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 3,484 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. This data set has been used for the analysis described in "10.1504/EJIM.2021.10031658". The uploaded data set is anonymized. If you need further assistant, please feel free to contact us.
1 Jan 2020