≡ Menu

Last year, I provided a starting point for my running back projections. The idea is pretty simple: some fantasy statistics are much more repeatable, or sticky, than others. Over at Footballguys.com, I used the following formula to help isolate those factors:

1) Rushing Yards (R^2 = 0.47). The best-fit formula to predict rushing yards is:

-731 + 3.73 * Rush Attempts + 180 * Yards/Rush

2) Receptions (R^2 = 0.42). The best-fit formula to predict receptions is:

11.1 + 0.39 * Receptions + 0.032 * Receiving Yards

3) Receiving Yards (R^2 = 0.38). The best-fit formula to predict receiving yards is:

83.7 + 1.65 * Receptions + 0.46 * Receiving Yards

4) Rushing Touchdowns (R^2 = 0.29). The best-fit formula to predict rushing touchdowns is:

0.1 + 0.0037 * Rushing Yards + 0.35 * Rushing Touchdowns

5) Receiving Touchdowns (R^2 = 0.23). The best-fit formula to predict receiving touchdowns is:

0.1 + 0.0022 * Receiving Yards + 0.25 * Receiving Touchdowns

Using these formulas, we can come up with a good starting point for your 2014 running back projections.

You can read the full article here.

{ 5 comments }
  • Bryan Frye May 31, 2014, 5:07 pm

    Forgive me if you have already done this, but can we expect an article on underperformers and overperformers based on these formulas? Obviously, the article wouldn’t come till the end of the season, but it could be an interesting read.

    Reply
    • Chase Stuart June 1, 2014, 11:14 am

      Hmm. Interesting. One could do that based on my projections from 2013, although I’d be a bit terrified to see those results :)

      Reply
      • Bryan Frye June 1, 2014, 2:23 pm

        Well, from super popular guys like Peter King, down to virtually unknown guys like me, it seems we are all wrong more often than we are right…about most things. Doesn’t mean it isn’t fun to delve deeper into the sport we love!

        Reply
        • Chase Stuart June 1, 2014, 9:13 pm

          Heh. We can put it on the ever-growing to-do list.

          Reply
  • Shrinidhi July 31, 2014, 4:33 pm

    This is very cool work. I would suggest adding age and career carries as predictive factors for your regressions, as they were shown to be highly significant in my own regression analysis of running back performance.(http://thebarstoolgm.wordpress.com/2014/07/31/additional-factors-impacting-running-back-performance/)

    Reply

Leave a Comment