I've been looking over the grades that other guys have been giving out, and mine seem to cluster more. My initial reaction to Bruce was:

Man you're a hard marker. (Can't speak to the Calgary game since I didn't see it, but...)
Maybe we'll need a ... SomethingCute number... to correct for things at the end of the year. Knock a point off my grades and/or add one to yours. ;)

Bruce pointed out that our averages are actually fairly close; mine for Ottawa was 5.8, his 5.4. My response:

I guess my reaction is more because I tend to cluster players around 5/6, with the occasional 7/8 and the occasional 4. Dunno if I've given 3 or less. You've got a bigger spread, whereas I'm inclined to think an average player will make mistakes. I don't think most Oilers have usually been individually BAD this year, but they've not been GOOD either, and that's the problem.

In comments on David Staples' blog somebody named Cathe had a few questions. My response there was:

Cathe, I've been trying not to think of previous games when rating the current one, I'm trying to have them be more or less standalone, but yeah, I've given some thought to the scoring system. I'm thinking next season we might like to try something a bit different. Part of the problem is that for a player who didn't suck out loud, but who wasn't great either, 5 and 6 are the only two real grades, and a 7 is a bit much. I've been cheating a bit with weak/strong, but it's not really satisfying. Maybe a bit more spread in the middle would be better - most NHL players are never going to turn in a 1 or a 2, nor a 9 or a 10.
I must think more about my grading strategy, although I guess for the balance of the season I'll need to stay the course, just so my own grades will be consistent.

Later I said that I would re-visit the grade descriptions. As I said, I don't think it's a good idea for me to change my methodology now, but it's a consideration for next year, should we continue the project. It's not clear to me that we're interpreting the grades the same way, although again, as David and Bruce pointed out, we tend to agree on who was better and who was worse any given game. Might be worth loading the grades into a spreadsheet for various computations though, just to play around.
For the record, here's a copy of the original marking guidelines suggested by Staples. Sorry for the (lack of) linewraps; download the file and open it in a text editor that inserts soft wraps or something.