Betting market efficiency: an examination of Australian sport

This study has two main purposes: (1) to investigate whether the sports betting markets of the Australian Football League, the National Rugby League and the A-League are efficient ; (2) to examine whether differences in incentives or competitive balance can explain efficiency differences. Efficiency is tested in three stages. Firstly, historical betting odds are analysed for arbitrage opportunities. Secondly, betting simulations are backtested to determine whether traditional biases such as the favourite-longshot bias exist. Thirdly, the extent of inefficiency is tested directly using a conditional logic model. The data consist of historical odds and league statistics from 2010 to 2015. Between eleven to fifty-five bookmakers are surveyed, depending on the match. The results provide evidence of many arbitrage opportunities, lending support for inefficiency at a rudimentary level. Statistical tesing reveals that average bookmaker odds are weakly efficient. The extent to which incentives and competitive balance can explain efficiency differences is varied.