Very good article... Pretty much emphasizes the fact that we really dont dont what we have in Tua because of the circumstances... Though I dont like the part about the split that focus on Tua throwing beyond the sticks when not pressured. This is a reason I dont like using analytics exclusively for specific cases, when Im going full analytics, Im focusing on trends.
Reason being is that in this specific instance, Tua throwing beyond the sticks/not pressured, I've actually watched his throws in those multiple times and know exactly what the environment was on those, I've got a bunch of screen shots as a result. When the Fins decided to take shots, they kept a bunch of people back to block, and given that defenses around the league had no problem getting pressure without blitzing, this resulted in the Fins having 6-7 blockers back there blocking 3-4-5 defenders on many occasions.
Here's an example...
View attachment 105289
Thats 8 Fins in the backfield at Tua's back of his drop against 4 rushers... What do you think downfield looks like? 3 guys out on routes vs 7 defenders...
Here's another one...
View attachment 105290
Here we got 6 vs 3, top of the pic, middle of the field looks crowded as **** right? Whats ****ed up about it is that its only half the defenders in coverage on that play. 4 vs 8/everyone is technically doubled and they still get pressure. What do we really expect from a 2nd year QB here?
Anyways... Liked the article as Im a big analytics guy, but I know where analytics have weaknesses... Short sample size in outlier scenarios... Again, thats why most of the research I do is on overall trends where sample size is plentiful.