Keith Goldner, Vice President of Data Science at FanDuel, led a workshop on the conference’s second day explaining how sports betting lines were made.
“Most people are just familiar with seeing bet lines on their phones; how does it actually work?” Goldner asked the crowd at the start of his presentation.
Goldner said the main responsibility of FanDuel data scientists is to put out probability and sports betting lines as accurately as possible. This task is accomplished through a series of data sets implemented in a specific model.
First, data scientists ask what target they are trying to measure for the sport. Are they looking to predict points? Wins? Performance?
The following impact variables are taken into account and measured through each simulation:
- Offensive efficiency
- Opponent defensive efficiency
- Weather
- Team pace
- Opponent pace
- Defense efficiency
- Opponent offensive efficiency
- Injuries
“The more simulations you run, the more accurate your predictions will be,” Goldner stated. More on the Making of sports betting lines