I know there have been lots of similar threads, but I wanted to post one in the Music theory for a more technical perspective. Let's keep this strictly nerdy and cut out the sentimental "voice of God" nonsense. It may be impossible (and some might argue detrimental to dissect music like that, but those shouldn't probably hang around Music Theory form), but we can try. Ironically though I am very much a fan, I find it almost easier to "argue" that it's not that great.
I think Septimal posted somewhere a fairly technical explanation about Mozart's greatness but I don't remember where.
Did you mean
this post? It talks a lot about interaction between opposite categories, motion vs. non-motion, texture 1 vs. texture 2, all both running alongside each other and interacting with each other.
Some people find Mozart cliche, probably because of the balanced phrasing, the half cadences or imperfect cadences, and the long cadential "padding" at the end of sections. But for me, those elements are (part of) what makes him so good! They give the medium/large scale harmonic motion of a movement a distinct identity and grace.
As far as Mozart's counterpoint goes, remember that it's usually a different kind of counterpoint from the imitative/canonic/fugal counterpoint. Of course, Mozart could write that kind of Bachian counterpoint, see the finales to the G major quartet, the D major quintet, the 19th piano concerto, or of course the Jupiter.
But more important than that (not to mention an extremely effective bassline rhythm and melodic inner voices, even if slow) is a slow counterpoint between
textures. Take the opening of the 27th piano concerto. The strings have a melodic texture. Then, the winds play a dotted rhythm descending arpeggio B flat -> F -> D. This happens again.
Then, the strings play a melody one more time, and full cadencing. But note: that wind arpeggio, especially from ending on the "weak" note D rather than the tonic note B flat, still has to be resolved! It's an independent entity, that progressed with the strings in parallel, that requires resolution too. In other words, two textures are progressing at the same time, resolving the needs each other.
And how is this wind arpeggio that I mentioned resolved? By the whole orchestra. The strings, where the first violins play a higher D with that very dotted rhythm, into a descending scale and full cadence, and the viola/cello play a dotted F, into a common perfect cadential pattern. The flutes then play that dotted F and go up to B flat, descending into cadence as the violins.
Only with such outwardly simple materials could Mozart achieve such a grammar of interaction of textural opposites, tonic and dominant opposites, rhythmic opposites, and others. But because music CD pamphlets tend to ignore these strengths and focus on how Mozart was proto-Beethoven or proto-romantic (really?), or how he was very chromatic for his time (This is true, but it's not completely right because there are chromaticisms that Haydn
would do in his expositions that Mozart wouldn't dare do, and it begs the question: why wasn't Mozart even more chromatic? And why do we still like the music of his that's quite diatonic? It turns out that these basic grammatical elements would be marred if Mozart did more chromaticism than he did.), or how "perfect" his music is (which doesn't help).
And, finally, as Mahlerian said, the irregular groupings of bar counts: these are largely possible (as are the textural shifts) because of such grounding in tonic and dominant. if Mozart was too "out there" harmonically, his music would be much worse.