I don't think the problem with games is that people are just copying other games they liked

if I had to pick a single core problem I'd say it's that development cycles are way, way, way too long. without a relatively tight feedback loop how is anyone supposed to learn what works and what doesn't? good games get made by accident and then they can't replicate their success because they didn't understand it

back in the NES/SNES and early PC days game studios were slamming out games in 6-18 months. now it's like 5 years or more and the next game is half done by a different team before they know if anyone is even going to like the new release. team leads are deep in their careers and they've only touched like three games, with zero/minor creative control over those in the first place

and that would segue into major problem #2: huge teams. but that's a different story