On Fri, 26 Jun 2009, Michael Smith wrote:
> This "1400%" pseudo-number is clearly going to become a Bloody Shirt.
It's not pseudo-number, Michael. It's perfectly standard procedure when comparing elections to consider the absolute number of votes rather than their shares precisely because, as you say, so many things change -- the field, the issues -- that's it's often hard to compare. But what doesn't change is that to increase your vote, you have to get people to vote for you who didn't the first time for one reason or another. You have to get those votes from somewhere, usually at some cost to your base. Sometimes people double their vote between elections in absolute terms. It's rare, and it's remarkable. But nobody ever 14-folds it.
> What happened -- according to the official stats -- is that a candidate
> went from 6% to 50-something %, in one province, in the space of four
> years
And that in itself is absurd. The 6% vote in an election he won is a sign that the guy was despised in this district. This is like Bush suddenly getting 50% of the black vote in Harlem four years after he gets 6%. It doesn't happen in real life.
It seems clear there is no anomalous electoral result so absurd that you are capable of recognizing it.
Michael