simply add a "in the pocket" variable to resolve scrambles that led to broken plays (which, on first glance, helps explain the absurd rate of success for running QBs in their adjusted completion percentage) and/or a target type (RB/WR/TE)

and some of their comments strike me as bad interpretations to fit a fun narrative (Barkley sucks, Klein is great!). Barkley's <10 yard passing completion was poor, but he was above average 15-25 yards. a couple bullet points down, they criticize Geno Smith for his subpar 15-25 yards, shouldn't they give barkley credit there then?

there's the obvious comparison to his replacement that shows barkley was better and that it might have been the other players/playbook that explains his poor short passing form, as Wittek was across the board worse than barkley.

and i'm not necessarily a barkley fan, but this is the concern that a flawed mechanism that appears to be an improvement in detail leads to bad conclusions--if one makes conclusions off of it.

-----------
It's only funny till someone gets mad. Then it's hilarious.