Performing Binary Classification of Contests Profitability for Draftkings Public
Downloadable Contentopen in viewer
In this Major Qualifying Project, we worked alongside the online daily fantasy sports company DraftKings to build an algorithm that would predict which of the company's contests would be profitable for them. Our goal was to detect contests at risk of not filling to their maximum number of entrants by four hours before the contest closed. We combined categorical and numerical header data provided by DraftKings for hundreds of thousands of contests using modern data science techniques such as ensemble methods. We then utilized parameter estimation techniques to model the time series data of entrants into a given contest. Finally these parameters were fed into a Random Forest algorithm with the header data that provided our final prediction as to whether a contest would fill or not.
- This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.
- Date created
- Resource type
- Rights statement
Permanent link to this page: https://digital.wpi.edu/show/pg15bh26d