Most publishing-house developmental/content/story/etc. editors don't mess with particulate-level correction, sticking instead primarily with story-level issues; typically, the only people dealing with those are the copyeditor and then the proofreader. They're very much after the fact of the truly creative phase of the book; while the author tends to have the final say-so on whether to accept or stet the edits made by the copyeditor and proofreader (and thereby sometimes adjust or reconfirm the artistic angle), the creative aspect is largely finished well before that point. That being said, I can tell you that a copyeditor who misses 1 of 5 errors simply won't have work for long; 80% is rotten for a professional copyeditor. 95% is better, but in my experience, there's a rough average of one-ish straight-up error (i.e., not a question of style or preference but of actual
mistake) per page, be it a typo, omitted/doubled word, that kind of thing); let's call it one in every two hundred words. Picking an arbitrary average of 100k words per novel (which feels about right across the eighty I've done in the past year and a half), that comes to about 500 pure errors per novel. Catch 95% of those and you still leave 25 errors in the novel...and that rate gets seen by
everyone, as even the least error-sensitive reader eventually notices at that rate of a clear mistake roughly every sixteen pages. No, 95% is still too much...hence the proofreader as a belt-and-suspenders step. And even with the proofer,
some errors will inevitably slip through, but by that point, they're infrequent enough to bother none but the most quiveringly indignant, eager-to-take offense reader.
Anyway, splitting the editing between developmental and copy, then finishing up with a proofread, is a good process, one that lets the big-picture story items be hashed out in their own time, making sure those are done and
set before that final superficial polish is done. But yeah, you're absolutely right that assembling a full editorial team can be pricey as hell (although my mind boggles at the figure of $25,000 you mentioned for editing; are you sure that doesn't include cover artist, internal artist, typesetter [or whatever the digital equivalent is now] and publicist, among others?). I'm not cheap—the cats have to eat, after all—and I can see $4000 easily spent on quality editorial work. That's one of the strongest arguments out there for going the traditional-publishing route instead of self-publishing: you're putting the financial burden for that upfront work on a much larger institution that can afford to shell it out (although even that isn't a given in terms of top quality, but YMMV).
Anyway, as I said, I know this is only peripheral to your original post; sorry for derailing. Just thought I'd clear up a touch about the one aspect I truly know intimately. Hope it helps.