International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 72 - Number 5 |
Year of Publication: 2013 |
Authors: Sameerchand Pudaruth, Nicolas Medard, Zaynah Bibi Dookhun |
10.5120/12493-9048 |
Sameerchand Pudaruth, Nicolas Medard, Zaynah Bibi Dookhun . Horse Racing Prediction at the Champ De Mars using a Weighted Probabilistic Approach. International Journal of Computer Applications. 72, 5 ( June 2013), 37-42. DOI=10.5120/12493-9048
Horse racing is a popular sport in Mauritius which attracts huge crowds to Champ de Mars. Nevertheless, bettors face many difficulties in predicting winning horses to make profit. The principal factors affecting a race were determined. Each factor, namely jockeys, new horses, favourite horses, previous performance, draw, type of horses, weight, rating and stable have been examined and appropriate weights have been assigned to each of them depending on their importance. Furthermore, data for the whole racing season of 2010 was considered. The results of 240 races of 2010 have been used to determine the degree to which each factor affect the chance of each horse. The weights were then summed up to predict winners. The system can predict winners with an accuracy of 58% which is 4. 7 out of 8 winners on average. The software outperformed the predictions made by the best professional tipsters in Mauritius who could forecast only 3. 6 winners out of 8 races.