Testing the efficiency of an 'in-play' sports betting market: A Monte Carlo approach

Norton, Hugh (2013). Testing the efficiency of an 'in-play' sports betting market: A Monte Carlo approach Honours Thesis, UQ Business School, The University of Queensland.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
Hugh_Norton_41747465_BCom_Hons.pdf Thesis full text application/pdf 1.13MB 6
Author Norton, Hugh
Thesis Title Testing the efficiency of an 'in-play' sports betting market: A Monte Carlo approach
School, Centre or Institute UQ Business School
Institution The University of Queensland
Publication date 2013-10
Thesis type Honours Thesis
Supervisor Stephen Gray
Total pages 122
Language eng
Subjects 1502 Banking, Finance and Investment
Abstract/Summary In this study, we develop a Monte Carlo simulation technique for estimating the probability of victory at any stage in the first or second innings. This model is then used to test market efficiency in the Betfair ‘in-play’ market for one-day international cricket matches. We find strong evidence of overreaction in the first innings. A trading strategy of betting on the batting team after the fall of a wicket under strict trading restriction results in a profit of 20.8% that is significant at the 1% level. We also find some evidence of underreaction in the second innings although it is less economically and statistically significant than the first innings overreaction. We also implement a series of more sophisticated strategies that aim to exploit any systematic mispricing throughout a match independent of the reaction to a particular ball. Trades are placed when the discrepancy between the probability of victory implied by current market odds differs substantially from the odds estimated by our Monte Carlo simulation. We document a number of trading strategies that yield large statistically significant positive returns in both the first and second innings, suggesting that inefficiencies exist in the ‘in-play’ market. We also use our Monte Carlo simulation technique to develop a rain rule alternative to the current Duckworth-Lewis system. Our method preserves the probability of victory across rain delays by resetting the target score such that the probability of the batting team winning the match after the rain delay is the same as it was prior to the interruption. Our method also outperforms prior models at predicting the winner of the match in the majority of situations.

Citation counts: Google Scholar Search Google Scholar
Created: Mon, 03 Feb 2014, 13:28:24 EST by Karen Morgan on behalf of UQ Business School