Sounds quite reasonable to me. Just as how I have argued that one cannot reason from "X is popular, therefore every individual component of X is popular," there is the reverse problem: we may not know which specific components are vital to the experience and which are optional or even detrimental to it. This, unfortunately, leads many designers (not just in TTRPGs) to "play it safe," failing to critically examine their work because it's risky to make changes. You see this sort of thing a lot in video games. When a particular game becomes very popular, it can actually ossify not just itself but its whole genre. World of Warcraft is a good example here for its genre (MMORPGs), a hegemony and reluctance to change that has dealt the game some harsh lessons recently, but another is Master of Orion, a series that casts a long, long shadow over the space 4X genre and which even the MoO games themselves (specifically MoO3 and the recent remake) have struggled to step out of.
It is a difficult thing to address flaws in a system. There are reasons Paizo gave it a decade with the slightly tweaked 3.5e they were running, and "because our customers ask for it" isn't the only one.
Hence why I have tried to make a clear distinction between bad design (which, as I've said, more or less means design that has to be replaced because it just doesn't work as is) and weak design (which does still function, but falls significantly short of what it could achieve if iterated upon, revised, or augmented by some other thing). 5e has very little outright bad design (IMO, the only part of it that is truly bad is the CR system because it's got exactly the same problems as 3e's CR system had.) But it has a lot of weak design, whether in the form of trying to serve conflicting goals with a single structure simultaneously, or setting a very reasonable design goal and then falling quite a bit short. And I lay much of the blame for its weak design elements on the amount of wasted time and silly choices made during the playtest period.
