Thinking more on this, I think this is getting at the crux of why I don't find using metascience principles particularly persuasive.
My work is very demanding. There have been many times where I've thought of quiting TTRPGs, at least for a while, and just relaxing with TV or something less taxing. But the running my campaign brings me a kind of enjoyment that other pastimes don't provide for me. And the games I choose for my main campaign are certainly not the most simple game systems. There are plenty of systems that are simpler than 5e and I actually ADD to the complexity by buying third-party rules or developing my own for subsystems I want to add to my campaign.
When my last campaign ended in December, I could have just started up a new 5e campaign, but moved to Warhammer Fantasy. Which is many ways is more complex than 5e. And THAT was after spending time reading through and understanding the rules of a several other systems so that I could pitch ideas to my group.
There is a kind of enjoyment for many in learning the rules of a game. If the rules are too simple, I find from my own experience and general (anecdotal) observations that people are less likely to invest their time in it over the long term.
I think what 5e did well, is it has simple to learn core starter rules that you can get for free. They are still complex enough to keep it interesting and build that feeling of system mastery for new players. Then they offer many more rules in the core books and supplements. And the system makes it very easy to customize and add to, whether additional options provided by WotC or by third-party publishers.
While I am a fan of stream-lining rules and have a simple and easy to get started with core, I find that for games I want to build a hobby that I will spend years with, I want that core to be design in such away as to support the building on of complexity in a way that doesn't break the entire system. Throughout its entire history, D&D has, for all its flaws and iterations, managed to do this well.