We're reaching the bitter end of the semester, that time when all projects come crashing down upon students and teachers with sudden cold deadline logic. In a few short weeks I will collect papers and give a final exam. Then I will have to grade papers and finals. I used to do traditional grading, where you actually read the papers and finals, but now I have a more scientific approach, or rather approaches, because you can't grade both papers and finals the same way.
For the papers, you need to do a calculation based upon length of paper, sources cited, and position in the stack. When those three numbers are combined, you then place the stack on a small coffee table and throw each paper across the room, noting whether it lands face up or face down. If it's face up, you take the length of pages and double it. If it's face down, you subtract the sources cited. Finally, you toss the paper back towards the location of the original stack. Anything staying on the table garners an A or a B, depending on the other scoring factors. Those papers falling off the table earn B's, C's, or D's, unless they manage to land on a predetermined spot marked out by a dinner plate; those papers earn an F.
Yes, it's elaborate, but on the other hand you are saving time by not reading the papers.
For the finals, I usually count the number of words in the first short response question, then double it to reach a final grade. It's that simple.
I've been using this system for about ten years now and let me tell you it takes the stress and worry out of the end of the semester. You may need to work out your own system -- I experimented before cementing mine -- but it really comes down to feeling comfortable with your criteria.