dkthom2018’s RB Predictivity Index

Hey, Reddit! I’m Dylan or dkthom2018 from Dynasty Nerds and I’ve been working on a project I wanted to share with y’all! Introducing the RB Predictivity Index.

I take 9 of fantasy football’s most predictive metrics year to year and see how well each of the Top 50 PPR RBs performed. Here are the 9 categories:

1. Generic Yards Created
2. Yards Created by @DavidZach16
3. Juke Rate
4. Draft Capital
5. Touches P/G
6. Yards Before Contact

AND some of @MiKeMeUpP ‘s TD regression data such as..

1. TD Distribution
2. Inside 10 TD Conv. Rate
3. No TD PPG

I gave each player a “Predictivty Score” of 1 - 5 according to how well they did against their fellow RB peers. The Top 1-10 got a 5, the top 11-20 performers got a 4, etc. the higher the score the better! I don’t think a higher score means Player X is better than Player Y or anything btw. It’s simply a fun data experiment. I’m honestly not sure how “predictive” a predictive score is anyways and its something I’m monitoring for the future. I simply treat this as a checklist. The more boxes a player checks off, the more safe they are relative to their ADP. I put all of this data into a google sheet and it has some graphs in it as well!

Here’s all of the data onto a graph: https://imgur.com/gallery/MQ0boIe

Here is the google sheet: https://docs.google.com/spreadsheets/d/1DmZSJv8lhe8A78WclqS5O1zNILRb4j2yK-lXs0S1C3Y/edit?usp=sharing

Please check out my twitter thread for a deeper dive and better explanation:

https://twitter.com/dkthom2018/status/1217528242850144259?s=21