The reason why people want to maintain the use of analog sticks is due to conditioning and familiarity.
But there is another aspect, which is that this method of control also makes perfect sense to the current era of games we're playing.
We live in a design philosophy where "context sensitivity" rules all of 3D game design. I can think of no major first or third person 3D game where context sensitivity is not the backbone of interacting.
Context sensitivity actions is essentially controlling your character, going up to a wall and the act of pressing a context sensitive button (like A) to interacte with it. Game designs approach is context driven in this manner in everything the player does. Talking to NPCs, picking up a weapon, opening a door, kicking a crate, petting a dog. The interface pops up and up and allows the gamer to interacte with whatever the player is close to.
But this sort of interaction doesn't really make sense as we move forward, and in VR you feel the disconnect because it makes intuitively sense for us to use all of our fingers, our hands and our entire arm. It doesn't make sense to "press A" to pick up a weapon in VR.
Star Citizen is trying to do something revolutionary that is removed from everything its doing. Regardless if this game fails or succeds, it's new interaction system shows how much more immersive gaming can be if we move past the age of context sensitivity. People like to think of context sensitivity actions in meme terms like "press X to pay your respect" or "Press AA AA AAA AA A to win", but really, the nonsensical design choice of this extends to many many great games.
The interaction we need in many games is more precision. You're not pressing A to accelerate the car, you're physically aiming (looking) at the ignition, gear shift, and steering wheel, doing multiple actions to start driving.
The problem has always been to translate many of these instantaneous actions we do effortlessly in real life to the digital interactive realm without it being a slog.