Wednesday, October 14, 2015

i'm pretty sure there's some people peeking in here, so let me say it.

i've been posting in a few places that edmonton and calgary are in play. like, the whole cities. guh? what? none of the models say they are.

aha. the models are shit. and, allow me to explain why.

first, let us look at the results of a riding poll released at midnight. and, understand that my argument says nothing of liberal or ndp support, it is only able to measure conservative support - and then argues we need polling for strategic voting.

the riding is edmonton-mill woods.

2011 results: 61%
cbc projection:45%
torstar projection: 46%
riding poll: 36%
my calculation: 36%

36% is in play. and, that would suggest that the other ridings in edmonton and calgary they won with 60% are also in play.

ok, so here's how this works.

the conservatives are down about 20% province wide. the models use a uniform swing that takes the same amount off of every riding. 61 - 0.2*61 = 48. i'm using rough numbers, so i'm less accurate, but you get the point.

but, this is daft. rural alberta has been voting in the same proportions for 100 years. it's ludicrous to suggest a uniform swing on an area that demonstrates total stasis. rather, it makes sense to suggest all of the movement is in the urban ridings. because calgary and edmonton are roughly half the population of alberta, you can roughly double the loss, there, to 40%. so, instead of saying they're down 20% across the whole province, i'm saying they're own 40% in the cities and 0% in the rest of the province.

and, nicely, 61 - .4*61 = 36.6.

now, if that calculation holds, we should expect roughly the following shifts in edmonton and calgary, for the conservatives:

55% ---> 55 - 55*.4 = 33
65% ---> 65 - 65*.4 = 39
75% ---> 75 - 75*.4 = 45

as we can see, many seats are in play.

but, upon calculating where the conservatives are likely running, i am none the wiser as to where the liberals or ndp are running, whether one is actually winning and which should be supported, strategically. that's up for the people that do these things to do.

i'm just pointing out what's in play.

http://www.mainstreetresearch.ca/friends-of-canadian-broadcasting/