The biggest lesson from TLJ & TLoU2 is that there are big, vocal groups within a particular category of consumer that do not want their beloved properties to run counter to a comfortable and "expected" experience, especially if that is inclusive of more modern perspective informing the story's plot, characterization, & subtext. Star Wars had the additional problem of flooding the market with multiple movies, all of which varied in both quality and, perhaps more importantly, seemed to suffer from a real lack of overarching executive vision.
But it's not really about "quality"... that falls apart when looking at some of the arguments which constantly come up in these comparisons (anyone claiming the prequels are better than TLJ aren't talking about quality even if they think they are, for instance). No, it's the presence of these kinds of elements which inform the stories -- the very existence of specific ideas and character decisions --- which point to why these respective "outrages" happened. That's about what people like and don't like, not good vs bad.
As far Druckmann as auteur, I don't think that kind of label really applies to game design and development as much as it does/can in film. Games, even cinematic narrative games, are much different than films, and that means the director is going to have less impact on the overall experience than a film director will have in dictating the audience's experience in a 2/3 hour film. But regardless of that, the idea that you could just plug and play a new director in a ND project, as if he's so replaceable on these games, to me that's a really limited view on the reality of the job of game director. Their fingerprints are all over every single element of their game/s.
Games being what they are the as a medium, the department directors/leads are probably more impactful than in film, at least in terms of having an impact on the audience's experience with the final product... but if that applies to Druckmann it applies to everyone else in the medium too. Possible exceptions could be "smaller" games, like Braid or Inside, which are two examples in which the label of "auteur" could possible apply in a more understandable way. IMO.