And then there is the knowledge graph & other flavors of scrape-n-displace, which is largely content recycled from elsewhere, given prominent positioning not based on merit or editorial quality, but based on who the publisher (or recycler) is.

Another parallel trend would be the confirmation bias / brand bias factors promoting older and staler sites. Or simplified "take" articles in the mainstream media rather than the original source articles on niche hobbyist blogs and forums or such.

And in taking broad sets of new niche intents and trying to guide those streams of users back down well worn paths. For example, sometimes when you want to find a particular news story about a broad & well-known web platform like Apple, Amazon, Facebook, or Google it can be hard to find sites other than the official site. And on some other longtail queries Google rewrites what is being searched for in a way that brings up some results that don't match the true searcher intent. Probably the best example I can come up with on this front is say you wanted a pair of shoes of a specific brand, size, width, and model number. If they are not the most recent and most heavily marketed versions it can be tough. Auto-generated internal search pages on trusted brand sites rank well, while a small retailer carrying that specific shoe might be penalized by Panda.