The importance of suspense and surprise in entertainment demand: Evidence from Wimbledon

Abstract

This paper empirically examines how suspense and surprise explain demand for entertainment. We use the Wimbledon Championships tennis tournament as a natural laboratory.This setting allows us to both operationalize suspense and surprise using the audience's belief about the final outcome of the match and observe of the demand for live entertainment using JEL Classification: D83, L82, L83

Unscripted Drama: Soccer Audience Response to Suspense, Surprise, and Shock

“…Results are exhibited in Table 3. In similar experimentation, Bizzozero, Flepp, and Franck (2016) found that the slopes of suspense and surprise increased in the later sets of a tennis match but we do not have analogous results here. A possible reason is that football viewers may make a firmer commitment of time to watching the entire event.…”

“…As in the study of Swiss television audiences for tennis (Bizzozero, Flepp, and Franck 2016), our modeling provides strong support for the hypothesis in Ely, Frankel, and Kamenica (2015) that suspense and surprise are ingredients in driving entertainment demand. And shock also seems to matter in the case of football.…”

“…Their use of the term is therefore not the same as in Ely, Frankel, and Kamenica (2015), who were of course writing several years later. In all the following text, we use the terms suspense and surprise consistently in the same sense as Ely, Frankel and Kamenica. Although related to a different sport, Bizzozero, Flepp, and Franck (2016) is by far the most relevant empirical paper for our study because these authors did offer formal, explicit testing of the significance of suspense and surprise (in the Ely-Frankel-Kamenica sense) by modeling the size of minute-by-minute television audience for sports matches. Their setting was the Wimbledon tennis championships and they analyzed Swiss audience data for 80 men's singles matches between 2009 and 2014 (which yielded 8,563 minute-observations).…”

“…Because the scenario is hypothetical, there are no odds quoted from which the probabilities required could be derived. Addressing this problem for tennis, Bizzozero, Flepp, and Franck (2016) forecast future hypothetical probabilities using a Markov chain method, where future probabilities of a player winning an individual point were set according to the server's career record of winning service points. But the method cannot readily be applied to football because it depends on a sports structure where the future will be divided into discrete plays.…”

“…Dynamic panel estimation, and in particular the Arrellano-Bond Generalized Method of Moments (GMM) technique, provides a way to deal with the problem, based on removing the fixed effects by means of the first difference transformation of the model together with the use of a complete set of lagged dependent variables as instruments, albeit there is a cost in terms of loss of efficiency. Indeed Bizzozero, Flepp, and Franck (2016) favor GMM in their minute-by-minute modeling of audiences for Wimbledon tennis.…”