A screenshot from the article.

This off-season I finally got around to reading The AccuPredict Method, an article by Cliff DeJong over on NASCAR Fantasy Racing Cheat Sheet. Wow, I sure wish I’d read it earlier.

DeJong is a research scientist who says he’s been crunching numbers his entire life. He’s also a NASCAR fan, and he’s brought his mathematics and statistics skills to bear in the NASCAR handicapping game. The result: AccuPredict, an algorithm that uses traditional and loop data statistics to predict NASCAR driver finish positions.

The article describes the value of stats and how to use statistics in NASCAR handicapping in a language anyone can understand. I too have approached NASCAR handicapping from a math/stats perspective, albeit minus the math degree, but I learned a lot from DeJong’s article, and I’ve taken away several ideas I’m integrating into my handicapping process this year.

Example: DeJong crunches the numbers to see how well the traditional track groupings work as a prediction tool (e.g., short-flat tracks, banked-cookie tracks, etc.), and surprise surprise, those traditional groupings are not optimum. DeJong runs more numbers to identify new groupings that significantly improve his prediction results. I’ll be using his groups this year in my process.

Oh, and nowhere in the article does DeJong discuss using that mystical element called “momentum” as a handicapping tool. Sweet validation!

Anyway, if you play fantasy NASCAR and/or bet NASCAR, you’ve gotta check this article out. At the least, it’ll give you some ideas to chew on.