Spectator Demand in Russian Premier League

Размещено на http://www.allbest.ru/

Размещено на http://www.allbest.ru/

Spectator Demand in Russian Premier League

• Abstract
• demand football club premier league
• Spectator demand for premier league's football matches in Russia is much lower comparing to top-10 European regular leagues. This thesis estimates the determinants, which influence the attendance, basing of the data from Russian Premier League's games. The choice of the determinants stems from the prior studies and availability of the data. The Ordinary Least Squares regression is used for the empirical part, with the stadium occupancy at match-by-match basis as the dependent variable. One thesis's hypothesis is confirmed, one is rejected, one is partially confirmed. The practical significance is the recommendations for managers of Russian football, basing on the aspects which are found significant in influencing the attendance. The scientific contribution of this study is in the demand-based study for the market of Russian football, which is not researched properly, focusing on the on-field performance of the football clubs. The research question is what determinants influence Russian Premier League matches' spectator demand the most significantly?
• Keywords: Russian Premier League, attendance, spectator demand, on-field performance

Introduction

Soccer nowadays is the world most popular sports, as, for example, the statistics of the 2018 FIFA World Cup shows: the overall audience of the tournament was more than 3.5 billion people. (FIFA.com., n.d.). However, still just a small number of football clubs experience the excessing of the demand over the supply in the regular domestic leagues, as it was during the World Cup, e.g. Bundesliga's Bayern Munich and Borussia Dortmund usually have sold outs for almost all matches during the season (Global attendances, n.d.). Simultaneously, the majority of the leagues experience the vice versa situation and the stadiums could remain incompletely full even for top league matches. Moreover, teams, which are the members of a league do not earn enough from attendance, like it often happens in the Russian Premier League. (Russian Premier Liga Statistics, n.d.; PwC, 2018a) That is the reason why it is fundamental to ascertain the determinants influencing the attendance, to be able to influence the increase of attendance.

From the very beginning, the domestic Russian league should be presented. Russian Premier League, which was called “Russian Football Premier League” until season 2018/2019, is a domestic premier league of the Russian championship. The league is holding the 6^th place in UEFA ranking of top world leagues. (UEFA.com, n.d.) Nowadays, the league is suffering from the problem of low attendance for the majority of matches. (Russian Premier Liga Statistics, n.d.)

According to PwC (2018a), in Russian Premier League, the income from gate attendance is just 4.9% out of the overall income of the RPL clubs., whereas the average index of this value in top-10 European championships is 15.5% (excluding Russia). It means that even the leagues which are lower in the ranking than Russian league usually earn more from attendance. This paper will contribute to the discerning of the determinants of the spectator demand, which specifically are applicable to the Russian market. So then, this research's results could be used by clubs as a basis for developing the new ticket sales strategies.

The motivation to research this field is to make Russian football clubs more competitive in the international arena, to make them less dependable on the sponsorship donations, to contribute to the development of Russian football. The way to reach these goals is to contribute to an increase in Russian Premier League's matches' attendance, which will increase the overall incomes of Russian clubs and increase the share of matchday revenue in these incomes structure. Moreover, attendance rates are directly dependent on the broadcasting licensing and sponsorship and other commercial income (Villar & Guerrero, 2009). These two income categories should be paid extra attention as the two most beneficial ones in top-10 European championships, excluding Russia, meaning, again, that these indexes should be improved. (PwC, 2018a) Suchwise, in general, the paper is done for contributing to the development of the market of Russian professional football.

In fact, the topic is noticeably popular among researchers. The prior studies have conducted analysis in order to understand whether this or that factor influences the attendance, developed models to link them and to define how the attendance influences the sport's economics. This study is contributing to the investigation of the Russian league, which is highly under-researched.

The main aim of this paper is to create a basis of understanding which of the determinants, defined in the prior studies, are significant for Russian Premier League's matches' attendance changes by the exploration of the dependence of the attendance on these determinants. This basis would serve for recommendation creation and could possibly be useful for future researches. Hence, the research question is formulated as what determinants influence Russian Premier League matches' spectator demand the most significantly?

To answer the research question, the research objectives are defined as follows:

- Study the current Russian Premier League matches' attendance situation

- Define the factors, influencing the attendance rates

- Create the recommendations for RPL clubs/ management of the league/ Russian Football Union how to increase attendance at league matches.

These objectives are achieved through critical analysis of the literature of the topic and data collection, processing and analyzing, which let find correlations and interdependences. Generally, the objectives are achieved by performing the tasks of the thesis, which are:

1. Examine literature of the study to explore the most significant determinants influencing the spectator demand defined by the prior studies

2. Collect and process the data on Russian Premier League attendance for 6 seasons, from 2013/ 2014 to 2018/2019

3. Analyze the data using the ordinary least squares regression analysis with the aim of defining correlations of different factors with the attendance

4. Basing on the results of the analysis conducted, define the most significant determinants, influencing the attendance in RPL

To conduct the analysis, the literature on the topic is to be reviewed and analyzed to determine the main variables defined in the previous researches. Then the empirical analysis should be provided to contribute to the understanding of what determinants specifically are important in dealing with attendance increase, which could consequently provide matchday revenue boost. (Szymanski & Smith, 1997; Borland & MacDonald, 2003).

The methodology of this study is as follows: the data is collected using the script written in the Jupyter Notebook in Python programming language. The processing of the data is also performed in this software. The ordinary least squares regression is used as the main method of data analysis, considering the stadium occupancy at game-level as the dependent variable. The analysis is provided in the same software, as the collection and processing. The packages used to perform the tasks are: “BeautifulSoup”, “Request”, “Pandas”, “StatsModel”. The data is scraped from the website WhoScored, the additional source of data is the official website of the Russian Premier League, which is generally used to fill the gaps in data and to gather data on seasonal basis attendance.

This paper lies into the sports economics theory, carrying out the demand-based study analyzing the determinants of demand for tickets in the Russian Premier League. The research is limited with determinants which can be researched, the global factors like political and economic situation are not considered, the methodology is limited with possibilities provided with the chosen tools.

This thesis's results are relevant for specific RPL team's managers as well as for the league's management in general, providing them with recommendations which could positively influence attendance if followed. Russian Football Union, as controlling institute is also the beneficiary of the thesis's results. Since the main objectives of teams' managers are either to maximize profits or to maximize the percentage of winning of a team seasonally, this paper is to satisfy both of them. Moreover, this thesis's results could become the standpoint for future researches in this field in Russia, in case that the most of the researches are conducted for European championships and just a few papers are done for Russian league, investigating the brand strength and the uncertainty of outcome influence.

The structure of the thesis is as follows: the introduction inducts in the research and presents the main standpoints of the study, then the theoretical foundation discloses the previous papers' main insights, further the research question is stated and discussed and the methodology of the research is specified, the results section interprets the main outcomes from the empirical analysis of the paper and the recommendation section follows, conclusion discusses the results of the study, references list finishes. Each section has its own conclusion.

1. Theoretical foundation

The demand for attending football matches is an extremely important topic to be investigated. The quantitative value of the spectator demand has direct dependence with the income that a football club is able to get from its home matches. (Buraimo, 2008) Therefore, the researchers have always been interested in how matches' attendance demand (spectator demand) is dependent on different factors, which can influence the intention of fans to attend football matches. Furthermore, the realization of different correlation trends can bring benefits not only to researches themselves and contribute to the scientific field, but it may also have practical relevance for leagues, clubs, and industry of football in general. This practical relevance is not just about incomes of clubs from matchday but it also is about teams' performance, marketing issues and the great number of other factors since they are interdependent (Dobson & Goddard, 2001; Borland & MacDonald, 2003; Ro?ca, 2018). However, a spectator demand is a preferable topic for researching basically from the economic perspective because it is one of the variables to generate clubs' revenue and to specify, and it used to be one of the most considerable ones from the income perspective during the previous century (Szymanski & Smith, 1997).

Rottenberg (1956) and Neale (1964) were first to conduct such researches in the field of professional sports' market regulation, they defined the essential basis of its functioning and described them. Rottenberg (1956) was the pioneer to characterize demand for sports events, investigating the issue of quality of a play, taking demand, preferences, income, and prices as usual determinants in the research. Rottenberg (1956) was also the first to introduce an uncertainty of outcome hypothesis as he hypothesized that fans are more interested in watching equally matched teams playing than teams with different level of quality. Neale (1964) contributed to the new findings of standing in which the teams are moving up and down. The author concluded that instability in standings is increasing fans interest to attend games, which is also correlating with the uncertainty. Thus, these two papers served as a foundation for further demand-based researches.

The authors of ulterior studies had been endeavoring to link together all the variables and factors influencing the demand on attending matches, and subsequently to define how the demand in its turn influences other indicators. Some papers were focused on the particular factor, e.g. European Cup performance, promotion/ relegation to a tier league (Forrest & Simmons, 2006), clubs' achievements, family and socialization needs (Jere & Mathidza, 2014), self-esteem etc. (Garcнa & Rodrнguez, 2002; Villar & Guerrero, 2009), but the majority of papers' authors took several factors into account. Since there is a great number of determinants, the division is essential to be used. Szymanski (2003) stated the factors, most influencing the demand as the quality of the team and quality of the opponent, but these factors are forming only one group, according to Villar and Guerrero (2009). The author defines four basic groups of determinants which influence the demand:

- economic aspects

- expected quality

- the uncertainty of outcome

- opportunity costs and other factors.

This division is a standpoint for an examination of variables in this thesis. As it is shown below, all these groups are closely interconnected but still, such division is convenient to separate different determinants from each other and refer them to some categories, that is why this division is used. The description of each category is following.

1.1 Economic aspects

Surely, it is essential to consider the difference in economies. Since the thesis is limited with investigating only the determinants a club/league can influence in changing, no economic factors are included in the analysis. Still, the economic situation should be taken into account in order to apply the most significant determinants for a specific league to keep the relevance of the findings. Intertwining developed and developing football market and economies, Buraimo, Tena and de la Piedra (2018) define four categories of markets and suggest that each market would experience its own conditions. Since Russia is a country with a developed economy (Medvedev, 2015) and developing football markets (UEFA.com, n.d.; Parshakov, 2017), all the researches' results conducted for Spanish, English, Italian, German and French leagues could not be completely applicable for Russian Premier League because these countries are with developed markets of sports and there are a lot of differences in developed and developing sports markets. Still, the researches of these markets could be considered in this thesis to define the possible determinants for empirical analysis.

As it is claimed by Downward, Dawson and Dejonghe (2009), the game or contest should be considered as a specific product which is to be sold to the fans (online streaming license or ticket). So, it is still arguable question whether football match is a normal or inferior good. Some studies claim that that the positive elasticity of demand for tickets proves that football is normal good (Feehan, Forrest & Simmons, 2003), some contradict and present proof of the inferiority of football matches. (Bird, 1982). Different results in these studies can be explained with different leagues to be under the investigation and this could be the reason because usually football matches in the developed football markets are considered normal good while in the developing - inferior. Still, the investigation of this factor is not essential for this thesis since the economic aspects are not considered.

1.2 Expected quality

Further, the quality of the match is to be taken into consideration. The basic formula to estimate the demand for team's matches is to relate ticket prices with the quality of a team, excluding general shifts in attendance affected by external factors like weather and a general decline in attendance (e.g. season or economic situation) (Szymanski & Smith, 1997; Demmert, 1973). Primarily, this approach is good to estimate demand approximately but surely it needs more determinants to be included in order to improve the accuracy of the measurement.

The competitiveness level between two teams playing also significantly impacts the attendance rates, e.g. high level of competitiveness between two close rivals positively influences the attendance of matches between these teams (as it is observed with Russian teams PFC CSKA Moscow and FC spartak Moscow) whereas it also lowers the intention of fans to attend other matches of these teams. (Downward, Dawson & Dejonghe, 2009)

Considering the strength of the brand name of a football club is very important and actually, the expected quality of a game depends on the brand of home and away teams, playing a match. Fans of a home team tend to prefer watching their team playing with a rival that is possible to win the championship in theory, basing not only on the live standings but also on the brand of a team in general. This confirms the conclusion of Noll (2002) about the competitiveness level importance.

As Coates, Naidenova and Parshakov (2017) state in their study, the brand of a sports club is extremely significant to be investigated. These conclusions are explained from the perspective that deeper disclosure provides an opportunity to maximize financial indexes, not taking the on-field performance into account. Whether it is poor or great, it improves the attendance rates because of the fans' loyalty to the brand of a football club. The study's results affirm that attendance of the RPL matches is dependent on the brand of the opponent and, what is interesting, the strength of the brand of the visiting team is not changing during the season. Of course, it is affected by winning trophies, buying new players and hiring a new coach, but the period of such effect is longer than 5 years (the authors used data for such period) and practically, the history of a club makes sense. For example, the history of CSKA Moscow starts in 1911 and the brand of CSKA as a guest team will definitely increase attendance in the city a game takes place. Also, a team's on-field performance decreases would not significantly change the case if they are shorter than 5 years. All in all, as Parshakov (2017) concluded from their work, the brand of a club can potentially be more important in estimating determinants of attendance than the uncertainty of outcome. The brand of a club has an extremely significant influence on the intention to attend the stadium at matchday, which was proved by a number of studies.

1.3 The uncertainty of outcome

Still, the influence of uncertainty of outcome cannot be omitted since it is one of the most significant determinants estimating the demand for tickets.

The great number of studies conducted on this topic proves this statement.

The research of Rottenberg (1956) was the first work investigated the uncertainty of outcome effect's connection with demand on sports matches tickets, specifically using data on baseball matches.

The main illation was that high competition level between two teams provides better attendances. The conclusion of their study was the formulation of the uncertainty of outcome hypothesis (UOH), which describe the dependence between attendance and uncertainty, and this gave a new topic to be studied for the researchers. Furthermore, many authors tested UOH, proving and rejecting the hypothesis for different contests.

The paper of Parshakov (2017) is one of the first studies of the Russian market, which estimates the uncertainty effect. It is being a foundation for this thesis's uncertainty of outcome subsection because it completely discloses the topic specifically for the Russian Premier League. The authors test uncertainty of outcome hypotheses basing on the data from RPL matches attendance on a match-by-match basis.

Specifically, the authors reviewed more than a dozen studies researching the UOH, dividing them into two categories: those which are based on differences in winning percentages and league standings and whose authors are researching match outcome uncertainty basing on the probability of the home team winning. The conclusions of these studies are very different, even the same authors find different results in their works, the data from one league can be asymmetrically interpreted by various authors. Basing on the previous studies, the authors chose the vector of their research. The usage of data on a match-by-match basis is explained by the possibility to detect instant demand changes. Measuring the uncertainty through betting odds is decided to be used because it provides the opportunity to investigate facing teams in previous matches and this approach provides a more relevant description of uncertainty.

Usage of the uncertainty of outcome hypothesis analysis regarding the Russian Premier League is justified by three arguments: demand for tickets is not limited by the capacity of the stadiums, the concentration index is low due to there are just 6 teams in RPL competing for the trophy and, finally, low commercial development of Russian league comparing to world top sports leagues provides low importance of commercial factors and an increase in uncertainty of outcome importance. Anyway, the uncertainty of outcome should be considered in the estimation of the demand just as one of the determinants and it is improperly to use it as the only variable. There are some disadvantages of using the uncertainty as a determinant of attendance:

- Absence of difference between fans who buy tickets for particular games and season ticket-holders

- Uncertainty is not distinguished from an interest in the home team winning of fans by attendance data

The main inference of this paper is that UOH is not interpreting the fans intention to attend RPL matches, providing the U-shaped dependence which is explained by visiting the team brand. Generally, fans intention on visiting home matches depends more on visiting team than on home and on the uncertainty of outcome. Providing the analysis, the authors found some valuable results. Omitting the expected results like the positive significance of temperature and negative effect of precipitation on attendance, there were some arguably results like playing a match, not at home stadium has no significance as well as the distance between cities.

1.4 Other factors

The literature in the field of the study describes a variety of factors, influencing the spectator demand, for example, playing history of a team forms the trend of fans intention to attend home matches of a team, which takes a lot of time to be changed (Noll, 2002), then it is better to analyze as longer period as possible to omit this factor. Buraimo (2008) concluded that day of the week when a match is taken place as well as the presence and the number of star players in a team is also very significant factors to be considered when estimating the reasons why people are able and wish to go to the stadium. According to Buraimo, Forrest, and Simmons (2009), home city attributes are essential to be taken into account, e.g. better conditions of the home city, like megalopolis with developed infrastructure, will definitely have positive influence on the optimal quality of the team, which is, according to Noll (2002), will impact increase in attendance. Buraimo et al. (2009) reinvestigated the effect of star players presence in a team, while Jere and Mathidza (2014) in turn research the impact of players from national country percentage out of all the players. There is a great number of other factors which were investigated, so this section reviews most of them.

Considering free on-air TV-broadcasting of the match, it is a very arguable point whether it really has a significant influence on the stadium attendance. As it was already mentioned, free live broadcasting of a match can have a negative influence on attendance (Buraimo et al., 2009; Buraimo, 2008), while the study of Martins and Cro (2016) affirms that TV broadcasting has no significant influence on attendance.

One of the most popular topics in the demand-based studies on sports economics is estimating ticket price correlation with attendance. The majority of authors conclude that it is definitely a very significant variable. Moreover, the study of Villar and Guerrero (2009) states that a diversity of additional costs should be considered to estimate the real price of attending a football game, like transportation to the stadium or parking price, which actually makes the real price significantly higher due to low prices for tickets in Russian Premier League considering with other European championships.

Another important determinant is stadium capacity, but as Demmert (1973) stands for, capacity is extremely important if the sellouts or near-sellouts are often for the league, but less significant if not. Surely, the situation in the market of Russian football presents a very few numbers of sellout matches and only a few teams experience a high level of attendance during the whole season, e.g. FC Zenit Saint-Petersburg. Moreover, sellout matches are very seldom in RPL and usually, less than 5% of matches are really close to sellouts, but the recent situation of attendance boost of clubs, which home cities were holding the FIFA World Cup 2018 at new stadiums makes the conclusions about the insignificance of such variable unclear, since all the new stadiums' capacity exceeds the capacity of previous ones in several times. So, the factor of new stadiums should definitely be considered since the majority of the researchers, who investigated this determinant found a positive impact of this determinant on attendance. (Villar & Guerrero, 2009). There are 5 new stadiums which are now used by teams of Russian Premier League, so this determinant surely needs to be taken into account.

Furthermore, the authors consider the performance of the clubs in the international arena. For instance, Ro?ca (2018) concluded in their study that domestic leagues' matches' attendance could be positively influenced by the participation of a club in European international leagues, like the UEFA Europa League and the UEFA Champions League. Obviously, the positive impact of the winnings in such tournaments is confirmed, but even poor performance in such tournaments positively influenced the attendance rates' increase.

There also are several variables which could not be changed by a team or a league, e.g. scheduling of other leagues. The case of the Division 1, the second football league of the UK's sports market, presents that the league, in general, has to play one of the world largest number of matches per season but matches of this league are usually scheduled in midweek evenings because of loss in audience due to the UEFA Champions League/ the UEFA Europa League matches with English Premier League participants on the weekends (Noll, 2002). Thus, the scheduling of other tournaments, like the Russian Cup could be significant but the estimation of this determinant is not the point of this research.

Also, Arnold and Veth (2018) investigated the culture of Russian active supporters, naming Russian fans one of the most racist and xenophobic and, since the research took place right before the FIFA World Cup 2018, the authors supposed the problems of fans from different countries with local supporters to appear during the World Cup. So, it could be considered that the attendance rates decrease is possibly impacted by local supporters' culture in the Russian Premier League's matches. However, no problems were observed during the FIFA WC 2018 hosting in Russia. The suppositions of the authors have not been confirmed. Moreover, it is difficult to consider such a factor in the analysis because of the subjectivity of value judgments in the estimation of the level of the quality of supporting each team. According to the authors, the observed incidents, connected with assaults of local racist supporters; took place in different regions constantly, so this factor can be estimated equal for all teams of the Russian Premier League.

The choice of the determinants for this thesis's analysis stems from the availability of the data in the Russian Premier League. As it was mentioned above, the factors which are not influenced by the management of the league are not considered. Since the market is just developing and the attitude to data collection is not formed, a great number of variables are not considered due to the current situation that such kind of data is not collected.

1.5 Conclusion

Summarizing, since sports economics is an extremely important topic among researchers, a great number of studies has been conducted. The majority of them are made for developed sports markets with high demand for matches. Developing sports contests are poorly investigated, as well as the Russian Premier League. Since the results of the studies on developed markets are not applicable to developing, there is a gap in researches. Also, since the attendance and profits from matchday in Russian league are much lower comparing with leading European regular leagues, such study is highly necessary to contribute to the improvement of these numbers.

Precisely for RPL, the topic is highly under-researched since only two studies of Russian domestic league's spectator demand investigation are published: Coates, Naidenova and Parshakov (2017) research the impact of clubs' brand to attendance changes, Parshakov (2017) applied foreign researches experience of the uncertainty of outcome hypothesis on Russian Premier League, investigating the influence of uncertainty on attendance rates. These papers are very significant for the professional sports area in Russia, but still, they are not covering the whole gap. Current research is linking a significant quantity of determinants together, so fundamentally new correlations' and dependences' detection could possibly improve the current situation. This thesis is more focused on investigating the variables of teams' on-field performance, which is providing the academic novelty not just for the market of Russian professional sports but also contributes to the whole field of the study.

2. Research question statement

The problem of low attendance at sport competitions is being researched since the middle of the previous century. As the general trend shows, the problem usually refers to developing sports markets, while the developed, like the UK's, are rid of such problem. (Global attendances, n.d.) A large volume of literature made for the developed markets several decades ago can be used as a basis for this research, but still, the topic of the Russian football market, which has a large number of specific features, needs to be investigated to provide the relevant analysis. The absence of a systematic approach to influencing the low attendance rates in RPL lead to non-systematic attempts of some football clubs to attract visitors to the stadiums. Finally, such attempts showed its inefficiency, for example, lowering the price of tickets or even giving them for free has not changed the case, which is obvious from the data analysis.

The problem of low attendance is a very topical concerning Russian Premier League, especially now, after FIFA World Cup 2018 has taken place in Russia. First of all, the revenues, which are generated by football clubs in Russia from matchday attendance are extremely low. During the 2016/2017 season, all the clubs from Russian Premier League earned 1,955,908,000 Russian rubles (30,204,739 US Dollars, the exchange rate is 64.67 Russian rubles per 1 US Dollar). (PwC, 2018b) For comparison, Manchester United, one of the top clubs from English Premier League, earned 635,000,000 US Dollars in the same season from gate attendance (Forbes, n.d.). Just one team from the developed football market earned 21 times more on match day than the whole league from the developing market. Moreover, comparing Russian championship with top-10 leagues indexes it is seen that profit from matchday is less than 5% out of all profit of the Russian clubs, while in top-10 leagues, excluding Russia, this index is 15.5% (PwC, 2018b). These numbers are definitely needed improvements and could be improved by increasing attendance rates and increasing the demand for tickets. When the demand for tickets would be close to supply, the demand for TV broadcasting will definitely increase, subsequently boosting profits from media rights, which are 4.1% in Russia comparing to 38.7% average in top-10 excluding Russia (PwC, 2018b; Salaga & Tainsky, 2015). Finally, there are a lot of new stadiums, specially built for the FIFA World Cup 2018. These stadiums are now given to the local football clubs for usage as the main arenas in the national championship but some of the clubs cannot even afford to use such infrastructure because the costs of playing at such large stadiums are higher than the profits from ticket sales. If the problem was not solved during 2-3 years, these stadiums would be left and would not be used for regular league's matches. That is the topicality of the research problem.

Thus, the research question of this thesis is formulated as the following: what determinants influence Russian Premier League matches' spectator demand the most significantly? Specifically, the study questions the people's intention to visit the football match and investigates what influences this intention.

Since the motivation of this thesis is to increase attendance rates in Russian Premier League, the main objective of the study is to give the recommendations, applicable for clubs' management, leagues' management and/ or the Russian Football Union. It is expected that implementing the given recommendations would positively influence the attendance of RPL matches.

The hypotheses, which are created basing on the research question and which are tested in this research are as follows:

H1: The matches played in December and March are attended with a smaller number of fans.

H2: A determinants of the day of the week and kick-off time are significant for attendance.

H3: Football fans' intention to attend the match is significantly influenced by the estimated performance of the team in the last 5 league matches.

The formulation of the hypotheses is based on the literature of the study area and is specified due to peculiar features of Russian championship and mentality, e.g. the scheduling of the league in general and absence of stadiums with the roof closing impacted the quality of the fields a lot, so the games played in early winter and in early spring are usually played on the fields of a poor condition, which impacts the on-field performance and subsequently could influence attendance rates (H1).

Thereby, the objective is to be achieved through performing the tasks: examining the literature of the study, collecting and processing the data on RPL attendance, analyzing the data using the ordinary least squares regression, defining the most significant determinants, influencing the attendance in the RPL. Examining the literature task means investigating the theoretical frame of the question. The topic of sports economics contains a great number of studies and the core of this task is to define the appropriate and relevant literature. Then this literature is to be studied and critically reviewed. The main conclusions of the authors are to be used as a basis for this thesis's definition of the determinants for analysis. Suchwise, it is essential to choose the appropriate literature from reliable sources because the whole research's relevance depends on this choice.

The second task intends finding reliable sources of data and collecting the data from these sources. Obviously, the highest quality and accuracy of the data is essential for the relevance of the research's results. It is necessary to examine all the available sources of data and chose the sources properly. Also, since the collection of the data is a pretty time-taking process, it is essential to choose the relevant tools and methods to perform this task for saving time and exclusion of possible errors and mistakes in data. Moreover, since the data collection methods are not perfect nowadays and some changings are necessary for making the dataset convenient for usage, the processing of the data is a crucial point to be done. Firstly, the data is collected from the defined source and it is necessary to make one perfect dataset with all the variables. Also, since the data is imperfect too, it is needed to fill in the gaps, fix some lines and columns and so on. Before the analysis will start, the dataset should be brought to perfection, as much as possible.

Analyzing the data is the most important part of the whole research because this step finally provides an opportunity to conclude the results. Undoubtedly, the choice of methodology for performing this task properly is essentially important. The wrong methods' choice could lead to inappropriate results and the research would lose its relevance. After the methods are chosen, the analysis is taking place. The majority of researches in the field of study use ordinary least squares regression analysis to define the dependence of several variables. As none of these researches are conducted for Russian sports leagues, but several are done for developing sports markets, the usage of the OLS regression is chosen. A justification of the methodology is presented in the “Methodology” section.

Finally, the results are to be analyzed. Finding correlations and dependences is fundamental to make relevant recommendations. Basing on the results of the data analysis, principles of fansы behavior can be disclosed. It is decisive that all the previous tasks are performed as perfect as possible because it provides relevance for the analysis of the final results. The recommendations formulation is performed on the basis of this analysis, providing not only the scientific but also the practical relevance of the study.

3. Methodology

This thesis's first objective is to study the current attendance case of the Russian Premier League. The objective intends the reaching of the indexes, which are demonstrated by the football clubs and management of organizations. Achieving this objective is performed by analysis of the literature to understand the indicators of estimation of the case and by collection and visualization of the data. The second objective of the study is to define the reasons for low attendance rates. This objective is reached through analysis of collected data and the justification of these analyses' results. The last but not the least objective's essence is in the creation of the recommendations for managers of Russian football, advising how to increase attendance rates. This objective is reached with the help of critical investigation of the Russian Premier League matches' data analysis. The research design of the empirical analysis is presented below.

3.1 Current attendance situation

The analysis of the nowadays situation of attendance is requesting the understanding of the championship's structure and the data on seasonal attendance. So, the RPL's structure is usual for the Premier Leagues in Europe. 16 teams (European leagues usually have a little bit more teams in one league) play with each other twice a season, once at home and once away, thereby providing 30 games each per season, home and away half-and-half. Each winning is awarded 3 points for a winning team, draw brings 1 point for each team, the loss is not awarded. Standings also consider goals scored, goals conceded, a difference of scored and conceded goals and H2H index (head to head, the result of games between two specific teams). The standings after 30 tours show the results of the championship.

The team which occupied the 1^st place becomes the champion and is awarded golden medals, the 2^nd place team is called vice-champion and gets the silver medals. Both these teams are qualified for the group stage of the UEFA Champions League. The team that captures the 3^rd place is awarded bronze medals and it qualifies for the second qualifying round of the UEFA Champions League. Teams which occupied the 4^th and the 5^th places are qualified to the third and the second qualifying round of the UEFA Europa League respectively. The cup winner is directly qualified for the group stage of the UEFA Europa League but since usually the cup winner is one of the top-5 teams in standings, then the division is changing and then the 4^th place's team is qualifying for the UEFA Europa League's group stage and the 5^th and the 6^th place teams go to the third and the second qualifying round respectively. (UEFA.com, 2018)

The two teams captured the last two places (15^th and 16^th) are relegated to the lower division, Football National League. The 13^th and 14^th place teams are playing two playoff matches with the 3^rd and the 4^th place teams of Football National League. Thus, in the light of the season maximum of 4 teams and the minimum of 2 are to be promoted to Russian Premier League from the lower division. (RPL, n.d.)

Russian domestic football championship, which is now called Russian Premier League, was organized after the collapse of the USSR in 1992 (previous names: the Highest League, the Highest Division, Russian Football Premier League). During 1992-2010 the championship was taking place from March to November. The season of 2011/2012 has become transitional of changing the system of football competition to the system “autumn - spring”, which is now usual for the most football leagues. The transitional season began in 2011 and continued till the middle of 2012, totally 44 games by each team were played instead of 30. The season 20112/2013 was the first in Russian season in the “autumn - spring” system.

Now the Russian Premier League is under control of Russian Football Union (RFU), which is the main organization controlling Russian football in general, from RPL to amateur leagues and national teams. This institute is aiming the objective of the development of football in Russia. Nowadays the RFU is the most significant organization that initiates new amendments, controls the execution of the laws and which is responsible for main indexes of Russian football performance. (Charyev, 2014)

The evaluation of the current attendance situation is based on the data on the seasonal attendance in Russian Premier League scraped from the RPL's official website. The visualization of the data is expected to provide an understanding of trends in attendance changing.

3.2 Data collection and processing

Data collection stage starts right after all the determinants to be investigated are defined based on the literature of the topic and availability of the data. The sample of this research is matches of football clubs playing in the Russian Premier League for 7 years, from 2013/ 2014 to 2018/2019 seasons. The season 2018/2019 is not finished at the moment of data collection, but it has no significance in demand estimation on a match-by-match basis. Hence, the dataset of this research consists of 1360 matches, which are played during 6 seasons by 24 different teams. The representativeness of the sample is proved by the comparison with the prior studies of the topic. The researchers either use the data on a match-by-match basis, considering the several seasons or investigate the attendance at a seasonal level, taking into account more seasons to provide the unbiased coefficients. Though it is very significant for the latter type to provide as much data as it is possible, the former type does not require using the large datasets for analysis. Since this research is game-level based, consideration of 6 seasons in the analysis provides the representativeness of the sample. The preliminary dataset with all the used variables is available at https://drive.google.com/open?id=1lzULxeePBdYgXWt9YyxIdi9SfkNuBFuH.

As it was already mentioned, the decision of the variables to be researched was based on the availability of the data. Since the management of RPL has not contributed to getting access to the data, only open sources were used in the data collection stage. Thus, the data, which are collected in this stage are as follows: teams playing, attendance, date of the match, day of the week when the match was played, statistics of the on-field performance of the team (number of scored goals, shots, aerial duel success, dribbles won, tackles), stadium capacity and the name of the stadium.

Date of the match is necessary to be considered for the relevant estimation of the seasonal dependence of the attendance (e.g. it could be lower in winter periods and higher in summer). Day of the week is also a very important determinant to be estimated since the scheduling of the leagues' matches could play a very significant role in attendance. (Noll, 2002) On-field performance statistics provide the opportunity to estimate the average indexes of a team's performance during the last several matches to correlate it with attendance rates changing. The capacity of the stadium allows estimating the share of sold tickets out of all the available number of tickets. The weather is not a very important factor since, first of all, the effect of this factor cannot be completely eliminated by the club and, secondly, because attending football match is mostly pre-paid, meaning the majority of fans buy tickets in advance, (Buraimo, Forrest & Simmons, 2009), and finally, some researchers found weather insignificant to attendance in soccer (Bird, 1982) so it is not collected for this research analysis. Generally, the data on global factors are not considered because the influence of these factors is impossible to fully eliminate. For example, building the stadium with a possibility to close the roof (“Gazprom Arena” in Saint-Petersburg) could save the fans from the unpreferable weather conditions and decrease weather conditions significance but people still need to get to the stadium first. The same for external factors like political and social.

Due to the unavailability of some data, a great number of variables are not considered in this research. Free on-air broadcasting of the match variable is not included due to the absence of the open accessed statistics of broadcasting. Also, it seems not possible to gather the data of ticket prices to the Russian Premier League's matches because such data is not collected by any reliable and openly accessible sources. Meanwhile, the football clubs in Russia are not obligated to publish the financial reports, the league is not keeping such data, the open sources of data collection are not interested in such data. The average price of tickets for all RPL the matches during the season is available (PwC, 2018a) but is not applicable for this study.

Also, there is the data which seems very labor intensive to be collected since the collection should be done manually. For example, star players variable, which seems to be, first of all, pretty subjective in consideration since the estimation was planned to be based on the price of the player in the transfer market, and also hard to be collected. Moreover, the data of the team's standings position before a match is not considered since there is no reliable official data, and the usage of unreliable sources could decrease the relevance of the research. This data can be collected from official mass media sources of the RPL, but it is mostly presented in pictures, which should be downloaded, described and added to the dataset. Betting odds could also be used to consider the uncertainty of outcome, but prior studies proved the insignificance of the uncertainty in estimating the behavioral pattern of attendees and, moreover, there is already the research which estimates the uncertainty effect in Russian Premier Leagues. Hence, such a given number of variables is providing the relevance of the analysis's results.

The data come from the secondary sources, all of them are websites, which collect and aggregate such data. At first, it was planned to get the data contacting the officials from the Russian Premier League management but the contacted manager did not provide any possibilities of getting the data from the league directly. Obviously, the official website of Russian Premier League was the next applicant to become the main source of the data and it actually is providing the statistics on leagues matches, the attendance at match-by-match basis, but the attempts of these data's extractions failed due to the improper and illogical structure of the website. The only possible way of gathering data from this source was to do it manually so this would have taken an incomparable amount of time.

Moreover, the data on the on-field performance of the team is not collected by RPL (in general, all the statistics gathering is outsourced to Sport Radar company), so the choice of the data source finally fell on the WhoScored website, which is one of the worlds' most popular resources connected with statistics on sports. This website was used as the main source of the data. Summarizing, WhoScored website is providing the statistics of the match-by-match, date of the match, day of the week, teams playing, scored goals number, the on-field performance of the team. RPL official website is used to gather data on stadium capacity and is used to control the reliability of the data from WhoScored website and to fill in the gaps appeared after the primary data collection. Moreover, the official website of the RPL is used to get the data on the seasonal attendance for seasons 2002 - 2018/2019 for estimation of the current attendance situation in Russia.

The process of data collection was automated since the manual gathering of the data would have taken a large amount of time. The choice between using the special software like “Web Scraper” as a browser application or using professional software like Python's “BeautifulSoup”, “Request” packages depended on the opportunities of the parsers. To collect data from the WhoScored website, the virtual user should be created. Using the browser applications is not applicable for this case since the gathering of the data from the particular website is impossible without virtual user imitation. However, professional software like Python allows the implementation of the selenium framework to create a virtual user to provide the automatization of the gathering data process, since the data on the chosen source is not placed on one web-page. From the other hand, the opportunities of the professional software are much wider in general and would be useful during the data processing stage, for example, Python's “BeautifulSoup” package is possible to provide the extraction of the data directly from the HTML markup automatically. So, finally, the Jupyter Notebook framework was used for data collection, the script was written using Python programming language. Package “Request” makes requests of the webpage and downloads it, “BeautifulSoup” package handles to HTML-markup and excludes the needed data from the markup.