The 50% of forums you're talking about, like you, trashes a global fixing patch (more than 80+ changes, bugfixes etc.) for 2 or 3 elements.
From someone who's playing this game from D1, saying "when is 1.6 coming out" the day after 1.5 has been released is really frustrating. I just think that the half of this forum you're talking about needs to refresh their memory and talk with other gamers. I copypasted the changelog of the latest 3 patches to a friend of mine and this is what let him choose to buy this game. You see the new patch and want to roll back. So i just say, read the changelog once more, since you've nothing to do. But after all i highly suggest just to play the game, and judge by your own. Yeah conquest was maybe more fun in 1.4 without coalitions, but what about the overall experience? You can judge that in 7 hours? I've played 480 hours and still cannot define that, and actually enjoyed every patch, yeah even 1.3.2, because i cared about other things than coalitions.
Not sure I "enjoyed" every patch, but I did give them all a chance before whining about them.
The only one that was a "fail" for me was 1.4, with 1.4.1 only improving it slightly. This actually had more to do with protectorates and colonies that coalitions. But the absence of coalitions did not make for a better game, IMHO. Ugly intra-HRE blobbing, for example, was the norm in 1.4. Sure you could do more conquering, more easily, and that's fun once or twice. But unrealistically rapid expansion gets old for me. Forcing conquest to be slower, even if its done in a ham-handed, kludgy fashion really does make for a better game. Yes, it'd be nice if the system were more elegant. Yes, it'd be nice if there were something to fill in the boring parts. But, no, nothing about 1.4 was better.
And, actually, is it my imagination, or were there some (undocumented) tweaks to the coalition system? It's still pretty kludgy and ham-handed, but I've had fewer WtF moments since 1.5. No illogical coalition joins. Some coalition members staying out of wars. Small sample size, so far, and could be my recollections are just sketchy, but it seems a bit more rational.