A simple linear regression w/ fantasy points per game (Y) and BMI (X) yields an R^2 of .016. That& #39;s 553 players, over the past 9 seasons. It, like every football stat, is not predictive. When you say X WRYs had BMI > Z, you& #39;re just describing who& #39;s been good, not who will be good
Since the 2012 season, the average WR1 (PPG) has had a BMI in the range of 26.9 : 27.5. Clearly we should be targeting guys in this BMI range, right? That& #39;s the logic? We look at who& #39;s been productive, and try to draft guys that look like that?
I mean this is an 8 year sample size, pretty good right!? 8*12 = 96...but there& #39;s only 42 unique WR1s over the past 8 seasons. So when we try to quantify what a & #39;WR1& #39; looks like, we& #39;re not only using a far smaller sample size than we think. We& #39;re also 2/3/4/5Xing some players
And this doesn& #39;t even include the natural size curve of players who play WR in the NFL. Point being: calc& #39;ing and interpreting a metric are two completely different skillsets. Take a deep breath before arguing that a certain metric definitively proves or disproves something
You can follow @dynasty_complex.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: