An Expected Goals on Target (xGOT) Metric as a New Metric for Analyzing Elite Soccer Player Performance

Abstract

Introduction: Football analysis is an applied research area that has seen a huge upsurge in recent years. More complex analysis to understand the soccer players’ or teams’ performances during matches is required. The objective of this study was to prove the usefulness of the expected goals on target (xGOT) metric, as a good indicator of a soccer team’s performance in professional Spanish football leagues, both in the women’s and men’s categories. Method: The data for the Spanish teams were collected from the statistical website Football Reference. The 2023/24 season was analyzed for Spanish leagues, both in the women’s and men’s categories (LigaF and LaLiga, respectively). For all teams, the following variables were calculated: goals, possession value (PV), expected goals (xG) and xGOT. All data obtained for each variable were normalized by match (90 min). A descriptive and correlational statistical analysis was carried out. Results: In the men’s league, this study found a high correlation between goals per match and xGOT (R² = 0.9248) while in the women’s league, there was a high correlation between goals per match (R² = 0.9820) and xG and between goals per match and xGOT (R² = 0.9574). Conclusions: In the LaLiga, the xGOT was the best metric that represented the match result while in the LigaF, the xG and the xGOT were the best metrics that represented the match score.

1. Introduction

Football analysis is an applied research area that has seen a huge upsurge in recent years. It has long focused on analyzing individual on ball actions [1] but it is increasingly requiring more complex analysis to understand the soccer players’ or teams’ performances during matches [2]. That is why the use of advanced data analysis to evaluate the team’s performance has begun to be considered a competitive advantage for many professional soccer teams [3]. Nowadays, the most prevalent metric to evaluate both a soccer team and a specific player’s performance are the goals that alone determine the match scores [4,5]. However, a lot of information is lost with this metric, giving more importance to results than to the process. In this line, the tools to collect information about the players’ and teams’ task during competitions for the technical team have evolved to more complex judgment metrics that have significantly changed the interpretation procedure of players’ performances.

A very wide-spread metric has been the possession value (PV) [6]. There is no clear definition of the PV within football rules [7,8,9], but this metric may be considered the time intervals in which a team controls the ball. However, the growing availability of data from football matches has led to an increase in interest in the characteristics of match performances. As a consequence, the role of shots as a successful proxy within different research studies in soccer was fortified [10]. That is why the expected goals (xG) metric emerged in 2013, which tried to calculate the probability of any given shot being transformed into a goal based on numerous different factors describing the shot [11]. The xG is well established in the soccer analytics community [12] and it is statistically validated by professional match analysts, among other experts. This metric represents the probability between 0 and 1 for each shot taken by a soccer team in a game (1 is considered a definite goal and 0 is considered no possibility of the shot becoming a goal). Experts [13,14] suggested that a shot-based metric, such as the xG, was a better measure to understand the randomness in soccer than a goal-based metric, since a goal is a much less common event than a shot. In this line, analysts may have a broader vision of the game so that they may calculate the probability of a shot becoming a goal. Expressly, they could determine how many goals both teams ‘should have’ scored given the chances they created. Initially, this was the best-performing metric and was based on hand-crafted features from synchronized positional and event data of more than one hundred thousand shots in the German Bundesliga. Today, the xG has expanded to the rest of the football leagues worldwide. Even so, to the best of our knowledge, the scientific literature is limited on the xG parameter applied to the analysis of football matches [8,15].

In 2018, a new metric appeared, known as expected goals on target (xGOT), that has also been popularized under the name post-shot expected goals (PSxG) [16]. This metric was an extension of xG that specifically focused on shots that are on target. While the xG represented the probability of a shot based on the situation when the shot was taken, the xGOT was based on the situation after the shot was taken, specifically where the shot ended. To date, researchers have only found one publication about the xGOT metric in an important scientific journal, recently published in January 2024 [17]. All other works could be considered “gray literature”.

Therefore, this study had three objectives: 1. To propose a new metric xGOT that should help create a better understanding of the performance of all soccer teams of the Spanish professional soccer championship (LaLiga and LigaF), including the accumulated match statistics obtained during the last competitive season (2023–2024). The authors hypothesized that this approach (xGOT) would allow us to evaluate team performance with greater precision than the traditional metrics. 2. To assess reference values for the previously studied variables in elite women’s and men’s soccer leagues. 3. To propose a model to analyze soccer performance, denominated the “chain on goals model in football” and that may serve as guide to soccer analysts, coaches or technical staff.

2. Materials and Methods

2.1. Sample

The sample included all professional football teams that competed in the first division of Spanish football during the 2023–2024 season, both in the women’s and men’s categories (16 teams in LigaF and 20 teams in LaLiga, respectively). A total of thirty matches for LigaF and thirty-eight for LaLiga were analyzed. The data used were obtained for convenience.

Data for Spanish teams was collected from the statistical website Football Reference https://www.fbref.com (accessed on 27 June 2024), which used a comprehensive dataset from Opta (Stats-Perform) https://www.fotmob.com (accessed on 27 June 2024). All data are freely accessible to anyone. The reliability of the OPTA system has been previously proven [18] and the reliability of TRACAB’s multi-camera tracking system was evaluated for the players’ positioning and physical performance [19].

2.2. Design and Ethics Committee

This was a descriptive and correlational study in which we made a comparative analysis of the end of season accumulated match statistics obtained during a competitive season of professional Spanish football teams competing in the league, both in the men’s (20 teams per season) and women’s (16 teams per season) categories. The local ethics committee approved this study.

2.3. Methodology

For all teams in both leagues, the following variables were analyzed: goals, PV, xG and xGOT. All variables were obtained from the website previously mentioned. Unfortunately, football event and position data and more complex metrics are rarely publicly available [20,21], with companies such as Opta and Stats Bomb collecting the data themselves and publishing them directly on their websites. Still, the reliability of the data has been previously demonstrated [18,19].

• -. Goals is the total number of goals scored by the team per season.

• -. Possession value (PV) measures the probability of a team scoring from their possession.

• -. Expected goals (xG) is the estimation of the probability of any given shot being converted to a goal based on various different factors describing the shot. Shot location is the most common factor, which depends on two variables: the distance and the angle toward the goal when the shot was taken.

• -. Expected goals on target (xGOT) measures the post-shot quality of on-target efforts at goal. It is an indicator of how well a player is shooting.

According to football analytics experts, all of the results obtained for each variable were normalized by match (90 min) for the best understanding and interpretation.

2.4. Statistical Analysis

The data were statistically analyzed using SPSS (version 18; SPSS Inc., Chicago, IL, USA). The Shapiro–Wilk test was used to perform normality analysis, and the result indicated a normal distribution. The statistical analysis performed a descriptive study of all variables expressed as the average and the standard deviations, as well as the percentiles (30th and 70th for LaLiga and 20th and 80th for LigaF). The difference in the calculation of the percentiles between the men’s and women’s football leagues is due to the fact that the LigaF has fewer teams and fewer competitions during the season. Pearson correlation coefficient (r) compared the interactions between variables, and R² also analyzed the correlation between goals per match and all variables studied.

This investigation applied the following criteria to determine the magnitude of the correlation (r): <0.1 trivial, 0.1 to 0.3 small, 0.3 to 0.5 moderate, 0.5 to 0.7 large, 0.7 to 0.9 very large and 0.9 to 1.0 almost perfect.

3. Results

Table 1 and Table 2 illustrate the values of all variables mentioned throughout the entire season, including the mean, standard deviation, range of values (lowest and highest) and percentile distribution, for both leagues. The data for both leagues were different, which means that the interpretation of these metrics in relation to the game situations in the respective leagues must be different. The defense system in the LaLiga was more important than the attack system, and the goalkeeper’s performance in the men’s league had a greater impact on the match result than the goalkeeper’s performance in the women’s league.

Table 3 establishes the relationship between all of the variables analyzed and the number of goals scored per match by each team in both leagues. The PV was the metric that had the lowest correlation with total goals, both in the men’s and women’s leagues (R² = 0.4570 and R² = 0.8442, respectively). In the men’s league, the xGOT was the best metric that represented the match result with a very high correlation level (R² = 0.9248) and in the women’s league, both the xG and xGOT were the best metrics that reflected the match outcome with a very high correlation (R² = 0.9820 and R² = 0.9574, respectively). The distribution of the point cloud in these relationships is appreciated in the figure. Figure 1 shows the relationship of goals per match with the PV, xG and xGOT for the women’s (Figure 1A–C) and men’s (Figure 1D–F) leagues.

4. Discussion

The main finding of this study was to present a new parameter (xGOT) for a better understanding of elite soccer matches in women’s and men’s leagues, specifically in the Spanish championships (LigaF and LaLiga). In addition, the study provided reference values for percentiles for all of the variables studied, such as total goals, PV, xG and xGOT, in both leagues (Table 2). Football has seen the creation of various novel, ubiquitous metrics used throughout clubs’ analytics departments. In this line, this is the first study that applied the xGOT metric to data from one of the main European leagues and demonstrated how this approach (xGOT) would allow us to improve the understanding of the football match outcome as well as evaluate the football players’ and teams’ performances more accurately than the traditional metrics.

In this article, the researchers would like to introduce a novel analysis model, focusing on the process rather than the results, for the best interpretation of the players’ or clubs’ performances during a match or season. This model is called the “chain on goals model” (Figure 2) and was based on the use of the four metrics analyzed in this study. The optimal model built in this project based on these four metrics may be more competitive when compared to previous metrics within the existing literature [22,23,24]. These previous studies have always focused on analyzing the players’ physical and tactical characteristics to understand the team’s game, and thus they could be able to justify the match outcome [2]; however, more complex analyses using new metrics, such as those in this study, highlighted the role of the shot, specifically the characteristics before and after the shot [13,14,17]. Following this line, football analysts may determine how many goals both teams ‘should’ have scored given the chances they created. In this study, we demonstrated that the calculation of the xG and xGOT metrics may predict or explain the match result in both leagues because they had a high correlation with goals scored.

The development of this model was based on two conditions: on the one hand, this model was described from conversations with professional match analysts who have performed and applied it over more than 10 years in the football match analysis; on the other hand, we have evaluated it through the correlations established between the game score and the three metrics studied (PV, xG and xGOT) (Table 3 and Figure 1). Indeed, this model may be considered a competitive advantage for many coaching staff members to obtain better sports results.

Traditionally, and still today, technical staff and journalists tried to correlate the match outcome with ball possession. In fact, this theory was supported by the scientific literature because there were still many studies that analyzed individuals’ and teams’ ball possession to understand the players’ or teams’ performances as well as the match result [25,26,27,28]. That is why scientists investigated the relationship between ball possession, physical demands [29] or team performance [30]. However, Wang et al. [26] carried out a systematic review in 2022 and they concluded that the match outcome was not related to ball possession percentage. The data of our study were in this line since the PV was the metric that had the lowest correlation with total goals, both in the men’s and women’s leagues (R² = 0.4570 and R² = 0.8442, respectively).

Regarding the other two metrics and focusing on the men’s league, the findings of this study indicated that the xGOT was the metric that best represented the match result with a very high correlation level (R² = 0.9248) (Figure 1). This meant that the xGOT better reflected the players’ and teams’ performances throughout the season, compared to the xG and the PV. In the case of the women’s league, our results showed that both the xG and the xGOT were the best metrics that reflected the match outcomes with a very high correlation (R² = 0.9820 and R² = 0.9574, respectively) (Figure 1). The authors suggest that this may be because the Spanish women’s soccer league is less competitive and, therefore, both metrics (xG and xGOT) had a high correlation level with total goals, unlike the Spanish men’s soccer league. According to Papalardo et al. [31], the competitive levels of women’s and men’s soccer in Spain were different. That is why women’s football should be considered a distinct sport and coaches should design tactics tailored to women’s teams. Based on these data, the main strength of this study was that the application of this model to football could help to understand the quality of player and team performance in relation to the match score for scientists, journalists, coaches and fans.

On the other hand, this study established useful reference values provided by elite football clubs that could be used for the formulation of tactical strategies (Table 2). To our knowledge, this is the first study that highlighted the usefulness of these two metrics, using data that belonged to high-level competitive clubs, such as the Spanish men’s and women’s soccer leagues. It is also notably important to give references for the new metrics proposed. To this extent, the percentile distribution showed that, in the case of the men’s league, xGOT values higher than 1.50 (p70) should be considered indicators of a strength index since these values reflected the score of this variable for the teams that competed at the European level until the last matchday, and xGOT values lower than 1.12 (p30) should be considered a weakness index since these values represented the score of the variable of the teams that are in relegation positions to another lower category league until the last matchday. In the case of women’s league, xGOT values higher than 1.70 (p80) should be considered indicators of a strength index and xGOT values lower than 0.85 (p20) should be considered a weakness index.

4.1. Limitation Section

This work has several limitations: 1. This study was conducted with data from one of the main European leagues, the Spanish soccer league. Applying this metric in other European leagues, even worldwide, might turn out to be interesting. 2. This study analyzed a very novel statistical metric. Future studies applying this parameter to practical situations may be important to know the potential of the xGOT metric. 3. The study was conducted using a static rather than a dynamic method [32,33,34]. However, this static method may be didactically useful in decision making during matches for coaches [35]. 4. The scientific literature on the xG and xGOT metrics is very limited. Future investigations are necessary to apply these new metrics to technical and tactical strategies, which will be very useful for coaches and technical teams.

4.2. Practical Application

First, researchers have described a new metric (xGOT) to facilitate football match analysis and the understanding of the soccer players’ or teams’ performances during matches. This measure was strongly consistent with the match result.

Second, this investigation provided reference values of variables studied for female and male soccer players through a percentile distribution. These references should be especially useful for the control of tactical strategies for a particular player or team performance during championships.

In weekly monitoring during the season, these findings may be used to plan the tactical strategy of the training season and the xGOT may be a good metric of how the performance of each player and soccer team could improve, and could aid in the adaptation of the team to the opposing team.

5. Conclusions

The application of the xGOT metric as an indicator of a soccer players’ or teams’ performance could improve the understanding of different results in soccer matches. Also, this study provided reference values of all of the variables studied in elite soccer players that may be very important to soccer clubs during the season, both in the women’s and men’s leagues.

Author Contributions

Conceptualization, A.R.-d.-A.-Q.; methodology, B.D.-l.-C.-T.; formal analysis, B.D.-l.-C.-T.; investigation, A.R.-d.-A.-Q. and B.D.-l.-C.-T.; data curation, A.R.-d.-A.-Q.; writing—original draft preparation, A.R.-d.-A.-Q. and B.D.-l.-C.-T.; writing—review and editing, A.R.-d.-A.-Q. and B.D.-l.-C.-T.; supervision, A.R.-d.-A.-Q. and B.D.-l.-C.-T. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of University of Seville (protocol code 2024-1326 and date of approval 26 June 2024).

Informed Consent Statement

Not applicable.

Data Availability Statement

The original data presented in the study are openly available on the following website: http://www.fbref.com (accessed on 26 August 2024).

Conflicts of Interest

The authors declare no conflicts of interest.

Footnotes

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.

Figures and Tables

Figure 1. R2 correlation of total goals with possession value (PV), expected goals (xG) and expected goals on target (xGOT) in LigaF (A–C) and LaLiga (D–F).

Figure 2. Chain on goals model in football.

Table 1

The values for all variables of the study in each league.

Variable	Mean		SD		Lowest		Highest
	LigaF	LaLiga	LigaF	LaLiga	LigaF	LaLiga	LigaF	LaLiga
Goals/90	1.57	1.32	0.94	0.48	0.67	0.66	4.57	2.29
PV/90	1.68	1.32	0.34	0.16	1.18	1.10	2.50	1.69
xG/90	1.39	1.30	0.73	0.31	0.65	0.87	3.74	2.04
xGOT/90	1.36	1.32	0.69	0.35	0.58	0.86	3.42	2.04

Abbreviations: PV, possession value; xG, expected goals; xGOT, expected goals on target; SD, standard deviation.

Table 2

The reference values for all variables in the study.

Percentile
Variable	LigaF		LaLiga
	Weakness Index(20th)	Strength Index(80th)	Weakness Index(30th)	Strength Index(70th)
Goals/90	0.93	2.01	1.02	1.53
PV/90	1.36	1.93	1.20	1.33
xG790	0.83	1.68	1.12	1.37
xGOT/90	0.85	1.70	1.12	1.50

Abbreviations: PV, possession value; xG, expected goals; xGOT, expected goals on target.

Table 3

The Pearson correlation coefficient (r) and R² for the exposed variables.

Goals/90	LigaF		LaLiga
	r	R2	r	R2
PV/90	0.9188	0.8442	0.6760	0.4570
xG/90	0.9910	0.9820	0.9326	0.8698
xGOT/90	0.9785	0.9574	0.9616	0.9248

Abbreviations: PV, possession value; xG, expected goals; xGOT, expected goals on target.

References

1. Pollard, R. Charles Reep (1904–2002): Pioneer of notational and performance analysis in football. J. Sports Sci.; 2002; 20, pp. 853-855. [DOI: https://dx.doi.org/10.1080/026404102320675684]

2. Lames, M.; Hermann, S.; Prüßner, R.; Meth, H. Football Match Dynamics Explored by Recurrence Analysis. Front. Psychol.; 2021; 12, 747058. [DOI: https://dx.doi.org/10.3389/fpsyg.2021.747058] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34630256]

3. Rein, R.; Memmert, D. Big data and tactical analysis in elite soccer: Future challenges and opportunities for sports science. SpringerPlus; 2016; 5, 1410. [DOI: https://dx.doi.org/10.1186/s40064-016-3108-2] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27610328]

4. Lucey, P.; Bialkowski, A.; Monfort, M.; Carr, P.; Matthews, I. “Quality vs Quantity”: Improved shot prediction in soccer using strategic features from spatiotemporal data. Proceedings of the 9th Annual MIT Sloan Sports Analytics Conference; Boston, MA, USA, 27–28 February 2015; pp. 1-9.

5. Tenga, A.; Ronglan, L.T.; Bahr, R. Measuring the effectiveness of offensive match-play in professional soccer. Eur. J. Sport Sci.; 2010; 10, pp. 269-277. [DOI: https://dx.doi.org/10.1080/17461390903515170]

6. Fernández, J.; Bornn, L.; Cervone, D. A framework for the fine-grained evaluation of the instantaneous expected value of soccer possessions. Mach. Learn.; 2021; 110, pp. 1389-1427. [DOI: https://dx.doi.org/10.1007/s10994-021-05989-6]

7. Power, P.; Ruiz, H.; Wei, X.; Lucey, P. Not all Passes Are Created Equal: Objectively Measuring the Risk and Reward of Passes in Soccer from Tracking Data. Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; Halifax, NS, Canada, 13–17 August 2017; pp. 1605-1613.

8. Decroos, T.; Brasen, L.; Van Haaren, J.; Davis, J. Actions speak louder than goals: Valuing players actions in soccer. Proceedings of the 25rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; New York, NY, USA, 4–8 August 2019; pp. 1851-1861.

9. Brasen, L.; Van Haaren, J. Measuring Football players on the-ball contributions from passes during games. Proceedings of the International Workshop on Machine Learning and Data Mining for Sports Analytics; Dublin, Ireland, 10 September 2018; Springer: Cham, Switzerland, 2018; pp. 3-15.

10. Spearman, W.; Basye, A.; Dick, G.; Hotovy, R.; Pop, P. Physics based modeling of pass probabilities in soccer. Proceedings of the MIT Sloan Sports Analytics Conference; Boston, MA, USA, 3–4 March 2017; pp. 1-14.

11. Anzer, G.; Bauer, P. A Goal Scoring Probability Model for shots based on synchronized positional and event data in football (Soccer). Front. Sports Act. Living; 2021; 3, 624475. [DOI: https://dx.doi.org/10.3389/fspor.2021.624475]

12. Davis, J.; Robberechts, P. How data availability affects the ability to learn good xG models. Proceedings of the 7th International Workshop of Machine Learning and Data Mining for Sports Analytics; Ghent, Belgium, 14–18 September 2020; pp. 17-27.

13. Brechot, M.; Flepp, R. Dealing with Randomness in Match Outcomes: How to Rethink Performance Evaluation in European Club Football Using Expected Goals. J. Sports Econ.; 2020; 21, pp. 335-362. [DOI: https://dx.doi.org/10.1177/1527002519897962]

14. Noordman, R. Improving the Estimation of Outcome Probabilities of Football Matches Using In-Game Information; Technical Report Amsterdam School of Economics: Amsterdam, The Netherlands, 2019; Available online: https://www.scisports.com/wp-content/uploads/2019/10/Noordman-Rogier-12366315-MSc-ETRICS.pdf (accessed on 27 June 2024).

15. Mead, J.; O’Hare, A.; McMenemy, P. Expected goals in football: Improving model performance and demonstrating value. PLoS ONE; 2023; 18, e0282295. [DOI: https://dx.doi.org/10.1371/journal.pone.0282295]

16. Goodman, M. A New Way to Measure Keepers’ Shot Stopping: Post-Shot Expected Goals; StatsBomb: Singapore, 2018; Available online: https://statsbomb.com/articles/soccer/a-new-way-to-measure-keepers-shot-stopping-post-shot-expected-goals/ (accessed on 28 June 2024).

17. Baron, E.; Sandholtz, N.; Pleuler, D.; Chan, T.C.Y. Miss it like Messi: Extracting value from off-target shots in soccer. J. Quant. Anal. Sports; 2024; 20, pp. 37-50. [DOI: https://dx.doi.org/10.1515/jqas-2022-0107]

18. Liu, H.; Hopkins, W.; Gómez-Ruano, M.A.; Molinuevo, J.S. Inter-operator reliability of live football match statistics from OPTA Sportsdata. Int. J. Perform. Anal. Sport; 2013; 13, pp. 803-821. [DOI: https://dx.doi.org/10.1080/24748668.2013.11868690]

19. Linke, D.; Link, D.; Lames, M. Football-specific validity of TRACAB’s optical video tracking systems. PLoS ONE; 2020; 15, e0230179. [DOI: https://dx.doi.org/10.1371/journal.pone.0230179] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32155220]

20. Madrero, P.; Fernández, J.; Arias, M. Creating a Model for Expected Goals in Football Using Qualitative Player Information; Technical Report Universitat Politècnica de Catalunya (UPC): Barcelona, Spain, 2020; Available online: https://upcommons.upc.edu/bitstream/handle/2117/328922/147841.pdf (accessed on 10 August 2024).

21. Pappalardo, L.; Cintia, P.; Rossi, A.; Massucco, E.; Ferragina, P.; Pedreschi, D.; Giannotti, F. A public data set of spatio-temporal match events in soccer competitions. Sci. Data; 2019; 6, 236. [DOI: https://dx.doi.org/10.1038/s41597-019-0247-7] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31659162]

22. Errekagorri, I.; Fernandez-Navarro, J.; López-Del Campo, R.; Resta, R.; Castellano, J. An eight-season analysis of the teams’ performance in the Spanish LaLiga according to the final league ranking. PLoS ONE; 2024; 19, e0299242. [DOI: https://dx.doi.org/10.1371/journal.pone.0299242] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38416760]

23. Errekagorri, I.; López-Del-Campo, R.; Resta, R.; Castellano, J. Performance Analysis of the Spanish Men’s Top and Second Professional Football Division Teams during Eight Consecutive Seasons. Sensors; 2023; 23, 9115. [DOI: https://dx.doi.org/10.3390/s23229115]

24. González-Rodenas, J.; Ferrandis, J.; Moreno-Pérez, V.; López-Del-Campo, R.; Resta, R.; Del-Coso, J. Differences in playing style and technical performance according to the team ranking in the Spanish football LaLiga. A thirteen seasons study. PLoS ONE; 2023; 18, e0293095. [DOI: https://dx.doi.org/10.1371/journal.pone.0293095]

25. Link, D.; Hoernig, M. Individual ball possession in soccer. PLoS ONE; 2017; 12, e0179953. [DOI: https://dx.doi.org/10.1371/journal.pone.0179953]

26. Wang, S.H.; Qin, Y.; Jia, Y.; Igor, K.E. A systematic review about the performance indicators related to ball possession. PLoS ONE; 2022; 17, e0265540. [DOI: https://dx.doi.org/10.1371/journal.pone.0265540]

27. Iván-Baragaño, I.; Maneiro, R.; Losada, J.; Ardá, A. Technical-tactical evolution of women’s football: A comparative analysis of ball possessions in the FIFA Women’s World Cup France 2019 and Australia & New Zealand 2023. Biol. Sport; 2024; 42, pp. 11-20.

28. Maneiro, R.; Losada, J.L.; Casal, C.A.; Ardá, A. Identification of explanatory variables in possession of the ball in High-Performance Women’s Football. Int. J. Environ. Res. Public Health; 2021; 18, 5922. [DOI: https://dx.doi.org/10.3390/ijerph18115922]

29. García-Calvo, T.; Ponce-Bordón, J.C.; Leo, F.M.; López-Del-Campo, R.; Nevado-Garrosa, F.; Pulido, J.J. How Does Ball Possession Affect the Physical Demands in Spanish LaLiga? A Multilevel Approach. Res. Q. Exerc. Sport; 2023; 94, pp. 931-939. [DOI: https://dx.doi.org/10.1080/02701367.2022.2072798]

30. Jerome, B.W.C.; Stoeckl, M.; Mackriell, B.; Dawson, C.W.; Fong, D.T.P.; Folland, J.P. Evidence for a new model of the complex interrelationship of ball possession, physical intensity and performance in elite soccer. Scand. J. Med. Sci. Sports; 2024; 34, e14546. [DOI: https://dx.doi.org/10.1111/sms.14546] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38059701]

31. Pappalardo, L.; Rossi, A.; Natilli, M.; Cintia, P. Explaining the difference between men’s and women’s football. PLoS ONE; 2021; 16, e0255407. [DOI: https://dx.doi.org/10.1371/journal.pone.0255407] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34347829]

32. Plakias, S.; Tsatalas, T.; Armatas, V.; Tsaopoulos, D.; Giakas, G. Tactical Situations and Playing Styles as Key Performance Indicators in Soccer. J. Funct. Morphol. Kinesiol.; 2024; 9, 88. [DOI: https://dx.doi.org/10.3390/jfmk9020088] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38804454]

33. Pratas, J.M.; Volossovitch, A.; Carita, A.I. Goal scoring in elite male football: A systematic review. J. Hum. Sport Exerc.; 2018; 13, pp. 218-230. [DOI: https://dx.doi.org/10.14198/jhse.2018.131.19]

34. Prieto, J.; Gómez, M.-Á.; Sampaio, J. From a static to a dynamic perspective in handball match analysis: A systematic review. Open Sports Sci. J.; 2015; 8, pp. 24-34. [DOI: https://dx.doi.org/10.2174/1875399X01508010025]

35. Garganta, J. Trends of tactical performance analysis in team sports: Bridging the gap between research, training and competition. Rev. Port. Cien. Desporto; 2009; 9, pp. 81-89.