Bravo, Gonzalo (2026) Modeling the NFL Draft probability with Generalized Linear Mixed Models based on the NFL Combine performance. Bachelor's Thesis, Mathematics.
|
Text
Bachelors-Thesis-GBravoBlanco-s5692202.pdf Download (980kB) | Preview |
Abstract
This thesis investigates the modeling of the National Football League (NFL) Draft probability through Generalized Linear Models (GLMs) and Generalized Linear Mixed Models (GLMMs) based on the performance of prospects at the NFL Combine. The Draft is a 7-round selection process with approximately 255 selection slots where teams can select players. Some of the players declaring for the Draft get invited to the Combine, where they participate in various physical drills. Models with three different types of random effects structures are analyzed: standard logistic regression models, random intercept models and models with random slopes and intercepts. It is of interest to study the predictive and discrimination power of the models. These will be tested by performing two types of cross-validation and obtaining the Receiver Operating Characteristic (ROC) curves. The model with the best Area Under the Curve (AUC) value was found to be that with the most complex random slope structure (AUC = 0.7422). The model with the 40 yard sprint and the BMI variables was the best performing among all under cross-validation (MSE = 0.1948).
| Item Type: | Thesis (Bachelor's Thesis) |
|---|---|
| Supervisor name: | Grzegorczyk, M.A. and Krijnen, W.P. |
| Degree programme: | Mathematics |
| Thesis type: | Bachelor's Thesis |
| Language: | English |
| Date Deposited: | 08 Jul 2026 13:37 |
| Last Modified: | 08 Jul 2026 13:37 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/37393 |
Actions (login required)
 |
View Item |