Monday, June 9, 2014

2014 Ontario election projection - June 9

June 9 Projection Map
We're now just three days away from Thursday's provincial election in Ontario. As expected, a whole slew of new polling data has rolled in, and yet we're still no clearer to knowing what's going to happen as we were last week. Last week, it looked like the Liberals were poised to break away from the Tories into a comfortable lead- perhaps enough to win a majority. However, all the most recent polls suggest the race is now a dead heat between the two parties.

For today's projection, I've inputted three new polls into my model, conducted by EKOS, Ipsos and Forum Research. EKOS Research's is the most recent poll in my model, released this morning as part of their daily tracking poll. They show the Liberals and Tories tied – exactly – right down to the decimal point (35.9 – 35.9). It also shows the NDP dropping back to 17.6%, a drop of 2.9% since their previous poll released, Saturday. Most of that drop has helped the Liberals, who are up 1.7% since Saturday, while the Tories are up 0.4% since Saturday. Meanwhile, Ipsos' poll, released Friday, also shows the two parties tied, at 35% a piece. However, they have the NDP with a much better standing, at 26%! And Forum Research alsoshows a close race, with the Liberals up by two points over the Tories (39% - 37%). Their NDP numbers are closer to EKOS's numbers, giving the party just 17%.

While Ipsos' methodology is questionable (opt-in online panel), their numbers appear to be more in tune to the other pollsters than previous polls they conducted (poll which showed the Tories ahead, while the other polling firms had the Liberals ahead). However, Ipsos' NDP numbers seem way too high compared to EKOS and Forum Research (a whole 9 point difference). Ipsos polled over 2000 people in their survey, which has meant that my model (which weights by sample size) is heavily reliant on their poll, which means my model's results may be slightly skewed towards their results.

In addition to their province-wide poll, Forum Research also published 18 riding polls on Friday, most of which were conducted the previous weekend (May 31-June 1). I've inputted these polls into my model, which has resulted in many of the my projection's seat changes since Friday. I inputted the riding polls with the polling conducted over the weekend they were conducted, so my model's numbers in those ridings will be slightly different from the polls themselves, taking into account polling trends over the last week.

Here are the results of Forum's 18 riding polls and my commentary:

Davenport: Liberal 44%, NDP 38%, PC 11%, Green 5%. This confirms that the NDP slide in Toronto has put their current seats in the city in jeopardy. However, it's still close, and thanks to poll movements over the last week, my model now shows the NDP back on top here.
Etobicoke-Lakeshore: Liberal 49%, PC 39%, NDP 7%, Green 4%. Recent rumours have indicated that the Liberals were leading in this riding, which the Tories won in a by-election last summer. This poll confirms these rumours. Without this riding, the Tories will likely be shutout in Toronto.
Trinity-Spadina: Liberal 37%, NDP 36%, PC 19%, Green 8%. I had thought this riding would be the first to go for the NDP if the Liberals were to gain any seats at their expense this election (close race in 2011). This poll shows the NDP may have a better shot to hold this seat than in Davenport.
Windsor West: NDP 41%, Liberal 34%, PC 18%, Green 5%. It's not all bad news for the NDP, as this would be a pickup from the Liberals, who currently hold this riding. With Windsor West being such a pro-incumbent riding, the best chance for the NDP to win the riding was in 2011, when the seat was open. But they are running a much stronger candidate in this election, and the Liberal incumbent has only been in office for one term.
Cambridge: Liberal 42%, PC 29%, NDP 16%, Green 9%. When a previous projection I made showed the Liberals ahead in Cambridge, I credited it as being a strange result, and quickly fixed it. However, I shouldn't have if this poll is correct. It would be a huge gain for the Liberals, who have never held the riding in its 40 year history. The result is possible, as the race between the Liberal candidate and the Tory incumbent is a re-match which was very close last time (2000 vote difference).
Kitchener-Waterloo: Liberal 35%, NDP 33%, PC 24%, Green 7%. Forum polled three of the ridings that the NDP picked up in recent by-elections, and this is the only one of those where the NDP is losing, but by just 2%. It looks like the Liberals are doing quite well in the Waterloo Region (as evidenced by the poll in Cambridge as well).
London West: NDP 35%, PC 31%, Liberal 29%, Green 3%. The big NDP by-election win in this riding last August seems to have been no fluke, but the race is a very close three way race. The Liberals polling at 29% seems to be a surprise, considering they only won 16% in the by-election, but they did win the seat in 2011.
Glengarry-Prescott-Russell: PC 45%, Liberal 41%, NDP 10%, Green 3%. This poll confirms the increase in Tory support in Eastern Ontario is for real, and it means PC pickups at the expense of the Liberals. So while the Liberals are making gains elsewhere in the province, they could lose up to 3 seats in the Ottawa area.
Ottawa South: Liberal 46%, PC 41%, NDP 9%, Green 4%. While the Tories are way up in Eastern Ontario, they aren't quite polling high enough to win this longtime Liberal riding, which was won by then-Premier Dalton McGuinty in 2011. Forum had the Tories way ahead in the August by-election in this riding, even though the Liberals ended up winning it. So it's possible this poll result indicates a Liberal landslide in Ottawa South!
Ottawa West-Nepean: PC 45%, Liberal 38%, NDP 12%, Green 4%. The Tory increase in Eastern Ontario has meant that they have enveloped this riding as well, which was extremely close in 2011.
Thunder Bay-Atikokan: Liberal 38%, PC 30%, NDP 19%, Green 9%. Forum also polled this riding earlier in the campaign, which showed the Liberals way ahead, but the NDP in 2nd. The Tories polling in 2nd place would be a huge surprise, considering recent close election results between the Liberals and NDP. I checked electionprediction.org to see if there was any evidence of a strong Tory campaign here, but most are suggesting the poll is a fluke. While I inputted this poll as is, I may try and keep the Liberals ahead within the margin of error based on this qualitative assumption (note how close it is in this projection due to recent polling!).
Sudbury: NDP 44%, Liberal 36%, PC 15%, Green 4%. This poll result confirms a previous poll conducted at the beginning of the campaign by OraclePoll which also showed the NDP winning. This seat was won by the Liberals last time, but is an open seat. The NDP winning would be no surprise, since the party holds it federally.
Pickering-Scarborough East: Liberal 49%, PC 35%, NDP 11%, Green 4%. No surprise here, as the riding is a safe Liberal seat. Not sure why Forum bothered to poll it.
Brampton-Springdale: Liberal 38%, PC 34%, NDP 19%, Green 4%. This riding is probably the PC's best shot at picking up a seat in the Peel Region (as it is currently an open seat, which the Conservatives won by a large margin in the federal election), but this poll shows they are trailing. The NDP has targeted the riding by running a popular member of the Sikh community. (The riding has similar demographics to neighbourhing Bramalea-Gore-Matlon which the NDP won in 2011 by running a popular member of the Sikh community). It's possible the NDP is under-polling here thanks to the low response rate among visible minorities to polls.
Burlington: Liberal 42%, PC 41%, NDP 13%, Green 4%. If the Liberals win in Burlington, it would be a huge pickup at the expense of the Tories. The Tories have never lost an election in Burlington since the riding was created in 1975. And if the Liberals are ahead here, it could mean other big gains in the 905 (like Newmarket-Aurora, Halton and Thornhill to name a few) where they have more of a history.
Mississauga-Erindale: Liberal 47%, PC 34%, NDP 13%, Green 3%. A fairly safe Liberal seat, so no surprise they are way ahead.
Niagara Falls: NDP 44%, PC 34%, Liberal 16%, Green 4%. If this poll stays true, it would end Niagara Falls' bellwether streak which dates back to 1985 (it has voted for the winning party in every election since then). The NDP picked up the seat in a close by-election earlier this year, so it was thought that out of all the ridings the NDP has picked up in recent by-elections, this would be the first to fall. However, this poll seems to indicate the opposite.
St. Catharines: Liberal 46%, PC 27%, NDP 21%, Green 5%. Liberal MPP Jim Bradley has held this seat since 1977, and this poll has him winning yet again. The race in 2011 was a close one, with Bradley defeating his PC opponent by just 1700 votes, but this poll says it wont be close again.



Projected seat changes (since Friday)

All of those riding polls threw a large wrench in my model, shifting the numbers in nearby ridings as well, creating for a plethora of seat changes in my model. Not to mention the regional polling changes as well. In all 12, ridings have changed hands compared to Friday, but the result was only a net gain of two seats for the NDP at the expense of the Liberals (no change for the Tories):

York Centre: PC to Liberal
Davenport: Liberal to NDP (riding poll was closer than expected, plus strong numbers for the NDP in the Ipsos poll in Toronto)
Etobicoke-Lakeshore: PC to Liberal (riding poll)
Trinity-Spadina: Liberal to NDP (riding poll had the Liberals up by just one)
Cambridge: PC to Liberal (riding poll)
Kitchener-Waterloo: NDP to Liberal (riding poll)
Windsor West: Liberal to NDP (riding poll)
Burlington: PC to Liberal (riding poll)
Halton: Liberal to PC
Newmarket-Aurora: Liberal to PC
Northumberland-Quinte West: Liberal to PC
Thornhill: Liberal to PC

If the Liberals end up winning Davenport and Trinity-Spadina, and pick up two of those 905 seats that flipped back to the Tories (Halton, Newmarket-Aurora, Northumberland-Quinte West and Thornhill), then they will have won a majority. This appears to be the most likely path to a majority right now for the Liberals. For the Tories, the path to a majority seems to be too difficult to attain (where are they going to win 18 more seats?). 

Here are my projected numbers for each seat. Ridings are coloured by how they voted in 2011:



Methodology
The basis of my model extrapolates the results of the 2011 election using regional polling averages from recent polls. If a party is polling at double the level they won in 2011, then that party would see its support double in each riding in the region my model (that is, if they won 5000 votes in a riding, they would be projected to win 10000). I have also taken into account recent by-elections (only in the case where a non-incumbent part won), but projecting the results of the by-elections backward to the last election based on poll numbers from around the time of each by-election. That is, if a party did 5 points better than polls indicated they would've won in a by-election, then I have made their 2011 result 5 points more than they actually received in 2011.


I've also made numerous tweaks in the model where I've felt appropriate, to boost particular candidates, or to use estimates based on recent federal election results. Also, I have inputted riding specific polling for riding that have been polled during the campaign. Please refer to previous blog posts for more details.

3 comments:

  1. I'm loving it!
    Mike (Timmy) Hudac needs to get a real job now.
    That should get his, fire 100,000 down to 99,999, for starters.
    Enjoy the evening, folks.
    D

    ReplyDelete
  2. I find myself constantly drawn back to the subject of Peel Region municipal elections 2014. Though Peel Region municipal elections 2014 is a favourite topic of discussion amongst monarchs, presidents and dictators, it is important to remember that ‘what goes up must come down.’ Inevitably Peel Region municipal elections 2014 is often misunderstood by socialists, many of whom blame the influence of television. With the primary aim of demonstrating my considerable intellect I will now demonstrate the complexity of the many faceted issue that is Peel Region municipal elections 2014.

    ReplyDelete