This project will compare the rates at which horses do not complete (fall or un-seat their jockey) in national hunt (jumps) races in Ireland, United Kingdom and France. These comparisons will take into account a number of risk factors such as the type of race, the length of the race and ground conditions. The data required for this project will need to be scraped, using appropriately programmed automation from commercial websites that provide race results and reports. The data will need to be re-structured, validated and possibly augmented. The main method of statistical analysis will be generalised linear models but there will be scope to apply other machine learning techniques. An expected output of this project will be the communication of the comparisons and the identified risk factors to a non-analytics audience through the development of an interactive dashboard display.