[rjyoungling]

>But odds are you won't make all that much. You might write far better than JK Rowling and still not sell very much because you write something that just doesn't catch the imagination of book buyers.

Really like that sentence. Very true.

>One of the things a publisher gives you, that many who choose to self-publish don't want, though, is an objective third party assessing if you've written something that they think is likely to sell. Of course they can be wrong, and frequently are. But they're investors, effectively, considering whether it would responsible for them to put their resources and time at risk to back a book or an author. They're selective because frankly most authors are not very good, and many one the ones that are good or even amazing are unlikely to sell or unlikely to sell to their audience.

I slightly disagree here. I don't know if this is true for fiction, but I don't think it's true for non-fiction.

Publishers can only asymmetrically help. They're able to somewhat accurately spot which books might work, but they can't do the inverse (spot which books won't).

[pbhjpbhj]

> if you've written something that they think is likely to sell //

Isn't it more "that they think they can sell and maximise their profit above other things they might choose to sell". It's close but has a couple of distinct differences. One is cost to them, the other is sellability (and by that publisher in particular).