Yeah, I think I came to the realization as I was finishing my post above that for most people, "best" and "biggest" really are incompatible, for precisely the reasons you mention. Once a project becomes sufficiently large, it can become a grind to work on; additionally, the scope is inevitably bigger, which means even more work to make sure it's hitting on all cylinders.

A small, quick project that does just one thing well is probably a lot easier to see as your "best" because there's just fewer edge cases for you to say "boy, I wish I would have done it this other way..."