Publicação
Live betting markets efficiency: the NBA case
| Resumo: | Several studies have been assessing the efficiency of sports betting markets, by comparing the pre-game prices with the actual outcomes of each event. While some have documented particular forms of inefficiency, as the favourite/longshot bias, an important part has been unable to reject the efficiency hypothesis, while identifying the betting volume and the event's notoriety as key factors for a market to be efficient. In this study, we seek to bridge a gap in the literature, by assessing the efficiency of betting markets as the inherent sports events are taking place. To this extent, we tested the in-play (live) betting markets efficiency of 4 NBA Finals games, by comparing, on a near second by second basis, the winning probability of the home team, implicitly expressed in its betting odds - the price element -, with a theoretical estimation of its winning probabilities - the information element -, generated through a logit regression based on a sample considering all plays and pre-game odds for the NBA seasons between 2007/2008 and 2014/2015. Our results show that, for the 4 games considered, the in-play betting markets are not efficient, as we reject the hypothesis that the difference between the price and the information element is zero. Although the testing framework and the limited set of considered games prevents us from validating the cause and persistence of our findings, we identify as possible mechanisms inducing these results the asymmetric valuation of game-related events by bettors and market rigidities preventing agents from reacting instantaneously to important events. |
|---|---|
| Autores principais: | Dias, André Cardoso |
| Assunto: | Betting markets efficiency NBA Live efficiency Logistic regression Economia financeira Desporto Investimento directo Risco financeiro Eficiência de mercado Regressão logística |
| Ano: | 2016 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | ISCTE |
| Idioma: | inglês |
| Origem: | Repositório ISCTE |
| Resumo: | Several studies have been assessing the efficiency of sports betting markets, by comparing the pre-game prices with the actual outcomes of each event. While some have documented particular forms of inefficiency, as the favourite/longshot bias, an important part has been unable to reject the efficiency hypothesis, while identifying the betting volume and the event's notoriety as key factors for a market to be efficient. In this study, we seek to bridge a gap in the literature, by assessing the efficiency of betting markets as the inherent sports events are taking place. To this extent, we tested the in-play (live) betting markets efficiency of 4 NBA Finals games, by comparing, on a near second by second basis, the winning probability of the home team, implicitly expressed in its betting odds - the price element -, with a theoretical estimation of its winning probabilities - the information element -, generated through a logit regression based on a sample considering all plays and pre-game odds for the NBA seasons between 2007/2008 and 2014/2015. Our results show that, for the 4 games considered, the in-play betting markets are not efficient, as we reject the hypothesis that the difference between the price and the information element is zero. Although the testing framework and the limited set of considered games prevents us from validating the cause and persistence of our findings, we identify as possible mechanisms inducing these results the asymmetric valuation of game-related events by bettors and market rigidities preventing agents from reacting instantaneously to important events. |
|---|