Agree, I am hesitant to read too far into that stat as I feel Butler's big play ability really skews that average upwards and Purdy is the benefactor. I would anticipate that average coming back down to earth this year if I was putting money on it.
Some of Butler's longest plays in his career (what we think about for his 'Big Play Ability') were in games that Purdy was not QB. 2018 vs Oklahoma (Noland), 2017 at Baylor (Noland), 2017 vs Iowa (Park), 2017 Liberty Bowl (Kempt)...
Plays like these did not inflate the QB numbers for Zeb, Kyle, or Jacob like they did for Purdy which shows he wasn't a bailout, but that Purdy was actually that accurate.