I’ll also create forms to handle those “typical problems” you mention; that is, in addition to the specific assessment of the individual paper, I’ll also note in a general way how it did with regard to broad technical issues – organization, spelling and typos, effectiveness of thesis, etc. So I can go through the checklist of typical paper problems fairly quickly, and then turn to the specific concerns of a particular student’s argument.

I think part of what makes grading so tedious is that it requires the same level of thought and care as other academic work, but with most of the creativity stripped out. As you note, unless a student does something startling or unexpected, most of one’s responses are going to be fairly predictable, and, as one gets more experienced, one’s ability to state the problem in succinct yet routinized ways improves. This indeed makes for boredom!

Indeed, I have found that, for my survey students, it’s more fruitful to do several small assignments than one large one, because the character of the students’ strengths and weaknesses is usually revealed within a paragraph or two – so unless part of the lesson is to write at length, it’s not worth having them write more than a page or so at a time. (And if the assignments can build on each other, so much the better. With the level I’ve been teaching, having them put all their eggs in one basket is a recipe for disaster for them.) So they get more feedback over the course of the semester, and because the assignments are short, they are less painful to grade.