This research examines the sentiment bias in the market for online tennis betting in the period 2000 - 2016. Google Trends data is used to measure the sentimental value of players in 6156 Grand Slam matches. A probit model combines this data with Bet365\'s decimal odds and a variable controlling for the home bias. Findings challenge market efficiency because bookmaker odds do not reflect the true outcome probability. Bookmakers strategically set the odds to rationally exploit sentimental bettor preferences. They underestimate the winning probability for the high-sentiment players in tournament finals and in matches with any of the Big Four. As a result of this reversed sentiment bias, bookmakers offer a relatively high price for bets on popular players. The strategy of placing bets on the relatively cheap low-sentiment player in tournament finals yields a return of -2.80%. It is an improvement on randomly betting, which makes the market for tennis betting weak-efficient. Additional analysis of the full sample revealed presence of the favorite-longshot bias. Overall, this study found sample evidence for market inefficiency of the market for tennis betting.

, , ,
T.L.P.R. Peeters
hdl.handle.net/2105/38118
Business Economics
Erasmus School of Economics

S. van Rheenen. (2017, May 23). The Sentiment Bias in the Market for Tennis Betting. Business Economics. Retrieved from http://hdl.handle.net/2105/38118