Hey, this is pretty interesting!

Some good work, indeed. I'd really be curious to know how the points are assigned, exactly, but this might be a bit picky for you to explain clearly and in details on here, especially if they are partly assigned subjectively.
Anyway, there's one question regarding the system's functioning I'd like to ask you though: Is PR'ing consistency rewarded? For example, does someone who achieves 1 PR/week during 26 weeks get more points than someone who would achieve the same number and kind of PR's within a single week?
On a different thought, I don't think the WR/PR ratio should be as high as 20:3. When I started working with such ranking systems in the past (a few of you might remember this
old site), I'd always be tempted to keep a considerable gap between #1 and #2 and sometimes down to #3-#5 in points assignation, but I don't think this is a good and representative thing really, because the line between a very good PR (top 5 for example) and a WR is just so thin, that it's not worth giving the WR holder that huge of an extra. (plus, WR's are being constantly beaten) Furthermore, a ranking reflecting people's recent PR'ing activity should give more importance to the quality of the PR improvements rather than the quality of the PR itself.
Anyway, as I'm still mostly ignorant of how the ranking works exactly, all these things I said above might seem obsolete to you. If you're already granting like ~15 points to a large PR cut or to a near-WR (in comparison to 20 for a WR, and 3 or something for an "ordinary" PR cut in the low ranks), then you can pretty much ignore what I said in the last paragraph; If not, however, then I think it may be worth a thought.
That being said, I'm surprised to see I'm ranked 2nd to you in the ranking for the April-October 2003 period. If I remember well though, that was a pretty quiet and inconsistent period for the NTSC activity, so that would explain everything.

Simon