1.60 | 17.31 | <0.001 * | 2.61 | 1.51 | 0.09 | 0.91 | 2.26 | 1.42 | 1.97 | 0.14 |
| Unbalanced | 2.62 | 1.56 | (0.01) | 2.57 | 1.54 | (0.00) | 2.32 | 1.48 | (0.01) | ||||
| Very unbalanced | 2.46 | 1.58 | 2.57 | 1.52 | 2.03 | 1.36 | |||||||
| Red cards | Equaled | 0.13 | 0.36 | 4.29 | 0.01 * | 0.12 | 0.38 | 0.05 | 0.95 | 0.09 | 0.30 | 0.20 | 0.82 |
| Unbalanced | 0.12 | 0.34 | (0.00) | 0.12 | 0.34 | (0.00) | 0.10 | 0.33 | (0.00) | ||||
| Very unbalanced | 0.09 | 0.30 | 0.11 | 0.31 | 0.08 | 0.27 | |||||||
| Ball possession | Equaled | 50.00 | 10.19 | 0.00 | 1.00 | 50.00 | 10.99 | 0.00 | 1.00 | 50.00 | 11.41 | 0.00 | 1.00 |
| Unbalanced | 50.00 | 10.64 | (0.00) | 50.00 | 10.73 | (0.00) | 50.00 | 13.42 | (0.00) | ||||
| Very unbalanced | 50.00 | 12.35 | 50.00 | 13.11 | 50.00 | 14.04 | |||||||
| Total shots | Equaled | 11.81 | 4.61 | 4.30 | 0.01 * | 11.31 | 4.50 | 0.99 | 0.37 | 10.45 | 4.24 | 1.26 | 0.29 |
| Unbalanced | 12.01 | 4.70 | (0.00) | 11.59 | 4.70 | (0.00) | 10.76 | 5.01 | (0.00) | ||||
| Very unbalanced | 12.38 | 5.22 | 10.93 | 4.56 | 11.12 | 4.70 | |||||||
| Shots on goal | Equaled | 4.17 | 2.31 | 6.56 | 0.00 * | 4.06 | 2.33 | 1.86 | 0.16 | 3.61 | 2.07 | 3.88 | 0.02 * |
| Unbalanced | 4.30 | 2.35 | (0.00) | 3.69 | 2.09 | (0.01) | 3.56 | 2.11 | (0.01) | ||||
| Very unbalanced | 4.53 | 2.76 | 3.93 | 2.27 | 4.13 | 2.33 | |||||||
| Shots off goal | Equaled | 7.64 | 3.57 | 1.03 | 0.36 | 7.25 | 3.50 | 3.41 | 0.03 * | 6.83 | 3.33 | 0.72 | 0.49 |
| Unbalanced | 7.70 | 3.65 | (0.00) | 7.90 | 3.86 | (0.01) | 7.19 | 3.91 | (0.00) | ||||
| Very unbalanced | 7.85 | 3.73 | 7.00 | 3.65 | 6.99 | 3.47 | |||||||
| Free kicks | Equaled | 16.07 | 4.55 | 7.74 | <0.001 * | 15.60 | 4.24 | 0.62 | 0.54 | 16.98 | 5.27 | 7.84 | <0.001 * |
| Unbalanced | 15.85 | 4.52 | (0.00) | 15.28 | 4.52 | (0.00) | 16.33 | 5.43 | (0.02) | ||||
| Very unbalanced | 15.34 | 4.50 | 15.76 | 4.65 | 15.02 | 4.76 | |||||||
| Corners | Equaled | 4.89 | 2.70 | 0.66 | 0.52 | 4.78 | 2.61 | 1.49 | 0.23 | 4.46 | 2.71 | 0.42 | 0.66 |
| Unbalanced | 4.91 | 2.85 | (0.00) | 4.52 | 2.48 | (0.00) | 4.27 | 2.54 | (0.00) | ||||
| Very unbalanced | 5.02 | 2.78 | 4.40 | 2.46 | 4.29 | 2.61 | |||||||
| Offsides | Equaled | 2.35 | 1.84 | 1.46 | 0.23 | 2.13 | 1.69 | 0.16 | 0.86 | 2.14 | 1.69 | 3.27 | 0.04 * |
| Unbalanced | 2.46 | 1.83 | (0.00) | 2.07 | 1.61 | (0.00) | 1.96 | 1.46 | (0.01) | ||||
| Very unbalanced | 2.45 | 2.02 | 2.16 | 1.91 | 1.76 | 1.62 | |||||||
| Goalkeeper saves | Equaled | 2.87 | 1.92 | 1.97 | 0.14 | 2.76 | 1.88 | 1.30 | 0.27 | 2.39 | 1.64 | 2.27 | 0.10 |
| Unbalanced | 2.97 | 1.87 | (0.00) | 2.52 | 1.76 | (0.00) | 2.40 | 1.76 | (0.01) | ||||
| Very unbalanced | 3.02 | 2.03 | 2.76 | 1.85 | 2.73 | 2.02 | |||||||
| Fouls committed | Equaled | 14.12 | 4.25 | 11.21 | <0.001 * | 13.78 | 3.93 | 0.24 | 0.79 | 13.66 | 4.17 | 7.02 | <0.001 * |
| Unbalanced | 13.84 | 4.22 | (0.01) | 13.60 | 4.28 | (0.00) | 13.30 | 4.46 | (0.02) | ||||
| Very unbalanced | 13.30 | 4.15 | 13.88 | 4.13 | 12.18 | 3.68 | |||||||
| Attacks | Equaled | 101.71 | 23.41 | 0.33 | 0.72 | 106.84 | 24.07 | 1.52 | 0.22 | 101.95 | 23.02 | 0.10 | 0.91 |
| Unbalanced | 100.83 | 22.68 | (0.00) | 106.94 | 21.58 | (0.00) | 101.53 | 27.77 | (0.00) | ||||
| Very unbalanced | 102.14 | 24.29 | 110.51 | 25.28 | 100.88 | 28.16 | |||||||
| Dangerous attacks | Equaled | 54.75 | 18.56 | 0.43 | 0.65 | 45.43 | 16.81 | 1.66 | 0.19 | 46.09 | 17.37 | 1.10 | 0.33 |
| Unbalanced | 55.82 | 19.42 | (0.00) | 46.91 | 15.72 | (0.00) | 46.16 | 19.85 | (0.00) | ||||
| Very unbalanced | 55.19 | 19.87 | 48.21 | 18.26 | 48.65 | 21.70 | |||||||
| Total passes | Equaled | 423.84 | 117.74 | 1.68 | 0.19 | 417.79 | 116.46 | 3.05 | 0.05 * | 424.83 | 123.22 | 3.71 | 0.03 * |
| Unbalanced | 425.26 | 108.06 | (0.00) | 416.53 | 116.81 | (0.01) | 428.33 | 133.94 | (0.01) | ||||
| Very unbalanced | 443.10 | 137.32 | 444.18 | 143.25 | 458.42 | 153.41 | |||||||
| Tackles | Equaled | 16.96 | 8.59 | 0.05 | 0.95 | 14.86 | 4.83 | 0.50 | 0.61 | 14.28 | 4.60 | 0.73 | 0.48 |
| Unbalanced | 17.05 | 5.19 | (0.00) | 14.91 | 4.31 | (0.00) | 14.22 | 4.21 | (0.00) | ||||
| Very unbalanced | 17.17 | 5.16 | 15.28 | 4.34 | 13.78 | 4.45 | |||||||
Note: equaled = 0–5 places in the league table; unbalanced = 6–10 places in the league table; very unbalanced = 11–19 places in the league table. * p ≤ 0.05.
Multiple comparisons between the game indicators, according to the partial ranking, are shown in Table 2. In the pre-COVID-19 period, the results indicated that equaled teams received more yellow cards, compared to unbalanced and very unbalanced teams. These first teams received more red cards, compared to very unbalanced teams, but they made fewer total shots and shots on goal, compared to very unbalanced teams. On the other hand, equaled and unbalanced teams took more free kicks, and committed more fouls than very unbalanced teams. In the COVID-19 period, unbalanced teams reported more shots off goal than very unbalanced teams. In the post-COVID-19 period, equaled and unbalanced teams reported fewer shots on goals than very unbalanced teams; in contrast, these same teams took more free kicks, and committed more fouls, compared to very unbalanced teams. Equaled teams committed more offsides, compared to very unbalanced teams; in contrast, they made fewer total passes than very unbalanced teams. Despite this, no effect (0.000–0.199) or a small effect (0.200–0.499) was reported on the significant game indicators.
The proposed model, a linear mixed model (period*partial ranking), reported significant statistical differences in three game indicators (Table 3): free kicks, offsides, and fouls committed. The inter-class correlation coefficient was close to 0; thus, the responses of each team were stable, and the repetition of the matches had no effect.
Multiple comparisons according to the period*partial ranking interaction are shown in Table 4. In general, during the post-COVID-19 period, equaled teams took more free kicks, compared to the pre-COVID-19 and COVID-19 periods. Moreover, very unbalanced teams committed fewer offsides and fouls during this pandemic period.
3.2. Final Ranking of the LaLiga Teams
Table 5 shows the differences in game indicators according to the final ranking. In the pre-COVID-19 period, the most prominent (p < 0.001) game indicators were as follows: the ball possession, shots on goal, and total passes, with an intermediate effect (0.060–0.139). In the COVID-19 period, the most prominent game indicators were as follows: the ball possession, and dangerous attacks, with an intermediate effect (0.060–0.139); and total passes, with a large effect (η2 > 0.140). In the post-COVID-19 period, the most prominent game indicators were as follows: the total shots, shots on goal, attacks, and dangerous attacks, with an intermediate effect (0.060–0.139); as well as the ball possession and total passes, with a large effect (η2 > 0.140).
Table 5.
Differences in game indicators according to the final ranking, independent of the pandemic period.
Table 5.
Differences in game indicators according to the final ranking, independent of the pandemic period.
| Game Indicator | Ranking | Pre-COVID-19 | COVID-19 | Post-COVID-19 | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M | SD | F | p (η2) | M | SD | F | p (η2) | M | SD | F | p (η2) | ||
| Yellow cards | Europe position | 2.47 | 1.61 | 19.53 | <0.001 * | 2.40 | 1.40 | 4.50 | 0.01 * | 2.05 | 1.39 | 6.14 | 0.00 * |
| Middle position | 2.77 | 1.55 | (0.01) | 2.63 | 1.57 | (0.01) | 2.41 | 1.48 | (0.02) | ||||
| Relegation position | 2.89 | 1.63 | 2.89 | 1.58 | 2.04 | 1.29 | |||||||
| Red cards | Europe position | 0.11 | 0.32 | 4.32 | 0.01 * | 0.07 | 0.27 | 4.17 | 0.02 * | 0.07 | 0.25 | 2.41 | 0.09 |
| Middle position | 0.11 | 0.34 | (0.00) | 0.14 | 0.39 | (0.01) | 0.11 | 0.34 | (0.01) | ||||
| Relegation position | 0.16 | 0.40 | 0.16 | 0.39 | 0.06 | 0.24 | |||||||
| Ball possession | Europe position | 53.77 | 10.97 | 140.17 | <0.001 * | 55.16 | 11.08 | 50.11 | <0.001 * | 57.63 | 10.80 | 96.19 | <0.001 * |
| Middle position | 48.43 | 10.28 | (0.07) | 47.83 | 10.49 | (0.12) | 45.37 | 11.73 | (0.20) | ||||
| Relegation position | 46.45 | 9.93 | 45.19 | 10.62 | 47.64 | 10.50 | |||||||
| Total shots | Europe position | 13.42 | 5.19 | 96.36 | <0.001 * | 12.45 | 4.64 | 13.68 | <0.001 * | 12.52 | 4.73 | 41.68 | <0.001 * |
| Middle position | 11.30 | 4.36 | (0.05) | 10.58 | 4.46 | (0.04) | 9.36 | 4.02 | (0.10) | ||||
| Relegation position | 11.00 | 4.39 | 11.09 | 4.29 | 10.76 | 4.44 | |||||||
| Shots on goal | Europe position | 5.19 | 2.71 | 152.47 | <0.001 * | 4.61 | 2.42 | 19.46 | <0.001 * | 4.61 | 2.37 | 40.51 | <0.001 * |
| Middle position | 3.86 | 2.13 | (0.07) | 3.57 | 2.13 | (0.05) | 3.16 | 1.85 | (0.10) | ||||
| Relegation position | 3.63 | 2.11 | 3.54 | 1.89 | 3.43 | 1.80 | |||||||
| Shots off goal | Europe position | 8.23 | 3.81 | 21.86 | <0.001 * | 7.85 | 3.86 | 4.21 | 0.02 * | 7.92 | 3.68 | 20.22 | <0.001 * |
| Middle position | 7.44 | 3.51 | (0.01) | 7.02 | 3.52 | (0.01) | 6.20 | 3.19 | (0.05) | ||||
| Relegation position | 7.37 | 3.47 | 7.54 | 3.46 | 7.33 | 3.70 | |||||||
| Free kicks | Europe position | 15.66 | 4.55 | 2.75 | 0.06 | 15.15 | 4.45 | 1.82 | 0.16 | 16.01 | 5.52 | 1.38 | 0.25 |
| Middle position | 16.01 | 4.43 | (0.00) | 15.82 | 4.49 | (0.05) | 16.49 | 5.02 | (0.00) | ||||
| Relegation position | 15.68 | 4.85 | 15.54 | 4.01 | 16.93 | 5.42 | |||||||
| Corners | Europe position | 5.47 | 2.99 | 40.96 | <0.001 * | 4.81 | 2.64 | 2.04 | 0.13 | 5.06 | 2.80 | 16.15 | <0.001 * |
| Middle position | 4.67 | 6.63 | (0.02) | 4.61 | 2.52 | (0.01) | 3.88 | 2.49 | (0.04) | ||||
| Relegation position | 4.50 | 2.42 | 4.24 | 2.36 | 4.38 | 2.41 | |||||||
| Offsides | Europe position | 2.69 | 2.00 | 29.39 | <0.001 * | 2.23 | 1.75 | 3.67 | 0.03 * | 2.18 | 1.69 | 3.80 | 0.02 * |
| Middle position | 2.31 | 1.81 | (0.02) | 2.16 | 1.72 | (0.01) | 1.85 | 1.52 | (0.01) | ||||
| Relegation position | 2.04 | 1.74 | 1.73 | 1.59 | 2.14 | 1.72 | |||||||
| Goalkeeper saves | Europe position | 2.71 | 1.79 | 24.76 | <0.001 * | 2.53 | 1.63 | 1.55 | 0.21 | 2.31 | 1.61 | 1.51 | 0.22 |
| Middle position | 2.95 | 1.97 | (0.01) | 2.76 | 1.97 | (0.00) | 2.55 | 1.85 | (0.00) | ||||
| Relegation position | 3.39 | 2.05 | 2.83 | 1.84 | 2.53 | 1.82 | |||||||
| Fouls committed | Europe position | 13.12 | 4.20 | 31.40 | <0.001 * | 13.21 | 3.66 | 5.97 | 0.00 * | 12.33 | 3.99 | 10.53 | <0.001 * |
| Middle position | 14.18 | 4.22 | (0.02) | 13.82 | 4.17 | (0.02) | 13.84 | 4.19 | (0.03) | ||||
| Relegation position | 14.42 | 4.07 | 14.75 | 4.49 | 13.44 | 4.33 | |||||||
| Attacks | Europe position | 104.43 | 23.60 | 10.97 | <0.001 * | 113.88 | 25.37 | 14.73 | <0.001 * | 110.11 | 25.84 | 28.50 | <0.001 * |
| Middle position | 101.27 | 23.73 | (0.01) | 104.51 | 22.27 | (0.04) | 95.28 | 24.11 | (0.07) | ||||
| Relegation position | 95.81 | 20.64 | 103.64 | 21.11 | 102.89 | 23.53 | |||||||
| Dangerous attacks | Europe position | 59.02 | 20.19 | 18.70 | <0.001 * | 52.23 | 19.40 | 25.64 | <0.001 * | 55.26 | 21.53 | 52.63 | <0.001 * |
| Middle position | 53.67 | 18.79 | (0.02) | 43.18 | 14.81 | (0.06) | 40.59 | 15.42 | (0.12) | ||||
| Relegation position | 51.05 | 15.82 | 43.94 | 12.90 | 46.64 | 16.00 | |||||||
| Total passes | Europe position | 472.26 | 142.90 | 32.77 | <0.001 * | 502.24 | 137.44 | 110.36 | <0.001 * | 542.73 | 124.02 | 220.36 | <0.001 * |
| Middle position | 413.26 | 103.45 | (0.08) | 385.08 | 90.10 | (0.23) | 369.87 | 98.26 | (0.37) | ||||
| Relegation position | 379.19 | 75.05 | 364.38 | 84.56 | 385.55 | 84.58 | |||||||
| Tackles | Europe position | 16.70 | 4.98 | 0.48 | 0.62 | 15.28 | 4.33 | 3.50 | 0.03 * | 14.27 | 4.26 | 0.32 | 0.73 |
| Middle position | 17.19 | 8.47 | (0.00) | 14.54 | 4.77 | (0.01) | 14.03 | 4.60 | (0.00) | ||||
| Relegation position | 17.32 | 5.74 | 15.63 | 4.41 | 14.32 | 4.46 | |||||||
Note: Europe position = 1–7 places in the league table; middle position = 8–17 places in the league table; relegation position = 18–20 places in the league table; light green color = intermediate effect size; dark green color = large effect size. Two colors have been used for the cells, to highlight the most significant results, and to help the reader understand them. * p ≤ 0.05.
Table 6 shows multiple comparisons in game indicators according to the final ranking. In general, the most prominent game indicators, in favor of Europe teams, were the possession of the ball and total passes, which resulted in more attacking actions. The trend was similar in all three periods, although these game indicators were particularly prominent in the COVID-19 and post-COVID-19 periods, with a large effect (0.800–≥1.000).
The proposed mixed model (period*final ranking) indicated significant statistical differences in ten game indicators (Table 7): the ball possession, total shots, shots on goal, shots off goal, free kicks, corners, offsides, attacks, dangerous attacks, and total passes. Except for in the total number of passes, the inter-class correlation coefficient was close to 0; thus, the responses of each team were stable, and the repetition of the matches had no effect.
Multiple comparisons according to the period*final ranking interaction are shown in Table 8. In general, the teams in a Europe position reported higher values in the game indicators in the three pandemic periods. The ball possession, total passes, and attacks by Europe-position teams were higher in the post-COVID-19 period, compared to the pre-COVID-19 and COVID-19 periods. On the contrary, the number of dangerous attacks by these teams was higher in the pre-COVID-19 period, compared to the COVID-19 and post-COVID-19 periods.
4. Discussion
Soccer is dominated by game indicators that can influence the performance of teams [1,3,24]. Therefore, this study analyzed, independently for the pre-COVID-19, COVID-19, and post-COVID-19 pandemic periods, the influence of game indicators on the partial (S1) and final (S2) ranking of LaLiga teams over seven consecutive seasons. Additionally, the statistical differences between pandemic periods were identified. The main results reported that, according to the partial ranking, there was no or a small effect on these game indicators that was statistically significant for each pandemic period. So, hypothesis 1 was rejected. According to the final ranking, an intermediate effect and a large effect were obtained in some game indicators that were statistically significant (pre-COVID-19 period: the ball possession, shots on goal, and total passes; COVID-19 period: the ball possession, dangerous attacks, and total passes; post-COVID-19 period: the total shots, shots on goal, attacks, dangerous attacks, ball possession, and total passes). For this reason, hypothesis 2 was partially accepted. In addition, the pandemic period*ranking interaction only reported intermediate and large effects in terms of the final ranking. Therefore, hypothesis 3 was partially accepted. The mentioned indicators were decisive to reaching a Europe position at the end of each season and pandemic period, especially the ball possession and total passes. They had a greater influence on the final ranking than on the partial ranking.
In the pre-COVID-19 period, our results reported that the most prominent game indicators, in favor of the Europe teams, were the ball possession, shots on goal and number of total passes, with an intermediate effect. In the same competition, Fernández-Cortés et al. [2] indicated that teams who won at home played with more ball possession, and made more shots on goal, among other factors. Ball possession has previously been identified as marking out the top-ranked teams in both the Chinese Soccer Super League [25] and the UEFA Champions League [26]. It is possible that teams ranked in Europe spend more time ahead on the scoreboard than their opponents [26]. This could mean that, when they are winning, they are not in a hurry to score a goal or reach the opponent’s area, assuming a strategy of keeping the ball and slowing down the game [27]. Moreover, this strategy involves making a greater number of passes [25]. The indicators of ball possession and total passes are positively associated with the shots on goal [28]. Ball possession leads to better collective play and optimal shooting positions, and also limit the opportunities of the opposing team [29].
During the COVID-19 pandemic period, the most prominent game indicators, also in favor of the Europe-ranked teams, were possession of the ball and dangerous attacks, with an intermediate effect; and the number of total passes, with a large effect. As can be seen, the first two pandemic periods followed a similar trend. This could be because the teams did not have time to adapt to the new changes in the way of playing. In general, Europe-ranked teams would have a higher technical level/quality [30]. This led to a higher ball possession, linked to the total number of key passes [25]. In the major European competitions (i.e., Ligue 1, Serie A, LaLiga, the Premier League, and Bundesliga) during the 2019/2020 season, with fans in the stadia (before the restrictions), home teams scored more goals than visitors [7], because the number of dangerous attacks was higher. In contrast, another study [14] comparing the pre-COVID-19 and COVID-19 period in European soccer matches showed a decrease in the goals scored by home teams in ghost matches, which also affected the points and ranking. Another indicator that was affected during the COVID-19 period, in similar studies [9,12,13], was the higher number of fouls committed by home teams which, in turn, resulted in more yellow cards. It was essential to work on set pieces because, given the increase in this game indicator during the COVID-19 period, they became of great importance to the result [31]. The absence of fans and noise caused coaches and players to adapt their behavior to the empty environment in which they operated [32].
In the post-COVID-19 period, the most prominent game indicators, again in favor of Europe-ranked teams, were the total shots, shots on goal, attacks, and dangerous attacks, with an intermediate effect; and the total passes and ball possession, with a large effect. In the three pandemic periods studied, two interrelated game indicators stood out: the ball possession and total passes. Both game indicators, together with the attacks, were higher in the post-COVID-19 period, compared to the other two pandemic periods. The dangerous attacks were higher pre-COVID-19, compared to during COVID-19 and post-COVID-19. This last pandemic period differed from the rest, because there was an even greater interest on the part of the teams in maintaining possession of the ball and avoiding direct play, associated with more attacks and, therefore, greater possibilities of shots [28]; or in retaining the ball when they were winning [27]. In this particular period, the higher-ranked teams controlled the ball more, and imposed their style of play [6,33,34]. In this regard, in the major European leagues (i.e., Ligue 1, Serie A, LaLiga, the Premier League, and Bundesliga), the ball possession, number of key passes, and scoring first were correlated with the accumulation of goals scored and the points at the end of the season [16]. In turn, Yang et al. [33] reported that lower-ranked teams performed more defensive actions, especially in unbalanced matches.
The main differences found in this study were in the attack indicators, and not in the defense ones. This could be because teams tend to opt for the strategy of keeping possession of the ball, to achieve victories. It has been shown that higher-ranked teams show higher ball possession [25,26], and lower ranked teams show lower ball possession [33]. Maintaining ball possession implies an increase in attacking actions [28]. Despite this, it requires precision on the part of the teams, and the teams with the best technical/quality level achieve this [30]. In addition, when in possession of the ball, the teams defend simultaneously. As a result of this study, soccer coaches are advised to choose a technically sound starting eleven that allows them to maintain ball possession, i.e., to have few losses per match. In this way, they will be able to stay in the top positions for a season. For managers, it is important, when recruiting players, to know that these game indicators are fundamental for a team to be in the upper part of the ranking. In addition, the more players with these profiles a team has, the easier it will be to obtain a better final ranking.
Limitations, Practical Applications, and Future Studies
Among the limitations, it should be noted that, among the major European leagues, this study collected data only from LaLiga. Despite the existing equality at the scorer level among the major European leagues [34], it is not possible to generalize our results to other leagues.
The notational analysis allowed the identification of the game indicators that should be reinforced to increase the probability of success (a better ranking). Following this line of thought, according to our results, LaLiga teams should play with more ball possession, related to the total passes. A greater possession of the ball allows them to take more shots (offensive actions), and thus to have a better chance of winning. This information is of interest to coaches and soccer assistants when preparing and directing matches.
Controlling the competitive balance in soccer is difficult, due to its rapid evolution. In this vein, team dynamics change between seasons, or even within a season [9]. Further research is needed to confirm the results obtained in this study. Moreover, it would also be interesting to analyze intensities according to the quality of the opponent.
5. Conclusions
The obtained study results indicate that the ball possession and total passes are two interrelated game indicators that significantly influence the final ranking. In all three pandemic periods, LaLiga teams ranked in a Europe position at the end of each season (the final ranking) reported higher ball possession and total passes. In turn, a greater possession of the ball allowed them to make more shots (offensive actions), and thus to have a better chance of winning. Similarly, these game indicators are higher in the post-COVID-19 period, compared to the pre-COVID-19 and COVID-19 periods. Based on the partial ranking, there is little or no effect on those game indicators that is statistically significant for each pandemic period. Thus, the game indicators analyzed have a greater influence on the final ranking, with an intermediate and large effect, than on the partial ranking.
Author Contributions
Conceptualization, J.F.-C. and S.J.I.; methodology, J.G.-R. and S.J.I.; formal analysis, J.F.-C., J.M.G.-C. and S.J.I.; investigation, J.F.-C.; data curation, J.F.-C., J.M.G.-C. and S.J.I.; writing—original draft preparation, J.M.G.-C.; writing—review and editing, J.F.-C., J.G.-R. and S.J.I.; visualization, J.M.G.-C.; supervision, J.G.-R. and S.J.I.; funding acquisition, S.J.I. All authors have read and agreed to the published version of the manuscript.
Funding
This study has been partially subsidized by the Aid for Research Groups (GR21149) from the Regional Government of Extremadura (Department of Economy, Science and Digital Agenda), with a contribution from the European Union from the European Funds for Regional Development. The author J.M.G.-C. was supported by a grant from the Universities Ministry of Spain and the European Union (NextGenerationUE) “Ayuda del Programa de Recualificación del Sistema Universitario Español, Modalidad de ayudas Margarita Salas para la formación de jóvenes doctores” (MS-01).
Institutional Review Board Statement
Not applicable.
Informed Consent Statement
Not applicable.
Data Availability Statement
Data will be available upon reasonable request to the corresponding author.
Conflicts of Interest
The authors declare no conflict of interest.
References
- Fernández-Cortés, J.; Cáceres-Sánchez, L.; Antúnez, A.; García-Rubio, J.; Ibáñez, S.J. Analysis of the influece of situational variables in professional soccer. Retos. Nuevas Tend. Educ. Física Deporte Recreación 2022, 46, 114–119. [Google Scholar] [CrossRef]
- Fernández-Cortés, J.; Escudero-Tena, A.; García-Rubio, J.; Ibáñez, S.J. The importance of the match location in the Spanish La Liga football league. J. Sport Health Res. 2020, 12, 285–294. [Google Scholar]
- García-Unanue, J.; Pérez-Gómez, J.; Giménez, J.-V.; Felipe, J.L.; Gómez-Pomares, S.; Gallardo, L.; Sánchez-Sánchez, J. Influence of contextual variables and the pressure to keep category on physical match performance in soccer players. PLoS ONE 2018, 13, e0204256. [Google Scholar] [CrossRef] [PubMed]
- Leite, W.; Pollard, R. Comparison of home advantage between level 1 and level 2 in women’s football leagues. J. Anthropol. Sport Phys. Educ. 2020, 4, 9–13. [Google Scholar]
- Medina, J.A.; San José, J.R.; Lorente, V.M. El gol como unidad de medida de rendimiento en futsal. Retos. Nuevas Tend. Educ. Física Deporte Recreación 2019, 36, 251–258. [Google Scholar]
- Ponce-Bordón, J.C.; Nobari, H.; Lobo-Triviño, D.; García-Calvo, T.; Vicente-Giménez, J.; López del Campo, R.; Resta, R.; Fernández-Navarro, J. Match Movement Profiles Differences in Spanish Soccer Competitive Leagues According to Opposition’s Team Ranking: A Comparison Study. Appl. Sci. 2022, 12, 12635. [Google Scholar] [CrossRef]
- Ferraresi, M.; Gucciardi, G. Team Performance and Audience: Experimental Evidence from the Football Sector; Working papers; ocietà Italiana di Economia Pubblica: Tokyo, Japan, 2020; Volume 94. [Google Scholar]
- Lastella, M.; Roach, G.D.; Sargent, C. Travel fatigue and sleep/wake behaviors of professional soccer players during international competition. Sleep Health 2019, 5, 141–147. [Google Scholar] [CrossRef]
- Leitner, M.C.; Richlan, F. No fans–no pressure: Referees in professional football during the COVID-19 pandemic. Front. Sports Act. Living 2021, 3, 720488. [Google Scholar] [CrossRef]
- The International Football Association Board Laws of the Game. Available online: https://www.theifab.com/es/ (accessed on 8 August 2022).
- Scoppa, V. Social pressure in the stadiums: Do agents change behavior without crowd support? J. Econ. Psychol. 2021, 82, 102344. [Google Scholar] [CrossRef]
- Richlan, F.; Leitner, M.C. No more advantage for the home team: Refereeing without social pressure in european elite football during the Covid-19 pandemic. Int. J. Sport Exerc. Psychol. 2021, 19, S272–S273. [Google Scholar]
- Bryson, A.; Dolton, P.; Reade, J.J.; Schreyer, D.; Singleton, C. Causal effects of an absent crowd on performances and refereeing decisions during Covid-19. Econ. Lett. 2020, 108, 212–214. [Google Scholar] [CrossRef]
- Lee, J.; Kim, J.; Kim, H.; Lee, J.-S. A Bayesian Approach to Predict Football Matches with Changed Home Advantage in Spectator-Free Matches after the COVID-19 Break. Entropy 2022, 24, 366. [Google Scholar] [CrossRef]
- Spitz, J.; Wagemans, J.; Memmert, D.; Williams, A.M.; Helsen, W.F. Video assistant referees (VAR): The impact of technology on decision making in association football referees. J. Sports Sci. 2021, 39, 147–153. [Google Scholar] [CrossRef]
- Ramos-Pérez, D.R.; Castellano, J.; Hernández-Mendo, A. Relationship between procedure and result indicators during a football season in the five major European leagues. e-Balonmano Com. 2021, 17, 59–72. [Google Scholar]
- Ato, M.; López, J.J.; Benavente, A. Un sistema de clasificación de los diseños de investigación en psicología. An. Psicol. 2013, 29, 1039–1059. [Google Scholar] [CrossRef] [Green Version]
- Landis, J.R.; Koch, G.G. The measurement of observer agreement for categorical data. Biometrics 1977, 33, 159–174. [Google Scholar] [CrossRef] [Green Version]
- Field, A. Discovering Statistics Using SPSS Statistics, 4th ed.; Sage Publications Ltd.: London, UK, 2013. [Google Scholar]
- Cohen, J. Statistical Power Analysis for the Behavioral Sciences, 2nd ed.; Lawrence Erlbaum Associates: Hillsdale, NJ, USA, 1988. [Google Scholar]
- Lenhard, W.; Lenhard, A. Calculation of Effect Sizes; Psychometrica: Dettelbach, Germany, 2016. [Google Scholar] [CrossRef]
- Gallucci, M. GAMLj: General Analyses for Linear Models [Jamovi Module] 2019. Available online: https://gamlj.github.io/ (accessed on 20 April 2023).
- The Jamovi Project. Jamovi (Version 2.3) [Computer Software]. 2022. Available online: https://www.jamovi.org (accessed on 20 April 2023).
- Maneiro, R.; Casal-Sanjurjo, C.A.; Ardá-Suárez, A.; Losada-López, J.L. Identification of significant variables in the corner kick in women’s football: Comparison with men’s football. e-Balonmano Com. 2019, 15, 91–106. [Google Scholar]
- Kong, L.; Zhang, T.; Zhou, C.; Gómez-Ruano, M.A.; Hu, Y.; Zhang, S. The evaluation of playing styles integrating with contextual variables in professional soccer. Front. Psychol. 2022, 13, 1002566. [Google Scholar] [CrossRef]
- Miñano-Espin, J.; Casáis, L.; Lago-Peñas, C.; Gómez-Ruano, M.Á. High speed running and sprinting profiles of elite soccer players. J. Hum. Kinet. 2017, 58, 169–176. [Google Scholar] [CrossRef] [Green Version]
- Bloomfield, J.; Polman, R.; Donoghue, P.G. Effects of score-line on intensity of play in midfield and forward players in FA Premier League. J. Sports Sci. 2005, 23, 191–192. [Google Scholar]
- Gong, B.; Cui, Y.; Zhang, S.; Zhou, C.; Yi, Q.; Gómez-Ruano, M.Á. Impact of technical and physical key performance indicators on ball possession in the Chinese Super League. Int. J. Perform. Anal. Sport 2021, 21, 909–921. [Google Scholar] [CrossRef]
- Redwood-Brown, A. Passing patterns before and after goal scoring in FA Premier League Soccer. Int. J. Perform. Anal. Sport 2008, 8, 172–182. [Google Scholar] [CrossRef]
- Gonçalves, B.; Coutinho, D.; Exel, J.; Travassos, B.; Lago, C.; Sampaio, J. Extracting spatial-temporal features that describe a team match demands when considering the effects of the quality of opposition in elite football. PLoS ONE 2019, 14, e0221368. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Fernández-Cortés, J.; Gómez-Ruano, M.A.; Mancha-Triguero, D.; Ibáñez, S.J.; García-Rubio, J. Evolution of Performance Indicators in Soccer during the Last Decade. Appl. Sci. 2022, 12, 12834. [Google Scholar] [CrossRef]
- Webb, T. The future of officiating: Analysing the impact of COVID-19 on referees in world football. Soccer Soc. 2021, 22, 12–18. [Google Scholar] [CrossRef]
- Yang, G.; Leicht, A.S.; Lago, C.; Gómez, M.Á. Key team physical and technical performance indicators indicative of team quality in the soccer Chinese super league. Res. Sports Med. 2018, 26, 158–167. [Google Scholar] [CrossRef]
- Martínez, F.D.; González-García, H. Analysis of goals in European football and the most influential part in final standings. e-Balonmano Com. 2018, 14, 89–98. [Google Scholar]
Table 2.
Post hoc comparisons between game indicators, according to the partial ranking, for each pandemic period.
Table 2.
Post hoc comparisons between game indicators, according to the partial ranking, for each pandemic period.
| Period | Game Indicator | Inferential Comparison | p-Bonferroni (d) |
|---|---|---|---|
| Pre-COVID-19 | Yellow cards | Equaled > Unbalanced | 0.00 * (0.13) |
| Equaled > Very unbalanced | <0.001 * (0.23) | ||
| Red cards | Equaled > Very unbalanced | 0.01 * (0.12) | |
| Total shots | Equaled < Very unbalanced | 0.01 * (−0.12) | |
| Shots on goal | Equaled < Very unbalanced | <0.001 * (−0.15) | |
| Free kicks | Equaled > Very unbalanced | <0.001 * (0.16) | |
| Unbalanced > Very unbalanced | 0.04 * (0.11) | ||
| Fouls committed | Equaled > Very unbalanced | <0.001 * (0.20) | |
| Unbalanced > Very unbalanced | 0.01 * (0.13) | ||
| COVID-19 | Shots off goal | Unbalanced > Very unbalanced | 0.05 * (0.24) |
| Total passes | No differences were observed in the comparisons | ||
| Post-COVID-19 | Shots on goal | Equaled < Very unbalanced | 0.03 * (−0.24) |
| Unbalanced < Very unbalanced | 0.04 * (−0.26) | ||
| Free kicks | Equaled > Very unbalanced | <0.001 * (0.38) | |
| Unbalanced > Very unbalanced | 0.05 * (0.25) | ||
| Offsides | Equaled > Very unbalanced | 0.04 * (0.23) | |
| Fouls committed | Equaled > Very unbalanced | <0.001 * (0.37) | |
| Unbalanced > Very unbalanced | 0.03 * (0.27) | ||
| Total passes | Equaled < Very unbalanced | 0.02 * (−0.25) | |
Note: equaled = 1–5 places in the league table; unbalanced = 6–10 places in the league table; very unbalanced = 11–19 places in the league table; d = Cohen’s d. Only statistically significant game indicators in the one-factor ANOVA test are shown in this table. * p ≤ 0.05.
Table 3.
Differences in period*partial ranking interaction.
| Game Indicator | F | p | R2 Conditional | BIC | ICC | LRT (p) |
|---|---|---|---|---|---|---|
| Free kicks | 2.99 | 0.02 * | 0.06 | 31242.72 | 0.06 | 231 (<0.001 *) |
| Offsides | 2.28 | 0.05 * | 0.04 | 21495.06 | 0.03 | 90.3 (<0.001 *) |
| Fouls committed | 2.39 | 0.05 * | 0.09 | 29999.63 | 0.09 | 434 (<0.001 *) |
Note: F = linear mixed model; BIC = Bayesian information criterion; ICC = interclass correlation coefficient; LRT = LRT test for random effect; Intercept|Team. Only statistically significant game indicators in the linear mixed model are shown in this table. * p ≤ 0.05.
Table 4.
Post hoc comparisons according to the period*partial ranking interaction.
| Game Indicator | Period | Ranking | Period | Ranking | t | p-Bonferroni |
|---|---|---|---|---|---|---|
| Free kicks | Pre-COVID-19 | Unbalanced | Post-COVID-19 | Equaled | −4.60 | <0.001 * |
| Pre-COVID-19 | Very unbalanced | Post-COVID-19 | Equaled | −5.89 | <0.001 * | |
| Pre-COVID-19 | Equaled | Post-COVID-19 | Equaled | −3.76 | 0.01 * | |
| COVID-19 | Equaled | Post-COVID-19 | Equaled | −3.86 | 0.00 * | |
| COVID-19 | Unbalanced | Post-COVID-19 | Equaled | −4.28 | <0.001 * | |
| Offsides | Pre-COVID-19 | Unbalanced | Post-COVID-19 | Very unbalanced | 5.20 | <0.001 * |
| Pre-COVID-19 | Equaled | Post-COVID-19 | Very unbalanced | 4.60 | <0.001 * | |
| Pre-COVID-19 | Very unbalanced | Post-COVID-19 | Very unbalanced | 4.59 | <0.001 * | |
| Fouls committed | Pre-COVID-19 | Unbalanced | Post-COVID-19 | Very unbalanced | 4.02 | 0.00 * |
| Pre-COVID-19 | Equaled | Post-COVID-19 | Very unbalanced | 4.91 | <0.001 * | |
| Pre-COVID-19 | Very unbalanced | Post-COVID-19 | Very unbalanced | 3.33 | 0.03 * | |
| COVID-19 | Equaled | Post-COVID-19 | Very unbalanced | 3.49 | 0.02 * | |
| COVID-19 | Very unbalanced | Post-COVID-19 | Very unbalanced | 3.57 | 0.01 * |
Note: Only statistically significant game indicators in the linear mixed model are shown in this table. * p ≤ 0.05.
Table 6.
Post hoc comparisons of game indicators according to the final ranking for each pandemic period.
Table 6.
Post hoc comparisons of game indicators according to the final ranking for each pandemic period.
| Period | Game Indicator | Inferential Comparison | p-Bonferroni (d) |
|---|---|---|---|
| Pre-COVID-19 | Yellow cards | Europe position < Middle position | <0.001 * (−0.19) |
| Europe position < Relegation position | <0.001 * (−0.26) | ||
| Red cards | Europe position < Relegation position | 0.01 * (−0.14) | |
| Middle position < Relegation position | 0.03 * (−0.12) | ||
| Ball possession | Europe position > Middle position | <0.001 * (0.51) | |
| Europe position > Relegation position | <0.001 * (0.69) | ||
| Middle position > Relegation position | <0.001 * (0.19) | ||
| Total shots | Europe position > Middle position | <0.001 * (0.45) | |
| Europe position > Relegation position | <0.001 * (0.49) | ||
| Shots on goal | Europe position > Middle position | <0.001 * (0.56) | |
| Europe position > Relegation position | <0.001 * (0.61) | ||
| Shots off goal | Europe position > Middle position | <0.001 * (0.22) | |
| Europe position > Relegation position | <0.001 * (0.23) | ||
| Corners | Europe position > Middle position | <0.001 * (0.29) | |
| Europe position > Relegation position | <0.001 * (0.34) | ||
| Offsides | Europe position > Middle position | <0.001 * (0.20) | |
| Europe position > Relegation position | <0.001 * (0.34) | ||
| Goalkeeper saves | Europe position < Relegation position | <0.001 * (−0.36) | |
| Middle position < Relegation position | <0.001 * (−0.22) | ||
| Fouls committed | Europe position < Middle position | <0.001 * (−0.25) | |
| Europe position < Relegation position | <0.001 * (−0.31) | ||
| Attacks | Europe position > Middle position | 0.05 * (0.13) | |
| Europe position > Relegation position | <0.001 * (0.38) | ||
| Middle position > Relegation position | 0.01 * (0.24) | ||
| Dangerous attacks | Europe position > Middle position | <0.001 * (0.28) | |
| Europe position > Relegation position | <0.001 * (0.42) | ||
| Total passes | Europe position > Middle position | <0.001 * (0.49) | |
| Europe position > Relegation position | <0.001 * (0.74) | ||
| Middle position > Relegation position | 0.02 * (0.40) | ||
| COVID-19 | Yellow cards | Europe position < Relegation position | 0.01 * (−0.34) |
| Red cards | Europe position < Middle position | 0.03 * (−0.29) | |
| Ball possession | Europe position > Middle position | <0.001 * (0.68) | |
| Europe position > Relegation position | <0.001 *(0.91) | ||
| Total shots | Europe position > Middle position | <0.001 * (0.41) | |
| Europe position > Relegation position | 0.02 * (0.30) | ||
| Shots on goal | Europe position > Middle position | <0.001 * (0.46) | |
| Europe position > Relegation position | <0.001 * (0.47) | ||
| Shots off goal | Europe position > Middle position | 0.01 * (0.23) | |
| Offsides | Europe position > Relegation position | 0.03 * (0.30) | |
| Fouls committed | Europe position < Relegation position | 0.00 * (−0.39) | |
| Attacks | Europe position > Middle position | <0.001 * (0.40) | |
| Europe position > Relegation position | <0.001 * (0.42) | ||
| Dangerous attacks | Europe position > Middle position | <0.001 * (0.54) | |
| Europe position > Relegation position | <0.001 * (0.47) | ||
| Total passes | Europe position > Middle position | <0.001 * (1.05) | |
| Europe position > Relegation position | <0.001 * (1.11) | ||
| Tackles | No differences were observed in the comparisons | ||
| Post-COVID-19 | Yellow cards | Europe position < Middle position | 0.01 * (−0.25) |
| Middle position > Relegation position | 0.05 * (0.25) | ||
| Ball possession | Europe position > Middle position | <0.001 * (1.08) | |
| Europe position > Relegation position | <0.001 * (0.93) | ||
| Total shots | Europe position > Middle position | <0.001 * (0.73) | |
| Europe position > Relegation position | <0.001 * (0.38) | ||
| Middle position < Relegation position | 0.01 * (−0.34) | ||
| Shots on goal | Europe position > Middle position | <0.001 * (0.70) | |
| Europe position > Relegation position | <0.001 * (0.53) | ||
| Shots off goal | Europe position > Middle position | <0.001 * (0.51) | |
| Middle position < Relegation position | 0.01 * (−0.34) | ||
| Corners | Europe position > Middle position | <0.001 * (0.45) | |
| Offsides | Europe position > Middle position | 0.03 * (0.21) | |
| Fouls committed | Europe position < Middle position | <0.001 * (−0.37) | |
| Europe position < Relegation position | 0.05 * (−0.27) | ||
| Attacks | Europe position > Middle position | <0.001 * (0.60) | |
| Europe position > Relegation position | 0.03 * (0.29) | ||
| Middle position < Relegation position | 0.01 * (−0.32) | ||
| Dangerous attacks | Europe position > Middle position | <0.001 * (0.81) | |
| Europe position > Relegation position | <0.001 * (0.43) | ||
| Middle position < Relegation position | 0.01 * (−0.39) | ||
| Total passes | Europe position > Middle position | <0.001 * (1.58) | |
| Europe position > Relegation position | <0.001 * (1.38) | ||
Note: Europe position = placed 1–7 in the league table; middle position = placed 8–17 in the league table; relegation position = placed 18–20 in the league table; light green color = intermediate effect size; dark green color = large effect size. Only statistically significant game indicators in the one-factor ANOVA test are shown in this table. Two colors have been used for the cells, to highlight the most significant results, and to help the reader understand them. * p ≤ 0.05.
Table 7.
Differences in the period*final ranking interaction.
| Game Indicator | F | p | R2 Conditional | BIC | ICC | LRT (p) |
|---|---|---|---|---|---|---|
| Ball possession | 2.98 | 0.02 * | 0.26 | 39,253.71 | 0.26 | 1062 (<0.001 *) |
| Total shots | 3.34 | 0.01 * | 0.01 | 31,112.48 | 0.08 | 318 (<0.001 *) |
| Shots on goal | 1.96 | 0.01 * | 0.11 | 23,623.16 | 0.10 | 375 (<0.005 *) |
| Shots off goal | 2.77 | 0.03 * | 0.04 | 28,690.35 | 0.03 | 108 (<0.001 *) |
| Free kicks | 2.51 | 0.04 * | 0.07 | 31,261.42 | 0.06 | 235 (<0.001 *) |
| Corners | 3.06 | 0.02 * | 0.04 | 25,632.37 | 0.03 | 80.9 (<0.001 *) |
| Offsides | 2.56 | 0.04 * | 0.03 | 21,481.62 | 0.02 | 37.2 (<0.001 *) |
| Attacks | 5.46 | <0.001 * | 0.16 | 27,478.17 | 0.14 | 317 (<0.001 *) |
| Dangerous attacks | 5.58 | <0.001 * | 0.21 | 25,848.70 | 0.17 | 317 (<0.001 *) |
| Total passes | 4.23 | 0.00 * | 0.52 | 26,788.22 | 0.51 | 1096 (<0.001 *) |
Note: F = linear mixed model; BIC = Bayesian information criterion; ICC = interclass correlation coefficient; LRT = LRT test for random effect; Intercept|Team. Only statistically significant game indicators in the linear mixed model are shown in this table. * p ≤ 0.05.
Table 8.
Post hoc comparisons according to the period*final ranking interaction.
| Game Indicator | Period | Ranking | Period | Ranking | t | p-Bonferroni |
|---|---|---|---|---|---|---|
| Ball possession | Pre-COVID-19 | Relegation | COVID-19 | Europe | −4.84 | <0.001 * |
| Pre-COVID-19 | Middle | COVID-19 | Europe | −3.54 | 0.02 * | |
| Pre-COVID-19 | Relegation | Post-COVID-19 | Europe | −6.04 | <0.001 * | |
| Pre-COVID-19 | Middle | Post-COVID-19 | Europe | −5.06 | <0.001 * | |
| Pre-COVID-19 | Europe | Post-COVID-19 | Europe | −4.29 | <0.001 * | |
| COVID-19 | Middle | Post-COVID-19 | Europe | −3.76 | 0.01 * | |
| Total shots | Pre-COVID-19 | Europe | COVID-19 | Europe | 3.34 | 0.03 * |
| Pre-COVID-19 | Europe | COVID-19 | Middle | 4.61 | <0.001 * | |
| Pre-COVID-19 | Europe | Post-COVID-19 | Europe | 3.41 | 0.02 * | |
| Pre-COVID-19 | Europe | Post-COVID-19 | Middle | 7.22 | <0.001 * | |
| Pre-COVID-19 | Middle | Post-COVID-19 | Middle | 5.60 | <0.001 * | |
| COVID-19 | Relegation | Post-COVID-19 | Middle | 3.23 | 0.05 * | |
| Shots on goals | Pre-COVID-19 | Europe | COVID-19 | Europe | 3.65 | 0.01 * |
| Pre-COVID-19 | Europe | COVID-19 | Middle | 4.83 | <0.001 * | |
| Pre-COVID-19 | Europe | Post-COVID-19 | Relegation | 3.60 | 0.01 * | |
| Pre-COVID-19 | Europe | Post-COVID-19 | Europe | 4.16 | 0.00 * | |
| Pre-COVID-19 | Europe | Post-COVID-19 | Middle | 6.28 | <0.001 * | |
| Pre-COVID-19 | Middle | Post-COVID-19 | Middle | 3.83 | 0.01 * | |
| Shots off goal | Pre-COVID-19 | Europe | Post-COVID-19 | Middle | 5.82 | <0.001 * |
| Pre-COVID-19 | Middle | Post-COVID-19 | Middle | 4.93 | <0.001 * | |
| COVID-19 | Relegation | Post-COVID-19 | Middle | 3.24 | 0.04 * | |
| Free kicks | Pre-COVID-19 | Relegation | Post-COVID-19 | Middle | −3.46 | 0.02 * |
| Pre-COVID-19 | Relegation | Post-COVID-19 | Relegation | −3.78 | 0.01 * | |
| Corners | Pre-COVID-19 | Europe | COVID-19 | Europe | 3.69 | 0.01 * |
| Pre-COVID-19 | Europe | Post-COVID-19 | Middle | 6.12 | <0.001 * | |
| Pre-COVID-19 | Middle | Post-COVID-19 | Middle | 4.14 | 0.00 * | |
| COVID-19 | Middle | Post-COVID-19 | Middle | 3.25 | 0.04 * | |
| Offsides | Pre-COVID-19 | Europe | COVID-19 | Europe | 3.91 | 0.00 * |
| Pre-COVID-19 | Europe | Post-COVID-19 | Europe | 4.09 | 0.00 * | |
| Pre-COVID-19 | Europe | Post-COVID-19 | Middle | 4.40 | <0.001 * | |
| Pre-COVID-19 | Middle | Post-COVID-19 | Middle | 3.98 | 0.00 * | |
| Attacks | Pre-COVID-19 | Middle | COVID-19 | Europe | −5.73 | <0.001 * |
| Pre-COVID-19 | Middle | Post-COVID-19 | Europe | −3.21 | 0.05 * | |
| Pre-COVID-19 | Europe | COVID-19 | Europe | −5.82 | <0.001 * | |
| COVID-19 | Relegation | Post-COVID-19 | Relegation | 3.42 | 0.02 * | |
| COVID-19 | Europe | Post-COVID-19 | Relegation | 5.00 | <0.001 * | |
| COVID-19 | Europe | Post-COVID-19 | Middle | 6.09 | <0.001 * | |
| COVID-19 | Relegation | Post-COVID-19 | Middle | 3.32 | 0.03 * | |
| Dangerous attacks | Pre-COVID-19 | Middle | COVID-19 | Europe | 4.73 | <0.001 * |
| Pre-COVID-19 | Middle | Post-COVID-19 | Europe | 3.62 | 0.01 * | |
| Pre-COVID-19 | Europe | COVID-19 | Europe | 4.58 | <0.001 * | |
| Pre-COVID-19 | Europe | COVID-19 | Middle | 6.57 | <0.001 * | |
| Pre-COVID-19 | Europe | Post-COVID-19 | Relegation | 3.28 | 0.04 * | |
| Pre-COVID-19 | Europe | Post-COVID-19 | Europe | 3.13 | 0.03 * | |
| Pre-COVID-19 | Europe | Post-COVID-19 | Middle | 5.95 | <0.001 * | |
| Pre-COVID-19 | Middle | COVID-19 | Middle | 9.38 | <0.001 * | |
| Pre-COVID-19 | Middle | Post-COVID-19 | Relegation | 4.25 | <0.001 * | |
| Pre-COVID-19 | Middle | Post-COVID-19 | Middle | 8.02 | <0.001 * | |
| Total passes | Pre-COVID-19 | Relegation | COVID-19 | Europe | −3.84 | 0.01 * |
| Pre-COVID-19 | Relegation | Post-COVID-19 | Europe | −4.45 | <0.001 * | |
| Pre-COVID-19 | Middle | Post-COVID-19 | Europe | −3.50 | 0.02 * | |
| Pre-COVID-19 | Relegation | Post-COVID-19 | Middle | −3.26 | 0.04 * | |
| Pre-COVID-19 | Europe | Post-COVID-19 | Europe | −4.46 | <0.001 * | |
| COVID-19 | Relegation | Post-COVID-19 | Europe | −4.06 | 0.00 * | |
| COVID-19 | Middle | Post-COVID-19 | Europe | −4.55 | <0.001 * |
Note: Only statistically significant game indicators in the linear mixed model are shown in this table. * p ≤ 0.05.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
MDPI and ACS Style
Fernández-Cortés, J.; García-Ceberino, J.M.; García-Rubio, J.; Ibáñez, S.J. Influence of Game Indicators on the Ranking of Teams in the Spanish Soccer League. Appl. Sci. 2023, 13, 8097. https://doi.org/10.3390/app13148097
AMA Style
Fernández-Cortés J, García-Ceberino JM, García-Rubio J, Ibáñez SJ. Influence of Game Indicators on the Ranking of Teams in the Spanish Soccer League. Applied Sciences. 2023; 13(14):8097. https://doi.org/10.3390/app13148097
Chicago/Turabian StyleFernández-Cortés, José, Juan M. García-Ceberino, Javier García-Rubio, and Sergio J. Ibáñez. 2023. "Influence of Game Indicators on the Ranking of Teams in the Spanish Soccer League" Applied Sciences 13, no. 14: 8097. https://doi.org/10.3390/app13148097
APA StyleFernández-Cortés, J., García-Ceberino, J. M., García-Rubio, J., & Ibáñez, S. J. (2023). Influence of Game Indicators on the Ranking of Teams in the Spanish Soccer League. Applied Sciences, 13(14), 8097. https://doi.org/10.3390/app13148097
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.
