Models and Statistics Monthly – 1/17/21 (Sunday)

Fellow Modelers, I am a new to the modeling world and have a few questions that I hope you can provide some help with.

1.) I’m currently using pandas and python to pull NFL data and save it as .csv files which are then are used in excel. I currently use pro football stats for the bulk of the data but I’m always interested in what other sources people may use ( I know this is usually a touchy subject for most so feel free to not say if not comfortable).

2.) Once I import the data I’ve been using it to make a linear regression model to determine the spread and the total with two different regressions both using similar or the same data variables. Is there a point where there are too many variables being used? Should you keep it simple?

3.) I’ve watched Sports Betting Truth’s YouTube channel to help get most of my simple understanding of modeling and I’ve been trying to combine the idea of using adjusted stats from a teams individual games of the season in the model. Not sure if this is something I should even be doing

4.) How often should you update the regression? Should you re-run it each week?

5.) How do you test the models? Is there a way to automate that? Is there an R^2 value that is considered good by the community? I know the higher the better.

6.) Thank you in advance for any advice you can give 🙂 ALSO If you have any questions for me please let me know and I’ll do my best to answer!

Latest posts