A few days ago, I asked which recording of Dvo?ák's Cello Concerto they liked best. Both responded saying they liked Rostropovich and the Berlin Philhamonic under Von Karajan. Neither really answered "why" they thought this, though I'm sure many others will agree that this recording is one of the "better" Dvo?ák's Cello Concerto's out there. So why is this answer so definitive? That is why do you music experts agree one recording is better than another? HOW can you agree on this? I fail to comprehend how a particular recording can be "better" than another. Shouldn't music, as with all arts, be opinionated?

Say I enjoyed listening to Helfgott's rendition of rach3 the most. Why is this so wrong? Why would this make me so ignorant to music if I prefered this over Horowitz/Ashkenazy?

Or what if I enjoyed listening and playing my Bach cello suites like I play my romantic pieces with full use of vibrato and wide range of emotions? One could argue this is "wrong" from a purely historica
What about when getting into composers such as Xenakis, or maybe even computer generated music? Can one really say one recording is better than the other? Based on what?

Or even more extereme - suppose I get the computer to play a piece. Each note is played with perfect precision, dynamic, rhytm, etc without any fluxuation (granted depending on the program, one can add randomness to it, but suppose we're not doing this). Say I told it to play the moonlight sonata, 1st mvt. Why is it so wrong to like this version more than one played by a human?
I also want to mention I've never actually heard Helfgott's version of rach 3, but I know his version is flamed upon quite a bit from "expert classical music reviewers"