Sunday, May 1, 2011

EKOS: Tories by 3.2

Another poll, this time from EKOS, suggests that while things have been stable for a few days, they are now moving again for the NDP, this time outside Québec.

This survey has the NDP surging to a large lead in Atlantic Canada, which some other polls have shown as well. In Québec, the Bloc has fallen to 22.8%, giving the NDP a 17.1% lead. In Ontario, the Tories have increased their lead over the Grits to 13.1%, and are 3% ahead in the GTA, which is also where my model has them.

While all previous polls have shown the Tories leading by at least 5% in BC, this one is the third one today to essentially show a tie in the province. In the end, this shift may be what costs Harper his majority.

In the projection, the NDP picks up two seats in BC, one from the Grits and one from the Tories:

CON - 153 152
NDP - 97 98
LIB - 45
BQ - 13

The Tory average national lead decreases slightly, to 7.0%.

EKOS will have another update around 10pm tonight with data from today, so we still have at least three polls to come.


Anonymous said...

The Atlantic numbers could cost the Cons to lose seats such as Central Nova.

ArnBrown said...

The Globe and Mail watch page is showing 8 Liberal held ridings leaning Tory. They also show 5 Liberal ridings too close to call with 4 Conservative too close to call and 1 NDP the same. With 143 seats going in, a majority seems well within reach. Of course Globe isn't the definitive source and the analysis on this blog probably closer, however, there are many ridings in play right now and much uncertainty. I know from what I see locally that the numbers in many ridings for the Liberals are falling daily. This does seem to be a Conservative/NDP dance.

Anonymous said...

I don't know. I almost refuse to believe that the vote splitting is this efficient for the Tories. There is this huge base in Alberta that drives the Conservative numbers, but the prairies as well as Ontario are under-represented in the HoC. In other words, they poll best in regions that get the least seats per capita

Yes, I am biased, but at under 35%, how can the Conservatives be so close to 155 seats?

(I understand the math, but still).

GSN said...

Chretien had a majority government with 38% of support when there was a major reform/Pc split. I think the numbers in this blog make sense.

Anonymous said...

The Cons had 42 seats in Atlantic Canada + Quebec + BC. They WILL lose some of them. They need to compensate for this somewhere. Where? Ontario? They are under 40% there. That's the same percentage as 2008. Can the vote splitting really give them enough seats to compensate the loss elsewhere and gain enough to reach 155? I doubt it.

Election Watcher said...

Anon 4:10p: Yes, Central Nova is very close to going NDP. I'm using 2004 and 2006 data to project that seat.

ArnBrown: Yeah, it's really hard to say now. I'd put the chances of a Tory majority at 40-45% based on the current projection. But if tonight's polls confirm the NDP's rise in BC, that will get lower.

Anon 4:20p: MB and SK are about 50% overrepresented in the commons. The Tory vote in NB, MB, SK and BC is very efficient. The splits in Ontario are really nasty for the Left. They would be nicer if I applied a uniform swing, but I'm assuming the swing is stronger in the GTA since EKOS has consistently shown that in almost every poll during the campaign.

Andy said...

Suppose they are actually around 40% in Ontario and, as appears to be the case, the other two parties are around 26% each. That is likely a substantially more favourable split than Bob Rae had in 1990 when he won 57% of the provincial seats with a split of 37.6 (NDP) - 32.4 (L) - 23.5 (PC). If you imagine Rae with 2.5% more and the PC's taking 4.5 from the Liberals, it wouldn't be implausible to suppose that he might have won a majority of 70-75%. So, conceivably, a 40-26-26 split could give Harper 75% or about 76-77 seats, which would be a gain of around 25 and would likely compensate for losses elsewhere and bring him close to a majority. Not that this is likely to happen -- but we could be looking at a sweeping NDP success tomorrow AND a Harper majority.

Anonymous said...

Hi. Is Ekos is using standard boundaries of Toronto=GTA? And are these GTA numbers below correct for 2008?

GTA 2008: CPC 33.5%, LPC 43.6%, NDP 19.1%
Ekos GTA: CPC 36.7%, LPC 33.7%, NDP 23.6%

Election Watcher said...

Andy: Thanks for the info! I have a 40-27-27 split right now in Ontario. The NDP vote there is very inefficient for their current level: lots of votes wasted in 15 or so really safe ridings, competitive in a dozen more, and uniformly low everywhere else. So even if the NDP surges to 32%, it likely still wouldn't get to 30 seats... (This also means that if they surge to 42%, that "uniformly low" becomes "uniformly high", which could deliver a near sweep, but it's unlikely to happen.)

Anonymous said...

I believe Nanos will have one poll left tonight but I suspect the current projections will almost reflect the outcome. The Conservatives will be close enough to majority that the Liberals would support them over the "unstable coalition". (Ironically, eh?) I had thought there could be a disintegration of BQ & Liberal vote (as happened to the PC and NDP vote in 1993). But, for all the leftist allegations of a "smear campaign", I believe that the Layton revelations will save the bacon of the Libs and BQ.

Anonymous said...

I have the Cons right now winnning 63 seats in Ontario, but still 11 short of majority, mainly because of 6 seats lost in Quebec and 4 in BC.

Election Watcher said...

Anon 4:42p: EKOS is using the Toronto CMA, which doesn't correspond neatly to riding boundaries. So I just did an approximate estimate of the Lib-Con gap in 2008, and it was about 8%. You NDP number seems too high.

Anonymous said...

My frustration (I'm that third poster) derives from the fact that the Tories would get a smaller popular vote, their lead over runner-up would be cut in half (or more) and yet they would win more seats according to you.

Now, you're the most generous to the Conservatives between all the seat prediction sites, but I still like this site the most as you update very frequently.

willge said...

Anyone else overloaded with polls? My memory might be fuzzy, but I don't remember such a rush of polls towards the end in 2008.

Election Watcher said...

Thanks, Anon 5:11. I'm glad you enjoy this site. I'm more generous to the Conservatives than other sites because:

1. I'm giving them a bump in the GTA (and docking them elsewhere in Ontario), because EKOS has consistently suggested that.

2. I'm trying to account for their vote being more reliable, and giving them what averages out to a 1.3% bump nationally over raw polls.

Without these, the Tories would be about 5 seats lower, and 148 would be very consistent with what others have.

Most projections have the Tories winning more seats due to vote splitting in Ontario.

Anonymous said...

Ekos is using CMA, darn. My 2008 figure for the NDP in the GTA was plucked from an article -- not sure how reliable. I am trying to pin down the distribution of new support for the NDP. If Ekos says the NDP are at 26.2% in Ontario up ~8%, but are at 23.6% in the CMA representing less than 40% of Ontario, then would that not mean the NDP are near 28% outside of CMA Toronto?

Sorry if you've answered this before but how does your model handle growth within the GTA - uniformly or what?

Election Watcher said...

Anon 5:39p: Just eyeballing it, the NDP was around 15% in Toronto CMA in 2008, so the EKOS numbers suggest that it rose by roughly the same amount inside the CMA and elsewhere in Ontario.

Within the GTA, it's uniform swing.

willge said...

What's the latest a poll can come out. Probably 10 or 10:30 ET because that would mean it's technically E-Day in NFLD right?

Election Watcher said...

Yes, I think it's 10:30 ET.

Anonymous said...

Your seat projections are way off--if the current poll numbers hold the tories won't break 140 seats. As one commentator noted, with a 2-3 point spread, well, alot of the Tory support is concentrated in Alberta (on the other hand, the ndp in quebec).
Over-estimating the seats for the tories, helps win votes for NDP, as Liberals (who are already probably considering switching their votes) who otherwise might not really want a NDP government, may switch to vote NDP just to prevent a Tory majority (strategizing for a Tory minority gov't where Harper needs to suck up to the remaining Liberal rump).

Dio said...

Here's my prediction:

BC 22 12 2
AB 27 1 0
SK 12 1 1
MB 9 4 1
ON 65 20 21
QC 8 30 6 31
NB 7 1 2
NS 3 4 4
PE 2 2
NL 1 2 4
Terr. 1 1 1

Total 157 78 44 31

Anonymous said...

I predict that from BC east we won't get enough sleep tomorrow night.

Election Watcher said...

Anon 7:41p: That's probably better than any projection I'm making :)

Anon 7:06p: Note that on average, polls show a 4-5% lead for the Tories, not 2-3%. I agree that if the Tories win by just 2-3%, they will not get close to a majority. As I explained at 5:32p, I make some adjustments that favour the Tories as well, which is what gets me above 150.