I won’t be satisfied with the development of machine learning until our buckets-of-bolts overlords can tell us once and for all if running back target share is a quarterback stat.
Forget that the machines are on the verge of making human beings obsolete, that machines know what you’re going to say before you say it, that the apocalyptic SkyNet system may already be up and running. I want to know whether QB tendencies and skills/limitations dictate running back targets.
My priorities? They’re nothing if not sensible.
Last summer I took a look at mobile quarterbacks and found -- to exactly no one’s surprise -- they don’t tend to pepper running backs with targets.