All true. Unsexy projects or not, I work on really big data problems and I never once considered it boring. I love Clojure, I wish Clojure had taken off, but it fizzled (sorry, Neal Ford). Groovy was #2 for me, but Java 8 killed it because Java 8's lambda is good enough to accomplish that. Groovy is just barely hanging on now and I don't think anyone is starting new projects with it. Same with Scala. Kotlin could rise, we'll see.
I don't love Java the language, I never have, and Sun and now Oracle have much to answer for over the years. But, it's dependable. I make very decent money and although I have had to mentor a lot of youthful developers (who can barely write a loop, as you say) I do that without involving my ego in the process and take the high road, not the low road. If someone wants to learn, I teach them. I've had the unique opportunity to have a very positive effect on so many people's careers because I didn't act like the typical rock star programmer asshole who brushed them off (which by the way is why I quit my previous job--too many rockstar personalities lacking compassion).
Realistically, for most of the US, you are looking at either the Oracle or the Microsoft ecosystems and the rest makes up some smaller sliver of the pie. Just go look at the TIOBE Index.