Classical can certainly have its share of recording flaws. True, I don't believe anything in my classical collection suffers from dynamic compression, but it's by no means universally perfect in other respects.
For instance, some of my organ recordings have clipping where the engineers clearly weren't prepared for the loudest levels. This doesn't surprise me, since as a class the acoustic organ is the loudest single non-electric instrument ever devised, and the larger examples have a tremendously wide dynamic range. I've heard a bit of overdrive in a few other places as well, including in a Mozart concerto for flute and harp (the flute player got overenthusiastic, it seems), and all over the place in Gulda's Well-Tempered Clavier (one knock on an otherwise exemplary set of performances, which I'm pretty sure I mentioned here). This is by no means an exhaustive list.
This is of course leaving off all the incidental noises (pages turning, chairs scraping, objects being dropped, conductors tapping their batons to keep the beat*, etc.) that are inevitable when you collect up lots of people in one space, and the occasional clicks and pops that creep into even digital recordings. It's amazing how much of this I never noticed until I had decent headphones, though these days I've gotten so used to this level of detail that these little flaws don't stick out as much anymore.
Apart from this, the overall tonal quality of the production varies widely, even among "acceptable" recordings, from fairly neutral (Decca, Dutoit/OSM, The Planets) to vibrant (DG, Boyd/COE, Brandenburg Concertos) to laid back (Telarc, Previn/RPO, The Planets) to strident and energetic (DG, Kleiber/WPO, Beethoven's 5th and 7th Symphonies) to mid-heavy (pretty much anything from Neville Marriner/ASMF), and probably some other classifications that haven't immediately come to mind.
Aside from the dynamic compression problem (which, given my preference for older mainstream music, doesn't come up as much for me), I couldn't say that, on average, the material in my classical collection is that much better recorded than the remainder of my library. Somewhat better on the whole, yes, but not light years ahead, and definitely not consistent.
*As ridiculous as this sounds, I've heard it in several different places, most notably at the end of the Dutoit/OSM recording of "Venus" from Holst's The Planets; this particular example sticks out to me because it's such a quiet, sparse section in an already subdued piece, and it always grates on me a bit.
