So Week 13 was a solid week for the model. It predicted most of this week's upsets correctly, and coming into Sunday night, had a 90% win accuracy and a 78% spread accuracy. Of course, the Chiefs keeping it close and the Steelers losing proved to be a hit on the accuracies.
This past week's weightings for the adaptive spread were the closest they have ever been to the original educated guess spread weightings. The success of Week 13 has lead to a minimal change to the weightings for this upcoming Week 14. There are no structural changes to the model this week, but if the model really messes up this week then I will have to investigate further. The win accuracy for this week increased by 4.58%, the general cover accuracy increased by 16.25%, and the adaptive spread accuracy (which was far ahead of the cover accuracy) increased by 3.75%.
I also have plans to create an adaptive model for the upcoming basketball season. Doing this in Excel takes a lot of manual input so I am planning on using Python for the NBA Model, as it has the ability to be more automated. My Python programming levels are far above my VBA skill levels, so this next sport will be a similar model in the way it's calculated but implemented differently.
Also, I have created a Tableau Dashboard to document my Model's progress as the season continues. Please feel free to check it out on the Sports Viz tab, or click here to go directly to it.
That's all for this week, I hope you guys have an awesome weekend!