Sportsbet 21: a successful application of statistical modeling

Universities are under increasing pressure to increase funding from non government sources. Swinburne University is no exception, and has as one of its strategic themes to be an entrepreneurial university and to encourage innovation and commercialisation of research results. As part of the then School of Mathematical Sciences, Swinburne Sports Statistics was an early entrant into the world of consulting. Computer predictions for AFL football were sold to the media as early as 1980. The expansion of sports betting from the traditional horse, harness and dog racing into a wide range of sporting events has created a market for sports prediction expertise. This paper discusses the development of Sportsbet 21, a Swinburne startup company which provides to bookmakers computer generated odds driven by a statistical model for ‘betting in running’- betting on events which occur within a sporting contest. Models have been developed for both cricket and tennis, and the product has been operated successfully by Ladbrokes in the UK for some time. The product has achieved profit targets on growing turnover, and demonstrated the robustness of the mathematical models.