This project is looking at soccer statistics and more particular transfer fees within women’s soccer. Over the past few years, the world of women’s soccer has seen a significant increase in money spent on transfer fees with the numbers from 2020-2021 being almost double. There has been very little done to look at the values of women’s transfer fees and what goes into creating that value. We wanted to answer the question of what goes into a transfer fee in women’s soccer and could we create a model that predicts transfer fees as many go unknown. This will help to provide more insight into what makes up a fee as well as can help create more accurate fees in the future. Using the coding language R, we have created linear regression models in hopes of predicting the transfer fee of a given player based on her in-game statistics, league, age, nationality, and other relevant information regarding the player. We created three models: one with all of our variables and data points, one without the outliers, and one without the dummy variables. The first model with all the points is the best model due to its larger r squared value. With our model, we were able to predict the uncertain transfer value of player Lena Oberdorf to match the estimated description of the fee. In the future we could use other modeling types such as random forest modeling to create a model that might explain some of the variability more and overall achieve a more accurate model. As business majors with sports analytics minors both students have a passion for the business of sports and both students love soccer. We wish to work in soccer analytics someday and this project has been a good way to gain knowledge and experience.
Authors: Laura LaRocca and Ben Marks
Advisor: Jarred Wang, Sports Leadership and Management
You must be logged in to post a comment.