Football or soccer (in this paper the term ‘football’ is used) is the most popular sports in the world. According to the “Big Count” study of FIFA [1] there are 270 million people involved in the match (players and referees). Moreover, football attracts millions of spectators around the world. For example, the global TV audience that followed the 2015 UEFA Champion’s League final between FC Barcelona and Juventus Turin was estimated to be 180 million people from more than 200 territories [2]. Due to its high popularity, football stands out among sports and games. In contrast

to games such as basketball or handball, football is a low scoring game, and scoring a goal is usually a rare event. For this reason, the final match score does not provide a clear picture of the teams’ technical and physical performances. To understand success factors in football, various other performance indicators next to goals scored have to be considered. Football is also a sport which has elements of chance but nevertheless, this does not mean successful teams are just luckier than others [3, 4].

To identify the factors which lead to success in football it is necessary to find performance indicators which significantly discriminate winners and losers. However, the identification of critical factors for successful performance poses a major challenge [5]. In 1912, Fullerton did the first work in this area of performance analysis for baseball [6]. In football, Reilly and Thomas [7] performed one of the first systematic notational analyses. They used hand notation and audio tapes to analyze in detail the movements of English First Division football players [8], and found out, inter alia, that a player is usually in touch with the ball for only two percent of the time. In another early performance analysis, Reep and Benjamin [9] developed a new approach to study 3,213 matches in England between 1953 and 1968 using frequency distributions. Their analysis revealed that about 80 percent of all goals are scored after three or fewer passes and about 10 shots are needed for one goal.

A milestone for science and football was the first World Congress of Science and Football which was held in Liverpool in 1987 [5]. Various themes were discussed such as team management, computer-aided performance analysis and decision-making by referees [10]. In the following years, the numbers of research papers concerning football and performance analysis increased steadily [11-15]. Hughes and Bartlett [16] reviewed and analyzed research on performance indicators in sports and defined a performance indicator as “… a selection, or combination of action variables that aims to define some or all aspects of a performance. Clearly, to be useful, performance indicators should relate to successful performance or outcome” (p. 739). Researchers also monitored match structures, summarized some performance indicators and utilized them (e.g., numbers of shots, passes, dribbles or ball possession) in various subsequent papers which provided more insight into possible success factors in football [6, 17].

In the context of this paper, two review studies regarding performance analysis in football are noteworthy. Mackenzie and Cushion [18] critically reviewed 60 articles (articles published up to 2010) with a focus on methodological approaches and concluded that there is an overemphasis of research on predictive and performance controlling variables (e.g., location, shots). They suggested an alternative approach that focuses on research that investigates athlete and coach learning to enhance our understanding of football performance. However, these factors cannot readily be operationalized as success factors. Sarmento, Marcelino, Anguera, Campanico, Matos and Leitao [19] systematically reviewed 53 articles (articles published up to 2011) with a focus on major research topics and methodologies. They concluded that most studies used a comparative analysis to analyze differences between players or teams. Unlike Mackenzie and Cushion, they identified a lack of predictive studies. While it was not the main focus of their research, they also identified some success factors for a team such as the number of shots and shots on goal. They concluded that match location, quality of the opposition, match status and match half seem to have a greater importance for success due to a large number of studies that focused on these aspects.

Both aforementioned reviews comprised a wide variety of possible outcomes in the included articles, such as physical conditions or contextual variables. In this study, we focus solely on predictive or comparative studies that considered success as an outcome (win/loss, league ranking, etc.). This allows a clear identification of the critical factors for success. Moreover, this review also considers studies published after 2011, addressing a large and growing body of recent research that has not been covered in previous reviews, and enables an assessment of the current state of the art.1 Not only has the amount of the articles related to performance analysis in football grown substantially since 2011, also various new methodological approaches have been utilized. For example, Grund [20] introduced network analysis into the research about success factors and Collet [21] revealed new insights into the effect of ball possession using an ordered-logit regression. Liu, Gomez, Lago-Penas and Sampaio [22] used a k-means cluster analysis and a cumulative logistic regression to reveal the factors that differentiate the between winning and losing teams. Overall, the aim of this study is to provide a systematic review of the available literature on performance analysis in elite male football concerning methodologies and results to find out critical factors for success in football and to provide guidance for future research².

The systematic review of performance indicators in elite men’s football was done in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) statement [23]. The last search was conducted on June 24^th, 2017.

To search for relevant publications and ensure the quality of the articles, the following databases were utilized: Web of Science (the modules “Core” and “Medline”), Scopus and PubMed. Articles that were published in 2016 or before and in English were considered. The search strategy comprised search terms that combined one of two primary keywords (soccer OR football) with a second keyword (e.g., success, win, loss) using the Boolean operator and all utilized search terms are presented in Table 1.

For inclusion, the articles had to meet the following criteria:

• The data had to deal with performance analysis in football.
• The variables of interest were linked to success (win/loss, goals, continuance in league/tournament, league ranking and points won).
• Adult elite football was investigated.
• The study was written in English.
• The study was published in an academic journal.
• The study design was comparative or predictive or focused on home advantage in football.

It should be noted that we included studies on home advantage in this review as a separate category besides comparative and predictive studies utilizing inferential statistics. Although most of the studies on home advantage used a descriptive approach to reveal the influence of home advantage, we considered these non-inferential studies because home advantage is one of the most investigated variables regarding success factors [18].

The initial search revealed 19,161 articles (Web of Science [Core and Medline]: 9,706; Scopus: 6,038; PubMed: 3,417). After excluding the duplicates 10,833 articles remained. The articles were screened based on an assessment of both the title and the abstract. All articles without a focus on the investigation and analysis of data on the conditions of competition results in elite adult football were excluded. In total, 185 articles were relevant for this review. These articles were read in detail and assessed for relevance and quality. Articles which did not meet the criteria were excluded. After this step, 53 articles remained. Subsequently, the literature references of these 53 articles were screened for more articles meeting the criteria. Fifteen additional articles were identified. Finally, 68 articles were included in the review (Fig. 1).

Then, the articles that met the inclusion criteria were indexed, and each article was summarized. The summaries comprised the study purpose and design, methods of data collection and analysis, and key findings. This enables an overview and comparison of the articles and allows an assessment of the current state of research on performance indicators in football.

Fig. (1). Flow diagram of this systematic review [23].

The identified articles were published between 1986 and 2016, covering a time span of 31 years. More than half of the articles (exact 61.8%; 42 articles) were published within the last seven years (2010-2016) of the searched time period, indicating that this field of research has recently gained momentum.

To organize the identified analyses, the articles were categorized following a system used by Sarmento et al. [19] and Marcelino, Mesquita, and Sampaio [24]. In the first step, the articles were assigned to [25, 26] comparative, [27] predictive or Home Advantage (HA) analyses [28]. In the second step, articles were assigned to one of the three types of analyses from above according to the different operationalization of success (i.e., win/loss, goals, continuance in league/tournament, league ranking and points won) (Table 2).

Of the articles, 30 were predictive analyses, 22 were comparative analyses, and 20 focused on the analysis of home advantage. One of the articles [15] covers both types of analyses (predictive and comparative). In total, 21 articles over all three types of analysis utilized “win/loss” as the success variable. “Goal difference” was used by seven articles, “goals” by eight, “league/tournament ranking” by 12, “points” by four and “continuance in league/tournament” by two.

In the following section, methods and major results of the identified articles will be presented within the three different categories of type of analysis. Finally, all findings will be summarized and the most frequent and significant variables regarding success factors in football will be discussed.

In seven of the 21 comparative analyses, researchers compared wins and losses. In three of the seven papers draws were also included, and in one instance the percentage of wins was considered alongside wins and losses (Table 3). In the three papers that compared only wins and losses [29-31] the authors tried to find variables that explain differences between winners and losers. Broich et al. [29] identified goal efficiency (number of goals divided by the number of shots), shots, passes and ball contacts as the most important team parameter for winning. Efficiency was also analyzed by Szwarc [31]. He showed that players of winning teams are more efficient than their opponents. As a result of the small sample (seven matches) only shots on goal (p<0.05) and shots defended by a goalkeeper (p<0.01) differed significantly between winners and losers. Kapidizic et al. [30] did not analyze efficiency but they also found that the numbers of shots within 16 meters (p<0.05) and accurate passes (p<0.01) are significant indicators for winning teams at the European Championship in 2008. Winners also scored more goals than losing teams in the Championship. Three more papers investigated the differences between wins, losses and draws [27, 32, 33]. These studies reported various significant differences between winning, drawing and losing teams. Winners have more entries into the penalty area (p<0.01) [33], more successful attacks (p=0.003) and passes (p=0.015) as well as a higher ball possession rate (p=0.001) [32]. Armatas et al. [27] revealed that 71.4 percent of teams that scored the first goal subsequently won the match (p<0.05). In contrast to the other studies, one study focused on the total winning percentage [34]. Another difference is the use of group cohesion as the independent variable. The authors showed a statistically significant relationship between individual attraction to the group-task and performance with a very high effect size of 1.94 (p<0.05). The higher the positive feelings of each group member to the group-task, that is, to play football successfully, the higher were the likelihood of winning.

In nine of the articles, the authors compared teams with different positions in the league/tournament ranking (Table 4). Luhtanen, Belinskij, Häyrinen and Vänttinen [35] investigated the influence of offensive and defensive variables on the final ranking of the European Championships in 1996 and 2000. In 1996, interceptions and the success rate of all defensive actions showed a significant correlation (p<0.05) with the final ranking. In 2000, significant correlations with the ranking were found for success rate in passes (p<0.05) and attempts (p<0.05) on goal. In the other papers, different football leagues were investigated and it was shown that better-ranked teams (top-teams) need less shots for a goal than worse ranked teams [15, 36, 37]. This parameter corresponds to Broich et al.’s [29] ‘goal efficiency’. It was also found that top teams have more successful attacks, complete their offensive attacks more frequently between zero and 11 meters in front of the goal [38], have more successful passes [15, 38, 39], score more goals [36, 37, 40], perform more crosses [15, 40], have more ball possession [37, 39], shoot more often on the goal [37, 39], have more assists [36, 37, 39] and take more shots [15, 37, 39, 40]. The best teams in the league also perform fewer fouls [15] and allow fewer shots and crosses [40]. The worst ranked teams have fewer counter attacks, have less possession with zero to four passes and have less possession longer than 12 seconds [41]. Worse teams also have more very high-intensity running, high-intensity running and total distance covered [39]. Better teams cover more total distance with the ball and very high-intensity running with the ball [39]. Furthermore, the top teams show a faster recovering (recapture is 1.3 to 1.7 seconds faster than mean times) of ball possession [42]. Obviously, top teams score more goals per match [15, 36, 37, 40]. The cited studies showed that a lot of factors influence success (operationalized as league ranking) in football. Overall, it appears that goal efficiency, passes and shots are the most important factors in this research area.

Six more studies used a comparative approach to investigate success factors operationalized differently to the articles discussed previously (Table 5). Two papers focused on goal difference [43, 44]. Bekris et al. [43] compared matches with one-goal differences (short range results) as well as matches with three-goal differences or more (wide range results). Their analysis showed that winners in wide range results have more ball possession, perform more passes, win more duels (overall and aerial), and have more shots, shots on target and a higher shot accuracy. In the short range results these differences were not found. A winner-winner comparison showed that wide range winners perform more passes, have a higher pass accuracy, more short distance shots and shots on-target. Yue et al. [44] used a similar approach. They analyzed matches with a difference of two or more goals and matches with a difference of three or more goals. Goal efficiency, shots, passes and ball contacts were found to be the most important factors for scoring a goal (in this order). Clemente [45] and Delgado-Bordonau, Domenech-Monforte, Guzmán & Mendez-Villanueva [46] operationalized success as a continuance in a tournament. They compared teams with a different number of matches respectively teams that got to the semifinal. Both analyzed matches of the World Cup 2010. Clemente [45] revealed that teams with more matches in a tournament (the successful ones) score more goals through open play, have more shots inside the penalty area and perform more passes. Delgado-Bordonau et al. [46] showed that successful teams perform more shots on-target, have a higher efficiency and concede fewer shots. They also revealed that the first goal in the match leads to a victory for 66.7 percent in the group stage and for 81.3 percent in the knockout stage. Hughes and Franks [17] used a new and different approach to analyze football. They normalized the data into “goals/shots per 1000 possessions” to analyze the relative importance of ball possession. The authors used this parameter to compare successful teams (getting to the quarterfinals) and unsuccessful teams (first round losers) in the 1990 World Cup. Accordingly, successful teams show a strong trend to be better in converting possession into shots on goal (no significant difference). For ball possessions with more than eight passes, there is a significantly higher chance for successful teams to create a shooting opportunity (p<0.05). In contrast, the necessary shots for a goal increase with more passes per possession [17]. Hoppe, Slomka, Baumgart, Weber & Freiwald [47] used the final points accumulated by each team during one season in the German Bundesliga. They analyzed the running performance with and without ball possession of the teams. Only total distance with ball possession was a significant predictor for final points (p<0.01). They concluded that not only running performance is important for success, but rather the relation to technical/tactical skill regarding ball possession [47].

Fourteen of the predictive analyses focused on differences between wins, draws and losses (two of these papers considered two groups: winners and non-winners) (Table 6). Four of these papers used a discriminant analysis to reveal the most discriminating factors [48-51]. Shots on goal was a discriminant factor in all four studies. Crosses, match location and ball possession [48, 49] as well as the quality of the opponent (similar to strength or team ability) [49] were other identified factors. Collet [21] and Harrop and Nevill [52] used a regression analysis/model and showed that higher pass accuracy is a good predictor for success. More shots, fewer passes, fewer dribbling and match location are further predictors [52]. Collet [21] investigated the influence of possession on success and showed that possession is not as relevant as assumed. If the strength of a team is controlled, the influence of possession on success will range from -5.7% (in German Bundesliga; significant (p<0.05)) to +1.8% (all national teams; not significant). The fact that possession has a potential negative link to success may be worth further examination. Efficiency measures seem to be better predictors for success [21, 29, 31, 44, 46]. Liu, Gomez, Lago-Penas and Sampaio [22], Liu, Hopkins and Gomez [53] and Mao, Peng, Liu and Gomez [54] used cumulative logistic-regression in a generalized linear model. They also divided the sample into close matches and unbalanced matches (a cluster analysis based on the goal difference was used) with a cluster analysis and cut-off values. In past research it appeared to be more likely in close matches that both teams play at their best [22, 55]. They showed that shots on goal, shot accuracy, tackles and aerial advantage have positive effects on winning [22, 54]. Liu et al. [53] also investigated the within-team effects (changes in team values between matches) and between-team effects (differences between average team values over all matches). Shots on target and total shots have positive within-team effects on winning. Game location showed a small positive within-team effect. Ball possession showed a small negative within-team effect but also a small positive between-team effect. Within-team effects varied depending on strength of team and opponent [53].

Gómez, Gómez-Lopez, Lago and Sampaio [56] used a factor analysis with several factors and the zone of the pitch. For the zone of the pitch they divided the field into five zones from goal to goal and into three to five subzones in each of these zones. They identified four factors. All factors are highest for winners. The best discrimination is given for ball recovery in zone two (2.1, 2.2 and 2.3) (penalty zone to center circle) and offensive actions with long passing sequences in zone 5.1 (six-yard box) and 5.2 (within penalty zone). Bar-Eli, Tenenbaum and Geister [57] and Mechtel et al. [25] investigated the impact of a player’s dismissal. Both found out that a sending-off decreases (sanctioned team) respectively increases (opponent) the chance of winning. Mechtel et al. [25] also identified strength (points earned in the last three seasons) and home advantage as success factors. Torgler [58] applied an economic win function to determine the influences on winning or not winning during the FIFA World Cup 2002. He showed that a higher number of shots on goal leads to a higher probability to win. He also revealed the negative effect of a player’s dismissal. Hosting the tournament was a strong advantage as well. It increases the chance of winning by 45 percentage points [58]. Hanau, Wicker and Soebbing [59] investigated the difference between the expected outcome of a football match and the actual outcome. They found out that the actual outcome is determined by the standing in the last season and home advantage.

The second most frequent kind of predictive analyses are studies that used goal scoring as the indicator of success (Table 7). Pollard and Reep [12] developed a quantitative variable, called the ‘yield’, defined as the probability of a goal being scored minus the probability of one being conceded. The yield for the penalty area as starting zone of ball possession and open play is 78.3 (per 1000 possessions you can expect 78.3 more goals scored than goals conceded). They also found that open play always has a higher yield than set play [12]. Carmichael and Thomas [26] established a match-based production function. They found that shots on goal, shots that hit woodwork, tackles, own goals and free kicks are significant predictive factors (p<0.05) for the home teams. Kapidžić, Bećirović and Imamović [60] also identified shots on goal as a significant predictor for goal scoring (p=0.027). Wright, Atkins, Polman, Jones and Sargeson [61] postulated position of attempt, goal keepers’ position and type of shot as the three predictors for goal scoring. Tenga, Holme, Ronglan and Bahr [62] and Tenga, Ronglan and Bahr [63] used the same data set with different methods for their analysis. Both papers showed that counter attacks are more effective than elaborated attacks in producing goals. Grund [20] used a network analysis to identify success factors. He revealed that networks with high intensity and low centralization have a better performance. An increased passing rate lead to a better performance in this study [20].

In the last group of predictive analyses three variables of interest were collected (Table 8). The most frequent variable is goal difference as utilized in five papers [13, 25, 64, 65]. In all articles match location is positively linked to goal difference. Quality of the opponent was also identified as a significant predictor (p<0.05) [25, 64, 65]. Moreover, Carmichael et al. [13] showed that passes, tackles, interceptions, clearances, blocks, interceptions, free kicks and ball caught by goalkeeper are significant predictors for a positive goal difference(p<0.05). A red card was associated with a negative goal difference [13, 25, 64]. Garcia-Rubio et al. [65] showed that scoring first is the strongest predictor for a positive goal difference. Lago-Penas, Gomez-Ruano, Megias-Navarro and Pollard [28] used a tree analysis to determine the effects of scoring first on the outcome of a match. They showed that the first scoring team scored 1.88 goals more than their opponent on average. This is influenced by the quality of the teams and the match period in which the first goal was scored [28]. Oberstone [15], Hall, Szymanski and Zimbalist [66], and Kringstad and Olsen [67] investigated relevant factors for the league ranking in a predictive design. Hall et al. [66] focused on the relationship between payroll and performance. They found that there is a higher winning probability of 0.614 for 50% more spending in payroll. The top level is more sensitive to spending. Oberstone [15] developed a regression model to predict the league ranking. He revealed six variables which are sufficient for predicting the league ranking (in terms of points earned). These six variables are the percentage of goals to shot (goals divided by shots), the percentage of goals outside penalty area (goals from outside penalty area divided by goals within penalty area), ratio of short to long passes, total crosses, average goals conceded per match and yellow cards. Kringstad and Olsen [67] studied budgeted revenue and success. They showed that budgeted revenues are a significant factor (p<0.05) but only for the bottom-half of the teams and not for the top-half of the teams. The remaining three papers focused on points as the variable of interest. Lago [68] defined performance as shots performed minus shots conceded, and found that this is a predictor for more points. Furthermore, he showed that the higher the FIFA ranking is, the higher the chance to win. Collet [21] focused on ball possession. His result was that more time with the ball leads to more points and goals, but if it is controlled by team strength a negative effect for possession can be observed. Passes and shot accuracy turned out to be better predictors for points. Coates, Frick and Jewell [69] investigated the relationship between salary structure and success. They revealed that salary inequality has a negative effect on success but the wage bill of a team has a positive relationship with success by a similar amount. This results support the cohesion theory [69].

The aim of this work was to review research in performance analysis relating to success factors in elite men’s football. In total, 68 articles were identified and clustered based on their study design with regard to comparative, predictive or home advantage analyses. It was found that the most influential variables are efficiency, shots on goal, ball possession, pass accuracy/successful passes, as well as the quality of opponent and match location. New statistical approaches, such as discriminant analysis, factor analysis, regression analysis and magnitude-based inferences reveal interactions between these variables.

Concerning study design, an increase of predictive studies was found. For future studies, we suggest considering more often one of the ‘Big 3’ leagues (Spain, England and Germany) or all of them to get more representative samples. Furthermore, the consideration of other influences on success such as psychological factors and/or weather conditions would be of interest. Additionally, new methodological ways of analyzing success factors in football could be beneficial. For example, Borrie, Johnson and Magnusson [96] presented a method to investigate time-based events in sports. Moreover, more advanced statistical methods should be applied to ensure a broader insight into the mechanisms of performance such as regressions and magnitude-based inferences [21, 22, 25].

Most of the studies did not consider the influence of contextual (e.g., home advantage, quality of opponent) and interactional variables (e.g., first goal scored by time of goal scoring). In some studies, the influence of variables is also computed without a clear definition of the investigated variables. This lack of operational definitions poses a problem and, inter alia, does not allow valid comparisons between the studies. In future research, variables should be clearly defined to enable comparable and reproducible results (see also Mackenzie & Cushion [18]; Sarmento et al. [19]). The consideration of interacting variables such as quality of opponent and match location should also be considered in future investigations to provide more insights. Future study designs should also make sure to take the differences between different competitions (e.g. leagues, cup competitions) into account, especially the differences between a league match and a knockout match.

Moreover, we found very different approaches regarding the sample size required for generalization. Sample sizes of considered matches varied between very low numbers and thousands of matches. A small sample size is clearly a limitation in some of the reviewed papers, resulting in no generalizability. Studies investigating league matches should consider at least a sample size of one season. Hence, our review supports the finding of Mackenzie and Cushion [18] with regard to small sample sizes that remains a major deficit of performance analyses in football. Additionally, future studies should use effect sizes to interpret the results properly (see also Broich et al. [29]). A last important aspect to consider when designing a study is the context of the analyzed sample. For example, the tactic that is used (e.g., counterattacks vs. elaborate attacks) could vary regarding the opponent.

Based on the idea that performance is a consequence of prior learning, inherent skills, situational factors and influence of the opposition [97], the assumption holds that future performance is to a large extent a consequence of previous performance. Again, this underlines the aforementioned importance of considering the context of a sample as well as the operational definition of the investigated variables. Prior learning and inherent skills are two variables that were not considered in research about success factors in football as defined in this review. Both are exciting new possibilities for future research.

Finally, we would like to point to two methodological approaches that might lead to new insights in analyzing football performance. First, social network analysis provides new methods to analyze different aspects utilizing relational data, (e.g., the passing network of football teams), that have the potential to contribute substantially to a better understanding of success [20, 98, 99]. Second, psychological factors could be taken into account for future research (e.g., reversal theory, see Apter [100]). The investigation of psychological factors is, in fact, more difficult than the analysis of statistical data. The operationalization of cohesion found in this review [34] is a good example of the use of psychological concepts.⁶

As this review has shown, generalizable knowledge about success factors in football can be a helpful resource for coaches to gain a better understanding of the match. While significant progress in the field of performance in football was made in the last years, the review identified various deficits that future research has to address to provide more valuable information about what determines success.