> Interesting. If a vote for Wyoming, Idaho, or Montana were looked
as a vote for "Rockies" and a vote for New Hampshire, Vermont, and
Rhode Island were looked at as a vote for "New England" then it would
look like this?
> New England - 35%
> Rockies - 31%
> Alaska - 9%
> Delaware - 8%
> Hawaii - 6%
> North Dakota - 3%
> South Dakota - 2%
> Does anyone think it might make sense to select a region before
> selecting the state within it?
This idea looks attractive to me, but I think that the cumulative
count has the potential to be even more accurate if people do it
For instance, if a person is considering the Rockies and New England
as their top pick but they think the FSP will have a higher chance of
success in the Rockies than New England, then they might vote:
The idea is that they can divide up their votes between regions as
well as between states. If they consider all of the Rocky states
together and all of the New England states together, then they can
divide their votes to reflect this.
Problems only arise when people vote in a way that does not reflect
their true preferences. For instance if I prefered NH slightly over
WY, but gave all 10 votes to NH because I want NH to win, information
would be lost. If I've divided my votes this way and NH falls by the
wayside, no one will know that I prefer WY over DE if they happen to
be the two front runners.
As long as we all understand that we can divide votes however we want
and that it's to our advantage to do it as accurately as possible, I
think a straight cumulative count will give us the best consensus.