I really like the Points Above Average stat you guys used. That's the best skill evaluation metric yet. Good job :)
Kaxon's idea ;) All I did was pull the right numbers out. Well numbers anyway.

What I'd like to put in as a request for the future, is a stat that tracks average %ile. e.g., if you finished #9 in a match with 60 total predictions and #7 in another match with 70, your average %ile would be 12.5% (or 87.5%, depending on which direction is used). The points above average metric does a good job of lowering the biases of changing contest difficulty and changing scoring systems, but I think this would be slightly better. But of course this isn't important, and given the work involved might not be worth it. It would be nice to see though.
That can go to the top of the Requests list. I could pull that off rather "easily". Which means you'll need to remind me halfway through Round 1 since I'll forget about it. As for my statistical opinion on the request, I think the score distribution is too skewed towards 50, amplifying the luck factor. I'll wait until I crunch the actual numbers before I affirm they're worthless though ;)

BTW, kawaiifan should have been #1 in the rankings :) Her points above average is just way too good.

Is there anywhere to see the rankings of that?
Bah, you modest predictor! I'll try to put something up tomorrow but I can't promise anything. It's not online but the page has to be here somewhere from when I did it.
---
SpC2k4 Triple Crown WINNER (Oracle, Betting, Spread Betting)
SC2k4 - 12th place