Assessment of a Coding Tool to Analyse Goals in Football (CODITAG) 

Rubén Sánchez López de Toro

Ibon Echeazarra

Julen Castellano Paulis

*Corresponding author: Rubén Sánchez-López

Original Language Spanish

Cite this article

Sánchez-López, R., Echeazarra, I. & Castellano, J. (2023). Assessment of a Coding Tool to Analyse Goals in Football (CODITAG). Apunts Educación Física y Deportes, 151, 58-69.



The most defining event in a football match is scoring a goal; therefore, the analysis of the dynamics of the game and the behaviours that lead to scoring goals can provide an important contribution to the identification of each team’s most decisive attacking sequences in order to propose an appropriate strategy. Thus, the aim of this study was to set up and assess an ad hoc coding tool to analyse goals in semi-professional, amateur and training football. Based on several action-research processes, and using a set of experts through the Delphi method, a field-format coding tool was designed and assessed where each criterion is an exhaustive and mutually exclusive system of categories. The results demonstrated excellent content validity values, estimated through the degree of agreement (9.37 out of 10) and importance (9.66 out of 10). Similarly, excellent intra-observer reliability values were obtained (k ≥ .87), and good to excellent inter-observer agreement values (k_ ≥ .62). The evaluation of generalisability using a three-facet design (observers, categories, goals) showed excellent reliability (G > .90), and perfect representativeness (r2 = 1), showing that the variability is explained by the categories facet (in isolation or in interaction with the other facets). In conclusion, this study provides a valid and reliable tool that allows for the identification of the teams’ most representative traits through the analysis of their goals.

Keywords: attacking phase, attacking sequence, football training, Metodología Observacional, Observational Methodology, sporting performance, Tactical Analysis.


The internal logic of play in socio-motor sports is influenced by the structure of scoring interactions, as they map out the ways in which scoring goals can take place, and define the way in which teams are able to change the score during game time, as well as the types of interaction that can take place between players and opposing teams (Parlebas, 2001). In the case of football, despite being able to interact with both teammates and opponents, the type of scoring interaction is antagonistic given that the scoreboard changes through scoring a goal, beating the opposition in the collective duel. Due to the nature of scoring interactions in football, only 1% of professional teams’ possessions lead to scoring (González et al., 2020). Scoring a goal, therefore, is the most successful action in football and, so, analysing the scenarios that lead to goals can provide an important contribution to the identification of the most emblematic or critical game actions in order to understand the factors that provoke perturbations or imbalances in the attack/defence balance (Hughes, 1996). Therefore, observing the sequences of attacking play from previously defined criteria and categories allows us to understand how the teams behave when they score a goal, either in the moment of attacking transitions or in the execution of their attacks in the attacking phase.

In recent years, different observational tools have been used in order to analyse goals or sequences leading to goal-scoring opportunities. Some of these tools focus on recording exactly each event from the moment the team initiates the attacking sequence after regaining possession of the ball until the end of the action (Aranda et al., 2019; Barreira et al., 2013; Castellano, 2000; Echeazarra, 2014; Ortega-Toro et al., 2019; Papadopoulos et al., 2021; Sarmento et al., 2010; Tenga et al., 2010), while others do not look at the entire sequence of play (Caro Muñoz & Caro Muñoz, 2016; Kubayi, 2020; Ugalde-Ramírez & Rodríguez-Porras, 2021). Spatial characterisation is another criterion that is involved in many of these tools. Some record from where sequences of play start and which areas of the pitch the ball travels across until the end of its completion (Barreira et al., 2013; Castellano, 2000; Echeazarra, 2014; Ortega-Toro et al., 2019; Sarmento et al., 2010). Others have only focused on spatially characterising the last kick (Papadopoulos et al., 2021; Ugalde-Ramírez & Rodríguez-Porras, 2021), or the area where possession is initiated (Aranda et al., 2019; Caro Muñoz & Caro Muñoz, 2016; Kubayi, 2020; Papadopoulos et al., 2021; Tenga et al., 2010). Other criteria studied have been the result of the match before the goal or scoring opportunity (Barreira et al., 2013; Sarmento et al., 2010), the type of attack (Aranda et al., 2019; Kubayi, 2020; Papadopoulos et al., 2021; Sarmento et al., 2010; Tenga et al., 2010; Ugalde-Ramírez & Rodríguez-Porras, 2021), the behaviours that appear during the development and completion (Aranda et al., 2019; Barreira et al., 2013; Echeazarra, 2014; Ortega-Toro et al., 2019; Sarmento et al., 2010), the surface used in the last contact (Echeazarra, 2014; Ortega-Toro et al., 2019; Papadopoulos et al., 2021), the number of contacts made by the player (Echeazarra, 2014; Ortega-Toro et al., 2019), the number of passes before completion (Aranda et al., 2019; Caro Muñoz & Caro Muñoz, 2016; Kubayi, 2020; Papadopoulos et al., 2021; Sarmento et al., 2010), the typology or direction of the passes (Echeazarra, 2014; Sarmento et al., 2010; Tenga et al., 2010), the temporality of goals or goal-scoring opportunities (Barreira et al., 2013; Kubayi, 2020; Papadopoulos et al., 2021; Sarmento et al, 2010; Ugalde-Ramírez & Rodríguez-Porras, 2021), whether the home or away team finishes (Echeazarra, 2014; Sarmento et al., 2010), the centre of play (Barreira et al., 2013), the context of interaction between the two teams (Barreira et al., 2013; Castellano, 2000; Echeazarra, 2014; Ortega-Toro et al., 2019) and the context of opposition applied by the opposing team (Aranda et al., 2019; Ortega-Toro et al., 2019; Tenga et al., 2010). 

Although some of these tools offer a high level of detail, they are often not sustainable in certain contexts, as they require high-quality recordings to record everything that it intends to. As a result, it would be of interest to design a tool that is sustainable in non-professional and training football and that, in addition to allowing delayed analysis, can be used live, and focuses in a particular way on what happens in the last seconds before the goal is scored; structured on the basis of some of the criteria considered and other criteria not yet considered in the scientific literature. 

In light of the above, the aim of this study was to set up and assess a coding tool designed ad hoc to specifically analyse goals in football. The results of the present study will allow the tool to be applied both in the scientific field and in the field of sports competition analysis. In order to achieve this objective, the degree of validity and reliability of the data collected was determined to ensure the quality of the coding tool.



The study presented responded to a follow-up observational, nomothetic and multidimensional design (Anguera et al., 2011). It was follow-up because the data collection was carried out in one season or competitive period over several matches; it was nomothetic because the data was recorded in a cross-sectional way (Hernández-Mendo & Molina, 2002), coding the goals of all the teams that participated in the same league, independently, without any link between them; and it was multi-dimensional because the goals were analysed according to several criteria. The data type was therefore concurrent and event-based, in other words, type II (Bakeman, 1978), because several dimensions were recorded in the same cluster, irrespective of the duration of the events. Therefore, data collection required the configuration of an ad hoc coding tool which, based on the structure of the observational design, was a combination of field format and category systems (Anguera & Blanco-Villaseñor, 2006).


To assess the tool, a total of 12 experts contributed their findings via Google Forms in two phases (n = 6, n = 6). Each of the selected experts met at least two of the following four requirements: (1) work as a coach or analyst with more than 10 years of experience; (2) be a coach with a minimum level 3 qualification; (3) have a degree in Physical Activity and Sport Sciences with a specialisation in football; (4) possess a Ph.D. on football.

For the design and optimisation of the coding tool, a sample of 477 goals corresponding to 18 match days of the 2019/20 season in the 3rd division of the RFEF group 7 (Madrid) was used. Subsequently, to calculate the reliability, a sample of 52 goals corresponding to match days 1 and 2 of the 2021/22 season in the 3rd RFEF Group 7 (Madrid) was used, with three observers for this process.

As no recordings were made directly on participants, no ethical consent was required as no invasive measures were taken to gather the data. The sequences of play were observed using the videos that the Madrid Football Federation offers openly and publicly on its website, in which all the goals are shown after each match day of 3rd RFEF group 7.

Coding Tool

The initial tool was developed using other previous tools aimed at the analysis of goals or attacking sequences as a reference (Barreira et al., 2013; Caro Muñoz & Caro Muñoz, 2016; Sarmento et al., 2010), and was used for the first time during the 2019/20 season. In that season, and alongside the recording of observed goals, the tool was modified to accommodate those criteria that could be analysed with their consequent exhaustive and mutually exclusive category systems. In the 2020/21 season, the tool was presented to the first group of experts (n = 6), who, using the Delphi method, provided their contributions in order to provide evidence of content validity. In the 2021/22 season, again following the Delphi method, a second group of experts (n = 6) assessed the tool, modifying it for analysing goals in football (CODITAG). The tool consists of a combination of a field format and exhaustive and mutually exclusive category systems, with 11 criteria, as shown in Table 1. It should be noted that 6 of these 11 criteria (match day, home team, away team, scoring team, minute of the goal, status of the scoreboard before the goal) do not depend on the analysis of the game action, as they can be verified without the need to observe the attacking sequence leading to the goal. Therefore, they have not been taken into account in the process of estimating validity and reliability. The criteria that had to undergo a process of estimating their validity and reliability for inclusion in the tool, due to their direct relationship with the observation of the game action, were: the type of attack, the contextualisation of the penultimate action, the contextualisation of the last action, the number of contacts of the last action and the surface used in the last contact. 

Figure 1
See Full Size
Recovery zones to record goals in transition.

Table 1

Criteria and categories of the coding tool for analysing goals in football -CODITAG-

See Table


The design of the tool and the processes to provide evidence of validity and reliability were carried out in six stages (see Figure 2): (a) literature review and design of the provisional coding tool, (b) action research processes through piloting to ensure that the criteria category systems were exhaustive and mutually exclusive, (c) optimisation and content validation of the coding tool using two expert groups, (d) development of the inter- and intra-observer reliability estimation processes, (e) data quality control based on the analysis of generalisability and, finally, (f) quality assessment of the whole process using the checklist of methodological quality for studies based on observational methodology —MQCOM— (Chacón-Moscoso et al., 2019).

Figure 2
See Full Size
Phases to design and assess CODITAG.

In the first stage, the selection of the observation categories that make up the tool was made through a literature review of a set of studies that had used coding tools for the analysis of goals in football (Caro Muñoz & Caro Muñoz, 2016; Sarmento et al., 2010; Tenga et al., 2010).

In the second stage, the ad hoc coding tool was implemented for the data collection process using the software Microsoft Excel 365 (Microsoft Corporation, Washington, USA). The use of other types of more specific software within observational studies was discarded, because several criteria of the coding tool had a large number of categories (e.g. the criterion “minute of goal” has 90 categories). In this stage, the coding tool was used to record and analyse 477 goals in the 3rd national division group 7 (Community of Madrid), during 18 match days of the 2019/20 season. Alongside this process, the problems that arose in assessing the criteria that made up the tool were resolved to ensure that the category systems were exhaustive and mutually exclusive.

The third stage consisted of two phases: optimisation of the tool and content validity. In the first phase, through the suggestions of the first group of experts (n = 6), 5 new categories were created and the definitions of 11 categories were reformulated. In the second phase, the content validity of the tool was established by the second group of experts (n = 6) through the content validity coefficient (CVC) (Hernández Nieto, 2002). For this, a 1-10 scale was used with the experts to facilitate their responses on two factors: the degree of agreement and the degree of importance of each of the criteria and categories. Subsequently, the data was remodelled to a 0.1-1 scale, averaging the experts’ responses for each criterion and category, according to factor. All of the tool’s categories had agreement and acceptance values above 0.8, with all categories being accepted (Bulger & Housner, 2007). This was because the tool had undergone many pilots before being presented to this second group of experts. Finally, the resulting values were remodelled back to the 1-10 scale. After the coding tool was approved, it was updated in Excel. 

In the fourth stage, the process of gathering evidence of the tool’s reliability — both intra-observer and inter-observer — was carried out. Following the procedures developed in other works (Barreira et al., 2013; Fernandes et al., 2019; Ortega-Toro et al., 2019; Sánchez-López et al., 2021), 3 observers performed the analysis of the goals (n = 52) of the first and second league match day of the 3rd RFEF group 7 (Community of Madrid), 2021/22 season. Using the software SPSS Statistics for Windows, v19 (IBM Corporation, New York, USA), as well as the Microsoft Excel 365 add-in XrealStats, inter-observer reliability was calculated, and the criteria analysed as different between observers were discussed and analysed again, with the first author of the article acting as observer-moderator. Six weeks later, using the test-retest reliability method, one of the observers repeated the analysis process and the results obtained were compared with their previous analysis in order to calculate intra-observer reliability. 

In the fifth stage, given the nature of the data analysed and in order to control their quality, the generalisability theory (Cronbach et al., 1972) was applied by modelling the different sources of variability or facets (observers [O], goals [G] and categories [C] of the taxonomic system), designing six possible models: [CG/O], [O/CG], [OG/C], [C/OG], [OC/G] and [G/OC].

In the sixth and final stage, the study was evaluated using the checklist of methodological quality for studies based on observational methodology —MQCOM— (Chacón-Moscoso et al., 2019), consisting of 20 items (1 point per item). The study scored 16.67 out of 18 points (2 items were not analysed).

Once the stages were completed, the coding tool could be used to code the goals, documenting the data in Excel. From this same application, the data is analysed descriptively through dashboards designed ad hoc to obtain the frequency and percentage of occurrence of each category within each criterion, as well as linking data between criteria in order to obtain more information. The following is an example (see figure 3) of the criterion “type of attack” on a sample of 273 goals corresponding to the first 12 match days of 3rd RFEF Group 7 (Community of Madrid) of the 2021/22 season.

Figure 3
See Full Size
Descriptive analysis of goals according to the criterion “type of attack” using the criteria home, away and scoring team.

Statistical Analysis

The coding tool was assessed in relation to the quality of the data. To achieve this, the content validity of the tool was reached qualitatively through the consensus agreement of a group of experts following the Delphi method and using the CVC (Hernández Nieto, 2002). The tool was also analysed quantitatively, calculating its intra-observer reliability using Cohen’s kappa coefficient, and its inter-observer reliability using Fleiss’ kappa coefficient for comparing more than 2 observers, as well as Cohen’s kappa coefficient for comparing pairs of observers.


Content Validity

To estimate the CVC (Hernández Nieto, 2002), the averages of the two factors used with the expert group were calculated following the Delphi method: the degree of agreement (9.37 out of 10), which reflects the clarity of the language (“do you think that the definition of the category is well elaborated and exclusive with respect to the other categories of the criterion?”), and the degree of importance or appropriateness (9.66 out of 10), which represents the theoretical and practical relevance (“do you think that the category should be part of the criterion?”). The scores obtained for both factors showed very high content validity. 

Table 2

Intra-observer reliability through the test-retest method with Cohen’s kappa coefficient. 

See Table

Intra-Observer Reliability

In order to calculate the intra-observer stability index, the test-retest method was used by applying Cohen’s kappa coefficient to data recorded by the same observer twelve weeks apart. The results showed agreement rates (k ≥ .87) that could be rated as very good (Altman, 1991) for the five criteria analysed, for a coding tool of these characteristics.

Inter-Observer Reliability

Inter-observer agreement was estimated from two perspectives. On the one hand, through Cohen’s kappa coefficient (k) between pairs of observers, and, on the other hand, using Fleiss’ kappa coefficient (k_) to calculate the total reliability for more than two observers (n = 3).  The results obtained show good (k > .60 and k_ > .60) and excellent (k > .80 and k_ > .80) agreement values between observers from the two perspectives.

Table 3

Inter-observer reliability calculated pairwise using Cohen’s kappa and for more than 2 observers using Fleiss’s kappa.

See Table

Generalisability Evaluation

The evaluation of generalisability was carried out using the software SAGT v1.0 build 218.0.1. (Hernández-Mendo et al., 2016). For this, three facets (observers [O], categories [C] and goals [G]) were used, resulting in six possible models (see table 4).

Table 4

Estimated values of the relative (ξρ2(δ)) and absolute (ξρ2(Δ)) generalisability coefficients for the designs: [CG/O], [O/CG], [OG/C], [C/OG], [OC/G] y [G/OC]. 

See Table

Based on the analysis, three aspects were addressed: reliability, variability and representativeness of the model.

Designs that used the “Categories” facet as a differentiating facet showed relative and absolute generalisability coefficients close to 1. It seems, then, that the observers agreed on their observations, linking this to a high reliability when discussing the generalisability of the results (close to 1). 

The possible sources of variance showed that most of the variability (70.84%) is explained when the facet “Categories” is linked to the facet “Goals”, the remaining part being explained by the facet “Categories” in isolation (15.38%) or in interaction with the other facets (13.78%). This reveals the heterogeneity shown by both the categories established and the goals observed, as well as the homogeneity in the observations, an ideal situation which means that the recording made by the observers has not influenced the values obtained, with no notable differences between the records (Usabiaga et al., 2013). Therefore, the categories can be considered as exclusive within the taxonomic system set up.

Lastly, the coefficient of determination (r2) was estimated using the following formula (r2 = SCE/SCT) as follows: with SCT being the total sum of squares, SCE the sum of squares explained, and SCR the residual sum of squares, STC = SCE + SCR (representing the ideal model STC = SCE, and SCR = 0). When using a three-facet design, there were seven possible sums of squares (each facet in isolation, the facets in pairs, and all three facets). Taking the four options where the facet “Categories” appears as a differentiating facet for the calculation of the SCE, it was obtained that SCR = 0, since the sum of squares of [O], [G] and [O][G] is 0.00, as shown in table 4. Therefore, r2 = 1.00. This would mean that the model is fully representative.


The aim of this study was to outline the steps taken to set up and assess an ad hoc coding tool for analysing goals in football. The survey provides a valid and reliable tool that allows for the collection of data in a rigorous and relevant, yet agile and simple manner.

This tool was purposely built to allow its sustainable use in semi-professional, amateur and training football teams where sometimes coaching staff face serious complications when properly recording their matches and obtaining records of opposing teams. Therefore, it differs from other tools in existing scientific literature in that it is composed of criteria aimed at analysing the events that take place in the last few seconds before a goal, with the possibility of collecting live and delayed data. It is even possible to use it in match summaries that do not offer a full visualisation of the game sequences, as was done in the process of obtaining evidence of reliability in this study.

The coding tool consists of 11 criteria, five of which depend on the observation of the game action (the type of attack, the contextualisation of the penultimate action, the contextualisation of the last action, the number of contacts of the last action and the surface used in the last contact). Thanks to the process of obtaining evidence of validity, for the criterion “type of attack”, a very broad classification of categories was developed to bring together all the expert contribution. Combination play was divided into positional and quick, in order to identify the strategic use of space (Amatria et al., 2019). In other words, in the positional combination play, width was prioritised over depth, while in the quick combination play, depth was prioritised over width. The counter-attacks were divided according to the recovery zone, and the set-piece was also distinguished according to the situation leading to the goal. For the criterion “contextualising the penultimate action” the large box was used as the main reference (inside-inside, inside-outside, outside-inside, outside-outside) to determine the type of pass, which greatly facilitated the classification of the play action. This idea was based on previous work (Echeazarra, 2014), in which the action of the player in possession of the ball is contextualised by taking the effective playing space as a reference (Castellano, 2000). When the penultimate action was not a pass as such, other categories were used (header, rebound, steal-interception, throw-in, no handball, other). For the criterion “contextualisation of the last action” the level of opposition (1v0, 1vP, rest) and again the large box (last contact inside or outside) were used as references. A distinction was also made as to whether the last action was with the ball in play or a set-piece (direct foul or penalty). Lastly, the criteria “number of contacts of the last action” and “surface area used in the last contact” were used, as seen in previous work as outlined in the introduction of this paper.

With regard to the contribution of the other six criteria, it can be noted that the criterion “match day” allows the tracking of goals by comparing their progression in a league championship. The criteria “home team” and “away team” allow the identification of patterns of attacking play in terms of scoring and conceding goals by teams when playing at home or away. This, undoubtedly, can be linked to the teams’ game model, since knowing these game patterns allows us to identify and define their most characteristic features (Martín Barrero et al., 2021), in order to incorporate a specific tactical intention that fits with that game idea when designing training tasks (Lapresa et al., 2020). The criterion “scoring team” identifies whether it is the home or away team that scores the goal. The criterion “minute of the goal” allows the goals to be analysed by the time that they were scored. As mentioned previously, it has 90 categories, from minute 1 to minute 90, with the intention of establishing post-observation time intervals, due to the fact that in training football the duration of matches depends on the age category. The criterion “scoreboard status before the goal” allows us to analyse the goals in terms of the result, looking at whether the goals scored serve to widen the lead on the scoreboard or are decisive in drawing or winning the match (Fernández-Hermógenes et al., 2017).

In terms of the applicability of the tool, several options can be found in two different ways: research and competition. At the research level, one could differentiate how goals are scored or conceded using some of the criteria that make up the observational tool as independent variables, as well as to compare goals from different leagues or sport contexts. On a competitive level, one could analyse and compare the goals scored by a club’s teams to establish differences and similarities between them, as well as compare the goals scored and conceded by one’s own team with respect to teams in its league. 

The coding tool is one that does not study the whole sequence of play. This is a limitation when it comes to going into detail on the action of the game, although it is true that it is an intentional fact that facilitates the recording of the goals in an easy way.

In terms of future prospects, the tool could include new criteria, with their respective category systems, which could support more in-depth analyses that would allow for other possible comparisons to be made. 


As a conclusion of the study, it is worth mentioning that the presented coding tool presents optimal validity and reliability values. This ensures its use in possible research projects or specific scientific studies; as well as by clubs, sports performance analysis departments and coaches in order to analyse and assess how goals are scored, improving their teaching and training processes.


The authors received no financial support for the research, authorship and/or publication of this article.


[1] Altman, D. (1991). Practical statistics for medical research (1st Edition). Chapman and Hall/CRC. New York.

[2] Amatria, M., Maneiro-Dios, R. & Anguera-Argilaga, M. T. (2019). Análisis del éxito de la Selección Española en la UEFA-Euro 2012. Apunts Educación Física y Deportes, 137.

[3] Anguera, M. T. & Blanco-Villaseñor, Á. (2006). ¿Cómo se lleva a cabo un registro observacional? Butilletí LaRecerca, 4.

[4] Anguera, M. T., Blanco-Villaseñor, Á., Hernández-Mendo, A. & Losada, J. L. (2011). Diseños observacionales: Ajuste y aplicación en psicología del deporte. Cuadernos de Psicología del Deporte, 11 (2), 63-76. Recuperado a partir de (Accessed 30 September 2021)

[5] Aranda, R., González-Ródenas, J., López-Bondia, I., Aranda-Malavés, R., Tudela-Desantes, A. & Anguera, M. T. (2019). "REOFUT" as an Observation Tool for Tactical Analysis on Offensive Performance in Soccer: Mixed Method Perspective. Frontiers in Psychology, 10.

[6] Bakeman, R. (1978). Untangling Streams of Behavior: Sequential Analysis of Observation Data. In G. P. Sackett (Ed.), Observing Behavior. Data collection and analysis methods (Vol. 2, pp. 63-78). Baltimore: University Park Press.

[7] Barreira, D., Garganta, J., Castellano, J. & Anguera, M. T. (2013). SoccerEye: A Software Solution to Observe and Record Behaviours in Sport Settings. The Open Sports Sciences Journal, 6(1), 47-55.

[8] Bulger, S. M. & Housner, L. D. (2007). Modified Delphi Investigation of Exercise Science in Physical Education Teacher Education. Journal of Teaching in Physical Education, 26, 57-80.

[9] Caro Muñoz, Ó. & Caro Muñoz, A. (2016). Aproximación a los modelos tácticos generales ofensivos mediante el análisis de los goles en fútbol profesional. Journal of sport and health research, 8(1), 1-12. (Accessed 14 October 2021)

[10] Castellano, J. (2000). Observación y análisis de juego en el fútbol. Tesis doctoral. Universidad del País Vasco.

[11] Chacón-Moscoso, S., Anguera, M. T., Sanduvete-Chaves, S., Losada, J. L., Lozano-Lozano, J. A. & Portell, M. (2019). Methodological quality checklist for studies based on observational methodology (MQCOM). Psicothema, 31(4).

[12] Cronbach, L. J., Gleser, G. C., Nanda, H. & Rajaratnam, N. (1972). The Dependability of Behavioral Measurements: Theory of Generalizability for Scores and Profiles. New York: J. W. and Sons.

[13] Echeazarra, I. (2014). Análisis de la respuesta física y del comportamiento motor en competición de futbolistas de categoría alevín, infantil y cadete. Tesis doctoral. Universidad del País Vasco.

[14] Fernandes, T., Camerino, O., Garganta, J., Pereira, R. & Barreira, D. (2019). Design and validation of an observational instrument for defence in soccer based on the Dynamical Systems Theory. International Journal of Sports Science and Coaching.

[15] Fernández-Hermógenes, D., Camerino, O. & García De Alcaraz, A. (2017). Acciones ofensivas a balón parado en el fútbol. Apunts Educacion Fisica y Deportes, 129.

[16] González, R., Aranda, R., Tudela, A., Sanz, E., Crespo, J. & Aranda, R. (2020). Past, present and future of goal scoring analysis in professional soccer. Retos, 37, 774-785.

[17] Hernández-Mendo, A., Blanco-Villaseñor, Á., Pastrana, J. L., Morales-Sánchez, V. & Ramos-Pérez, F. J. (2016). Aplicación informática para análisis de generalizabilidad. Revista Iberoamericana de Psicología del Ejercicio y el Deporte, 11(1),77-89.

[18] Hernández-Mendo, A. & Molina, M. (2002). Cómo usar la observación en la psicología del deporte: principios metodológicos. Efdeportes, 8(49).

[19] Hernández Nieto, H. (2002). Contributions to statistical analysis. Mérida: Universidad de Los Andes.

[20] Hughes, M. (1996). Notational analysis. En T. Reilly (Ed.), Science and Soccer (pp. 343-361). London: Routledge.

[21] Kubayi, A. (2020). Analysis of Goal Scoring Patterns in the 2018 FIFA World Cup. Journal of Human Kinetics, 71(1).

[22] Lapresa, D., Blanco, F., Amatria, M., Arana, J. & Teresa Anguera, M. (2020). Observational Analysis of the Execution of the “Control” Core Technical/Tactical Concept by Sergio Busquets. Apunts Educación Física y Deportes, 140.

[23] Martín Barrero, A., Marcos Gutierrez, I. & Falces Prieto, M. (2021). Analysis of the game model in a professional football team in the German First Division. Case study. Retos, 39, 628-634.

[24] Ortega-Toro, E., García-Angulo, A., Giménez-Egido, J. M., García-Angulo, F. J. & Palao, J. M. (2019). Design, Validation, and Reliability of an Observation Instrument for Technical and Tactical Actions of the Offense Phase in Soccer. Frontiers in Psychology, 10.

[25] Papadopoulos, S., Papadimitriou, K., Konstantinidou, X., Matsouka, O., Pafis, G. & Papadopoulos, D. (2021). Factors Leading to Goal Scoring in the Spanish and Italian Soccer Leagues. Sport Mont, 19(1), 13-18.

[26] Parlebas, P. (2001). Juegos, Deporte y Sociedad. Léxico de praxiología motriz. Barcelona: Paidotribo.

[27] Sánchez-López, R., Echeazarra, I. & Castellano, J. (2021). Validation of a Football Competence Observation System (FOCOS), Linked to Procedural Tactical Knowledge. Sustainability, 13(12), 6780.

[28] Sarmento, H., Anguera, T., Campaniço, J. & Leitão, J. (2010). Development and validation of a notational system to study the offensive process in football. Medicina, 46(6).

[29] Tenga, A., Holme, I., Ronglan, L. T. & Bahr, R. (2010). Effect of playing tactics on achieving score-box possessions in a random series of team possessions from Norwegian professional soccer matches. Journal of Sports Sciences, 28(3), 245-255.

[30] Ugalde-Ramírez, A. & Rodríguez-Porras, L. (2021). Análisis de los patrones de anotación y su asociación con el resultado del partido en la Copa Oro 2019. SPORT TK-Revista EuroAmericana de Ciencias del Deporte, 10(2).

[31] Usabiaga, O., Castellano, J., Blanco-Villaseñor, Á. & Casamichana, D. (2013). La Teoría de la Generalizabilidad en las primeras fases del método observacional aplicado en el Ámbito de la iniciación deportiva: calidad del dato y estimación de la muestra. Revista de Psicología del Deporte, 22(1). 103-109.

ISSN: 2014-0983

Received: May 3, 2022

Accepted: July 25, 2022

Published: January 1, 2023