Ontario 2018 election

[DL]

Just looking at our numbers now. I'm not going to give the topline, but since we probably won't have regional #s out, I can say the NDP have a decent lead in Toronto. They're up 50-24 in central (Liberals in third!) and down 38-40 in the outer region. (margin of error for both regions is around 9).

I'm assuming that "central" means Parkdale-High Park, York South-Weston, Davenport, University-Rosedale, Spadina-Fort York, Toronto Centre, Toronto-Danforth, Beaches-East York and St. Paul's and that "outer" means all the Scarborough, Etobicoke and North York seats, but where do you classify Eglinton-Lawrence and Don Valley West?

If the NDP is essentially dead even with the PCs in outer Toronto it would suggest that the NDP would likely win at least a few seats in Scarborough and North York...and if Ford himself were not running Etobicoke North would be a likely pick up

[DL]

For the fun of it, I plugged all the results from yesterday and today into the TCTC simulator.  As you know, I don't think much of the simulator, but did it anyway.  This is what the result in parliamentary seats looks like:

QP MPP Seats (63 needed for majority)               
TCTC Simulator               
                          PC  NDP   Liberal   Government
               
Ipsos Reid           75   45   6      PC Majority
Mainstreet          71   53   0      PC Majority
Pollara                44   80   0      NDP Majority
Inovative           67   51   6      PC Majority
Angus Reid        78   46   0      PC Majority

I wouldn't draw too many conclusions, but the trend is pretty clear.  Unless NDP opens up at least a 5%+ lead, PC win.         

The Ipsos poll had the PCs leading the NDP by 3 points and you say it leads to 75 PCs and 45 NDP and the Angus Reid poll has the NDP leading the PCs by 2 points and you say that yields 78 PCs and 46 NDP? that doesnt make sense

[DL]

BTW, the 6 "safest" Liberal seats according to the simulator are:

Don Valley East
Don Valley West
Mississauga Centre
Ottawa Vanier
Toronto - St. Paul's
Vaughan - Woodbridge

Of those six seats, I would bet any amount of money that the Liberals will lose Vaughan-Woodbridge to the PCs and that it won't even be close. In fact LIUNA released a poll of that riding that had the Liberals losing by 25 points

Mainstreet has the Liberals running THIRD in Don Valley East and has the Liberals in a dead heat with the NDP in St. Paul's...I have not seen any polling specifically in Mississauga Centre but the swings against the Liberals across Peel are so strong its hard to see them winning there

[DL]

I certainly love reading about the riding polls by Mainstreet but it should be noted that it is extremely challenging to do riding polls these days - especially in urban ridings - because it is literally impossible to get cell phone numbers at the riding level. That means that if you see a poll of 500 people in St. Paul's (for example) it is a poll of 500 people with directory listed land lines in that riding...there is likely almost no one under the age of 40 in the sample apart from a few millennials living with their parents who happen to pick up the phone. You can weight by age...but its tough when you have such a tiny number of young people in the sample to begin with.

[DL]

Is Fedeli behind in that poll?

No, but he only has a modest single digit lead over the NDP in a seat that he won by a big margin in 2014

[DL]

So the latest Ekos is out and has it about tied NDP 38.4, PCs 37.9 and Liberals 19.1%...The EKOS polls always have kinda skimpy sample sizes by IVR standards (e.g. n of 1,000 over three days while Mainstreet has n of 2,400)...the marginal cost of adding sample in an IVR poll is so low i wonder why they don't do more

[DL]

So the latest Ekos is out and has it about tied NDP 38.4, PCs 37.9 and Liberals 19.1%...The EKOS polls always have kinda skimpy sample sizes by IVR standards (e.g. n of 1,000 over three days while Mainstreet has n of 2,400)...the marginal cost of adding sample in an IVR poll is so low i wonder why they don't do more

The trendlines in that EKOS poll, over the past two days, are interesting:
http://www.ekospolitics.com/index.php/2018/05/race-deadlocked/

Numbers always bounce around from night to night...last Wednesday Ekos had a 10 point NDP lead and then after Thursday's results were added it was back to a dead heat...then it looks like the same has happened over the past few days.

[DL]

Mainstreet's riding polls are boring and predictable at this point.

Any theory as to why they seem to be starting with some relatively boring seats? Most of the ridings they have pollsed so far seem to be pretty rural or exurban seats where you would expect a big PC edge...apart from Toronto Centre they have not touched any ridings that would be considered NDP/Liberal battles.

That being said, if i look at the glass as being half full I wonder what to make of how they have the NDP surprisingly competitive in some bedrock PC seats like York-Simcoe or Bruce-Grey-Owen Sound?

[DL]

At long last a real, live CATI poll by Innovative. This is the first one of the campaign and interestingly they did an online poll at the exact same time so as to be able to compare methodologies and (drumroll please) the results are identical!

NDP 37% (CATI) 36% (online)
PCs 34% (CATI) 34% (online)
Libs 21% (CATI) 22% (online)

https://innovativeresearch.ca/wp-content/uploads/2018/05/OntarioThisMonth_WhatDoThePhonesSay.pdf

[DL]

I wouldnt be surprised if the NDP was surprisingly close to the PCs in both Barrie seats. They are surprisingly competitive next door in York Simcoe which is a much more rural seat. With the increasing urban/rural split in Ontario I could see Barrie being more competitive since it is a bigger and bigger city and could start to vote more like similar sized places like Peterborough or Brantford  

[DL]

Before people get too agitated about these riding polls let's keep in mind that riding polls do not have a very good record - and tend to have HUGE problems properly sampling young people and ethnic communities.

In the last BC election Mainstreet did a whole slew of riding polls and every single one of them were dead wrong and drastically overestimated BC Liberal support and underestimated NDP - especially in some Surrey seats (which are very similar demograpgics to Brampton) 

[DL]

Quito Maggi predicts a Doug majority.

I think a week before the election when the popular vote is this close it’s a bit premature to be so categorical about who will win the a majority. If I were Quito I would save that posting for next Tuesday or Wednesday

[DL]

FYI, yesterday Mainstreet initially neglected to post their riding poll for Pickering-Uxbridge. It is currently a Liberal seat but one that most woulkd expect the PCs to easily pickup - the PCs are in the lead but the NDP is surprisingly competitive there in the mid-30s....we have seen this pattern in a lot of the riding polls. For whatever reason, Mainstreet has chosen to mostly (with a few exceptions) poll in ridings that would normally be NDP dead zones in York region and rural Ontario...they haven't had the NDP actually ahead in many of those places but they do show the NDP within single digits of the PCs in such unlikely places as Grey-Bruce-Owen Sound, Kitchener-Conestoga, York-Simcoe and Pickering-Uxbridge. Of course coming close only counts in horseshoes...

[DL]

They'll be posting some more NDP-friendly ridings today including:

Ottawa Centre
Spadina-Fort York
Niagara Falls
Thunder Bay-Superior North
Kitchener Centre

Whatever they show for Spadina-Fort York you can throw in the garbage, as who in that riding even owns a landline?

Maggi hinted last night that his riding polling yesterday was convincing him of a PC majority. Does this mean the NDP is trailing in these ridings? These should all be going NDP except maybe TBSN.

Nah...the fact is none of the ridings listed above would be likely to go PC even if they won 80 seats. They are mostly NDP/OLP contests...and Maggi seems to expect the NDP to get close to 50 seats. To go from 20 seats to 50 seats you have to be winning most if not all of those NDP/Liberal contests. I think that when Maggi says he expects the PCs to get a majority it is more based on what he sees in all the riding polls he has done in 905 that seem to show the PCs running the tables there

[DL]

In Germany the "liberal" party is the FDP - and they position themselves as liberal on social policy and laissez faire on economic policy...their nickname (which the Ontario Liberals better get used to) is "the party of doctors and dentists"

[DL]

They're also polling Humber River-Black Creek (née York West), which will be a giant crap shoot too.

From experience I an VERY sceptical of anyone's ability to IVR poll in urban ridings that are largely condos with young populations or in very low income heavily ethnic ridings...the response rates tend to be very very very low and very unrepredsentative

[DL]

but just like the left in other parts of the world, the NDP's base is becoming more and more higher educated while the working class is becoming more and more attracted to right wing populism.

It is certainly true that the Liberal base is now very much the highly educated. The Ontario NDP still wins a good chunk of working class votes that the Liberals can no longer access at all...that is how they win in places like Hamilton and Niagara Falls etc...

[DL]

I have a hypothesis that the geography of PC support will be less "Ford Nation" and more the PCs doing well where they traditionally do well when they get strong support across the province. I think Ford has been a disappointment during the campaign and the coalition of low income multiethnic people who supported his brother en masse are starting to realize that Doug is not Rob and that in fact he is very much a traditional Tory

[DL]

Keep in mind that a lot of this demographic is socially very conservative and many have experiences that cause them not to trust politicians at all. From their point of view, Ford is at least promising them something.

Socially conservative voters may also be dismayed at how Ford has positioned himself as the "vice" candidate. He seems to want more gambling, more drinking, more drugs...just waiting for him to propose a red light district so all his "incel" followers can get some sex too. A friend who is married to a Muslim woman told me that at his in-laws mosque, people are getting turned off of Ford because oif all this

[DL]

Just for the hell of it I decided to go back and dig up the riding polls that Mainstreet (and others) did in the 2017 BC election to see how accurate they were. The answer - NOT VERY. In almost every single case they overestimated BC Liberal support by a lot and underestimated BC NDP support. I know that BC is not Ontario and that Ontario ridings are larger etc... but I think its still worth looking at

Saanich North
BC Liberals 40% (they got 27%)
NDP           25% (they got 30%)
Green        35% (they got 42%)

Fraser Nicola
BC Liberals 54% (they got 42%)
NDP           26% (they got 38%)
Green        20% (they got 16%)

Delta North
BC Liberals 41% (they got 40%)
NDP           39% (they got 48%)
Green        20% (they got 11%)

Surrey-Fleetwood
BC Liberals 48% (they got 36%)
NDP           39% (they got 53%)
Green        13% (they got 11%)

Another company - Oracle did a bunch of riding polls on Vancouver Island. Same pattern, almost all were dead wrong and again they generally overestimated the BC Liberals and underestimated the NDP and in these cases the Greens

Courtenay-Comox
BC Liberals   44% (they got 37%)
NDP             29% (they got 37%)
Green          27% (they got 18%)

Saanich North and The Islands
NDP             34% (they got 30%)
BC Liberal    34% (they got 27%)
Green          32% (they got 42%)

Esquimalt-Metchosin
NDP            40% (they got 46%)
BC Liberals  32% (they got 28%)
Green         28% (they got 25%)

Cowichan Valley
BC Liberals  37% (they got 28%)
NDP            35% (they got 31%)
Green         24% (they got 37%)
 

[DL]

Makes me wonder about York South-Weston, a riding with an almost identical vote share to the riding in question in 2014 and an almost identical mayoral vote share distribution from 2014.

There are some key differences though. YSW is in an older part of the city and even has some rapidly gentrifying areas, whereas HVBC is all pretty dismal postwar bungalows and housing projects

[DL]

As horrible as it would be to have Ford become premier, I console myself that the worst case scenario for the NDP is now wayyyy better than the best case scenario appeared to be just a few months ago!

[DL]

The reason riding polls can be so unreliable is that it’s is impossible to get riding specific cell phone sample especially in urban ridings and that mean that no matter how much weighting you try to do you are still only surveying people with listed landline numbers are are not representative of the population

[DL]

Poll out today: http://www.ekospolitics.com/index.php/2018/06/without-significant-changes-doug-ford-on-track-for-majority-government/

39-35-19

Pardon m’y schadenfreude that the OLP is stuck at 19%. Mainstreet seems to be the only one showing any recovery at all for them. I’m still optimistic and that a lot of soft Liberals and Greens will go NDP in the end

[DL]

For all the buzz about the Liberals showing “signs of life” and getting a “bounce” after the debate...looking at EKOS and Mainstreet and Research.co polls out today....looks to my like the dead cat bounce is DEAD