buddyholly posted this in the Madrid thread, it sounds like a good idea to me
maybe we can have a vote on it or if it will be problematic rankings-wise to do it mid-season maybe discuss it in the off-season. I've always thought it's a lot more impressive to predict that, say, Gil will make a hardcourt semifinal or Hernych will beat Youzhny than to predict nadal as champion of a clay tournament.
I should not bring this up, because it seems I may win two weeks in a row, but I think there is a flaw in our scoring system. In the real world the champion gets 1000 points for winning the tournament. In our FITD the champion gets 1000 points for winning the final, 600 points for winning his semifinal, 360 points for winning his quarterfinal, etc. Thus our results are heavily skewed towards the tournament winner and the points gained for picking winners in the early rounds become irrelevant. The scoring should really accumulate, thus today's four players should have accumulated a total of 360 points each by now and the winners today can add 240 points each, bringing their total to 600.
For example, chowdahead could win this week if Federer wins the final, but under actual points awarded his total draw would not have got the most points. OK, I know the ATP Challenge does the same thing, but I do think picking the winner of a match in round one is often harder than picking the winner in the final. Any opinions?