I somewhat disagree. Obviously I am biased on this topic as I am one of these armchair theorycrafters (being on xbox) who is none the less commenting.

My reasons are as follows:
1) I think it is possible to form educated opinions on changes based on theorycrafting, however in my own personal case I do always try to make it explicit in my posts that it is only theory and that if anyone actually tests it (as often happens, so thanks to all the actual PTS testers for doing this sort of thing) I will then gladly accept that result.
2) There are often cases where a non-PTS player may ask for for something to be tested which may have so far have been overlooked, which is then done so by someone actually on the PTS. So in a way non-PTS players may enhance the process by contributing to the direction of testing.
3) I think it is generally already reasonably clear when someone has actually tested something vs not, as generally they will describe the test and how they used it to form an opinion. To me, this is the most important part of adding weight to their feedback.
4) With the idea given of an email/code etc obtained when logging in to the PTS - a player could simply log in once, get this code, and then freely comment even though they may do no actual testing. This is no different to a non-PTS player commenting. Again, the most important part of adding weight to feedback that is actually gained from testing is describing the test performed as I mention in 3).

That said, I do agree that there is potentially an issue with the quality of the feedback that may be present on this forum due to this issue. As even with PTS players describing the tests it can be lost amongst all the other comments made.

A possible alternative would be to somehow have different sections to allow both non-PTS and PTS players to give feedback, or have a restriction apply on certain discussions only, or have some way of identifying actual PTS players when they make comments as a means of filtering out actual tested feedback vs speculation (although these still don't address the issue I raised in 4) ).
It is then up to ZOS (as it would be anyway) to decide how much weight they give to these two types of feedback.

In a way, this is already the case as described by

Souterain wrote: »

Mr Lambert recently stated in an interview that in-game bug reports are read while forum stuff...not so much. I expect the only threads in pts forum they will take an active interest in are those started by themselves where they are looking only for specific feedback on that often very narrow subject. So, OP, from that perspective the only ones listened to would be those on the pts that file in-game/pts feedback.

Judging from comments made here over the years the forum, even pts, is pretty much largely ignored.