In play, I've yet to really see the supposed power differences make that much difference.
The trouble is Ralif, whilst I don't doubt you're sincere when you say this, that "I haven't seen any problems!" always seems a bit facepalm-y because of the lengthy usage it saw in 3.XE, with the LF/QW problem. If you ever mentioned LF/QW back then, or even any suggestion that maybe Fighters fell behind compared to full casters, especially as you got above L5, in 3.XE, someone would chime in with "Well I haven't seen any problems with it in my game!". And the sad thing is, I suspect the vast majority of them
were being honest. Either they played games where it didn't matter for some reason mechanically (often because they just didn't play higher levels much, whether by design - i.e. E6 and the like, or because people just tended to restart/start new campaigns a lot), or where the players or DM just played in a way that avoided it (usually meaning they were anti-optimized in certain ways, but whatever, nobody has to play optimized - but equally the DM could have been a really high encounters/day guy with a lot of lower-end encounters).
But let's be real, 5E isn't 3.XE. The difference between the weaker classes, with sensible play and stat placement, and the stronger, is not gigantic. Someone did a detailed Tier analysis with points and everything and the spread between the highest and lowest was just like, 20% or something, and that feels kinda right.
At the same time, I have seen a subtle thing where people gradually get disenchanted with characters that just can't do the stuff other people are doing, and I've seen it with Sorcerers. It's not like 3.XE or some games, where the PC obviously sucks, but when they don't really have a cool gimmick (and metamagic tends not to cut it, because again you don't really have enough choices or enough sorcery points in single-digit levels to really make it sing), they don't have access to a breadth of spells, and yet they also don't have deadly cantrips of Warlock, so they just seem a bit second rate. And this wears players down, in my experience.
It's particularly bad if the player is one of those ones who both likes a strong concept and roleplaying, but likes to be mechanically effective, and not sidelined. And that's exactly the kind of person who picks Sorcerer, in my experience. They get this strong concept and it works for RP, and when we're all doing RP and stuff they have fun - but the more mechanical stuff gets, the less fun they have, and the more they seem overshadowed and to have no "good tricks", not "Omg make the Sorcerer do that thing!". This is especially true if they want to be effective but aren't willing to actually powergame it. That works for a lot of classes (Paladins and Warlocks, for example, where its hard not to be effective), but not Sorcerers.
I say this because I've seen it before with these sort of players. With a character they like and which is mechanically effective, they're fully engaged in all areas of the game and have a great time. With one that they just like the idea of, but which doesn't really deliver, and/or has mechanical meh-ness, they gradually get less engaged. Then either they are the one who keeps turning up late or not being able to make it, or they make a new character they like better, in my experience.