![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
ABSTRACT
Research question
Studies considering demand for professional boxing are almost completely absent from the Sport Management/Sports Economics literature. Little is known about consumer preferences for a sport which attracts global attention but is unique insofar as it is exempt from standard sporting institutions maintaining competitive balance. We use two new datasets to ask, what are the determinants of main event viewership (Nielsen ratings) and pay-per-views buys? In addressing this question we offer new insights on the uncertainty-of-outcome hypothesis and extend research on direct demand for individual sport.
Research methods
The datasets comprise of 210 HBO and Showtime broadcasts from 2006 to 2018. We estimate generalised linear models, controlling for economic determinants, bout features, boxer popularity and scheduling factors.
Results and Findings
For main events, we find conflicting evidence to the uncertainty-of-outcome hypothesis. Fans show a preference for rematches, domestic boxers and heavier divisions. NFL broadcasts and earlier scheduling negatively impact viewership. For pay-per-views, we find a positive price effect. Expectedly, Floyd Mayweather Jr increases buys. Both categories exhibit a negative trend with viewership and PPV buys declining over the sampling frame.
Implications
The results offer new evidence that is contrary to the uncertainty-of-outcome hypothesis for an individual sport. The findings point to differences in consumer preferences between two sources of demand for the same sport. Practically, these can inform the strategic decision-making of broadcasters, promoters, advertisers and potential new broadcasting entrants.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 Past research has focussed on the incentive effects related to contracts and rematches (Amegashie & Kutsoati, Citation2005; Tenorio, Citation2000). One unpublished study explored attendance and revenue (Balbien et al., Citation1981).
2 YouTube began experimenting with live streaming of PPV bouts in September 2017. In September 2018 the sports streaming platform DAZN entered the US market.
3 Like many elite sports, broadcasting revenue has become increasingly important relative to ticket sales, merchandising or sponsorship. This revenue growth is demonstrated by the increased frequency of PPVs. From 1975 to 1990, we estimate that HBO and Showtime broadcast just seven PPV events. The popularity of the PPV model grew in the 1990’s. Forty-six PPVs took place from 1990 to 2005, increasing to fifty-seven from 2005 to January 2019.
4 An example of this was seen when Floyd Mayweather faced Andre Berto. This was the last of a six-bout deal with Showtime and was allegedly Mayweather’s final contest prior to retirement. Achieving a comfortable victory was important as Mayweather was then free to negotiate the sale of his 50th bout to beat Rocky Marciano’s record of 49 victories.
5 Promoters typically earn the residual once a PPVs fixed costs are covered. Baglio (Citation1999) outlines the complexities of these negotiations. At times, the interests of boxers and their promoters are not compatible in a PPV setting as promotors may incur the costs of a higher purse.
6 Although Nielsen ratings are high quality, this data has limitations. We understanding that our Nielsen data did not capture viewership via smart phones, tablets or other digital devices. The data measures private consumption and does not account for viewers in out-of-home settings.
7 oddsportal.com is a betting odds monitoring service that importantly provides draw odds for boxing. It has become increasingly used in sports research (Butler et al., Citation2021).
8 This index is based on Theil (Citation1967).
9 Points are determined based on the value of the result (v: 0-1), the clarity of the decision (cd: 0-1) and the strength of previous opponents. For example, a knockout or technical knockout leads to the winning boxer earning v = 1, cd = 1. For a points win, disqualification or unanimous decision, v = 1, cd ≤ 1. When there is ambiguity (e.g. split decision), the winner earns v = 1, cd ≤ 0.5. For a draw v = 1, cd = 0. Points are awarded proportional to the rounds boxed and are weighted by the ratings difference prior to a bout. Ratings decline if a boxer is inactive, or does not box an opponent within at least 50 percentage points of his points total in the eighteen months previous. The ratings also control for transfers across weight divisions.
10 Popularity proxies such as social media supporters (e.g. Twitter followers) either did not exist at the time of the PPV or, in many cases, were only used by a limited number of boxers. At best, social media metrics would be unbalanced across boxers and likely crude given the years of the sampling frame. Other measures of popularity such as internet search volume tools also had limitations. For example, more recent boxers and bouts were disproportionally popular when we accessed search volume data. Furthermore, search volume tools often provided ambiguous data, based on average monthly searches or relative search volume.
11 This is the case when bout quality is included in the PPV model; the measure reports a VIF of 74.15. (11.34 for the complete model). This raises collinearity concerns.
12 Mayweather Jr was ranked as the highest-paid athlete in the world by Forbes in 2012, 2014, 2015 and 2018. Other exclusive PPV boxers over the sampling frame were experimented with but were not included in the final specification such as Oscar De La Hoya and Shane Mosley.
13 Past research has included PPV price as exogenous but noted the limitations (Reams & Shapiro, Citation2017; Tainsky et al., Citation2013; Watanabe, Citation2015).