Testing the effect of kick-off time in the UEFA Europa League

ABSTRACT

Research Question: The paper analyzes whether different kick-off times have an effect on attendance and performance of teams in the group stage of the UEFA Europa League.

Research Methods: I use data on all of the games from the group stage of the UEFA Europa League between the 2009/10 and 2015/16 seasons that took place in time zones CET and CET + 1. I estimate the average treatment effect of the kick-off time by using the distance-weighted radius matching approach with bias adjustment.

Results and Findings: There is less attendance as a share of the capacity of the stadiums in games that kick off at 21:05 CET compared to games that start at 19:00 CET. In addition, in games that begin later (21:05) a team with a lower UEFA ranking has a significantly lower home advantage compared to games that begin earlier (19:00).

Implications: The results of this study suggest that a lower ranked team that competes three times at home at 21:05 CET loses on average about 270,000 Euros from lower revenue from tickets and UEFA prizes. Additionally, a lower number of points reduces the chances of qualifying for the next round of competition, which is a much larger loss of at least 800,000 Euros, not counting all other revenue from tickets and advertising in more advanced stages of competition. Hence, it seems to be an important organizational task to adjust kick-off times in a way that eliminates any advantage driven by the schedule.

KEYWORDS:

• European soccer
• schedule
• fairness
• home advantage

Acknowledgements

I would like to thank Daniel Goller, Michael Knaus, Michael Lechner, participants of the sports management research seminar at Molde University College, and three anonymous referees for their helpful comments and suggestions. All errors are my own.

Disclosure statement

No potential conflict of interest was reported by the author.

ORCID

Alex Krumer http://orcid.org/0000-0002-3274-8078

Notes

1 With the exception of Eastern European countries where the time zone is later than CET + 2.

2 According to the 2016/17 financial report of the UEFA, the total broadcasting revenue in 2016/17 season was 2.12 billion Euros in the Champions League and 356 million Euros in the Europa League (available at https://www.uefa.com/MultimediaFiles/Download/OfficialDocument/uefaorg/Finance/02/54/02/87/2540287_DOWNLOAD.pdf).

3 Games that took place in time zones CET and CET + 1 represent 82.4% of the total number of games. For a detailed discussion on the choice of time zones, see the section that describes the data and variables.

4 In addition, Meier, Konjer, and Leinwather (2016) and Hogan, Massey, and Massey (2017) studied the attendance at women’s soccer leagues in Germany and rugby games respectively. They used schedule related variables in their regression analysis without discussing their roles in explaining the attendance.

5 It is worth mentioning Krumer and Lechner (2017) who found that the schedule of games in the group stage of the FIFA World Cups and the UEFA European Championships has an effect on the probability of qualifying for the next round.

6 See also Huber, Lechner, and Steinmayr (2015) who described this approach in detail and its implementation in various software packages such as Gauss, Stata and R.

7 Higher and lower ranked teams are defined according to the UEFA coefficients prior to the beginning of the tournament. For additional details, see the next section.

8 Over the years there have been many examples of the lowest seeded teams that played three home games at 21:05 CET. The examples include Molde FK from Norway, Timishoara and CFR Cluj from Romania, Thun from Switzerland, Venstpils from Latvia, Sheriff from Moldova and Asteras from Greece.

9 Note that the winning team receives three points, while the losing team gets no points. In case of a draw, each team gets one point.

10 The analysis was executed using the radiusmatch command in Stata 15.

11 One possible concern would be the risk of bias from censoring, since there are observations with the maximal share of capacity. However, there are only 7 observations (0.85%) with a share of capacity of 1. Therefore, there is no serious risk of bias from censoring.

12 For example, according to Novak (2014), Partizan Belgrad, one of the best Serbian clubs, has an annual budget of around 10 million Euros.