Title: From data to dollar: using the wisdom of an online tipster community to improve sports betting returns
Authors: Arne Gruettner; Thiemo Wambsganss; Andrea Back
Addresses: Institute of Information Management, University of St. Gallen, St. Gallen, Switzerland ' Institute of Information Management, University of St. Gallen, St. Gallen, Switzerland ' Institute of Information Management, University of St. Gallen, St. Gallen, Switzerland
Abstract: With thousands of (online) bookmakers accepting wagers on sporting events, sports betting has become a billion-dollar business worldwide. Therefore, researchers and practitioners have gathered interest in investigating the "wisdom-of-crowds" effect in online tipster communities to predict the outcomes of sports events. We extracted 1,534,041 tips of 3484 tipsters from Blogabet.com and used this user-generated content to investigate whether there is wisdom in online tipster communities that can be used to improve betting returns. We applied state-of-the-art data mining and natural language processing techniques and tested our hypotheses using quantitative research methods. Our results demonstrate that there is indeed wisdom in such online tipster communities that can improve sports betting returns. Tipsters won 3.29% more tips than the implied win probability set by bookmakers and produced averaged yields of 3.97%. We further identified four characteristics that are significant indicators for smarter sub-crowds within the overall crowd of an online tipster community.
Keywords: data mining; natural language processing; online communities; sports betting; sports innovation; user-generated content; wisdom-of-crowds.
European Journal of International Management, 2021 Vol.15 No.2/3, pp.314 - 338
Received: 30 Dec 2019
Accepted: 14 Feb 2020
Published online: 30 Jan 2021 *