Bachelor Thesis
Martin Smilevski
(11646721)
22.06.2020
BSc Business Administration
University of Amsterdam
Supervisor: drs. Rob van Hemert
Choking during Shootouts: How playing at home
influences the Outcome of Penalty Shootouts in
Football. Looking into Crowd, Location and
Cognitive Factors.
Statement of Originality
This document is written by Student Martin Smilevski who declares to take full
responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that
no sources other than those mentioned in the text and its references have been used
in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of
completion of the work, not for the contents.
Abstract
Although the phenomenon of home advantage is widely accepted in football, it is unclear whether it also holds for penalty shootouts. In order to explain whether there is a home advantage in penalty shootouts, I will draw on previous research in the field of home advantage and apply it to shootouts. I propose that both crowd and non-crowd factors (location and cognitive factors) which cause home advantage during regular football games, are also applicable to penalty shootouts and show a similar effect. More specifically, I hypothesise that attendance, crowd density, elevation, cultural differences, the round of the competition and the age of the players enhance the effect of home advantage in shootouts. These hypotheses are tested by analysing data from 238 shootouts (2566 penalty kicks) from 21 professional competitions. The hypotheses were not supported, suggesting that there is no home advantage in penalty shootouts. Instead, the results indicate that the away team is more likely to win the shootout.
1. Introduction
Football is one of the most-watched sports in the world. Around 3.5 billion people watched the FIFA World Cup 2018 held in Russia (FIFA, 2018) and 306.4 million people witnessed the underdog team and host Russia (ranked 70th 1) eliminating the 2010 World Champions Spain (ranked 10th1) in a series of penalty shootouts in front of 78,011 people and moving on to the Quarterfinals (FIFA, 2020a) (FIFA, 2020b). Was this historic moment based on luck, or was there a potential home advantage? If so, what caused this home advantage?
While the effect of home advantage has been researched thoroughly in the past, less attention has been paid to penalty shootouts. As the example above shows, we live in an age where underdog teams find a way to stall the game over 120 minutes and therefore make it to penalty shootouts. Football has evolved to be more strategic and defensive, which leads to fewer goals being scored and therefore, penalty shootouts occur more often (Geisler & Leith, 1997). For instance, before the World Cup in 1982, there was not a single penalty shootout. Ever since they have occurred in every single World Cup and in 1994 as well as 2006 the finals were decided in a penalty shootout. Also, the frequency of penalty shootouts during a single tournament has increased over time. Three out of the four past FIFA World Cups 1 Based on Coca Cola/FIFA Ranking on June 7th 2018
included four penalty shootouts respectively, resulting in 25% of the respective World Cup knock-out games being decided through penalties.
Besides, the prize pool of football tournaments has increased significantly. To demonstrate this, at the 1990 FIFA World Cup in Italy, the prize pool was 54 million US Dollars, and it has increased in every upcoming tournament (Statista, 2018). Just between the 2014 tournament in Brazil and the most current one in Russia, the prize pool increased from 576 million to its current peak of 791 million US Dollars, leading to an increase of a 215 million US Dollars. With this trend in mind, it is expected that unseen amounts of money will be at stake in future tournaments, and football teams will try to increase their chances of getting a share of the prize by excelling in every possible field of football, including penalty shots.
The findings will have managerial implications for football federations and associations to determine who gets the right to host a game. There are two types of tournaments that might be influenced by this effect. There are tournaments where there is only one game between the teams (e.g. Dutch KNVB Cup), and there are tournaments, where teams play two games, one at home and one away (e.g. Champions League). In the latter, it is to determine who gets to play at home in the second leg match (the one where the penalty shootout might occur). These procedures are not only used in football but also in many other sports (e.g. basketball, handball, ice hockey). If a home advantage exists and is strengthened by adjustable factors such as location factors, federations might have to adjust their policies for determining the host to make games more balanced, if this is what they aim for. Likewise, this would have implications for the teams themselves who then might have to adjust their strategies when playing away (e.g. playing riskier to avoid penalty shootouts). It will be interesting to find out what this research can teach us about the causes of home advantages in sports in general and whether there are solutions to make competitions more balanced or intentionally giving weaker teams the chance to get further by letting them host games or tournaments. Likewise, the findings have broader implications for performing in front of a hostile or friendly crowd (e.g. concerts, talent shows, speeches).
Consequently, it is essential to explore whether there is a home advantage during penalty shootouts, more importantly, since some federations give weaker teams the right to host the game (e.g. German DFB Pokal). This paper will answer how playing at home in connection to crowd and non-crowd factors (e.g. location and cognitive factors) influence the
outcome of penalty shootouts in football. First, I will outline findings on which factors have an impact on penalty shootouts. Afterwards, I will define home advantage and present which factors have been found in previous research. Later, the procedure of analysis will be
described, followed by a presentation of the results and its implications.
2. Literature Review
2.1 Penalty Shootouts
During the knock-out stages of tournaments, teams occasionally participate in a penalty shootout, which occurs if no team has been able to score more goals than their opponent after 120 minutes of play (Kocher, Lenz & Sutter, 2012). A penalty kick has an average success rate of 74.9 per cent (Chiappori, Levitt & Groseclose, 2002). Previous research examined which factors influence the outcome of shootouts.
Jordet et al. (2007) stated that the outcome of penalties is associated with the
importance of the kick, which is a proxy for stress, rather than by other factors such as skill (measured through position and age) and fatigue (exhaustion depending on how long a player has been on the field). The more important the kick was, the less likely they were to convert it. Furthermore, Geisler and Leith (1997) proposed that the success depends on how much players are concerned about how they present themselves and that penalty shootouts are deemed as mental skill, and not as a routine ability. Baumeister (1984) supports this as people switch from an automated state to a controlled state, which is not preferred during penalty kicks as they normally are an automated process. Furthermore, this effect is moderated by dispositional self-consciousness. Also, he suggested that pressure, which can be caused by factors such as rivalry, audience and money, decreases their performance. Both rivalry and audience factors have been researched in the past and have also been linked to the effect of home advantage.
Plenty of research has been conducted on the topic of possible home advantage. Then again, these studies focused on home advantages during football games as well as penalties that are taken during the game rather than during a penalty shootout (Jamieson, 2010; Pollard, 1986; Pollard & Gomez, 2014; Nevill & Holder, 1999; Schwartz & Barsky, 1977; Goumas, 2013; Nevill, Newell & Gale, 1996; Edwards & Archambault, 1979; Pollard, 2006; Pollard, 2008; Dohmen, 2008). This research will examine whether the home advantage effect is also present during shootouts and whether previously observed home advantage factors hold in these. The following section will outline these past findings.
2.2 Home Advantage
Home advantage has been defined as: “The consistent finding that home teams in sports competitions win over 50% of the games played under a balanced home and away schedule.” (Courneya & Carron, 1992, p.13). As Jamieson (2010) states, the phenomenon of home advantage seems to appear in every sport in the world (average approx. 60%), though football shows the strongest home advantage at 67.4% compared to other sports worldwide.
Meanwhile, Pollard (1986) observed that teams playing at home get approximately 64 per cent of the distributed points. This research was based on data of the English Football League (now English Premier League). While their numbers differ, researchers agree that there is a significant home advantage of at least 60% in football. Furthermore, both Jamieson (2010) and Pollard (1986), and later Pollard and Gomez (2014), stated that home advantages exist in all parts of the globe, no matter how strong or weak the competition is. Nevertheless, there are substantial differences in the strength of this effect across regions (Pollard & Gomez, 2014).
It is not clear whether this underexamined effect holds for penalty shootouts though. If a crowd can influence a player’s behaviour negatively during the game, where players are actively participating in the game and thus perceive less of the crowd, then there is a possibility that crowds can also affect the shooter’s behaviour during penalty shootouts, where players are waiting a long time before they get to shoot. Thus, they get confronted with the situation more than during in-game scenarios. This pattern would also support the finding of Baumeister in 1984.
However, some findings are not in support of home advantage, mainly when applied to penalty shootouts. For example, research by Pollard (1986) suggested that players might be putting in more effort when playing at home. Nonetheless, while trying harder during the game is plausible, it is not possible to try harder during a penalty shootout since it is one narrowly defined task where effort might not lead to an improvement. Moreover, it could also have a reverse effect on home players due to increased pressure. For instance, Harb-Wu and Krumer (2019) observed that competing in the presence of a supportive crowd leads to choking, a phenomenon which they label as “overcautiousness”. Supportive of this, Dohmen (2008) found that football players have a higher likelihood of underperforming during penalty attempts when they play at home, which has been linked to distractions when being at home such as family and social contacts (Harb-Wu & Krumer, 2019).
Also, during a penalty shootout, the game switches from a group competition to an individual competition (kicker versus goalkeeper). As a result, one could argue that the examined effect of home advantage during the game cannot be applied on penalty shootouts. Nevertheless, this is not of importance according to Jamieson (2010), as home advantages are the same on both the group and individual level, and consequently, the same effect should hold for penalty shootouts. Therefore, this research aims to answer the research question of whether there is a home advantage in penalty shootouts. Besides, possible interaction effects by crowd, location and cognitive factors will be examined. The following sections will look into factors that try to explain the circumstances under which the home advantage might occur.
2.3 Causes of the Home Advantage
The exact source of the home advantage is controversial. Most researchers suggested that this effect is mainly caused by factors associated with the crowd, but some findings hypothesised non-crowd-related factors to be of significance. Pollard and Pollard (2005), for example, suggested that there are many factors such as the crowd, territoriality and travel that cause home advantage. Nonetheless, they argue that psychological factors interact with all the above factors, and the more a player believes in the home advantage and its causes, the stronger the overall advantage. Additionally, some researchers argued that there are cognitive and
hormonal factors that play a role in explaining home advantage. Neave and Wolfson (2003), for instance, reported that testosterone concentrations were higher for home players than for the visiting team before the competition. Carré et al. (2006) are in line with these findings, as they found dissimilarities in mental and hormonal conditions between home and away players before the game. According to them, home players were more self-assured and away players had higher levels of perceived nervousness. If the theories of self-presentational concerns (Geisler & Leith, 1997) and self-consciousness (Baumeister, 1984) hold, this should lead to home players being more successful in penalty shootouts than away players. Thus, I
hypothesise that there is a home advantage during penalty shootouts in football.
H1: There is a Home Advantage during Penalty Shootouts in Football.
In the following sections, I will determine moderators based on previous findings on home advantage. The goal is to find out which factors influence home advantage during penalty shootouts. These factors have been split into the categories of crowd factors and non-crowd factors.
2.3.1 Crowd Factors
Crowd factors are the factors that are related to the audience physically present at the stadium. 2.3.1.1 Attendance
According to Nevill and Holder (1999), these are the most relevant factors of home advantage. Schwartz & Barsky (1977) found that especially for home teams, attempts are more successful when the number of spectators is higher. Pollard and Gomez (2014) support this idea, as they named the official FIFA ranking of the team to be the significant factor that influences advantages during home games. The ranking is an indicator variable for crowd support, and they argue that the higher a team was ranked, the bigger the crowd size and their respective support. Goumas (2013) came to a similar conclusion as he stated that the home advantage is becoming larger with the size of the audience, however, this increase was only observed up to an attendance of 20.000. Additionally, research on in-game penalties, rather than penalties during shootouts, by Nevill, Newell and Gale (1996) also supports the argument that the home advantage is significantly related to the average attendance. According to them, penalty outcomes were more successful for home teams with higher attendance. One possible explanation for this effect, as presented in their paper, is that a larger amount of people is more capable of inciting an away player, who then consequently tends to show irresponsible behaviour on the field than a smaller crowd is capable of. The idea is that a larger crowd is more supportive in favour of the home team.
Conversely, other observations have shown that the size of the audience is not relevant for the outcome of a penalty shot, even though excitement increased with the audience size. Geisler and Leith (1997) found that the outcome of penalty shots was not affected by the audience. Nevertheless, their study only involved a maximum of five staged audience members, rather than real game scenarios with supporting/hostile crowds. Football players regularly practice penalties during their practice while their team members and staff are watching. This number alone is already higher than the audience in Geisler and Leith’s experiment. During games, however, the number of spectators reaches a different level. Therefore, it is likely that an effect is only to be observed once higher amounts of people are within the crowd.
Furthermore, Jordet et al. (2007) found that the more notable a competition was (e.g. FIFA World Cup), the less goals were scored. This, however, does not necessarily have to be linked to the importance and the underlying stress of the tournament but could be due to the usage of bigger stadiums for bigger tournaments. As a result, more people attend these events.
As Pollard (1986) observed, bigger tournaments had more prominent home advantage effects which he based on bigger tournaments having bigger crowds. The recent constructing of new stadiums in future host countries of the FIFA World Cup, such as the United States and Qatar, emphasises this trend (Conn, 2018). Baumeister, Hamilton and Tice (1985) proposed that due to the crowd’s high hopes, outcomes are negatively affected in jobs that depend on skill. Larger amounts of spectators should be able to have an actual impact on individual players. Additionally, recent statistics of the German Bundesliga provide evidence that the audience is crucial to home advantage. The competition had to continue without any spectators due to the Covid-19 breakout, and home teams won only 12 out of 56 games, resulting in a 21.4%-win rate (Grohmann, 2020). Accordingly, attendance seems to be a factor of home advantage.
H2: Home Advantage during Penalty Shootouts increases with the size of the audience.
However, as stated earlier, researchers suggested that there are other factors besides crowd size that might play a role and therefore the next sections will present alternative explanations (Pollard, 2008; Pollard & Pollard, 2005).
2.3.1.2 Crowd Density
Crowd density, which is the degree to how exhausted the stadium capacity is, is another possible contributors to the effect of home advantage, as mentioned by both Jamieson (2010) and Edwards and Archambault (1979). It could be argued, that the fuller the stadium looks, the more supported the home team feels and the more distracted the away players feel.
H3: High Crowd Density increases Home Advantage during Penalty Shootouts.
2.3.2 Non-Crowd Factors
Pollard (1986) took a different stance and explained that the reason behind this effect is associated with the advantages of playing at home, which are not easy to measure, rather than straightforward factors such as the number of spectators. Different researchers have tried to find possible explanations which are not linked to the number of attendees.
2.3.2.1 Location Factors
Moreover, Pollard (2006), and later Pollard and Gomez (2014), found that more elevated regions (e.g. Andes compared to BeNeLux) and countries with a history of major conflicts (e.g. Balkans/Soviet States compared to Scandinavia) were more likely to win games at home. Pollard (2006) argues that regions that fall under these categories have a higher sense of “territoriality” and thus, players feel like they must win their home games. Additionally, it could be argued that the fans are more patriotic and more supportive of the home team.
H5: The Home Advantage is higher in Western European Cultures than in Eastern European Cultures.
2.3.2.2 Cognitive Factors
The following section will include non-crowd factors which are not dependent on the location, but instead, have to do with the circumstances under which a game takes place and subsequently cannot be influenced.
2.3.2.2 A) Stage
Jamieson (2010) found that games that are classified as “high-pressure games” yield a greater home advantage effect compared to “lower pressure games”. Pressure and importance can be associated with the stage of the competition. Consequently, Pollard (1986) found that the home advantage was more prominent, the later the phase of the tournament. This finding was based on data of the European Cup, which is equivalent to today’s UEFA Champions League.
H6: The Home Advantage is higher the later the Stage of the Competition.
2.3.2.2 B) Age
On the word of Tice, Buder and Baumeister (1985), teenagers performed worse than young adults when being watched. Even though Jordet et al. (2007) observed that age has no effect on penalty outcomes, age might moderate the effect of home advantage as younger players perceive more pressure than older players and consequently choke more when playing against a hostile crowd. Moreover, current trends show that football clubs are emphasising youth development and pay previously unseen amounts for transfers of young players. Together with their increased presence on social media and the press, young players are more likely to be in the spotlight than older players and therefore, might perceive more pressure. This could especially be the case during penalty shootouts, where performance is easier to quantify than judging them based on their game performance. Research on self-focus by Enright, Shukla and Lapsley (1979) suggests that adolescents become more focused on the environment and others while they grow. It could be argued that younger players are still transitioning between these stages and therefore might pay too much attention on how they perform on the penalty kick, rather than perceiving the home fans and seeing them as an advantage. Together with Pollard and Pollard’s (2005) finding on believing in the home advantage and its causes and Baumeister’s (1984) self-consciousness theory, it could be hypothesised that younger players do not make as much use of the home advantage as their older team members.
3. Methods
3.1 Data
This research used data from the knockout-stages from the Round of 16 onwards from 17 European domestic competitions, two intra-continental club competitions (Champions League and Europe League) and both the European Championship and World Cup during the period from 2008/09 until 2019/2020 (see Table 1). During this period there have been 238 penalty shootouts and a total of 2566 individual penalty attempts. Ten Bachelor’s Business students from the University of Amsterdam and their supervisor have allocated each person
competitions for data collection and afterwards contributed their data to create a dataset. The data was mainly collected from www.transfermarkt.com, and in some cases backed up by video material from www.youtube.com.
Furthermore, the dataset has been split into two separate datasets. The first dataset is used to analyse the team’s performance in the penalty shootout and consequently determine the home advantage effect. The second dataset includes all individual penalty attempts and will be used to determine the effect on the penalty kick performance. In both datasets, games that were played on neutral ground (44 shootouts) have been excluded in order only to analyse games that have a designated home team. Each hypothesis and its respective
performance outcome will be examined both under the individual penalty kick and the penalty shootout condition.
3.2 Variables
The dependent variables were scored (used for individual attempts) and winner (used for penalty shootouts). The independent variable was home (1 = home team, 0 = away team) to signal whether the team/player was on playing on home ground. Six moderator variables were examined regarding their influence on the home advantage relationship. The first moderator was attendance which showed how many people attended the game. The second moderator was the ratio crowd density and was computed by dividing attendance by capacity. It shows how full or empty the stadium was in relation to its maximum capacity. In eight cases, the outcome was higher than 1.00. Their attendance was higher than the official capacity of the stadium, and accordingly, was adjusted downwards to match the stadium capacity. The third moderator was elevation. It was based on the average elevation of the country in which the competition took place (see Table 2). The fourth moderator region grouped games into
“West” and “East” according to the culture of the country in which the competition took place (see Figure 3). Given that the competitions Europe League, Champions League, European
Championship and World Cup are all international competitions, these competitions will be excluded for the analysis involving the latter two variables. The fifth moderator was stage where the games have been split into the stages “Last Sixteen”, “Quarter Finals”, “Semi-Finals” and ““Semi-Finals”. Age was the sixth moderator, and analysis has been conducted under the individual condition only.
3.3 Analysis/Procedure
Crosstabulations was used to calculate the home winning percentage and the home/away player conversion rates. Furthermore, descriptives and correlations were analysed through SPSS. Logistic regressions via SPSS was used to examine univariate associations between the independent and dependent variables. Also, multivariate logistic regressions were analysed to rule out effects by control variables. For the team-based condition analysis, two control variables (favourite, and starting) were used. For the individual-based condition, 13 control variables were used in total. Besides the ones that have been used on the group level, the following additional ones have been used: stage (with dummy variables for last sixteen, quarters, semis, and finals), attendance, capacity, exhaustion, kick number, age, left-footed, field goal, momentum (with three dummies), negative valence and positive valence. Finally, multiple moderator analyses were conducted to examine whether the proposed factors enhance the effect of home advantage.
4. Results
4.1 Descriptives
During the examined 12-year period, 194 penalty shootouts (2115 penalty kicks) have taken place on non-neutral ground in the analysed competitions. The mean attendance was
18,923.56 (SD = 19,601.93; N = 186) and the mean crowd density 0.615 (SD = 0.301; N = 186). On average, a player taking a penalty kick was 27.39 years old (SD = 4.415; N = 2074). The most shootouts took place in the English League Cup (26), while the Portuguese Taca did not include any games on a team’s home ground. However, most penalty kicks were taken in the Russian Kubok (278). The descriptives for each respective competition can be found in Table 4.
The home win percentage was 46.91% and varied under different conditions. The highest home win percentage was to be found in flat countries (53.33%), while the lowest one was present in Semi-Finals (30.77%). The different win percentages can be found in Table 5. On average, 73.1% of the penalties have been converted. The lowest home player conversion
rate (64.95%) was also found in Semi-Final games. However, most home penalty kicks were converted (74.62%) when the player was younger than 23. The home and away player conversion rates under different circumstances can be found in Table 6. Under the team condition, there were no significant correlations between the variables (see Table 7a). Under the player condition there were some significant results as can be seen in Table 7b. There is a very weak positive relationship between scoring and the Round of 16 and a very weak
negative relationship between scoring and Semi-Finals. 4.2 Logistic Regression
Home Advantage (HA) Effect:
In hypothesis 1, I predicted that the effect of home advantage would hold for penalty
shootouts. Table 5 shows a slight disadvantage for home teams (46.91%-win rate). However, a logistic regression analysis under the team condition does not show a significant (OR = 0.768, P = 0.374) effect for an advantage for either team (see Table 8a). Under the individual condition, when playing at home, a home player converts 71.51% of the penalties in a
shootout, while the away player converts 74.67%. Table 9a shows that there is a significant difference in shot performance between home and away players (OR = 0.777, P = 0.027). This means that players perform worse on penalty kicks in shootouts when playing at home. Hypothesis 1 is not supported under both conditions.
4.3 Moderations
The following sections will present the results for the moderation analyses, which can be found in Table 8b (for team condition) and Table 9b (for player condition) respectively.
Moderation Attendance on HA:
To examine hypothesis 2 with attendance as moderator on home advantage, shootouts with less than 30,000 spectators and games with 30,000 or more spectators were compared. Under both the team and player condition, home is not significant in the first model (Team: OR = 1.114, P = 0.752; Player: OR = 0.826, P = 0.139), but in the second one it is (Team: OR = 0.297, P = 0.086; Player: OR = 0.561, P = 0.029). This means that playing at home is
negatively related to performance when there is a large crowd. Therefore, Hypothesis 2 is not supported.
Moderation Crowd Density on HA:
To examine hypothesis 3 with crowd density as moderator on home advantage, shootouts with less than 85 per cent stadium capacity exhaustion and games with 85 per cent or more have been compared.
(Team: OR = 1.198, P = 0.630; Player: OR = 0.830, P = 0.175), but in the second one it is (Team: OR = 0.344, P = 0.074; Player: OR = 0.570, P = 0.014). This means that playing at home is negatively related to penalty shot performance under both conditions when the crowd density is bigger. Thus, Hypothesis 3 is not supported.
Moderation Elevation on HA:
To examine hypothesis 4 with elevation as moderator on home advantage, shootouts in
competitions that fall under different elevation categories (flat, moderate, and high) have been compared. Under the team condition, home is not significant in both the first model (OR = 1.446, P = 0.409) and the second one (OR = 0.344, P = 0.125), but in the third one it is significant (OR = 0.300, P = 0.084). This means that playing at home is negatively related to team performance when the competition is in a highly elevated country.
Under the player condition, home is not significant in the first model (OR = 0.947, P = 0.753), but it is significant in both the second (OR = 0.599, P = 0.070) and third model (OR = 0.560, P = 0.017). This means that playing at home is negatively related to penalty shot
performance when the competition is in a moderately and highly elevated country. Hypothesis 4 is not supported under either condition.
Moderation Region on HA:
To examine hypothesis 5 with region as moderator on home advantage, shootouts in competitions that are within countries of different cultures (West and East) have been compared. Under both the team and player condition, home is not significant in the first model (Team: OR = 1.009, P = 0.981; Player: OR = 0.812, P = 0.137), but in the second one it is (Team: OR = 0.334, P = 0.096; Player: OR = 0.597, P = 0.038). This means that playing at home is negatively related to performance under both conditions in Eastern European
competitions. Hypothesis 5 is not supported.
Moderation Stage on HA:
To examine hypothesis 6 with stage as moderator on home advantage, shootouts in different stages of the tournament have been compared. Again, under both the team and player condition, home is not significant in the first model (Team: OR = 1.031, P = 0.861; Player: OR = 0.871, P = 0.302), but in the second one it is (Team: OR = 0.196, P = 0.023; Player: OR = 0.554, P = 0.015). This means that playing at home is negatively related to performance under both conditions in Semi-Finals. Hypothesis 6 is not supported.
Moderation Age on HA:
groups have been compared. In the first model (OR = 1.981; P = 0.042) and in the second one (OR = 0.661; P = 0.007) home is significant. However, in the third model it is not significant (OR = 0.836, P = 0.425). This means that players in the youngest age group (up to 22 years) are performing better when playing at home than players from the middle age group (23 – 29 years). Players from the oldest age group (30 and above) do not perform better or worse when playing at home compared to the other two groups.
5. Discussion
As can be seen in the results, all proposed hypotheses did not find support. However, under specific conditions, effects in the opposite direction were observed and thus leading to an “away advantage” instead. Penalty shootouts seem to be very different when it comes to home advantage compared to games that are decided beforehand. Still, the factors that cause a home advantage in regular games seem to play a role. Next, I will go through the hypotheses and offer possible reasoning for the observed effect. Afterwards, I will discuss the limitations of this study. Then, future research and practical implications will be suggested. Lastly, I will answer the research question of whether there is a home advantage in penalty shootouts.
The first hypothesis suggested that there is a home advantage during penalty
shootouts. Our results show that home teams won 46.91% of the penalty shootouts, which is significantly lower than the in-game home advantage that lies within the 60-65% range according to previous findings (Jamieson, 2010; Pollard, 1986). This supports Harb-Wu and Krumer’s (2019) observation that performing at home can also have the reverse effect of a home advantage due to “overcautiousness”. Moreover, this undermines Baumeister’s (1984) argumentation that during penalty kicks the game switches from an automated state to a controlled one, although different from what was expected. Instead of putting the away players under pressure, the home fans become a pressure for the home players and thus their performance decreases. Also, it could be argued that the team did not make use of the home advantage during the game, and hence is already underperforming on the matchday itself. Additionally, they might feel like they have disappointed their fans by not winning during regular time. The observed winning chance and conversion rate of both home and away players in this study fluctuated under different conditions. These conditions will be discussed in the following sections to understand the overall finding better.
In hypothesis 2 and 3, I considered crowd factors to play an essential role in enhancing the effect of home advantage. As can be seen though, the home winning chance is only
35.90% in games with an attendance of more than 30,000 viewers and 39.66% in games where the stadium is filled by at least 85 per cent. By looking at Table 6, it can be implied that the decrease in winning chance for the home team is due to the home players
underperforming on the individual level, rather than the away players overperforming. While the away players’ scoring outcomes stay relatively stable (maximum variance from average conversion rate of 1.75% in either direction), the conversion rate for home players drops significantly from 72.30% to 69.71% (H2) and 73.43% to 67.38% (H3) respectively. These findings contradict with the findings by Geisler and Leith (1997), who stated that the audience does not have an impact on penalty outcomes. Furthermore, research which suggested that increasing crowd factors would lead to a higher home advantage (Nevill & Holder, 1999; Pollard & Gomez, 2014; Goumas, 2013; Pollard, 1986) seems to not be applicable in penalty shootouts, at least when it is about enhancing the home advantage effect. Besides, Schwartz and Barsky’s (1977) observation that penalty kicks during the game were more successful the larger the crowd was is not translatable to penalty shootouts. This suggests, that especially when many fans are spectating or when the usual home ground is almost fully sold out, home players are under pressure not to disappoint their fans on home ground. Consequently, it can be assumed that psychological factors might play an important role in penalty shootouts.
Hypothesis 4 predicted that in higher elevated countries, the home advantage would be higher than in low or moderately elevated countries. Nevertheless, the highest home
advantage of this study was observed for flat countries (53.33%), while this percentage dropped for moderately (42.55%) and high elevated countries (38.64%). It seems that Pollard and Gomez’s (2014) finding has a reverse effect and the proposed territoriality puts the home players under fear to lose in front of their patriotic home crowd, at least when trying to explain the underperformance of home players in highly elevated countries (68.46%). Moreover, it can be argued that physiological factors that would usually lead to a home advantage in elevated countries, such as it being harder to breathe for the away players, are being overshadowed by the fear of losing at home. However, under the moderately elevated condition, the decrease in home advantage is not due to home players underperforming, but instead away players being very successful (home players: 72.27%; away: 78.38%). Reasons for this overperformance of away players under this condition have yet to be found.
Pollard’s (2006) reasoning is further supported through the finding for Hypothesis 5, where Eastern European cultures showed a lower home win percentage than Western
territoriality, which would explain these results. As can be seen in Figure 3, the sample does not include data from other regions such as Scandinavia, the Baltics or Balkans. Also, non-European competitions (with the World Cup being an exemption) were not examined in this study. Therefore, further research should include other regions to tell us more about home advantage differences among other regions about penalty shootouts.
Hypothesis 6 is about the change in home advantage during different stages of the tournament. As the results indicate, home teams are less likely to win penalty shootouts during semi-final games compared to other stages. In fact, the home winning percentage in semi-finals is the lowest one observed in this study (30.77%). Consequently, these findings conflict with Pollard’s (1986) and Jamieson’s (2010) research and undermine that home advantage in penalty shootouts is different from regular games. Similarly to the previous reasoning, in this round, the pressure to not disappoint the home crowd is even higher than in previous stages. In many cases, the final is played on neutral ground, and these stages usually take place after the regular league competition has ended. Thus, it is likely to be one of the last home games of the season for the respective team, and there is additional pressure to perform well. Additionally, it might also be one of the last game of players that are switching clubs after the season. In order to see if this could explain the results, further research needs to be conducted. This could include analysis involving the date of the games and whether a player is leaving after the season. Also, exhaustion towards the end of the season does not explain this effect, as away players are performing around the average level (74.19%), while home players underperform (64.95%). A limitation to this moderator is that final games, which should theoretically put even more pressure on players, are excluded in the analysis concerning stage as moderator. However, as stated above, most finals are played on neutral ground, and therefore the sample was too small to be considered.
Finally, Hypothesis 7 compared whether different age groups converted more
penalties when playing at home. Indeed, the results indicated significant results, however not supportive of the hypothesis. Players under the age of 23 performed better at home than their older team members in the age group 23 to 29. This finding is not in line with Tice, Buder and Baumeister’s (1985) observation. One possible explanation for the effect is that younger players get selected for shooting this early in their career only when they are known to be precise shooters. To examine this, further research could be conducted by differentiating between young players that took one of the first five penalty kicks (and thus were selected) and the ones that took later penalty kicks. Another explanation is that in today’s world
younger players get used to pressure earlier due to the press and social media. However, this can be ruled out as Table 6 shows that the young players outperform young away players, and therefore there are differences between young home and away players. Lastly, older players are more likely to have been part of the club for a longer time and therefore feel more pressure to perform well. Nonetheless, this does not explain why players above 30 years do not perform differently when playing at home. Consequently, additional research with data on tenure is suggested.
As mentioned earlier, one weakness of this study is that the sample is not representative of all competitions and regions. Nevertheless, this sample size is larger compared to previous research on penalty shootouts such as Jordet et al. (2007) (41 Shootouts/409 Kicks). Research by Kocher, Lenz & Sutter (2012) showed how crucial the sample size could be when they rejected previous findings of a supposed first-mover advantage during shootouts by increasing the sample size. Especially in football, a minimal change in the predicted outcome can have serious implications. Thus, with the increased sample size, new insights were gained. Another weakness is that some of the variable
outcomes were missing. Thus some cases were not considered in the analysis, yet again even after excluding these, the sample was larger than in earlier research. Furthermore, for the analysis concerning elevation as moderator, the average elevations of the countries were used, which does not have to be reflective of the height of the stadiums in which the teams compete. Lastly, the different conditions that were compared used strict cut-off points and choosing other thresholds might have led to slightly different results. Therefore, the results in this study should be used with caution when inferring to greater effects. Nonetheless, the results still provide new insights on the topic. They can be used as a basis for future research, especially since home advantage concerning penalty shootouts is still an under-examined field.
Some suggestions for further research were given earlier in this section. In addition to these, it seems like there is still a lot to learn about this field since quantitative measures do not explain everything. Therefore, it is also recommended to conduct qualitative studies to find out more about the personality of the players that take the kick. Also, finding out what they think not only while taking the penalty kick, but also what they perceive while they wait and how it is possibly different from during the game, could be of assistance in learning more about penalty shootouts and home advantages. Additionally, further research on crowd factors could consider the position of the home and away fans within the stadium in future analysis.
There might be differences depending on whether a hostile or friendly crowd is in front of the player when taking a penalty kick.
This research shows that both crowd and non-crowd factors have an impact on the home team’s winning chances. Clubs should make sure to offer adequate psychological support for their players, as it seems that players face much pressure during shootouts. Staff should try to explain to them that playing at home can lead to advantage and that they should perceive it as an opportunity rather than being under pressure. As the results show, young players are better able to deal with the situation perform well under pressure. It is therefore recommended to work closely with the young players to find out what makes them perform better and help their older team members out. This is especially important regarding factors that cannot be influenced, such as the stage of the tournament. Whether or not associations should act and use these findings as a basis to pick the location for knock-out games, depends on their objective. This means they will have to decide on whether they want a more balanced competition or if they want to let smaller teams have an increased chance at succeeding. Also, the coaching staff might have to adjust their tactics during a game and decide whether they want to take more risks to avoid a penalty shootout or if they have an advantage coming into the shootout and therefore play more defensive.
Overall, the data shows a slight advantage for the away team. Under the above- examined conditions, the proposed moderators significantly enhance the advantage for away teams. The results show that in most cases this is due to the home players underperforming, rather than the away players overperforming and therefore the term “away advantage” does not entirely reflect this matter either. Nevertheless, it becomes evident that there is no home advantage during penalty shootouts in soccer, and there is still a lot to learn about the differences in home advantage between penalty shootouts and regular games.
References:
Baumeister, R.F. (1984). Choking Under Pressure: Self-Consciousness and Paradoxical Effects of Incentives on Skillful Performance. Journal of Personality and Social Psychology, 46(3), 610- 620.
Baumeister, R. F., Hamilton, J. C., & Tice, D. M. (1985). Public versus private expectancy of success: Confidence booster or performance pressure?. Journal of personality and social psychology, 48(6), 1447.
Carré, J., Muir, C., Belanger, J., & Putnam, S. K. (2006). Pre-competition hormonal and psychological levels of elite hockey players: relationship to the ‘home advantage’. Physiology & behavior, 89(3), 392-398.
Chiappori, P. A., Levitt, S., & Groseclose, T. (2002). Testing mixed-strategy equilibria when players are heterogeneous: the case of penalty kicks in soccer. The American Economic Review, 92(4), 1138-1151.
Conn, D. (2018, November 21). Qatar 2022: £40 a week to build the World Cup stadiums. The Guardian, Retrieved from
https://www.theguardian.com/football/2018/nov/21/qatar-2022- 40-a-week-to-build-the-world-cup-stadiums
Courneya, K. S., & Carron, A. V. (1992). The home advantage in sport competitions: a literature review. Journal of Sport & Exercise Psychology, 14(1), 13-27.
Dohmen, T. J. (2008). Do professionals choke under pressure?. Journal of economic behavior & organization, 65(3-4), 636-653.
Edwards, J., & Archambault, D. (1979). The home field advantage. Sports, games, and play:Social and psychological viewpoints, 409-438.
Enright, R. D., Shukla, D. G., & Lapsley, D. K. (1980). Adolescent egocentrism sociocentrism and self-consciousness. Journal of youth and adolescence, 9(2), 101-116.
FIFA. (2018). 2018 FIFA World Cup Russia Global broadcast and audience summary. Retrieved from https://resources.fifa.com/image/upload/2018-fifa-world-cup-russia- global-broadcast-and-audience-executive-
summary.pdf?cloudid=njqsntrvdvqv8ho1dag5
FIFA. (2020a). FIFA/Coca-Cola World Ranking. Retrieved from https://www.fifa.com/fifa- world-ranking/ranking-table/men/rank/id12210
FIFA. (2020b). 2018 FIFA World Cup Match Report Round of 16 Spain v Russia. Retrieved from https://resources.fifa.com/image/upload/eng-51-0701-esp-rus-fulltime-pdf- 2979634.pdf?cloudid=rix0h2ufiablhohc1hsj
Geisler, G. W., & Leith, L. M. (1997). The effects of self-esteem, self efficacy, and audience presence on soccer penalty shot performance. Journal of Sport Behavior, 20(3), 322.
Goumas, C. (2013). Home Advantage and Crowd Size in Soccer: A Worldwide Study. Journal of Sport Behavior, 36(4).
Grohmann, K. (2020, June 15). Premier League hosts beware – German data shows end of home advantage. Reuters. Retrieved from
https://www.reuters.com/article/uk-soccer- england-bundesliga/premier-league-hosts-beware-german-data-shows-end-of-home-advantage-idUKKBN23M2DS
Harb-Wu, K., & Krumer, A. (2019). Choking under pressure in front of a supportive audience: Evidence from professional biathlon. Journal of Economic Behavior & Organization, 166, 246-262.
Jamieson, J. P. (2010). The home field advantage in athletics: A meta‐analysis. Journal of Applied Social Psychology, 40(7), 1819-1848.
Jordet, G., Hartman, E., Visscher, C., & Lemmink, K. A. (2007). Kicks from the penalty mark in soccer: The roles of stress, skill, and fatigue for kick outcomes. Journal of Sports Sciences, 25(2), 121-129.
Kocher, M.G., Lenz, M.V. and Sutter, M. (2012). Psychological Pressure in Competitive Environments: New Evidence from Randomized Natural Experiments. Management Science, 58(8), 1585-1591.
Neave, N., & Wolfson, S. (2003). Testosterone, territoriality, and the ‘home advantage’. Physiology & behavior, 78(2), 269-275.
Nevill, A. M., & Holder, R. L. (1999). Home advantage in sport. Sports Medicine, 28(4), 221-236.
Nevill, A. M., Newell, S. M., & Gale, S. (1996). Factors associated with home advantage in English and Scottish soccer matches. Journal of sports sciences, 14(2), 181-186.
Pollard, R. (1986). Home advantage in soccer: A retrospective analysis. Journal of sports sciences, 4(3), 237-248.
Pollard, R. (2006). Worldwide regional variations in home advantage in association football. Journal of sports sciences, 24(3), 231-240.
Pollard, R. (2008). Home advantage in football: A current review of an unsolved puzzle. The open sports sciences journal, 1(1).
Pollard, R., & Gómez, M. A. (2014). Comparison of home advantage in men's and women's football leagues in Europe. European journal of sport science, 14(1), 77-83.
Pollard, R., & Pollard, G. (2005). Home advantage in soccer: A review of its existence and causes.
Portland State University. (2020). Country Geography Data. College of Urban and
Public Affairs. Retrieved from https://www.pdx.edu/econ/country-geography-data.
Schwartz, B., & Barsky, S. F. (1977). The home advantage. Social forces, 55(3), 641-661.
Statista. (2018). Total prize money for the FIFA World Cup from 1982 to 2018.
Retrieved from https://www.statista.com/statistics/328497/fifa-world-cup-prize-money/
Tice, D. M., Buder, J., & Baumeister, R. F. (1985). Development of self-consciousness: At what age does audience pressure disrupt performance?. Adolescence, 20(78), 301.
Appendix:
Table 1. List of included Competitions (2008/09 – 2019/2020) 1. TOTO KNVB Beker (The Netherlands)
2. FA Cup (England) 3. EFL Cup (England) 4. DFB Pokal (Germany) 5. Coppa Italia (Italy) 6. Copa del Rey (Spain) 7. Coupe de France (France) 8. Beker van Belgie (Belgium) 9. SFA Cup (Scotland)
10. Taca de Portugal Placard (Portugal) 11. Kubok Russij (Russia)
12. Türkiye Kupasi (Turkey) 13. Kypello Elladas (Greece) 14. Totolek Polish Cup (Poland) 15. MOL Cup (Czech Republic) 16. ÖFB Cup (Austria)
17. Helvetia Schweizer Cup (Switzerland) 18. UEFA Europe League (European Clubs) 19. UEFA Champions League (European Clubs)
20. UEFA European Championship (European National Teams) 21. FIFA World Cup (International National Team)
Table 2. Average Elevation per Country.
Country Average Elevation (in Meters)
Switzerland 1,350 Turkey 1,132 Austria 910 Spain 660 Russia 600 Italy 538 Greece 498 Czech Republic 433 France 375 Portugal 372 Germany 263 Belgium 181 Poland 173
United Kingdom (England & Scotland) 162
The Netherlands 30
Note. Data was obtained from “Country Geography Data” by College of Urban & Public Affairs: Economics, Portland State University (2020).
Figure 3. Regional separation by culture.
Blue: Western Red: Eastern
Table 4. Descriptives of the examined competitions.
Competition Number of Penalty
Shootouts
Number of Penalty Kicks
The Netherlands 18 230 England (FA) 3 30 England (LC) 26 248 Germany 13 136 Italy 16 168 Spain 3 26 France 19 196 Belgium 23 264 Scotland 5 44 Portugal 0 0 Russia 23 278 Turkey 10 103 Greece 10 133 Poland 2 20 Czech Republic 2 18 Austria 2 16 Switzerland 6 65 Europe League 6 62 Champions League 5 60 European Championship /World Cup 2 18 Total 194 2115
Table 5. Home Team Winning Percentages computed via Crosstabulations.
Variable Home Team Winning Percentage
All Observations (N = 194) 46.91 CROWD Attendance (N = 186) under 30,000 (N = 147) 30,000 and above (N = 39) 51.70 35.90 Crowd Density (N = 186)
under 85 per cent (N = 128) 85 per cent and above (N = 58)
LOCATION 52.34 39.66 Elevation (N = 181) flat (N = 90) moderate (N = 47) high (N = 44) 53.33 42.55 38.64 Region (N = 181) West (N = 134) East (N = 47) COGNITIVE 49.25 40.43 Stage (N = 190) Non-Semi (N = 151) Semi-Finals (N = 39) 50.99 30.77
Table 6. Conversion Rate Home and Away Players computed via Crosstabulations.
Variable Home Player Conversion Rate
(in %)
Away Player Conversion Rate (in %) All Observations CROWD 71.5 (N = 1,053) 74.67 (N = 1,062) Attendance under 30,000 30,000 and above 72.30 (N = 805) 69.71 (N = 208) 74.01 (N = 808) 76.42 (N = 212) Crowd Density
under 85 per cent 85 per cent and above
LOCATION 73.43 (N = 734) 67.38 (N = 279) 75.03 (N = 737) 73.14 (N = 283) Elevation flat moderate high 72.28 (N = 487) 72.27 (N = 256) 68.46 (N = 241) 72.78 (N = 485) 78.38 (N = 259) 74.49 (N = 247) Region West East COGNITIVE 72.01 (N = 711) 69.60 (N = 273) 74.44 (N = 712) 75.27 (N = 279) Stage
Last Sixteen/Quarter Final Semi-Finals 73.18 (N = 809) 64.95 (N = 214) 74.63 (N = 816) 74.19 (N = 217) Age Under 23 23 – 29 30 and above 74.62 (N = 142) 69.76 (N = 603) 72.76 (N = 295) 69.72 (N = 130) 76.45 (N = 542) 73.22 (N = 312)
Table 7a. Correlation matrix under team condition. Mean SD 1 2 3 4 1. Winner .48 .501 - 2. Home .56 .497 -.059 - 3. Favourite .46 .500 .082 .000 - 4. Starting .99 .101 .099 .116 -.079 -
Note. No significant correlations found at the P < 0.10 level (2-tailed).
Table 7b. Correlation matrix under individual condition.
Mean SD 1 2 3 4 5 6 7 8 9 10 1.Scored .73 .444 - 2 Home .50 .500 -.036 - 3.Last 16 .52 .500 .039* -.001 - 4.Quarter .23 .434 -.006 .000 -.577*** - 5.Semi .20 .403 -.040* -.001 -.528*** -.280*** - 6. Attend. 18535 18888 -.034 -.002 -.170*** -.038* .177*** - 7.Density .60 .295 -.020 -.001 -.207*** .108*** .159*** .567*** - 8.Elev. .75 .824 -.004 -.006 -.169*** .014 .117*** -.247*** -.280*** - 9. Region .28 .449 -.008 -.005 -.107*** -.048** .057** -.166*** -.280*** .679*** - 10. Age 27.39 4.415 .019 .014 -.030 .025 -.004 .129*** .076*** -.008 .032 -
*** Correlation significant at the P < 0.01 level. ** Correlation significant at the P < 0.05 level. * Correlation significant at the P < 0.10 level.
Table 8a. Multivariate associations regarding team performance.
Variable Odds Ratio (P-Value)
Home 0.768 (0.374)
Favourite 1.411 (0.242)
Starting 0.964 (0.903)
Table 8b. Moderation Analysis (Team Condition) on the Relationship (1) Playing at Home on (2) Winning
Moderator Winning Odds Ratio (P-Value) for Home
CROWD Attendance (N = 184) Under 30,000 30,000 and above 1.114 (0.752) 0.297 (0.086) Crowd Density (N = 184)
Under 85 per cent 85 per cent and above
1.198 (0.630) 0.344 (0.074) LOCATION Elevation (N = 177) Flat Moderate High 1.446 (0.409) 0.344 (0.125) 0.300 (0.084) Region (N = 177) West East 1.009 (0.981) 0.334 (0.096) COGNITIVE Stage (N = 186)
Last Sixteen/Quarter Final Semi-Final
1.061 (0.861) 0.196 (0.023) Note. All odds ratios are for the variable home in reference to away.
Table 9a. Multivariate associations regarding individual penalty kick performance
Variable Odds Ratio (P-Value)
Home 0.777 (0.027) Last Sixteen 1.543 (0.350) Quarter Final 1.312 (0.565) Semi Final 1.196 (0.705) Final 1.742 (0.337) Favourite 1.134 (0.264) Starting 0.996 (0.974) Attendance 1.000 (0.652) Capacity 1.000 (0.725) Crowd Density 1.054 (0.880) Kick Number 1.008 (0.798) Age 1.010 (0.432) Left-Footed 1.014 (0.911)
Scored Field Goal 1.002 (0.992)
Negative Valence 0.981 (0.917) Positive Valence 0.899 (0.623) Momentum One 0.941 (0.689) Momentum Two 1.066 (0.669) Momentum Three 1.104 (0.707) N = 1612
Table 9b. Moderation Analysis (Player Condition) on the Relationship (1) Playing at Home on (2) Scored
Moderator Scored Odds Ratio (P-Value) for Home
CROWD Attendance (N = 1612) Under 30,000 30,000 and above 0.826 (0.139) 0.561 (0.029) Crowd Density (N = 1612)
Under 85 per cent 85 per cent and above
0.830 (0.175) 0.570 (0.014) LOCATION Elevation (N = 1498) Flat Moderate High 0.947 (0.753) 0.599 (0.070) 0.560 (0.017) Region (N = 1498) West East 0.812 (0.137) 0.597 (0.038) COGNITIVE Stage (N = 1561)
Last Sixteen/Quarter Final Semi-Final 0.871 (0.302) 0.554 (0.015) Age (N = 1612) Under 23 1.981 (0.042) 23 – 29 0.661 (0.007) 30 and above 0.836 (0.425)
