In this Major Qualifying Project (MQP), we partnered with the online sports betting company DraftKings to develop an algorithm that assisted in setting odds for propositional bets. With a focus on baseball, and more specifically the outcome of an at-bat, the goal of our project was to predict the likelihood of an at-bat outcome before assigning odds to that event. Working with data retrieved from the MLB API, our finalized dataset contained 16 features with nearly 80 thousand rows of data contained in each of them. Utilizing this dataset, machine learning models such as a Random Forest Classifier (RFC) were applied to predict an at-bat result, before evaluating the model’s overall prediction. After fine-tuning this model, K-Nearest Neighbors (KNN) was used on the testing dataset after an RFC trained the dataset to create accurate lines with a formula. This report is a set of preliminary findings submitted before the anticipated deadline of the project for the early completion of degree requirements by one of the group members.

• 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.

Creator

• Delphia, Hailey

Subject

• Computing
• Innovation
• Project-Based Learning
• Industry

Publisher

• Worcester Polytechnic Institute

Identifier

• E-project-032522-092915
• 53941

Keyword

• Moneyline
• Machine Learning
• DraftKings
• Prop Bets
• Baseball
• Data Analysis

Advisor

• Brown, Donald R.
• Paffenroth, Randy Clinton

Year

• 2022

Sponsor

• DraftKings

UN Sustainable Development Goals

• 9 - Industry, Innovation and Infrastructure

Date created

• 2022-03-25

Resource type

• Major Qualifying Project

Major

• Architectural Engineering

Rights statement

• In Copyright

License

• Creative Commons BY-NC Attribution-NonCommercial 4.0 International