Yep, I made that comment prematurely. This was a pretty good idea, although in this case, I think the 10% risk is useless (unless the semis really suggest otherwise, I doubt Link beats Mario with less than 55% or more than 65%). I also have a hard time picturing any of the remaining perfects being crazy enough gamblers to risk a 63-match streak on 1% accuracy. The 5% risks will probably separate the top prizes from everyone else.
--
Yoblazer:
http://oi52.tinypic.com/ad21i1.jpg
Watch and you'll see... someday I'll be... part of your world!