Abstract
This study examines the determinants of away fan attendance in Germany's top three football divisions over six seasons and assesses its impact on match outcomes. The analysis reveals that, after conditioning on home and away teams, distance and kick-off time are the most important predictors of away fan turnout. Moreover, away support is found to have a statistically significant positive effect on team performance: an additional 500 away fans are associated with a 2.3% to 3.8% increase in the probability of the away team winning the match. These findings suggest that league organizers should consider fan logistics and the potential influence of away fan presence when scheduling fixtures, in order to maintain sporting equity. At the club level, the results underscore the value of fostering fan engagement and indicate that strategies aimed at increasing away attendance may contribute to improved sporting outcomes.
Keywords
Introduction
Almost every weekend, tens of thousands of football fans travel across Germany to follow their favorite clubs and provide vocal support during away fixtures, aiming to contribute to their team's success. On matchday 25 of the 2024/25 season, for example, an estimated 25,000 fans of second-division club FC Schalke 04 traveled over 500 km from the Ruhr area to Berlin to attend their match against Hertha BSC (kicker, 2025). However, despite the prominent role that away support plays in German football culture, it remains unclear whether these journeys merely offer entertainment for fans or whether the presence of a substantial contingent of away supporters exerts a measurable impact on match outcomes. Moreover, empirical evidence on the determinants of away support is limited, as most prior studies concentrate on general attendance and do not distinguish between fan groups (e.g., Besters et al., 2019; Bond & Addesa, 2020; Cox, 2018; Martins & Cró, 2018).
This study addresses these gaps by examining the factors that influence the number of away supporters present at football matches and by assessing the extent to which the size of the traveling fan base affects team performance. The analysis reveals that the most significant predictors of away fan turnout are the geographical distance between the competing teams and the timing of the match, particularly the day of the week and kick-off time. In contrast, after conditioning on the home and away teams, recent team performances and competitive balance between the two sides exhibit only marginal effects on the willingness of supporters to travel. Furthermore, the findings indicate a statistically significant impact of away fan attendance on match outcomes. Specifically, an increase of 500 away supporters is associated with an approximate 1-percentage-point increase in the likelihood of the away team securing a win. Given an average away win rate of around 30%, this corresponds to a relative improvement of around 3.3%. This effect appears to be mediated by an increase of about 0.019 in the number of goals scored and a reduction of about 0.023 in the number of goals conceded.
Taken together, the findings demonstrate that away support has a considerable influence on match outcomes — and is to a significant extent shaped by factors determined by league organizers, such as the day and time of the fixture. For instance, the results suggest that scheduling an away match on a midweek date can reduce the away team's probability of winning by up to 2 percentage points, due to a reduction in away supporter turnout. These adverse effects are amplified when teams are required to travel long distances at inconvenient times, such as Friday or Sunday evenings or during the working week. These insights carry important implications for league organizers, who should account for the role of away fan support and the constraints imposed by travel distance and kick-off times when scheduling matches to maintain and improve competitive fairness.
Moreover, the findings underscore the broader significance of fan support and engagement from a club management perspective. They suggest that measures undertaken by clubs to facilitate and encourage travel to away matches, such as organized fan travel, or selective ticket price subsidies, could yield positive returns in terms of sporting performance. At the same time, clubs should be aware of the competitive implications when playing at home. Strengthening home-fan engagement and ensuring strong attendance at home fixtures can help minimize the presence of visiting supporters outside the designated guest section, preserving the traditional home advantage. In addition, clubs might consider measures to prevent away supporters from entering home sections, such as restricting ticket resale to visiting fans or prohibiting clothing that identifies the away team. From a longer-term perspective, the results also point to potential unintended consequences of oversized stadiums whose capacity exceeds local fan demand, as surplus seating may allow more away supporters into the stands and, in turn, increase the visiting team's chances of success.
With its findings, this study adds to the growing literature on the role of crowd support in professional football. By distinguishing between home and away supporters, and utilizing variation in fan attendance across clubs and over time, it allows for a granular assessment of fan effects on match outcomes at the intensive margin and highlights the importance of crowd composition rather than sheer crowd size. Previous research has primarily focused on the role of overall attendance, often exploiting matches without spectators, for instance during Covid-19 restrictions (e.g., Cross & Uhrig, 2023; Ferraresi & Gucciardi, 2023; Fischer & Haucap, 2021; Scoppa, 2021), following violent stadium riots (Singleton et al., 2023) or as disciplinary sanctions (Reade et al., 2022). 1 While matches played under the exclusion of fans offer exogenous variation in crowd size, they typically do not allow for more nuanced analyses, as they eliminate variation within the crowd entirely.
In contrast, this study focuses on a relatively small yet influential group of spectators, away fans, whose presence is expected to counteract the home advantage and who tend to display particularly high levels of commitment and dedication. Similar to Belchior (2020), it also complements prior research comparing games with and without spectators by analyzing continuous variation in supporter numbers across matches. It is most closely related to Humphreys et al. (2024), who incorporate away fan numbers into their analysis of fan attendance and match outcomes in English football, and Fioravanti et al. (2025), who exploit the 2013 ban on away supporters in Argentina's first division.
The existing evidence on fan effects suggests that the overall presence of spectators tends to reinforce the home advantage of clubs in terms of match outcomes, performance indicators, and referee decisions (e.g., Cross & Uhrig, 2023; Ponzo & Scoppa, 2018; Scoppa, 2021; Singleton et al., 2023; van Ours, 2024), although effects can vary across countries (Benz & Lopez, 2023; Bryson et al., 2021), and divisions (Fischer & Haucap, 2021; Lyhagen, 2025). Fans are thought to influence matches through crowd noise, chants and vocal support, which may pressure referees towards more favorable decisions (Endrich & Gesche, 2020; Reade et al., 2022; Scoppa, 2021), and directly improve performances of the supported players (Chacón-Fernández et al., 2025; Ferraresi & Gucciardi, 2023) while possibly impairing those of the opposition (Destefanis et al., 2022; Ferraresi & Gucciardi, 2023). However, these effects are likely heterogeneous across team allegiances and fan types, a fact which is typically masked by the aggregate effects reported in most studies. Consistent with this, the limited evidence on away support indicates that restricting away attendance can reduce the away team's chances (Fioravanti et al., 2025), whereas greater away presence may improve them (Humphreys et al., 2024).
The results presented here extend these findings by further emphasizing the importance of crowd composition, as both fan contingents appear to exert significant but opposing influences on match outcomes. Additionally, the results show that even relatively small groups of supporters can meaningfully affect team and player performances, underscoring the significance of fan heterogeneity. Away supporters, in particular, often consist to a large extent of members of organized fan clubs who are typically the most dedicated and vocal followers. This suggests that crowd composition matters not only in terms of team allegiance but also in the intensity of fan commitment. Supporting this interpretation, Singleton et al. (2023) find no significant increase in home advantage in Egypt after the readmission of spectators and the dissolution of major ultra groups.
Finally, this study also contributes to the literature on the determinants of fan attendance by studying factors influencing away supporter turnout. Previous literature has predominantly focused on general spectator attendance, highlighting the importance of match quality aspects, such as player quality and local rivalries, and the expectation of favorable outcomes for the home team in shaping fan attendance (Besters et al., 2019; Bond & Addesa, 2020; Cox, 2018; Martins & Cró, 2018; Schreyer et al., 2019; Serrano et al., 2015). These insights, however, are not easily transferable to away supporters, whose attendance entails greater planning and cost, leading to a self-selection of more committed fans. Consistent with Hernández et al. (2023), the present findings suggest that match quality and competitive intensity play a much smaller role for away attendance. Instead, factors related to travel convenience, in particular match scheduling and travel distance, emerge as central determinants. While home crowds are of greater financial importance in terms of ticket sales and match-day revenues than the much smaller group of away fans, the considerable influence of away supporters on sporting outcomes underscores the importance of understanding what drives their turnout.
In the following, Sections 2 and 3 describe the data and empirical approach, before Sections 4 and 5 present the results. Section 6 concludes.
Data
Data Set
I collect match-level data for the three highest German football divisions, 1. Bundesliga, 2. Bundesliga, 3. Liga, for eight seasons, from the 2017/18 season to the 2024/25 season from fussballmafia.de, a website dedicated to German football fan culture. 2 The data contain information on the day and time of each match, its final result, the road network distance between the two clubs, total attendance, the number and share of away fans, and the number of tickets allocated to the away section. The attendance figures published on fussballmafia.de are sourced directly from football clubs, though the platform also permits users to submit suggestions for revisions. In most instances, the figures reported by the clubs are not precise counts but rather estimates derived from factors such as ticket sales. This raises the question of the extent to which these estimates reflect actual spectator attendance.
Recent research has found a significant discrepancy between the number of tickets sold and actual fan attendance, a gap that appears to be primarily driven by season ticket holders of the home team (Schreyer, 2019; Schreyer et al., 2019). Although empirical evidence on the no-show behavior of away supporters remains limited, this issue is arguably less pronounced for away fixtures than for matches played at home. Away supporters constitute a smaller, more dedicated and committed subgroup of fans who invest considerable time and resources to attend matches, reducing the likelihood of spontaneous cancellations. Moreover, not all clubs offer season tickets for away fixtures, and where they do, these are often subject to strict monitoring and eligibility requirements (Eintracht Braunschweig, 2025; FC Köln, 2025; VfL Bochum, 2025). As a result, for most clubs, only a very small, highly committed subgroup of supporters is able to hold such tickets, making no-shows among away season ticket holders unlikely. In light of this, the difference between ticket sales and actual attendance should be much smaller for away fans than for home crowds. Consequently, the figures reported by fussballmafia.de can be expected to approximate actual away attendance reasonably well. Importantly, any remaining discrepancy between reported away fan numbers and actual attendance would bias estimates of the effect of away supporters on match outcomes downward, rendering the results rather conservative.
To complement the match-level data, I gather club standings after each matchday from kicker, a German sports magazine, as well as team market value data after each season's summer and winter transfer window from transfermarkt.de, and add information on whether each club was promoted or relegated in the previous season. Furthermore, I obtain daily information on maximum temperature and precipitation from the German Climate Data Center (DWD Climate Data Center, 2025a, 2025b), and, for supplementary analyses, gather data on betting odds as well as match-level disciplinary outcomes (fouls, yellow cards, and red cards) for the first and second divisions from football-data.co.uk, a free provider of football betting odds and match statistics. For the 2024/25 season, I additionally collect auxiliary data on the stadium capacities and fan membership figures from fussballmafia.de.
The dataset comprises a total of 7,936 matches involving 78 different teams. However, 2,276 of these matches were affected by COVID-19-related restrictions that either prohibited or significantly limited attendance for both home and away supporters. The first matchday impacted by such measures was the 26th of the 2019/20 season. Thereafter, the entire 2020/21 season was played behind closed doors, with no spectators permitted. Limited attendance was reintroduced at the start of the 2021/22 season, and restrictions were gradually eased throughout the season. By the 28th matchday, matches were again played in front of full-capacity crowds. To ensure consistency in fan attendance conditions, I exclude all matches from the 25th matchday of the 2019/20 season through to the first matchday of the 2022/23 season. This results in a final sample of 5,660 matches across six seasons involving 75 different teams. Of these, 1,755 games were played in each of the first and second divisions, which both include 18 clubs, while 2,150 matches took place in the third division, which comprises 20 teams. In each season, teams play every other team twice, once at home and once away.
Descriptive Statistics
Table 1 contains descriptive statistics for the entire sample and for the first, second, and third division separately. During the study period, an average of 23,500 spectators at- tended each match, with approximately 2,000 of them supporting the visiting team. This means that, on average, around 9.4% of the total crowd consisted of away fans. As expected, attendance figures vary considerably across divisions. The 1. Bundesliga, home to most of the country's large clubs and stadiums, attracts the highest numbers. 3 On average, matches in the top tier were attended by 41,800 spectators, including about 3,400 away fans — equivalent to 8% of the crowd. In contrast, the 2. Bundesliga and 3. Liga reported lower average attendances of 23,400 and 8,800 respectively, with corresponding away fan figures of approximately 2,100 and 800 per match.
Descriptive Statistics.
Note: The table reports descriptive statistics for the final analysis sample. The sample comprises 5,660 matches, including 1,755 from both the first and second divisions, and 2,150 from the third division.
While away fan attendance reflects strong supporter engagement, it is subject to structural features of stadium design and ticketing policies. In most stadiums, a designated share of total capacity, approximately 12% on average across all divisions, is reserved for away supporters. In practice, these allocations are not always fully utilized. In approximately 62% of all matches, the designated away section was not sold out. On average, guest sector usage stands at 65%, with notable variation across leagues: 83% in the 1. Bundesliga, 71% in the 2. Bundesliga, and just 40% in the 3. Liga. These figures indicate that in many matches, additional away fans could have attended within the designated away section. Moreover, the away allocation does not constitute a strict upper bound on away attendance. Visiting fans can, in principle, also purchase tickets in general seating areas, although access to these tickets is typically easier for home supporters, who may benefit from priority booking or club memberships. Consistent with this, in 1,313 matches, observed away attendance exceeded the nominal guest allocation, indicating that fans successfully accessed tickets beyond the designated section.
Across all divisions, fans traveling to support their club away from home see their team win in approximately 30% of matches. The probability of a home team victory is notably higher at 43%, resulting in a 57% likelihood that the away team either wins or draws. These outcome probabilities vary only slightly across divisions: the away win rate ranges from 29% to 31%, while the probability of an away team avoiding defeat remains stable at around 56% to 57%. This home advantage is also reflected in goal statistics. Overall, home teams score an average of 1.61 goals per match, compared to 1.31 goals by away teams. This amounts to an average difference of 0.30 goals per match, or roughly 23% more goals in favor of the home team.
The average distance that away fans travel to attend matches is approximately 388 km, with only minor variation across divisions. Travel distances are strongly influenced by a team's geographic location within Germany and the regional distribution of other clubs within the same division. For example, centrally located teams such as Rot-Weiß Erfurt and Eintracht Frankfurt benefit from comparatively short average travel distances, both around 290 km. Similarly, clubs based in North Rhine-Westphalia, the most densely populated federal state with numerous professional teams, also enjoy shorter travel distances due to the concentration of nearby opponents. In contrast, teams located on the geographic periphery of the country face considerably longer journeys. Fans of northern clubs such as Hansa Rostock and VfB Lubeck travel the furthest, averaging up to 580 km per away game. Likewise, teams in the far south or east, including Bayern Munich and Union Berlin, also face substantial average distances of 500 km and 480 km, respectively.
In contrast to travel distances that remain relatively consistent across divisions, match scheduling varies substantially, with kick-off times deliberately staggered to avoid overlap between leagues. Saturday afternoon is traditionally reserved for 1. Bundesliga matches, with about half of all games played in this time slot. The top match of the week typically takes place on early Saturday evening (5%) and the rest spreading across Friday evening (10%) and Sunday afternoon (16%) and evening (7%). The 2. Bundesliga mainly plays at Saturday and Sunday noon (60%), supplemented by Friday evening games (20%) and one Saturday evening fixture after the conclusion of the first division's game (6%). In the 3. Liga, most matches occur on Saturday noon (54%) and Sunday noon (14%). As in higher divisions, Friday evening often marks the start of the matchday (9%). Notably, some matchdays are scheduled for Tuesday or Wednesday evenings during the week. 4 Midweek fixtures are rare in the top two divisions, 6% in the first and 7% in the second, but more common in the third (13%) due to its larger number of teams and matches.
Empirical Approach
The following analysis is two-fold. First, I investigate the factors that determine the number of fans who support their team away from home by attending matches in opposing stadiums. After that, I analyze the impact of away fan numbers on match outcomes. Both analyses follow a nearly identical empirical approach, which is based on linear fixed effects regressions that allow accounting for unobserved heterogeneity across clubs, divisions, seasons, and matchdays.
For the analysis of the determinants of away supporter turnout, I closely follow previous related research and model fan attendance as a function of club-specific features, match quality aspects, scheduling decisions, weather conditions, and travel distance (e.g., Cox, 2018; García & Rodríguez, 2002; Hernández et al., 2023). To account for (largely) time-invariant club-specific characteristics, I include club fixed effects for both the hosting and visiting team. Such features that are frequently considered to exert a positive influence on spectator turnout are a club's market size and its regional economic conditions (Bond & Addesa, 2020; Hernández et al., 2023) as well as its general popularity, tradition, brand strength (Coates et al., 2017, 2025) and stadium capacity. Although the fixed effects to a large extent also capture overall team quality, I additionally include time-varying squad market values which reflect changes in team quality over time (Schreyer et al., 2019; Serrano et al., 2015).
Further match-specific quality aspects I consider are recent sporting performances of both teams, captured by each club's current position in the league table (Bond & Addesa, 2020; García & Rodríguez, 2002) and the number of wins over the past five matches (Forrest & Simmons, 2006), relegation or promotion in the previous season (Besters et al., 2019; Forrest & Simmons, 2006) as well as regional rivalries between clubs which render matches particularly intense (Besters et al., 2019; Martins & Cró, 2018). To proxy competitive intensity and uncertainty of outcome in the absence of data on betting odds for the full sample, I employ an approach similar to Forrest and Simmons (2006) and Hernández et al. (2023), and utilize the point differential between the home and away teams, taking on negative values if the away team is positioned above the home club, as well as the absolute difference in points between both teams. 5
Moreover, I include factors regarding away fans’ opportunity costs and convenience when traveling, such as the weekday and time of the game (Besters et al., 2019; Cox, 2018; Hernández et al., 2023), temperature and precipitation (Bond & Addesa, 2020; Pawlowski & Nalbantis, 2015), both at the venue and in the city of the visiting club, as well as the scheduling of consecutive away fixtures (Schreyer et al., 2019). Furthermore, I incorporate the distance to travel, which has already been found to play a significant role in general fan attendance decisions (Bond & Addesa, 2020; Cox, 2018) but can be expected to be substantially more relevant for traveling away fans (Hernández et al., 2023; Humphreys et al., 2024). Lastly, I also include season, matchday and division fixed effects to account for evolving fan interest within and between seasons as well as heterogeneity between divisions (Hernández et al., 2023; Pawlowski & Nalbantis, 2015; Schreyer et al., 2019). Monetary factors such as ticket and beer prices are not included in the analysis because no reliable data for the entire study period is available. However, I regard them as secondary influences since prices tend to vary only modestly within each division and thus do not meaningfully affect the overall cost of attending one away match over another. For example, during the 2024/25 season in the 1. Bundesliga, the price of a half-liter of beer ranged from €4.50 to €5.50, while tickets for standing areas were typically priced between €14 and €18. 6
To examine the determinants of away fan attendance, I estimate the following fixed effects regression:
To identify the impact of the number of guest fans on match outcomes, I estimate a similar fixed effects regression which is given by
The central challenge in both analyses lies in isolating the effect of interest, the determinants of away support and the impact of away fans on match outcomes, from confounding factors that simultaneously influence both the dependent and explanatory variables. The main threat arises from unobserved differences in club characteristics, such as a team's historical success, team quality, the size and intensity of its organized supporter base, fan culture, stadium capacity, overall popularity, or the attractiveness of the host city as a travel destination. These attributes are only partially captured by observable indicators such as market values, league standings, membership numbers, or city population. To mitigate this issue, I include fixed effects for both the home and away teams, which control for time-invariant club-specific factors that might otherwise bias the estimation. Moreover, these fixed effects are interacted with division indicators to account for the possibility that a club's (perceived) attributes vary depending on the division in which it currently competes. Clubs that are promoted or relegated often experience changes in fan engagement, competitiveness, and media attention, which in turn may influence both match outcomes and fan behavior.
In the context of explaining the number of away supporters, the fixed effects adjust for structural advantages or disadvantages across clubs. For instance, Eintracht Frankfurt is widely known for its passionate fan culture and benefits from a geographically central location in Germany, resulting in relatively short average travel distances for away matches. Without proper controls, this could upwardly bias the estimated effect of travel distance on away fan attendance. Likewise, high-profile clubs such as Bayern Munich or Borussia Dortmund, which command large national followings, are frequently scheduled in Saturday evening time slots for television coverage and regularly draw substantial away support. This could confound the effect of kick-off time if not properly accounted for.
In the analysis of match outcomes, the fixed-effects structure accounts for factors that influence both guest fan attendance and team performance. For example, prominent and successful clubs such as Bayern Munich not only travel with large numbers of fans but are also more likely to win matches due to superior squad quality. Failing to control for such characteristics would risk overestimating the impact of away support. To further control for confounding influences, I include a comprehensive set of match-specific variables from the analysis of guest fan number antecedents, as they are expected to influence fan attendance decisions, while at the same time possibly correlating with player performances and match outcomes. These variables include the geographic distance between clubs, a dummy indicating the scheduling of consecutive away fixtures, and the match's kick-off time which not only affect the likelihood of away fan attendance but may also influence player performance via travel-related fatigue or circadian rhythms. I also control for point differences between the home and away club, as well as each club's average player market value, recent form and table position to capture variation in quality, momentum and competitive intensity. Additionally, I include indicators for promotion and relegation in the previous season, since changes in divisional status may alter both relative team quality and the composition of the fan base. Finally, I account for the number of home supporters, as full stadiums may mechanically limit the number of away tickets available and prior studies suggest that a larger home crowd can positively affect team performance (Cross & Uhrig, 2023; van Ours, 2024). 7 By controlling for these factors and applying a rich fixed effects structure, the empirical strategy seeks to isolate the effect of away fan presence and to produce estimates that are robust to latent club-specific or match-specific sources of bias.
Determinants of Away Fan Attendance
Table 2 presents the results from estimating Equation (1), using the logarithm of the number of guest fans as the dependent variable. Column 1 shows the estimates for the full sample, while columns 2 to 4 report the results separately for the first, second, and third divisions. The findings indicate that geographic distance between the two clubs is the most important determinant of away fan attendance. On average, an additional 100 km of travel reduces the number of away fans by nearly 23%. This substantial effect is largely driven by the third division, where away attendance drops by 29% per 100 km. The effect is smaller in the second and first divisions, 21% and 15%, respectively, suggesting that fans of higher-tier clubs are more willing to travel longer distances than those supporting lower-tier teams.
The Determinants of Away Fan Attendance.
Note: The table analyzes the determinants of away fan attendance based on Equation (1), using the number of away fans in logarithmic form as dependent variable. All regressions include home team-division, away team-division, season, and matchday fixed effects. Standard errors are clustered at the away team level (in parentheses). ** and * denote statistical significance at the 1% and 5% level, respectively.
To capture potential non-linearities in the relationship between distance and away fan turnout, I re-estimate Equation (1) using a categorical distance variable. Figure 1 plots the estimated effects relative to matches requiring travel of less than 100 km. The results show that at shorter distances, away attendance declines steeply with increasing distance. However, beyond around 400 km the marginal effect of distance diminishes. In both the first and second divisions, attendance levels remain relatively stable beyond this point, implying that fans willing to travel 400 km are largely undeterred by additional distance. A similar pattern holds in the third division, where the decline in away fan numbers slows considerably after the 400 km mark. The stark decline in away fan turnout at shorter distances may be partly explained by rivalry matches and local derbies between geographically close clubs, which tend to attract a large number of away fans.
The Effect of Distance on Away Fan Attendance. Note: The figure displays distance estimates and 95% confidence intervals from Equation (1), using a categorical distance variable in 100 km brackets. The reference category is away games requiring less than 100 km of travel. The regression includes home team-division, away team-division, season, and matchday fixed effects as well as the full set of covariates, except for the derby-indicator. Standard errors are clustered at the away team level.
Similar to travel distance, kick-off time plays a crucial role in determining whether fans attend away games. In columns 1 and 2 of Table 2, the effects of different kick-off times are shown relative to Saturday afternoon, when most 1. Bundesliga matches take place. In columns 3 and 4, Saturday noon serves as the reference category, reflecting the typical match time in the second and third divisions. Most coefficients are statistically significant and negative, indicating that matches scheduled outside of the standard Saturday slot attract fewer away fans, highlighting a strong preference for Saturday games among traveling supporters. The estimates for Saturday evening are small and statistically insignificant, suggesting that fans are generally willing to travel on Saturday evenings — likely because Sunday is a free day for most people in Germany, allowing for a more relaxed return. In contrast, games scheduled on Sunday lead to substantially lower away attendance. In the 1. Bundesliga, Sunday afternoon matches see a 12% reduction compared to Saturday afternoon, while in the second division, Sunday noon matches are associated with a 10% decline relative to Saturday noon. The effect is even more pronounced for Sunday evening fixtures, with away attendance dropping by up to 25%, likely due to the inconvenience of late-night travel before the start of the work week. Similarly, games played on Friday evening or during the week are affected by practical constraints. Specifically, Friday evening fixtures reduce away attendance by 22% on average, while matches scheduled during the week lead to a 43% decline, likely due to work obligations and other weekday responsibilities that limit fans’ ability to travel longer distances in the evening. While fans of 1. Bundesliga clubs appear somewhat less discouraged by Friday night travel than fans in the 2. Bundesliga and 3. Liga, they are particularly sensitive to midweek games, with a 49% drop in away attendance.
Since both geographic distance and kick-off time have been identified as key determinants of away fan turnout, I extend Equation (1) by including interaction terms between distance and various kick-off times. The corresponding estimates are presented in Table A1. The results indicate that the negative effect of distance on away attendance is least pronounced for matches played on Saturday afternoon or noon, highlighting these time slots as the most favorable for traveling fans. For Saturday evening matches, the negative distance effect increases slightly, likely due to the inconvenience of returning home late at night. Across all divisions, Sunday matches show significantly stronger distance penalties. Interaction effects range from −5% to −11% per additional 100 km, suggesting that travel becomes more burdensome when matches are scheduled on Sundays, possibly due to work obligations the following day. Similarly, on Friday evenings, away fan turnout declines by an additional 4% in the first division, 9% in the second, and 8% in the third division for every 100 km traveled, compared to Saturday afternoon/noon. The effect is most pronounced during the week, where each additional 100 km leads to a 14–15% larger reduction in away supporter numbers across all divisions, underscoring the compounding impact of long travel and midweek scheduling.
While distance and kick-off time clearly play a central role in determining away fan attendance, the results suggest that fans do not respond sensitively to temperature, precipitation, or the scheduling of consecutive away fixtures. Similarly, squad quality, competitive intensity and recent sporting performance appear to be of secondary importance after accounting for club-specific and division-level characteristics. Across all divisions, most coefficients related to point differences between teams, the number of wins in the last five matches, current league standings, and player market values are statistically insignificant. In contrast, a club's relegation or promotion in the previous season has a notable and statistically significant impact. On average, away fan turnout increases by approximately 12% following a relegation and by 20% after a promotion. The increase in away support following a promotion is intuitive: a successful season often generates renewed excitement and optimism around the club, motivating more fans to travel. The rise in attendance after a relegation, however, is somewhat less expected. One possible explanation is that the adversity of a relegation season may strengthen the sense of community and loyalty among supporters, leading to greater commitment, including at away games. Additionally, away fans seem to respond not only to their own team's recent history but also to that of the opponent. Matches against recently promoted home teams attract significantly more traveling supporters — by around 16%. This may reflect the appeal of facing a “new” opponent in the division, adding novelty or prestige to the away match experience. Lastly, the estimates indicate that the home team's quality, proxied by the average player value, has a significant influence on away fans’ willingness to travel, even though the direction of the effect varies across divisions.
The results remain largely consistent when alternative measures of away fan engagement are used as dependent variables. Columns 2, 3, and 4 in Table A2 show the results of Equation (1) using the share of away fans, the away fan capacity usage and the number of away fans per club member as dependent variable, respectively. Moreover, in column 1, Table A2 presents the effects of distance, kick-off time, competitive intensity, player quality, and recent sporting performance on home fan attendance. The results reveal that after conditioning on home and away teams, home supporters respond positively to a stronger position in the league table and relegation from a higher division in the previous season. Notably, player market values and promotion to a higher league do not appear to significantly influence home attendance. Most coefficients related to the away team's performance, such as form, league standing, or recent promotion/relegation, are statistically insignificant, indicating that home fans are generally indifferent to the sporting status of the visiting side. One exception is the negative influence of the visiting club's average player value on the number of home fans, which hints at a preference for weaker opponents. In line with this notion, matches where the home team is ranked above the away team in the standings see a modest but statistically significant increase in attendance, possibly reflecting higher expectations of a favorable result. Furthermore, local rivalries or derby matches appear to generate more excitement and attract larger home crowds, suggesting an increased fan interest when geographically close clubs visit. Finally, unlike the substantial impact of kick-off time on away attendance, home supporter turnout appears largely unaffected by scheduling. The substantially lower and statistically insignificant coefficients for alternative time slots, including Friday, Sunday, and midweek evenings, suggest that local fans are more flexible, likely because they face shorter travel distances and less logistical burden in attending matches.
Effects of Away Fan Support
Table 3 presents the estimated effects of away fan support on match outcomes: specifically, the probability of winning the away fixture (column 1), avoiding a loss (column 2), and the number of goals scored (column 3) and conceded (column 4) by the away team. The table reports results from three sets of regressions, each using a different measure of away fan support: the total number of away fans (top panel), the share of guest fans in the total crowd (middle panel), and away fan numbers relative to the away section's capacity (bottom panel). The results suggest that away fan support has a statistically significant and meaningful impact on match performance. Specifically, on average, an additional 500 away fans increase the likelihood of winning the match by 1.2 percentage points and the probability of avoiding a loss (win or draw) by 1.3 percentage points. Given an average away win rate of 30%, this corresponds to a 3.8% increase in the probability of winning. Similarly, with an average no-loss rate of 57%, the effect translates into a 2.3% increase in the probability of avoiding defeat. 8
The Effect of Away Fan Support on Match Outcomes.
Note: The table analyzes the effect of away fan attendance on match outcomes based on Equation (2), using alternative measures for away fan support. The top panel measures away fan support by the absolute number of fans, the middle panel uses the share of away fans, and the bottom panel employs the number of away fans relative to the capacity of the designated guest fan section. All regressions include home team-division, away team-division, season, matchday, and day-time fixed effects as well as the full set of covariates. Standard errors are clustered at the away team level (in parentheses). ** and * denote statistical significance at the 1% and 5% level, respectively.
The rise in the probability to draw or win the away game is mediated by both an increase in the number of goals scored and a reduction in goals conceded by the away team. Both coefficients are statistically significant and amount to an average increase of 0.019 in the number of goals scored and a simultaneous reduction of 0.023 goals conceded associated with an additional 500 guest supporters. Taken together, this leads to an improvement of 0.042 in the total score difference between the home and away team. Given an average number of 1.3 goals scored and 1.6 goals conceded, the estimates translate to an increase in the number of scored goals of 1.4% and a reduction in the number of goals conceded by 1.4% per 500 fans.
As a complementary perspective that also accounts for the importance of away support relative to the home crowd, I estimate Equation (2) using relative measures of away fan presence that reflect the balance between visiting and home fan presence. The results indicate that a one percentage point increase in the share of guest fans is associated with a 0.33 percentage point increase in the likelihood of an away win, and a 0.40 percentage point increase in the probability of avoiding a loss. Additionally, each percentage point increase in the away fan share is linked to 0.007 more goals scored and 0.009 fewer goals conceded by the visiting team. Similarly, when measuring support by guest fan capacity utilization, a 10 percentage point increase is associated with a 0.7 percentage point higher likelihood of an away win, and a 1.0 percentage point increase in the probability of avoiding a loss. The corresponding effects on goals scored and conceded are +0.012 and −0.021.
To compare these figures with the ones reported in absolute numbers, I scale them to increases in away fan turnout of 500 additional fans. The average match attendance in the sample is approximately 23,500, meaning a one percentage point increase in guest share corresponds to roughly 235 additional away fans. 9 Scaling the effects accordingly, an increase of 500 away fans implies a 0.7 percentage point higher probability of winning, a 0.9 percentage point increase in the chance of avoiding a loss, and changes in goals scored and conceded of +0.015 and −0.019, respectively. Given that the average away fan section holds 3,200 spectators, a 10 percentage point increase in away fan capacity utilization represents an increase of 320 fans. Scaling the effects to 500 additional away fans, the results imply an increase in win and non-loss probabilities of 1.1 and 1.4 percentage points, respectively, along with an improvement in goal difference through +0.019 goals scored and −0.033 goals conceded.
The results discussed above capture average effects across all three divisions. However, given differences in overall attendance between the 1. Bundesliga, 2. Bundesliga, and 3. Liga, effects are likely to vary across divisions. Table 4 shows that the impact of away supporters is strongest in the third division, where an increase of 500 away fans raises the likelihood of an away win by approximately 2.6 percentage points. In contrast, the effect is smallest in the second division, where the same increase in away fan numbers is associated with only a 0.9 percentage point rise in the probability of an away win. In the first division, the estimated impact is closely aligned with the overall results presented in Table 3, suggesting a 1.1 percentage point increase in the chance of an away victory for each additional 500 away supporters.
The Effect of Away Fan Numbers on Match Outcomes by Division.
Note: The table analyzes the effect of away fan numbers on match outcomes separately for the first, second, and third divisions. The top panel shows the results of Equation (2) for the 1. Bundesliga, the middle panel for the 2. Bundesliga, and the bottom panel for the 3. Liga. All regressions include home team-division, away team-division, season, matchday, and day-time fixed effects as well as the full set of covariates. Standard errors are clustered at the away team level (in parentheses). ** and * denote statistical significance at the 1% and 5% level, respectively.
These heterogeneous effects, in particular the larger estimates in the third division, are consistent with pronounced differences in baseline attendance levels across leagues. In the 3. Liga, average away attendance is approximately 760 spectators, so that an increase of 500 fans constitutes a more than proportional rise in visiting support relative to the typical number of guest fans at this level. In contrast, average away crowds in the 1. Bundesliga and 2. Bundesliga amount to roughly 3,400 and 2,100 supporters, respectively, making the same absolute increase substantially smaller in relative terms. Notably, in the first division, away support appears to influence both the number of home and away goals. By contrast, in the second division away fans seem to primarily enhance the team's defensive stability, reflected in fewer goals conceded, whereas in the third division, fan support predominantly provides an offensive boost, increasing the number of goals scored.
Beyond the effects of away fan support, Equation (2) includes several additional covariates capturing team characteristics, travel distance, and home attendance. Table A3 presents the estimates for the remaining variables of Equation (2) using the entire sample. The results indicate that the number of home supporters negatively affects the away team's chances. Specifically, on average, an additional 500 home fans increase the probability of a home win by 0.36 percentage points and raise the number of goals scored by the home team by 0.008. However, the magnitude of the estimated effects is substantially smaller than those associated with away fan numbers. One possible explanation is self-selection: fans who choose to travel to away matches may be more committed and vocal, providing more intense support than the average home attendee, who faces fewer logistical and financial barriers to attend. In addition, the coefficients for distance are positive, suggesting that away team performance improves with greater travel distance, even though only the effects on the likelihood of winning/not losing the game reache statistical significance. Finally, the coefficients for player values, recent team form, point differentials, and league standings are generally statistically insignificant, implying that, once general differences between the home and away teams are accounted for, these factors do not meaningfully predict match outcomes.
The main results remain consistent when alternative model specifications are employed. First, Table A4 presents estimates from a logit model for match outcomes and a Poisson regression for goal counts. Using the logit model, an additional 500 away fans is associated with a 1.1 percentage point increase in the probability of an away win and a 1.5 percentage point decrease in the probability of a home win. The Poisson regression suggests that 500 additional away supporters lead to an increase of approximately 0.017 goals scored and a reduction of 0.023 goals conceded by the visiting team. Furthermore, the main findings hold when varying the fixed effects structure or controlling only for variables that were found to have a significant impact on away fan numbers, as shown in Table A5. Lastly, I assess the robustness of the results to the inclusion of betting odds as control variables. To this end, I restrict the sample to matches played in the first and second divisions and augment the baseline specification with match outcome probabilities based on betting odds from B365 and Bet & Win. The estimated effect of away supporter attendance remains statistically significant and of similar magnitude, indicating that the main results are not driven by differences in ex ante match expectations (Table A6).
So far, the analysis has focused on the overall impact of away fan support on match outcomes and goal-based performance measures, i.e., the number of goals scored and conceded. However, a large body of prior work suggests that crowd support may also affect referee behavior, leading to decisions that favor the supported team (e.g., Endrich & Gesche, 2020; Reade et al., 2022; Scoppa, 2021). To assess whether the performance effects documented above operate through such a channel, Table A7 examines the relationship between away fan attendance and referee decisions. Specifically, it reports the estimated effect of 500 additional away supporters on fouls committed, yellow cards received, and the probability of a red card, separately for the away and home teams, using a sample of matches from the 1. Bundesliga and 2. Bundesliga. With the exception of the coefficient on the probability that the away team receives a red card, all estimates are statistically insignificant, suggesting that away fan support does not systematically affect referee decisions.
Instead, the increase in the probability of a red card for the visiting club of approximately 0.4 percentage points may indicate that stronger away fan support encourages visiting players to adopt a more aggressive style of play, which in turn raises the likelihood of a sending-off. Overall, these findings stand in contrast to previous studies documenting a favorable effect of crowd support on referee decisions (e.g., Endrich & Gesche, 2020; Reade et al., 2022; Scoppa, 2021). A possible explanation is that, in most matches, the number of away supporters remains small relative to the home crowd, such that variation in away fan attendance is insufficient to exert a meaningful influence on referee behavior.
Conclusion
This study analyzes the determinants of away fan attendance in German professional football over six seasons and three divisions and assesses the impact of away support on match outcomes. The findings show that the number of traveling fans is primarily shaped by the geographic distance to the away venue and the scheduled kick-off time. Importantly, there are strong interaction effects, as fans respond more sensitively to distance when matches are held at less favorable times, such as during the week, on Friday evenings, or on Sundays. In turn, the results reveal that away fan turnout, measured in both absolute and relative terms, has a significant impact on match outcomes. Specifically, an additional 500 away fans increase the likelihood of winning the match by approximately 0.7 to 1.2 percentage points and reduce the probability of a home victory by 0.9 to 1.4 percentage points. Given an average away win rate of 30% and a no-loss rate of 57%, these effects correspond to an approximate 3.3% increase in the probability of an away win and a 2.3% increase in the likelihood that the away team does not lose the match.
With its findings, this study contributes to the literature on crowd support by providing novel evidence on the importance of crowd composition, emphasizing the role of fan allegiance in shaping sporting outcomes. The comparatively large effect sizes observed for a relatively small group of away supporters, among whom the share of organized fan clubs and ultra groups is particularly high, may further indicate that differences in fan composition, especially in terms of the intensity and commitment of support, play an important role. However, since the available data does not distinguish between different types of fans or between seating and standing areas, where much of the atmosphere typically originates, more granular analyses of crowd composition are not yet possible. Future research could address these limitations by employing stadium admission data, as in Schreyer et al. (2019) and Schreyer et al. (2019), to explore in greater detail how the characteristics of fans shape match outcomes.
The considerable influence of away fan support on match outcomes implies a competitive advantage for clubs with large and engaged fan bases, as they are able to attract more support at away fixtures, thereby improving their performance. This underscores the potential benefits of fostering fan engagement, and suggests that, where feasible, club-initiated measures aimed at boosting away attendance, such as organized transportation or charter trains for long-distance matches, may translate into improved sporting performance. However, as data on such interventions were not available, their specific impact remains unclear.
These insights also have direct implications for league organizers, who are responsible for match scheduling. The evidence highlights that the timing and location of fixtures can systematically influence team performance by affecting away fan support. Ensuring a more equitable distribution of favorable time slots and minimizing long-distance travel during the week would contribute to fairer competition. For example, while twelve teams in the 2024/25 1. Bundesliga season played eleven or more away matches on Saturdays, the most accessible time slot, FSV Mainz 05 had only seven Saturday fixtures, alongside six Sunday games, three Friday evening matches, and one midweek away trip to Leverkusen. Several of these matches required travel over long distances, including fixtures in Kiel (630 km), Berlin (600 km), and Bremen (490 km).
While the scheduling of matches is complex and must balance multiple factors, including international competitions, safety considerations, and broadcasting demands, the findings of this study underscore the need to consider fan logistics systematically. Overall, the results suggest that the organization of matchdays can have unintended yet significant consequences for sporting fairness, and that more thoughtful scheduling could enhance both sporting equity and fan engagement across the league.
Footnotes
Acknowledgements
I am thankful for valuable feedback and comments from Sven Werenbeck-Ueding, Thomas Bauer, Max Schäfer, and two anonymous reviewers. This research is self-funded.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
Notes
Author Biography
Appendix
The Effects of Away fan Numbers on Referee Decisions.
| Away Team | Home Team | |||||
|---|---|---|---|---|---|---|
| Fouls | Yellow cards | Red card | Fouls | Yellow cards | Red card | |
| Guest fans (500) | −0.0196 | 0.0005 | 0.0033* | 0.0203 | 0.0120 | 0.0009 |
| (0.0198) | (0.0097) | (0.0016) | (0.0181) | (0.0079) | (0.0013) | |
| No. Observations | 3,506 | 3,506 | 3,506 | 3,506 | 3,506 | 3,506 |
| Adj. R2 | 0.195 | 0.063 | 0.001 | 0.187 | 0.061 | 0.003 |
Note: The table analyzes the effect of the number of away fans on the number of fouls, yellow cards, and the probability of a red card for the away team (columns 1, 2, and 3), and for the home team (columns 4, 5, and 6). The analysis sample is restricted to matches played in the first and second divisions. All regressions include home team-division, away team-division, season, matchday, and day-time fixed effects as well as the full set of control variables used in the main analysis. Standard errors are clustered at the away team level (in parentheses). ** and * denote statistical significance at the 1% and 5% level, respectively.