A total of four different kernels were used to try and create the best Support Vector Machines model. These kernels were: linear, sigmoid, rbf (gaussian), and polynomial. For each type of kernel regularization parameters of 1.0, 1.5, and 2.0 were used in an attempt to see if changing the parameter increased accuracy. Out of these different models only two performed well enough to be considered a success. The polynomial (degree = 3) kernel model had an accuracy of 52.8% and the polynomial (degree = 4) kernel model had an accuracy of 53.5%. These models performed above the cutoff point of 52.4% to be profitable. However, only predicting with accuracy 0.4% and 1.1% higher than necessary leaves a very small margin for error and won’t generate meaningful profits unless the sports bettor is placing very large wagers every time. So, overall Support Vector Machines could be considered a slight success in this project.