I’ve recorded a new video on “How to build an NFL sports betting model.” Doing my best to keep it simple (at least in the beginning), trying not to over-intellectualize all the concepts and rabbit holes, like math nerds sometimes to do. The reality is, it’s not that complicated. Shout out to one of sports analytics OG’s Wayne Winston. He taught a class at UH on this topic, using data from 2013 season. I’ll continue to use his examples while updating the raw data for 2017 games.
1. How to preform Regression Analysis on 2017 NFL data
What factors actually contribute to winning in the NFL? To effort the answer, we use Regression analysis. It sounds complicated, but you will be surprised how easy it can be done (under 5 minutes). Here are the steps in excel:
2. How to build NFL Team (Power) Ratings based on Final Scores (using 2017 NFL data)
This video shows how to start rating NFL teams based on final score results. This is also a basic way of calculating a projected point spread for each game based on “power ratings.” How would this model have done betting the playoffs last season? We review the results at the end (hint: pretty dang well).
3. How to build NFL Strength of Schedule Ratings (using 2017 NFL data)
How do you calculate strength of schedule? We use the team ratings from the previous video to identify the average strength of opponent faced.
4. How to build individual Offense & Defense Ratings for each team (using 2017 NFL data)
It can also be useful to get a separate Offensive / Defensive rating for each team. This also will allow for a team total projection for each game.
Please complete the form below if you are interested in joining our study group!