"To be a great guitar player can be defined differently from person to person. A great cover player doesn't need to understand it but an original artist needs to, no matter what their skill level. If they don't, then how do they know what sounds good and what doesn't on a moments notice?"

By ear, of course - and playing experience. Theory won't tell you what "sounds good". It will only tell you what most other people thought sounded OK, in the past, in a general way.

"In the end, isn't everything you play on a guitar theory based? "

Well yes, it can all be DESCRIBED in theoretical terms. The question is whether you need those terms, or find them useful. Of course, most people do. And it's very few players who can do everything by intuition.
My point is that the music comes first, the theory comes later. Theory is only a route (and only one route) back to the music.
It's a little like understanding the grammar of a foreign language you want to speak. You could listen to natives and copy them, pick things up that way. That will get you there eventually. But will probably help to read a book of grammar, or a dictionary, too. (Having said that, the language of music is not actually foreign to any of us. We've been listening to it all our lives. We know how it's supposed to sound. The challenge of becoming a performer is more of a technical one than a theoretical one.)

Theory gives you a conceptual framework within which to build an understanding of how music works, and which helps you communicate (verbally) with other musicians.
Every musician needs a conceptual framework of some kind, a mental image of what they're hearing or aiming at, that helps them organise the sonic information. To actually PLAY, they don't need a standard or conventional framework; they could invent their own if they wanted. (They will get the same musical results, because they are listening to the same music, and aiming for the same targets.)
But to talk to others, they DO need the conventional framework. So it may as well be the conventional one that they learn.

So I'm not exactly disagreeing with you. I like theory myself, and find it hugely useful. But I do know it hasn't made me a better player. Well, maybe a tiny bit... But that's not really the point of it.

I agree totally with your first paragraph, but the rest doesn't follow.

We can of course argue about who is a "great guitarist" ;-).
Eg, I think the greatest jazz guitarists were Django Reinhardt and Wes Montgomery. They famously knew no theory, in the intellectual academic sense - but of course knew it intuitively, having studied their craft by ear and experimentation.
(Do you know the story of John McLaughlin on Miles Davis "Silent Way" session? He struggled and sweated for several takes to find the right thing to play, until Miles said to him "play like you don't know how to play the guitar". This baffled McLaughlin - but the next take was the successful one. ;-) )

As I said above, music is a language, basically. The goal is to be able to speak it like a native, like our mother tongue. We can do it by "going native" (listening and copying), or by reading books - or ideally I guess with a combination of both.
But it's wrong to suggest that an intuitive player (like Townshend, say) will never be as good as a trained one. It all depends on how good your ear is - and of course on what music you choose to play. The Who would not have been as good with Pat Metheny or John Williams on guitar, now would they??.. ;-).)