While studying rookies this spring, I encountered several critics warning us to beware that Jayden Daniels wouldn’t — or at least didn’t — throw to the middle of the field. If a sharp-tongued pundit snaps off something like that and moves on with all the confidence of a peacock, the rest of us tend to shuffle right along with their hands on the smalls of our backs; clearly, that issue has been solved. And the more I pored through scouting reports on different players this March and April, the more I noticed the same line rattled off about others: Russell Wilson, Kyler Murray, Deshaun Watson. It’s become on trend — a way to know-it-all harder than the other know-it-alls. And, honestly, the more I hear it, the more I feel like I am being had. It seems suspiciously like subterfuge; or at its most benign, perhaps grandeur.
Does it matter where yardage is being completed on the field? Are QBs consistent on this year after year? Are play-callers? Are there plausible explanations for those who don’t throw it over the middle? Are there any areas of the field that do matter for compiling or predicting fantasy points? These are all questions I intend to investigate today.
SETTING UP THE CORRELATIONS
I wanted a five-year sample with descriptive and predictive correlations, so I needed to set them up for same-season scoring (PPR) and following-year scoring (PPR.N1). I used the RotoViz Screener to compile passing yardage (paYDS) and PPR dating back to 2018. I then went to the RotoViz Stat Explorer, where I searched each player, toggled to the desired year, and manually entered the yardage for each of the nine areas of the field (deep left, intermediate left, shallow left, deep middle, intermediate middle, shallow middle, deep right, intermediate right, shallow right). I used 1000 paYDS as a minimum for my sample.
As an example, below is the field area table from the Stat Explorer for Dak Prescott in 2023:
I then ran correlations to PPR and PPR.N1. I did this by year, then compiled every line onto a master worksheet and re-ran correlations en masse. I also re-ran them organized by deep, intermediate, and shallow; and again as the left column, middle column, and right column of the field. In each case, I recorded the paYDS and converted them into percentages of total paYDS.
