Agnarr was PoP-locked. People seemed to think that was a great idea, and maybe it was, maybe not. I think they have a reasonable amount of people still playing since it hit its cap 5(?) years ago. Not like, thousands, but like a few hundred people at peak times and there's multiple guilds that still raid and shit.
I think there's a lot of cool stuff in the later expacs most EQ players never played, but there's also a bunch of garbage and the flavor of the game definitely changes. I think EQ TLP's know their $$$$ is nostalgia related though so really they can unofficially kill off any server after OoW with very little blowblack.
I almost think they don't want to do another Mischief yet because it's a guinea pig. They want to get the item drops/tier lists correct for stuff deeeeep in the game and Mischief has to run through it first and they'll tune it up. Once they feel comfortable with a good amount of balanced drops/bosses/tiers for as long as they THINK people will play it (House of Thule? Secrets of Faydwer?) they can launch another one using what they learned to make it smoother.
The issue with a locked server that never ends is that you can't make money off of it each year. It's a one and done idea, and at a specific point it becomes too top heavy.
It works on emulator because there's no profit incentive or requirement to pay people, and on a commercial game, there's no consistent revenue from doing so compared to re-iterating the same game with slightly different rules.
I used to believe that SOE / Daybreak could make a ton of money off of the $15 a month that people would pay to play a PoP-locked server, but reality didn't consider the financial cost of staffing a project like that as I lacked the experience to know how much personnel costed and the numbers behind servers.
I also, for the longest time until entering industry, mistakenly assumed that subscription costs would be enough to sustain their current profits and that very few spent much in the cash shops back then (This was about 2013, 2014?), but failed to realize they need to constantly improve and adapt to market trends so that they don't fall out of favor with investors who want a rather large return on their investment.
So the model works for them because they profit greatly from the repeated server launches.
I still agree now with the assessment that launching these and nixing support at a specific expansion is a reasonable ask if it means that other types of servers see more attention and quality increases, as that won't really affect production costs - and merging those characters into a live server afterwards means you can reprovision the server's hardware following its stagnation sooner.