,
12 tweets,
5 min read
Read on Twitter
Ran some quick numbers on Team PFF Grades vs. Win Percentage, and this is what I've come up with.
No surprise, the passing grade is highly correlated with winning football games. Do you want to win? Get yourself a top tier QB.
R^2 of 0.3999
No surprise, the passing grade is highly correlated with winning football games. Do you want to win? Get yourself a top tier QB.
R^2 of 0.3999
Team Running Grade vs. Win Percentage?
Presenting evidence #100200302402486t924691821468 that #RunningBacksDontMatter
R^2 of 0.06112
Presenting evidence #100200302402486t924691821468 that #RunningBacksDontMatter
R^2 of 0.06112
How about receivers? This obviously doesn't distinguish between RB, WR, or TE.
R^2 of 0.3184
Highly correlated to winning
R^2 of 0.3184
Highly correlated to winning
Team Pass Block Grade vs. Win Percentage
R^2 of 0.03153
VERY surprised this isn't higher. Relates to another surprise later on.
R^2 of 0.03153
VERY surprised this isn't higher. Relates to another surprise later on.
Team Run Block Grade correlates (ever so slightly) better to winning than pass blocking? 🤔
R^2 of 0.04013
R^2 of 0.04013
Team Tackling Grade vs. Win Percentage
R^2 of 0.04037
R^2 of 0.04037
Team Pass Rush Grade vs. Win Percentage
This REALLY surprises me, considering how much the pass rush theoretically impacts the passing game. Very little correlation to winning, apparently.
R^2 of 0.03113
This REALLY surprises me, considering how much the pass rush theoretically impacts the passing game. Very little correlation to winning, apparently.
R^2 of 0.03113
Team Coverage Grade vs. Win Percentage
Coverage is king, it seems. Highly (!!!) correlated to winning football games
R^2 of 0.2922
Coverage is king, it seems. Highly (!!!) correlated to winning football games
R^2 of 0.2922
Many have asked about how it breaks down between positions in terms of Coverage Grade vs. Win%, and this is what I've found with the data I've been working with today.
CB > S >>> LB
As @SeanFromSeabeck have shown, safeties seem to be getting increasingly important, tho.
CB > S >>> LB
As @SeanFromSeabeck have shown, safeties seem to be getting increasingly important, tho.
And the same numbers, but with a snap-weighted mean instead.
Still CB > S > LB
Still CB > S > LB
PFF Coverage Grade by position (snap weighted) vs. Points Allowed Per Drive (2011-2018).
With these numbers, it is:
CB >>> S, LB
With these numbers, it is:
CB >>> S, LB
