Major League Soccer Match Predictor

BRETT KWAK

Major League Soccer Match Predictor's purpose is to accurately predict the outcome of MLS matches for the 2018 season. The program is coded from scratch in Java using NetBeans and utilizes multiple csv files for data storage. The evaluation of each team is based on a team’s attacking strength, defensive strength, form, goals for, goals against, home field advantage, and rivalry. Each attribute is weighted depending on their in game importance and fed through my algorithm. The algorithm analyses the data and returns a predicted score line for the match. I then take that data, compare it to the actual outcomes, and compile the results in an Excel spreadsheet. The two numbers that I am focusing on in my results are the percent of correct outcomes and the percent of correct score lines. My goal is to have at least 50% correctness after 5 weeks and at least 60% correctness after 8 weeks.