One possible anaysis might go:
With the net and digital, production and distribution costs of publishing music are almost nil. So millions (and millions) of recordings of new music are now published annually as against certainly no more than 10,000 a year 50 years ago.
Not all music is of high quality or worth publishing at all. Most of it would not be published if publication cost what it used to. We can’t all be great composers and performers.
The fallacious conclusion is that means more music to choose from. But the correct analysis is that you are more than 100 times less likely to find you like a piece of music you try. Therefore it can be said that trying new music is an unrewarding experience, at least in terms of time spent.
The problem I see with your analysis is, it seems to assume that one is parsing all of those millions and millions of recordings, in order to find the exceedingly small number of new recordings of new music that one may like.
But it is quite easy to drastically decrease the numbers one has to look through by, avoiding mainstream sources, such as Billboard, or Grammy nominated artists, etc.
Or avoiding types or genres of music, that one knows has attributes one does not like.
My batting average in sampling new music, that ends up being something I like, is very high. And I sample and buy a lot!