Tuesday, November 12, 2013

Projection Update - Conservative win by 1 seat

I'm still working out some of the kinks in the newest ElectoMatic, but I wanted to update the projection that it's provided me. To show that it is not biased towards or against any parties, I also wanted to compare it to a projection based on the same numbers provided by 308.com.

He also has the Liberals winning 4 seats in Alberta. I could run you guys through the numbers as to why 3 of our seats are in Calgary if anyone wants me to.

Either way, here are the results with some commentary.

ATLANTIC:
The Liberal vote here is very very strong. Poll averages, and, in fact, individual polls keep showing the Liberals taking more votes than both the NDP and Tories combined. This is what the Atlantic looks like with these vote totals.



ALBERTA:
Figuring out what Independents win their seats is not easy. The current numbers show Rathgeber just holding on. The Liberals retain their 3 projected earlier in Calgary. Why is this? In short, it is because in the last election, the Liberals did far better in Calgary than they did in Edmonton, in part, because the NDP took much of the "progressive" vote in Edmonton. This is not really a one-off thing. Looking at results from 1993 to 2011 we can see a clear trend of growing Liberal strength in Calgary, vs stagnation and even some losses in Edmonton.



ONTARIO
Ontario is one of the reasons I'm not quite yet comfortable releasing the full version of the ElectoMatic. Some of these results still seem a bit odd to me. In particular, when you add up all the ridings, you end up with the Tories taking more votes in the province. This is not that odd, but when the Liberals begin with such an advantage it does seem strange. The ElectoMatic was designed this way since ridings can vary by size and that can have a huge impact on how poll translate into seats. I will continue to delve into the numbers and figure out exactly why this is happening and if it should happen.


BRITISH COLUMBIA
BC remains largely unchanged from the last projection, and now nearly identically matches the projection from 308. The Tories are simply doing poorly in the province (being just about tied with the NDP and Liberals) and the only "big" question is if the Greens have enough strength to take Victoria or not. (On these numbers, they do not)

PRAIRIES
Liberal strength in the area is, as usual, concentrated in Manitoba. Winnipeg could be effectively swept by the party, where we can manage 5 out of 8 seats. We also, on the numbers, lead in Churchill; but that strikes me as wrong so I will take a closer look at this and determine if it needs an over-ride or not. The good news for the NDP is that at these support levels they are stable in taking not only a Regina seat but a Saskatoon seat as well. Remember too that the Speaker's seat is half in Regina (in terms of voters) which is why the little outline of the city was left on the map. Thus it is not accurate to say these results would shut the Tories out of Regina.


QUEBEC (the rest of)
What a mess. The numbers are quite clear though; this is no mistake. On these current poll levels, this is what this part of the province would look like. The Tories are able to retain their seats, and, despite low polling numbers, would be able to pick up a couple seats on the splits. The Bloc would also be able to make a comeback in some areas; good news for the party if they plan to remain relevant. The NDP would be in trouble as many of it's seats will be lost. The 3 nearest Montreal are perhaps their most stable seats, as both of their rural seats north of Quebec City depend on some possibly weak calculations, one because of Andrew Arthur, and the other because of a member who crossed the floor. Most interesting is the stable level of polling that causes the prediction in Quebec City. It's likely that at these support levels we will see all 4 big parties represented from the city, despite the fact that it only has 5 ridings.

MONTREAL AREA
This is where the polling breakdowns matter. Once taken into account, the vote patterns for the NDP and Liberals shift. Currently the Liberals are doing far better within Montreal than within the Corona. Only one poll shows us leading outside the city while nearly every poll show us with a huge lead inside the city. What is bad news for the NDP is what happens on the Island of Montreal. While regional polls show them ahead of the Bloc, all province-wide polls show the Bloc far ahead of their combined total from the regionals. This means the Bloc must therefore be ahead of that. Considering the high share of the Bloc vote coming from Montreal, this could spell disaster for the NDP. In fact, Mulcair only holds on to his seat because of the fake 'boost' I gave him for being leader. Considering what would happen in a real election with real polls coming into the party HQ, I consider this realistic. If it looked like an island-wide shut out, the party would indeed dump its money into saving it's leader. Despite the boost however, should the NDP drop 1 or 2 more points in the province, even this seat becomes vulnerable.

What is perhaps most outstanding is the NDP strength in the Corona. Outside of Laval, they are expected to hold on to just about every seat. The biggest problem for the NDP is that this area of Quebec tends to like big swings, and I'm not talking on the playground. Given what has happened to the ADQ, and what polls show happened to the CAQ, it is quite possible that this entire region could shift en masse away from the NDP. Right now the party holds a solid lead, but the NDP needs to edge back into Montreal if they want real staying power in the province.

WIN OR LOSE
This projection has Harper up by a single seat. This is partly based on those Ontario numbers that I want to look deeper into. What we see is the NDP is sitting on almost 70 seats. 26 of those come from Quebec, which leaves 44 from outside the province. Given the seat increase in the big 3 English provinces this means, in short, that the only reason the NDP is doing so much better than historically, is their seats from Quebec. I know the media and some political analysts and bloggers like to play it like two dozen NDP MPs from Quebec means the party is doing awful in the province, but when you consider that a 14% popular vote in the province was seen as a "high water mark" for many many years, the NDP, in fact, is doing extremely well in Quebec. Their 22 seats in the current 106 seat Ontario means they won 21% of seats in the province. A similar share would net them 17 seats from Quebec. If they can manage to hold on to these sorts of numbers for 2 elections, we will finally have a real 3-party nationwide battle for government every time we go to the polls. I don't know about you, but I consider that a win for democracy.

No comments:

Post a Comment