I've built courses which finish away from the clubhouse two or three times in the past, and I'm about to do it again on a new project, I think.
The routing for Black Forest evolved over a period of time. Originally the client was thinking he would build a separate clubhouse for the second course, and we were trying to route all 18 holes down in the valley, but coming up a bit short. I started exploring up on the hill closer to the clubhouse, and found a couple of holes that I liked -- but then found out that some of that property had been sold as 10-acre undeveloped lots years beforehand.
After some thought, the owner bought back three lots so he could get the starting hole close enough to punt on a new clubhouse for the short term [and twenty years later, he's still not built one]. But, I decided not to make a long cart ride up the hill after 15 or 16 holes in order to play three finishing holes on the flats. My reasoning was that the holes were better down below, that you could actually walk the course and play a tournament as long as you didn't have to make the climb back up within the round [you could just put the scoreboard down below by #18 green], and that it was better not to break up the round with a big cart ride late in the game.
The new routing I'm working on now has some similar issues, so I suggested to the client that we build an "open-jawed" routing that doesn't start and finish in the same place. It is amazing how many more options this gives you in terms of finding 18 holes. It isn't appropriate for most projects, but occasionally, it fits fine. We thought about the same thing for Rock Creek, so people wouldn't have to make the trek from the clubhouse all the way to the high point of the course carrying their clubs, but eventually we realized a remote start would convince MORE people to take a cart, and we found a relatively gentle routing for the holes that get you up the hill.