
Originally Posted by
Mathman
When they first started working on the CoP, my understanding is that they looked at a bunch of old competitions where the question of "who is the best skater" had been decided by time-tested ordinal judging. Then they broke it down as to just what it was that the best skaters were doing that made them the best. Then they tested the provisional new system against the old and tinkered with it until they arrived at a system where the already acknowledged best skaters actually got the most number of points.
For instance, in one initial version quads were so highly scored that Tim Goebel, with three quads in the LP, would have won the 2002 Olympics over Yagudin and Plushenko. So they fiddled with the scoring system until Yags got back on top.
So I think it is not impossible to approach the problem from the direction of, first decide what the "correct" values ought to be, where "correct" means, "the best skaters win."
Bookmarks