Women’s volleyball demand across different distribution channels

ABSTRACT

Research question

In this study, we add to the slowly emerging literature on modeling the demand for women's sports by exploring the robustness of the determinants for women's volleyball – previously a largely neglected team sport – across three different distribution channels, i.e. online streaming, TV broadcasts, and the stadium experience.

Research methods

We exploit a data set containing information on consumer interest in the German Women's Volleyball League (VBL) in 2019. In line with the literature on modeling consumer interest in sports, we estimate on-/offline audiences and spectator demand, in terms of semi-elasticity, by Ordinary Least Squares (OLS) regressions.

Results

We find that volleyball demand is relatively stable across two of the three distribution channels (i.e. online streaming and TV broadcasts) and, to a lesser degree, the stadium experience. This pattern is particularly problematic for a niche sport's online streaming offerings that, often initiated with high hopes, seem to attract only a fraction of the otherwise relatively strong consumer interest in the sport's TV offerings.

Implications

While we believe that future research is well-advised to add nuance to the analysis of (women’s) sports demand across different distribution channels, our results also suggest that sporting bodies should carefully weigh the pros and cons of an exclusively online distribution strategy.

KEYWORDS:

• Attendance
• outcome uncertainty
• niche sports
• streaming
• television

Acknowledgments

This research is based on a chapter in the doctoral dissertation of the first author under the supervision of the third and fourth author. We would like to thank Daam van Reeth for helpful comments on an earlier draft of the manuscript, as well as participants in the Reading Online Sport Seminars (ROSES) in December 2021. Further, we gratefully acknowledge the excellent comments and suggestions from two reviewers, as well as an associate editor, all of which have helped us improve the manuscript’s quality. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Availability of data

The data that support the findings of this study are available on request from the corresponding author.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Unlike traditional media services distributed over the air, OTT media services are offered directly to viewers via the internet, bypassing broadcast, cable, or satellite TV platforms.

2 However, as one reviewer has pointed out, this does not mean that factors affecting niche sports demand or, even broader, growth strategies have not been discussed, to some degree, previously. In this regard, Meier et al. (2020a), for instance, have discussed the potential role of aggregation as a remedy for the decline of niche sports broadcasting in times of digital plentitude. Other exceptions include Jensen et al. (2014) and Tassiopoulos and Haydam (2008).

3 In 2020, due to the COVID-19 pandemic, there was no champion crowned.

4 VBL did not announce the actual license fee paid by the broadcasters.

5 Further, clubs benefit from reduced production costs as they no longer have to cover them.

6 We purchased TV data from the Gesellschaft für Konsumforschung, which conducts representative TV ratings for the German TV industry. The data is/are generated from a representative sample of 5,640 German and European households. In contrast, Sporttotal shared data free of charge. Unfortunately, our data does not capture the demand by the minute. Only a few authors have exploited such nuanced data, including Alavy et al. (2010), Chung et al. (2016), and Rodríguez-Gutiérrez and Fernández-Blanco (2017). Others have worked with 15-minute intervals (e.g. Sung et al., 2017).

7 We had to drop observations, i.e. the matches between Rote Raben Vilsbiburg and MTV Stuttgart (January 9th) and Dresdner SC and SC Potsdam (October 5th), due to technical problems by our data provider, which, unfortunately, resulted in missing viewer ratings.

8 Accordingly, in an additional, though omitted, specification, a simple dummy capturing whether a fixture was shown on Sport1 or Sporttotal explains about 90 percent of the variance in audiences’ demand.

9 Although we observe sold-out matches, and estimating an OLS might, therefore, produce biased estimates (e.g. Amemiya, 1973), these sell-outs are relatively scarce – only about eight percent of all 139 VBL matches.

10 Initially using a two-stage Heckman model (e.g. Buraimo & Simmons, 2015), we found no evidence for selection bias due to the leading broadcaster’s choice.

11 To calculate APD, we transform betting odds into adjusted probabilities, excluding the bookmakers’ margin (cf. Benz et al., 2009).

12 After every match, the three best players are awarded a gold, silver, and bronze medal. Based on these awards, an overall league MVP ranking is calculated at the end of the season.

13 We conducted additional analyses to examine the robustness of our findings. Specifically, we operationalized a match's start time as a factor variable, categorized as before/at/after 7 pm. Consistent with our original results, this alternative approach, however, did not result in any significant changes to the size or direction of our estimates.