We have to be really careful about ascribing causes to small changes in low-probability events like INTs. There is statistically no difference between throwing 8 INTs in 581 pass attempts and 11 in 570 even though it looks like a lot more. That kind of random variation is totally normal. The fact that we can point to specific ones as especially fluky reinforces the point that stuff just happens and we read too much into small fluctuations in low-probability events at our own peril. QB performance varies far more year-to-year than people want to admit. Matthew Stafford suffered a far more dramatic dropoff than Brady, at the age of 30 when he should be right in his prime. Carson Wentz' TD rate dropped more. Case Keenum, Ryan Tannehill (from 2016), Blake Bortles, Jameis Winston, all had similar or worse declines in ANY/A. There is a lot of variation. We ascribe narratives to it where convenient (Wentz, Tannehill returning from injury; Keenum changing teams; age for Brady; regression from Bortles), and those may be true at least in part, but there's also just a lot of fluctuation here. We understand this a lot better in baseball; no one would feel the need to force a narrative on Mookie hitting, say, .305/.390/.525 next year instead of reprising his MVP campaign. Why is it so different for QBs? EDIT: I also don't want to dismiss the possibility there is age-related decline. The red zone performance in particular is a concern I think, though there are team-wide issues there (they might be the worst goal-line rushing team in the league). I'm agnostic. I do think the statistical case laid out by Tanier is pretty weak.