Latest national poll median date: September 17
Projections reflect recent polling graciously made publicly available by pollsters and media organizations. I am not a pollster, and derive no income from this blog.

Thursday, September 19, 2019

Trends 2019: Third Instalment

For new readers: these trends are different - and I would say more meaningful - then the ones you might see elsewhere, since they are retroactively recomputed* when old data becomes available. Thus, my "Trends" posts avoid showing "fake" variations caused by polls coming out with different lags. More details were provided in the first instalment of this series.

Now seems like a good time to take stock of how the campaign was going, before a completely new trend *might* possibly develop. As usual, all numbers in this post reflect my turnout adjustment of CON +1.5, LIB +0.5, NDP -1, GRN -1.


For the main parties, the takeaways are quite boring:

- The Conservatives have been between 35.8% and 36.8% for the entire time. One might think that this could be due to the poll averaging formula putting high weight on old data, but that is far from the case. For example, polls conducted since the campaign started currently have a weight of 97.7% for the national average. The observed variations are well within what a normal amount of random noise can generate with a perfectly flat trend. The seat trend is almost equally unexciting, having been within 3 seats of 145 since July 29.

- The Liberals have also been relatively flat, except for a step up from around 34% to 34.5-35% in late August, when they brought out same-sex marriage and abortion. (The step occurs at the inclusion of a Léger and an EKOS poll, and sticked despite those polls now combining for marginally negative weight in the average.) The seat projection shows a similar step up, from around 155 seats through most of August to around 160 seats in the past 3 weeks. Of course, this step is very small, and only noticeable because so little else has been going on.

Things get more interesting for the smaller parties:

- The NDP went through a very rough period in late August and early September, dropping from 13% to around 11.5%, and from around 20 seats to around 15. In the past week, however, it has stopped the bleeding, right around the threshold of 12 for official party status - the model says just above, but it could well be just below as the remaining NDP ridings are rather atypical and may not move in accordance with provincial swing.

- As I've pointed out in earlier posts, the Greens were on a gentle slide through most of August. They may have partially recovered during the NDP drop, but now that the NDP appears to have steadied itself, the Greens seem to be sliding again. The first week of the campaign has not been kind to Elizabeth May, who positioned herself far to the left at the Maclean's debate (at odds with her campaign slogan), and then with the "separatist" vs. "sovereigntist" argument making her sound rather ridiculous.

- The Bloc appears to have gained some ground vote-wise, but the seat trend has kept bouncing around 10-12. The reason is that the Liberals, their main rivals, have also been gaining ground in QC. (In fact, if I were to look at regional trends, I'd probably find that the aforementioned Liberal step up basically happened in QC.)

Are these trends about to shift dramatically? We'll get a better idea over the next few days, when pollsters will probably be very active.

I end by pointing out that if these trends are more stable than those you might see elsewhere, it might be because (in addition to the recomputations) my poll weighting method attempts to minimize variations due to changes in pollster composition: because pollsters are in the field at different times, their relative weights shift in the average, which can cause spurious variations. One example of how this helps is that in today's projection updates, both 338Canada and the CBC Poll Tracker showed the Liberal seat lead shrinking by 14-15, while mine showed a change of just 5, despite actually weighting recent polls more heavily than the CBC Poll Tracker. This happened (at least partly) because my model assigned negative weight to old Angus Reid polls after the new one came out - and Angus Reid consistently produces poor numbers for the Liberals.

*I did not recompute projections before Aug. 28 to account for the revision in the Lac-Saint-Jean baseline adjustment because the net effect is under 0.1 seats for all parties, and polling averages are unaffected.

Projection Update: Innovative Research 9/10-15 and Mainstreet 9/16-18

The following polls have been added to the model:
Innovative's 9/10-15 national poll (current weight among national polls: 5%)
Mainstreet's 9/16-18 national poll (national results only; current weight among national polls: 29%)
For a full list of included polls, see previous Projection posts.

These polls affected the projection in opposite directions: Mainstreet helped the LIBs as their lead widened from yesterday, while the model saw Innovative Research's poll as decent news for the Tories who shrank the gap in ON (from the late July IRG poll conducted with similar methodology) from 14% to 8%.

Projection as of the latest national poll (midpoint: September 17)
LIB - 159.6 (35.1%)
CON - 147.4 (36.8%)
NDP - 14.9 (11.6%)
BQ - 11.9 (4.6%)
GRN - 3.2 (8.8%)

IND - 0.5
PPC - 0.5
If you're new to my 2019 projections, view key interpretive information here.

Seat changing hands since last night's projection:
- In ON, CONs regain Northumberland--Peterborough South and St. Catharines from LIBs.
- In AB, the NDP gets Edmonton Strathcona back from CONs.

Projection Update: Nanos 9/16-18

The following poll has been added to the model:
Nanos' 9/16-18 national poll (national results only; current weight among national polls: 21%)
For a full list of included polls, see previous Projection posts.

Nanos shows the Tories up 3.4% over the past 3 days, with all other major parties down, especially the NDP.

Note that Nanos calls from 6:30pm to 9:30pm local. Therefore, in Eastern Canada, most of yesterday's respondents probably weren't aware of the Trudeau blackface/brownface news, while in Western Canada, many probably were.

Projection as of the latest national poll (midpoint: September 17)
LIB - 158.7 (35.1%)
CON - 148.7 (37.1%)
NDP - 14.3 (11.5%)
BQ - 12.3 (4.6%)
GRN - 3.1 (8.5%)

IND - 0.5
PPC - 0.5
If you're new to my 2019 projections, view key interpretive information here.

I will provide the seats changing hands with the Mainstreet update this afternoon.

Wednesday, September 18, 2019

Projection Update: Angus Reid 9/13-16

The following poll has been added to the model:
Angus Reid's 9/13-16 national poll (current weight among national polls: 19%)
For a full list of included polls, see previous Projection posts.

This poll shows very little change from the August Angus Reid poll.

Projection as of the latest national poll (midpoint: September 16)
LIB - 160.8 (35.2%)
CON - 145.9 (36.7%)
NDP - 15.2 (11.8%)
BQ - 12.0 (4.6%)
GRN - 3.1 (8.6%)

IND - 0.5
PPC - 0.5
If you're new to my 2019 projections, view key interpretive information here.

Seat changing hands since the last projection:
- In PE, CONs retake Egmont and Charlottetown from LIBs. (Note that a Mainstreet poll showing LIBs leading in Charlottetown is not factored into the model as it is behind a paywall. If it is an indication that the LIBs are holding on better in Charlottetown than elsewhere in PEI, it could be that Malpeque flips instead.)
- In QC, LIBs gain Trois-Rivières and Jonquière from CONs.
- In AB, CONs take Edmonton Strathcona from the NDP.

Projection Update: Léger 9/13-17, Mainstreet 9/15-17 and Nanos 9/15-17

The following polls have been added to the model:
Léger's 9/13-17 national poll (partial results from article; current weight among national polls: 19%)
Mainstreet's 9/14-16 15-17 national poll (national results only; current weight among national polls: 25%)
Nanos' 9/15-17 national poll (national results only; current weight among national polls: 18%)
For a full list of included polls, see previous Projection posts.
Update Sept. 19: Corrected dates for Mainstreet, which were misreported in the iPolitics article.

Overall, these numbers show stability in the race: no party moves in the same direction in all three polls. There is nothing particularly noteworthy in the regional numbers from the Léger poll released by The Canadian Press.

There was also a Mainstreet riding poll in Portneuf--Jacques-Cartier showing a comfortable Conservative lead, as expected. I revised the regional adjustment for this riding to match the population split of 58% in Quebec City CMA and 42% outside (previously I had simply included this riding in the Quebec City CMA area). This made the projection almost exactly match the poll, so no riding-specific adjustment was necessary.

Projection as of the latest national poll (midpoint: September 16)
LIB - 162.2 (35.2%)
CON - 145.1 (36.8%)
NDP - 14.9 (11.7%)
BQ - 11.6 (4.5%)
GRN - 3.1 (8.6%)

IND - 0.5
PPC - 0.5
If you're new to my 2019 projections, view key interpretive information here.

Seat changing hands since the last projection:
- In ON, CONs regain Vaughan--Woodbridge from LIBs.

Tuesday, September 17, 2019

Projection Update: Mainstreet 9/14-16

The following poll has been added to the model:
Mainstreet's 9/13-15 14-16 national poll (national results only; current weight among national polls: 29 30%)
For a full list of included polls, see previous Projection posts.
Update Sept. 17: Now includes actual decided+leaning numbers and count of decided+leaning voters (previous numbers apportioned undecided proportionally and estimated the sample size of decided+leaning voters). Note also that approximate 9/12-14 13-15 numbers (inferred from iPolitics article) will be included in the future when non-overlapping with more recent samples.
Update Sept. 19: Corrected dates for Mainstreet, which were misreported in the iPolitics article.

This poll is a mild improvement for the Liberals over Mainstreet's 9/11-13 results. The Greens are down, just like in the Nanos tracking.

Projection as of the latest national poll (midpoint: September 15)
LIB - 160.3 161.2 (34.9 35.0%)
CON - 147.4 146.2 (37.1 36.9%)
NDP - 14.9 15.1 (11.7 11.8%)
BQ - 11.5 11.4 (4.4%)
GRN - 3.1 (8.6%)

IND - 0.5
PPC - 0.4
If you're new to my 2019 projections, view key interpretive information here.

No seat lead Seat that changed hands since the last projection:
- In ON, LIBs get Vaughan--Woodbridge back from CONs.

Projection Update: Nanos 9/14-16, Ipsos 9/11-13 and Mainstreet 9/10 Lac-Saint-Jean Poll

The following polls have been added to the model:
Nanos' 9/14-16 national poll (national results only; current weight among national polls: 26%)
Ipsos' 9/11-13 national poll (national, QC, BC and partial ON results only; current weight among national polls: 17%)
Mainstreet's 9/10 Lac-Saint-Jean poll
For a full list of included polls, see previous Projection posts.

The Nanos poll shows increases from yesterday for the two main parties at the expense of the NDP (back roughly to where they were three days ago) and Greens (down almost a third of their support from three days ago).

Coming from Ipsos, the national numbers are fairly encouraging for the Liberals (and poor for the NDP). The Ipsos poll is good for the Liberals in QC, casting some doubt on the Mainstreet release with the same field dates showing tightening there. However, these polls were mostly conducted before Trudeau started campaigning in QC, where he is dogged by the secularism debate, and before the controversy around the incomprehensible French version of the Liberal campaign song. The poll is also good news for the Liberals in ON - not relative to other polls, but relative to summer Ipsos polls that had been outliers. In BC, it's a good poll for the Tories: the third one in a week giving them a 10-to-12-point lead here. Yes, other polls have it closer, but only one poll from mid-July to early September showed this kind of BC results.

The Lac-Saint-Jean poll made me review the baseline adjustment and calibrate it more carefully; this led to a CON-LIB gap quite close to the one in the poll. I did not make a further adjustment based purely on the poll, even though the BQ's low result might warrant so. This is partly because the BQ is a distant third anyway in my projection, and partly because, given that the PQ did decently well in the region in the 2018 provincial election, I have doubts whether the BQ is really as low as the Mainstreet poll suggests. We'll find out in 5 weeks!

For now, I will not be making any adjustment based on the debate commission's polls because, while 15-20% (certain and likely) are strong results for the PPC in those four ridings (Etobicoke North, Nipissing--Timiskaming, Pickering--Uxbridge and Charleswood--St. James--Assiniboia--Headingley), that still puts them at least 10% away from actually winning a seat. This does mean my PPC expected seat count (which is just Beauce) is probably a few tenths too low. I may reverse this decision if the PPC takes off in the coming weeks.

There is also a poll in Bécancour--Nicolet--Saurel, and a QC regional breakdown from Ipsos. The former is a bit higher than the Liberals than expected, but not enough to warrant an adjustment. The latter shows some deviations from adjustments based on Léger and Forum polls (CON down less in Quebec City; LIB up more on Island of Montreal, but not doing that well in the suburbs), but due to the limited sample size, I have not changed those adjustments for now.

The Ipsos poll helps the Liberals in the projection, reversing most (but not all) of their seat losses from yesterday.

Projection as of the latest national poll (midpoint: September 15)
LIB - 159.8 (34.8%)
CON - 147.6 (37.1%)
NDP - 14.9 (11.8%)
BQ - 11.6 (4.4%)
GRN - 3.1 (8.6%)

IND - 0.5
PPC - 0.4
If you're new to my 2019 projections, view key interpretive information here.

Seats changing hands since the last projection:
- In PE, LIBs get Egmont back from CONs.
- In ON, LIBs get St. Catharines and Northumberland--Peterborough South back from CONs.
- In AB, the NDP gets Edmonton Strathcona back from CONs.

Monday, September 16, 2019

Projection Update: Nanos 9/13-15

The following poll has been added to the model:
Nanos' 9/13-15 national poll (national results only; current weight among national polls: 25%)
For a full list of included polls, see previous Projection posts.

This is a noticeable one-day increase for the Tories and drop for the Greens. It can also be inferred that the Bloc gained about 3 points in QC from yesterday. The impact on the projection is meaningful - almost half of the Liberal seat lead is gone.

The latest (freely available) national results from each pollster in September are now:
Innovative C+1
Campaign =
Léger C+1
Abacus C+2
Forum C+3.9
DART C+3
Mainstreet =
Nanos C+0.4
It looks like the Liberal momentum from late August has at least stalled.

Projection as of the latest national poll (midpoint: September 14)
LIB - 156.8 (34.4%)
CON - 149.2 (37.1%)
NDP - 15.8 (11.9%)
BQ - 12.0 (4.4%)
GRN - 3.3 (9.0%)

IND - 0.5
PPC - 0.4
If you're new to my 2019 projections, view key interpretive information here.

Seats changing hands since the last projection:
- In PE, CONs regain Egmont from LIBs.
- In ON, CONs regain St. Catharines from LIBs.
- In AB, CONs regain Edmonton Strathcona from the NDP.
- In BC, CONs regain Cloverdale--Langley City from LIBs.

Sunday, September 15, 2019

Projection Update: Nanos 9/12-14 and Mainstreet 9/11-13

The following polls have been added to the model:
Nanos' 9/12-14 national poll (national results only; current weight among national polls: 21 20%)
Mainstreet's 9/11-13 national poll (QC and partial national results only; current weight among national polls: 26 24%)
For a full list of included polls, see previous Projection posts.
Update Sept. 15: Fixed small spreadsheet issue related to change in age formula at writs issuance.

Note on rolling polls (such as Nanos): Only non-overlapping samples will be used. Thus, this Nanos poll replaces the 9/11-13 one; the latter will be used when the most recent Nanos poll is 9/14-16, 9/17-19, etc. 
Note on Mainstreet poll: National results for NDP and Greens were estimated from QC results and "little change in other provinces."

As the Le Soleil article states, the big news in the Mainstreet poll is the Liberal drop in QC. Has the Liberal momentum in QC reversed because Trudeau wouldn't rule out intervening on the Bill 21 court case in the future? We will, of course, have to wait for more polls to be sure. But Quebecers' vote intentions often shift massively late in the campaign, so this is a reminder of the potential for a bad "October surprise" for the Liberals in QC.

Projection as of the latest national poll (midpoint: September 13)
LIB - 160.4 160.6 (34.6%)
CON - 146.2 146.1 (36.6%)
NDP - 15.8 15.7 (11.7%)
BQ - 11.1 (4.3%)
GRN - 3.5 (9.5%)

IND - 0.5
PPC - 0.4 0.5
If you're new to my 2019 projections, view key interpretive information here.

Seats changing hands since the last projection:
- In QC, CONs retake Jonquière and Trois-Rivières from LIBs.
- In QC, the BQ gets Pierre-Boucher--Les Patriotes--Verchères back from LIBs.
- In ON, LIBs get St. Catharines back from CONs.

Saturday, September 14, 2019

Projection Update: Abacus 9/6-10 and Nanos 9/11-13

The following polls have been added to the model:
Abacus' 9/6-10 national poll (current weight among national polls: 12%)
Nanos' 9/11-13 national poll (national results only; current weight among national polls: 21%)
Update Sept. 14: I forgot to mention that this update also incorporates national results from the Nanos summer poll ending August 9 (which replaces those from the overlapping poll ending August 2), and that the poll ending July 12 was re-added; both of these polls currently have negative weight. The results from polls ending August 16 and 23 have also been added to the spreadsheet, and will factor into the next update of projection trends (they do not count going forward as they overlap with the last Nanos summer poll ending September 6). These numbers were made available by CTV News.
For a full list of included polls, see previous Projection posts.

As usual, Abacus' numbers are very stable. It continues to see a near tie in ON and BC, unlike some other pollsters, and continues to see a smaller-than-average (but still large) CON lead in MB/SK. Also, it continues to have the NDP fairly strong, although the NDP has dropped a few points in QC and ON since the last Abacus poll. The facts that this poll is stronger for the NDP than others but weaker than the previous Abacus poll essentially cancel out for the projection.

So why did the NDP gain (modestly)? It's due to the Nanos poll. Just like Abacus, Nanos sees the NDP down a bit from its previous poll, but still stronger than most other pollsters. The difference is that because the previous Abacus poll had a very large sample, reducing its weight had a large effect; this is not the case for Nanos (whose previous poll is also considered older than Abacus' because it was conducted over 4 weeks and I use the midpoint date). Therefore, the weight of Nanos in the projection has increased noticeably, which helps the NDP and hurts the Tories and the Bloc.

I've previously mentioned that Nanos weights were fairly different from population proportions. That appears to no longer be the case, so this and future Nanos polls are perhaps not strictly comparable to the summer ones. Also, I have added Nanos back to the list of pollsters on the right as the national numbers are now available from the Nanos website.

Projection as of the latest national poll (midpoint: September 12)
LIB - 161.4 (34.7%)
CON - 145.2 (36.4%)
NDP - 16.6 (12.0%)
BQ - 10.4 (4.3%)
GRN - 3.4 (9.4%)

IND - 0.5
PPC - 0.5
If you're new to my 2019 projections, view key interpretive information here.

Seats changing hands since the last projection:
- In PE, LIBs get Egmont back from CONs.
- In QC, LIBs regain Pierre-Boucher--Les Patriotes--Verchères from the BQ.
- In ON, CONs regain St. Catharines from LIBs.
- In BC, Wilson-Raybould retakes Vancouver Granville from LIBs.

Friday, September 13, 2019

Projection Update: DART 9/11

The following poll has been added to the model:
DART's 9/11 national poll (current weight among national polls: 23%)
For a full list of included polls, see previous Projection posts.

Unlike the mid-August DART poll, this one is not a CON-friendly outlier in terms of national numbers. As a result, even though the headline number favours the Tories, the projection actually moves against them as a consequence of the lower (actually negative) weight now applied to the previous DART poll.

The main beneficiary of this change is the NDP, for whom this poll is excellent everywhere except in QC - best result since mid-July in Atlantic Canada and all summer in BC (which puts them back into third place there). This poll was conducted on the same day as the Forum poll, which was disastrous for the NDP. This is a stark illustration of the IVR vs. online difference, which has only grown since I wrote about it: the three IVR pollsters now have the NDP at 7.2-8.6% in their most recent poll, while for the online pollsters, it's 11-18%.

The Bloc also did well in this poll. Conversely, this is a bad poll for the Greens - fourth in the Atlantic and distant fourth in BC.

Getting back to the two big parties, the Tories did well and the Liberals did poorly in QC, leading to the parties being within 5 points. This was also the case in the previous DART polls, but not the case in any of the 21 other most recent results in QC. So there appears to be a strong DART house effect in QC. Elsewhere, the Liberal lead in ON is pretty good news for them in relation to both this poll's national results and the previous DART poll's ON results.

Projection as of the latest national poll (midpoint: September 11)
LIB - 161.1 (34.7%)
CON - 146.2 (36.6%)
NDP - 15.1 (11.6%)
BQ - 11.1 (4.4%)
GRN - 3.5 (9.6%)

IND - 0.5
PPC - 0.5
If you're new to my 2019 projections, view key interpretive information here.

Seats changing hands since the last projection:
- In NL, the NDP retakes St. John's East from CONs.
- In PE, CONs regain Egmont from LIBs.
- In QC, the BQ gets La Pointe-de-l'Île and Pierre-Boucher--Les Patriotes--Verchères back from LIBs.
- In ON, LIBs get St. Catharines back from CONs.
- In AB, the NDP gets Edmonton Strathcona back from CONs.
- In BC, LIBs get Cloverdale--Langley City back from CONs.

Thursday, September 12, 2019

Mini Projection Update: Léger 9/6-9's QC breakdown

The following information has been added to the model:
QC Breakdown from Léger's 9/6-9 national poll
For a full list of included polls, see previous Projection posts.

I have made small changes to some QC regional adjustments in response to the information above. They are detailed in an update to this post.

Projection as of the latest national poll (midpoint: September 11)
LIB - 162.4 (35.0%)
CON - 148.7 (37.0%)
NDP - 12.7 (10.8%)
BQ - 9.7 (4.2%)
GRN - 3.5 (9.7%)

PPC - 0.5
IND - 0.4

If you're new to my 2019 projections, view key interpretive information here.

Seat changing hands since the last projection:
- The BQ gets Terrebonne back from LIBs.

Wednesday, September 11, 2019

Projection Update: Forum 9/11

The following poll has been added to the model:
Forum's 9/11 national poll (current weight among national polls: 25%)
For a full list of included polls, see previous Projection posts.

This poll is solid across the board for the Conservatives - except in QC where it is excellent for the Liberals. In ON, this poll has the Tories with a lead that, while marginal, is (if you round it) nevertheless the largest since at least early July in a non-Ipsos poll.

The Greens get their first result above 20% in BC in about a month, while the NDP records its worst BC number of the summer (and perhaps in years?). This is enough to push the Greens ahead of the NDP in the BC polling average, as you can see on the left. In fact, this whole poll is another disaster for the NDP.

Projection as of the latest national poll (midpoint: September 11)
LIB - 162.4 (35.0%)
CON - 149.0 (37.0%)
NDP - 12.7 (10.8%)
BQ - 9.5 (4.2%)
GRN - 3.5 (9.7%)

PPC - 0.5
IND - 0.4

If you're new to my 2019 projections, view key interpretive information here.

This is the Tories' highest poll average (though not seat count) since my 2019 tracking started - I'll have to update the trends I just posted a few hours ago.

Seats changing hands since the last projection:
- In NL, CONs retake St. John's East from the NDP.
- In QC, LIBs take Terrebonne and La Pointe-de-l'Île from the BQ. (I have doubts about this - these seats get the Montreal adjustment due to their location, but they're much more francophone than other Montreal seats.)
- In ON, CONs retake Northumberland--Peterborough South and St. Catharines from LIBs.
- In BC, CONs take Cloverdale--Langley City from LIBs.

Trends 2019: Second Instalment

Update Sept. 12: Extended the trend lines to include the Sept. 11 Forum poll.
Update Sept. 13: Modified the extension to include the Sept. 11 DART poll.

The writs have been issued! Now is a good time to take a look back at the pre-campaign period, especially since we got many older pieces of data in the past few days, such as Nanos 4-week polls, two EKOS July regional breakdowns, a MQO NS poll, etc.

For new readers: these trends are different - and I would say more meaningful - then the ones you might see elsewhere, since they are retroactively recomputed when old data becomes available. Thus, my "Trends" posts avoid showing "fake" variations caused by polls coming out with different lags. More details were provided in the first instalment of this series.


Here are my takeaways:

- The Conservatives have basically not moved at all since late July. At all 13 14 points at which a polling average was computed, they got between 35.9% and 36.7 37.0 36.7%. This is quite striking, as the sample has effectively been renewed 1.5 times. Accordingly, their seat count has consistently been around 145. What looked like a late-August drop in the previous "Trends" post proved to be a blip. It will be interesting to see if the latest bump also ends up being a blip. Well, it didn't take long for the DART poll to mostly eliminate the bump.

- The Liberals have been on an upward trend, gaining almost 2 about 1.5 points since late July. Adding the older data did not affect the conclusion that the Dion report had no apparent impact on Liberal support - it came smack in the middle of the period of Liberal stability in the first 3-4 weeks of August. The size of the Liberal gains in the past two weeks is uncertain, especially in terms of seats: as you can see, the projection has become less stable recently due to some pollster disagreements.

- The NDP was pretty flat during most of August: around 20 seats and 12-13%. However, it has dropped rather quickly in the past two weeks, which probably explains the steeper Liberal increase during this time. Will it be able to recover? And remember that things might actually be much worse than what the model is showing.

- The Bloc's trend is interesting. The previous instalment showed the Bloc gaining, but as it turns out, adding the retroactive data erased most of that trend - it probably wasn't real. What remained of it was wiped out by the latest Mainstreet poll, so the Bloc is back where it was in late July - around 10 seats.

- Finally, the Green trends are interesting too: as the previous instalment showed, it had been on a gentle month-long slide in August. In terms of voting intentions, that appears to have stopped in the past two weeks, and the party may even have regained some of the lost ground; this could be thanks to the NDP's struggles. However, while the seat projection has continued to drop stopped dropping, it has not turned up either, as the Greens' numbers in BC and Atlantic Canada, their most promising regions, continue to be unimpressive.

This will be an exciting campaign for many reasons. Just focusing on seats and vote shares:

1. Things are very tight between the top two parties. A 20-seat swing between them corresponds to only about a 2-point swing, which is well within the range of aggregate polling errors and potential turnout effects. Thus, even if the election were today, the outcome would be highly uncertain. And as we've seen many times, the Canadian electorate can be extremely volatile.

2. The Liberals appear to have a large map advantage: they have been consistently behind in the vote projection, but ahead in the seat projection. We may end up with a split outcome, and there's even a small chance of a Liberal majority while the Tories win the vote count. This (majority while losing vote count) has happened only once in a Canadian federal election, and it was in 1896.

3. There are two races for third place: NDP/BQ for seats, and NDP/GRN for vote share.

4. The NDP and the BQ happen to be right around the crucial 12-seat threshold for official party status. Moreover, neither of them has many very safe seats, so a near wipeout is a real possibility for both (the BQ has already experienced that in 2011).

So buckle up, and follow along!

Projection Update: Léger 9/6-9, Forum QC 8/27-28 and Mainstreet QC Riding Polls

The following polls have been added to the model:
Léger's 9/6-9 national poll (current weight among national polls: 23%)
Forum's 8/27-28 QC poll
Mainstreet's 9/3 QC riding polls: Beloeil--ChamblyTrois-Rivières, Sherbrooke, Louis-Hébert
For a full list of included polls, see previous Projection posts.

The Léger poll is fairly neutral for the two main parties - slightly positive for the Conservatives, but it's not much. The Conservative increase from the last Léger poll comes from MB/SK (not very useful) and BC (quite low in the last poll). The Liberals got an excellent number in Atlantic Canada. Léger continues to see the BQ above 20% in QC, but this time, they have the Liberals above 35% there, which partly neutralizes the good BQ number. The Green numbers are terrible in BC and the Atlantic, but excellent (for them) in ON - this isn't good news in terms of the distribution of their support.

The Forum QC poll is a strong one for the Liberals. The Mainstreet riding polls were roughly in line with the model's expectations, except in Sherbrooke, where the Liberals are doing better than expected. Adjustments resulting from these polls are described in an update to this post.

Projection as of the latest national poll (midpoint: September 7.5)
LIB - 164.8 (35.2%)
CON - 144.5 (36.4%)
NDP - 14.2 (11.4%)
BQ - 10.1 (4.2%)
GRN - 3.4 (9.6%)

PPC - 0.5
IND - 0.4

If you're new to my 2019 projections, view key interpretive information here.

Seats changing hands since the last projection:
- In NL, the NDP regains St. John's East from CONs.
- In QC, CONs get Beauce back from the PPC.
- In QC, LIBs take Beauport--Limoilou from CONs.
- In ON, CONs retake Vaughan--Woodbridge from LIBs.
- In ON, the NDP gets Windsor West back from LIBs.

Tuesday, September 10, 2019

Mapped Projection Update: Nanos Poll Ending 9/6

The following polls have been added to the model:
Nanos' national poll ending 9/6 (current weight among national polls: 3%)
For a full list of included polls, see previous Projection posts.
Update Sept. 10: The full regional breakdown for EKOS' 7/23-25 poll has also been added. There is no effect on this post, though some of regional polling averages changed by 0.1 point.

Verbatim from last week's Nanos update:
CTV has posted some national Nanos numbers. Their inclusion in the average has little effect because:
- they are conducted over 4 weeks, and I use poll midpoint dates,
- their samples sizes are small, and
- only the national numbers are available, so they do not impact differences between regions.

Projection as of the latest national poll (midpoint: September 7)
LIB - 165.6 (35.3%)
CON - 144.0 (36.4%)
NDP - 14.0 (11.2%)
BQ - 9.8 (4.1%)
GRN - 3.5 (9.7%)

PPC - 0.5
IND - 0.4

If you're new to my 2019 projections, view key interpretive information here.

No seat lead changed hands. Here's a map of ridings currently projected ahead and potential seating plan (before Speaker election).


Projection Update: MQO Summer Polls

The following polls have been added to the model:
MQO Summer 2019 polls (7/25-31 NL7/31-8/6 PEI8/6-14 NB8/20-27 NS)
For a full list of included polls, see previous Projection posts.

This set of polls shows the Conservatives higher and the Greens lower than Narrative's August Atlantic poll. As a result, the Liberal lead in the Atlantic shrinks.

I have updated my Atlantic provincial adjustments, both in response to the new NS numbers and with a couple of tweaks to the methodology.

Projection as of the latest national poll (midpoint: September 7)
LIB - 165.8 (35.4%)
CON - 144.0 (36.4%)
NDP - 13.9 (11.2%)
BQ - 9.8 (4.1%)
GRN - 3.5 (9.6%)

PPC - 0.5
IND - 0.4

If you're new to my 2019 projections, view key interpretive information here.

Seat changing hands since the last projection:
- In NL, CONs take St. John's East from the NDP. (The model has the three main parties within 3 points.)

Projection Update: Mainstreet 9/6-8

The following poll has been added to the model:
Mainstreet's 9/6-8 national poll (current weight among national polls: 36%)
Update Sept. 10: Partial regional figures from this article and this article included.
New update Sept. 10: Full regional breakdown (and exact decided+leaning sample sizes) now included.
For a full list of included polls, see previous Projection posts.

Only the national numbers and partial regional figures are publicly available, so this update moves parties up (mainly LIB, a bit GRN) or down (mainly NDP and BQ, a bit CON) more or less uniformly across Canada. The ON numbers are great for the Liberals, but adding them did not move the needle much more than just adding the national numbers, probably because Mainstreet's two previous polls also had very strong ON numbers for the Liberals. The good Prairies and BC numbers for the Tories and good Atlantic and QC numbers for the Liberals, on the other hand, were news to the model.

As a whole, the poll has a significant impact on the projection.

Projection as of the latest national poll (midpoint: September 7)
LIB - 166.3 166.7 166.5 (35.4%)
CON - 142.7 142.6 143.2 (36.3 36.4%)
NDP - 14.1 14.0 (11.2%)
BQ - 10.4 10.1 9.8 (4.2 4.1%)
GRN - 3.6 3.5 (9.7%)

PPC - 0.5
IND - 0.4

If you're new to my 2019 projections, view key interpretive information here.

It will be interesting to see if this shift, the largest in a month for a single projection, holds.

Seats changing hands since the last projection:
- In QC, LIBs take Jonquière and Trois-Rivières from CONs.
- In QC, LIBs take Pierre-Boucher--Les Patriotes--Verchères from the BQ.
- In ON, LIBs take St. Catharines, Northumberland--Peterborough South and Vaughan--Woodbridge from CONs.
- In ON, LIBs take Windsor West from the NDP.
- In AB, CONs take Edmonton Strathcona from the NDP.
- In BC, LIBs regain Burnaby North--Seymour from CONs.
- In BC, CONs take Cowichan--Malahat--Langford from GRNs.

Comparison of IRG Riding Clusters and Current Projection

In its latest release, Innovative Research Group divided the country into 14 riding clusters and provided voting intentions for each of them, based on its three summer polls, which totaled approximately 5,700 decided respondents (7,555 respondents total). That's roughly 17 decided respondents per riding. Therefore, for most clusters, these results have sample sizes roughly on par or smaller than riding polls.

Just like for riding polls, it's interesting to compare what IRG is seeing to what my model is showing. For each of the clusters, I have computed the average support for each party across the cluster's ridings. Note that these averages are NOT weighted by population.

To facilitate comparison, I am deviating from my policy of not reproducing poll results on this blog, but of course, you're highly encouraged to click through to IRG's website and see its full release (which contains traditional national and regional top line numbers) and analysis.

The numbers before party names are IRG's; the ones after are my model averages.

Main takeaways:
- The NDP is doing better than expected in LIB/CON races, and worse than expected where it actually counts for them. It's so bad that they could be at 7 seats rather than the current projection of 17.
- Crucial LIB/CON areas are near where the model sees them.
- LIBs and CONs are not winning by as much as expected in their respective strongholds - sign of discontent about both Trudeau and Scheer?
- There is no evidence that GRN support is more efficient than the model projects.

Atlantic
LIB-CON Atlantic (14 ridings)
39 CON 40.3
37 LIB 38.7
13 GRN 13.4
  7 NDP 4.8
IRG's numbers line up remarkably well with mine in competitive Atlantic races.

Quebec
BQ (18)
28 BQ 30.0
24 LIB 29.9
22 CON 20.8
14 GRN 9.0
  7 NDP 7.5

LIB-NDP QC (27, all held by LIBs)
44 LIB 44.6
17 CON 19.9
15 BQ 16.7
12 GRN 9.0
10 NDP 7.2

NDP QC (14)
30 LIB 27.6
26 BQ 22.0
20 CON 21.5
11 NDP 16.4
10 GRN 10.1

IRG is seeing the Liberals doing slightly worse than expected (and the Greens better than expected) in the Bloc's most promising ridings, while the NDP appears to be bleeding somewhat more than expected in its remaining QC ridings to the Liberals and the Bloc. However, given the sample sizes, these could just be noise.

Ontario
LIB-CON GTA (33)
41 LIB 43.2
35 CON 39.0
12 NDP 8.2
  7 GRN 6.1

LIB-CON Rest of ON (34)
32 CON 39.4
39 LIB 37.9
10 NDP 9.8
15 GRN 9.8

Here, it's important to remember that IRG's blended sample has the Liberals up about 8 points in ON, while my projection average has them up about 4 points. Given this, the LIB-CON gap is roughly as expected in these competitive ON ridings - perhaps a bit better for the Liberals outside the GTA, but the difference is easily explained by noise.

This is also evidence that my recent adjustment (based on Corbett and Abacus sub-provincial breakdowns) giving the NDP a boost in the Toronto suburbs, an area that largely overlaps with IRG's LIB-CON GTA area, was not aggressive enough. This is bad news for the NDP: they have little chance of winning these seats, so those votes are needed elsewhere in ON. If this trend is further confirmed, I will increase the adjustment - it sure looks pretty robust now, with all three pollsters with ON breakdowns showing the same thing.

BC
LIB-CON Lower Mainland (11)
33 CON 40.1
35 LIB 37.4
10 GRN 9.2
16 NDP 7.5
In BC's LIB-CON races, it's much the same story as in ON: considering that IRG's blended sample has the Tories up by 0-1 point in BC, while my average has them up by 3-4 points, these numbers are very close to expectations. And once again, the NDP appears to be having more support than expected in useless places.

National
LIB-CON QC and West, except Lower Mainland (19)
40 CON 41.0
33 LIB 37.6
15 NDP 8.6
  7 GRN 7.6
  2 BQ 1.7
These LIB-CON numbers are again broadly in line with model projections. Here, it's the CONs that appear to be doing a bit better than expected, but just like the LIB-CON observations above, this is easily explained by noise. And again we see the NDP doing better than expected in essentially lost areas.

LIB-NDP non-QC (26)
43 LIB 42.1
18 NDP 23.8
25 CON 21.0
12 GRN 9.9

NDP non-QC (16)
26 NDP 35.3
30 CON 28.5
28 LIB 23.8
  9 GRN 8.8

All that better-than-expected NDP support in LIB-CON seats? This is where the cost shows up: the NDP is dropping more than expected in seats where it could actually be competitive. The "NDP non-QC" cluster is the most important to watch: 12 of the 14 non-QC seats where the NDP is currently projected ahead are in this group. (The only exceptions are St. John's East and Vancouver East.) If the NDP is indeed in a 3-way tie in this group, it will likely get only around 5 of these seats, or a total of about 7 outside QC.

Three-way (14)
29 CON 36.1
25 LIB 27.6
23 NDP 19.6
13 GRN 9.3
  4 BQ 3.7
Perhaps this is where the NDP might get a surprise seat or two? Given the sample size though, this could again just be random variation.

Green (15)
30 LIB 29.1
24 CON 25.5
20 GRN 24.8
18 NDP 15.4
  3 BQ 2.2
The IRG data isn't much more encouraging for the Greens: it does not look like their support has greater concentration - and therefore greater potential to generate seats - than projected by the model. So if IRG is right, it will take a significant increase in Green fortunes to get them anywhere near official party status.

Strong CON (65)
51 CON 58.1
24 LIB 19.7
10 NDP 10.2
  8 GRN 6.9
  1 BQ 1.0

Strong LIB (32)
46 LIB 53.3
25 CON 25.1
  9 GRN 9.4
13 NDP 7.9
  3 BQ 1.3

Interestingly, both main parties are doing worse than expected in their respective strongholds. This could be a sign that neither the Liberal base nor the Conservative base likes their leader, and both may be thinking about staying home or casting a protest vote in safe ridings.

Monday, September 9, 2019

Projection Update: Innovative 8/30-9/5 and EKOS 7/16-18 Regional Breakdown

The following poll has been added to the model:
Innovative Research Group's 8/30-9/5 national poll (current weight among national polls: 21%)
Full regional breakdown from EKOS Research's 7/16-18 national poll
For a full list of included polls, see previous Projection posts.

Note: I have not yet delved into the seat cluster analysis by IRG. Once I have, I'll either update this post or make a separate one. Update Sept. 10: See this separate post.

This new IRG poll is mildly positive for the Bloc: a combination of it being above 20% and the Liberals below 35% in QC. It is also a good one for the Tories in ON, as the poll shows a near tie there. The Liberals get decent news in the Atlantic, MB/SK and BC. While the national number is OK for the Greens, the diffuse distribution of the support is disappointing news for them.

The most interesting feature of this poll (other than the seat cluster analysis, which combines the data from three polls) is that it is completely different from other IRG polls: different panel provider and very different poll question. Therefore, I am treating this poll as being from a separate "IRG-2" pollster. As we'll see below, this makes a BIG difference.

EKOS provided regional breakdowns for several polls; however, only one of the breakdowns was previously unavailable, and that poll has almost zero weight due to its age.

Projection as of the latest national poll (midpoint: September 4)
LIB - 158.0 (34.5%)
CON - 145.9 (36.4%)
NDP - 17.0 (12.1%)
BQ - 12.7 (4.4%)
GRN - 3.5 (9.3%)

PPC - 0.5
IND - 0.4

If you're new to my 2019 projections, view key interpretive information here.

If I had coded this poll as from the "regular" IRG, the Liberals would be at 152.5, the Tories at 150.6, and the vote share gap would be 2.9% instead of 1.9%. Why such a big difference? The reason is that "regular" IRG has been leaning fairly noticeably toward the Liberals. The weighting model would have put a higher weight on the most recent poll, and put negative weight on the older polls showing the Liberals up 3-4 points. The upshot is that this poll would have been treated roughly as a C+5 poll! In a sense, without explicitly computing a pollster house effect and stripping it out, the model would have effectively done so. I view this as a sign that things are working as they should.

Seats changing hands since the last projection:
- In QC, the PPC takes Beauce from CONs.
- In ON, CONs take St. Catharines from LIBs.

Mini Projection Update: Regional Breakdown from Nanos 8/30

The regional breakdown from Nanos' national poll ending 8/30 has been added to the projection.
For a full list of included polls, see previous Projection posts.

This breakdown is good for the Tories in the Atlantic (slightly ahead), and good for the Liberals on QC (huge lead) and BC (lead). The effects roughly cancel out, so there is little impact on the aggregate projection.

Projection as of the latest national poll (midpoint: September 4)
LIB - 158.8 (34.6%)
CON - 145.9 (36.5%)
NDP - 17.0 (12.1%)
BQ - 11.8 (4.3%)
GRN - 3.5 (9.2%)

PPC - 0.5
IND - 0.5

If you're new to my 2019 projections, view key interpretive information here.

The projected lead has not changed in any seat.

Methodological note: I've noticed, while making this update, that Nanos has weird regional weights, so the regional breakdown does not quite add up to the national numbers. As a result:
- I previously inferred the BQ number in QC from the national number, based on Quebec's population proportion (22.7%). But Nanos appears give QC a weight of 24.8%, meaning that those inferred numbers were overestimates. Correcting this, along with a strong Liberal number in QC, accounts for the BQ's slight decrease.
- For Nanos polls, when the regional breakdown is available, I will not taking the national numbers into account (usually, I combine both sets of numbers to minimize the effect of rounding). The national support levels implied by the regional breakdown in this poll show a Liberal lead of 5.4% instead of 4.1%. But this doesn't actually help the Liberals much, since low Tory numbers in the Prairies have little consequence.

Sunday, September 8, 2019

Projection Update: Campaign Research 9/3-5 and Abacus ON Breakdown

The following poll has been added to the model:
Campaign Research's 9/3-5 national poll (current weight among national polls: 43%)
For a full list of included polls, see previous Projection posts.

43% weight on a single poll?! Yup. It's due to a combination of the following factors:
- By midpoint date, this poll is one full week more recent than any other publicly available poll.
- The model assumes that voting intentions can move more quickly in September, which puts a greater premium on recency going forward.
- This poll has a sample size larger than most (though much smaller than the Abacus poll published on Thursday).
The weighting model does put a negative weight on Campaign Research's previous polls, which brings the weight of this pollster down to a more reasonable 20%.

This poll is:
- good for the Conservatives in ON,
- good for the Liberals in the Atlantic (though not as good as previous Campaign Research polls, making the effect on the polling average rather neutral), QC and MB/SK,
- good for the Greens in the Atlantic, and
- weak for the NDP in BC.

The projection moves toward the Tories thanks to their strength in ON. Note that this update also includes sub-provincial adjustments in ON in response to the breakdown posted by Abacus; these cost just under 1 expected seat to the NDP, which appears to be doing worse than expected exactly where it needs to do well, and benefited the Liberals and Conservatives roughly evenly. (The update that not include EKOS' 4-week roll of ON polling, as it includes ON figures published as part of national polls.)

Update Sept. 8: Removed the provisional baseline adjustment for Thunder Bay--Superior North due to Bruce Hyer running again.

Projection as of the latest national poll (midpoint: September 4)
LIB - 158.9 158.8 (34.6%)
CON - 145.7 145.8 (36.5%)
NDP - 16.9 (12.1%)
BQ - 12.0 (4.3%)
GRN - 3.5 3.6 (9.2%)

PPC - 0.5
IND - 0.5

If you're new to my 2019 projections, view key interpretive information here.

Seats changing hands since the last projection:
- In ON, CONs retake Cambridge, Kitchener South--Hespeler and Vaughan--Woodbridge from LIBs.
- In ON, CONs take Essex from the NDP.
- In BC, CONs retake Burnaby South from the NDP.

Friday, September 6, 2019

Mini Projection Update: Justason's Aug/Sept Polls in Vancouver Granville and Vancouver

The following polls have been added to the model:
Justason's 8/28-30 Vancouver Granville and 9/3 Vancouver polls
For a full list of included polls, see previous Projection posts.

The Vancouver Granville poll overlapped with Mainstreet's latest Vancouver Granville poll, but shows a different result: Jody Wilson-Raybould has a moderate lead. The resulting update in the adjustment for this riding makes her inch ahead in the projection. The City of Vancouver poll did not result in an adjustment. Details are provided in an update to this post.

Update Sept. 6: Lean Tossup has pointed out that this polling was sponsored by the Wilson-Raybould campaign. As a result, it will not be part of the projection.

Also, this update includes a tiny change in Atlantic Canada (+0.1 seat CON, -0.1 seat LIB) due to fixing a small spreadsheet error related to rounding.

Projection as of the latest national poll (midpoint: August 28)
LIB - 159.9 160.0 (34.6%)
CON - 142.0 (36.1%)
NDP - 19.1 19.2 (12.6%)
BQ - 12.4 (4.4%)
GRN - 3.4 (9.1%)

IND - 0.6 0.5
PPC - 0.5
If you're new to my 2019 projections, view key interpretive information here.

Thursday, September 5, 2019

Projection Update: Abacus 8/23-29

The following poll has been added to the model:
Abacus' 8/23-29 national poll (current weight among national polls: 33%)
For a full list of included polls, see previous Projection posts.

The first thing to note about this poll is that it was in the field at the same time and before the Léger and EKOS polls published last weekend. Therefore, it should be viewed in conjunction with those polls, and not as a subsequent one. This poll also has a very large sample size, which gives it high weight despite it being a bit older than the other two surveys.

Due to the sample size (and perhaps also Abacus' penchant for stability), there is nothing particularly surprising in this poll. However, across the board, it is not a great one for the Liberals:
- their QC lead is smaller than in the previous Abacus poll;
- they get an ON lead (the previous Abacus poll was tied there), but only a small one; and
- they marginally trail in BC, unlike in the previous Abacus poll.
Abacus also continues to see decent NDP numbers, which should throw some cold water on the notion that the NDP is collapsing. Unsurprisingly, all this caused a Liberal drop in the projection, to the benefit of the Conservatives and NDP - but the changes are fairly small. The Greens continue their gentle slide.

Note that this projection also includes a change in how I treat incomplete regional breakdowns from national polls. Previously, I only included regions where a full breakdown of the main parties is available. However, given that I adjust the interim regional averages (only counting breakdowns from national polls) to match the interim national averages (only counting those same polls) anyway, also using partial regional figures would not bias the overall projection.

This mainly affects Global News/Ipsos polls, as only figures from Global News articles are fair game per my poll inclusion criteria, and those figures are often partial (e.g. only CON and LIB figures provided in a region). Currently, the effect of this is negative for the BQ (due to its 14% in the latest Global News/Ipsos poll), and mildly positive for the CONs (gains in ON, losses in QC) and the LIBs (gains in QC, losses in ON). In fact, the BQ drop in the projection is entirely due to this change rather than the Abacus poll.

Projection as of the latest national poll (midpoint: August 28)
LIB - 160.1 (34.6%)
CON - 141.9 (36.1%)
NDP - 19.2 (12.7%)
BQ - 12.4 (4.4%)
GRN - 3.4 (9.1%)

PPC - 0.5
IND - 0.5

If you're new to my 2019 projections, view key interpretive information here.

Seats changing hands since the last projection:
- LIBs regain Laurier--Sainte-Marie and Laurentides--Labelle from the BQ.
- CONs retake Oakville North--Burlington, Burlington and Richmond Hill from LIBs.
- The NDP retakes Burnaby South from CONs.
- Wilson-Raybould retakes Vancouver Granville from LIBs. Update Sept. 5: Since the Abacus poll was contemporaneous with the Mainstreet riding poll released last week, I have marginally modified the riding adjustment.

Tuesday, September 3, 2019

Projection Update: Nanos Polls Ending 8/2 and 8/30

The following polls have been added to the model:
Nanos' national polls ending 8/2 and 8/30 (current weights among national polls: 1.6% and 5.7%)
For a full list of included polls, see previous Projection posts.

CTV has posted some national Nanos numbers. Their inclusion in the average has little effect because:
- they are conducted over 4 weeks, and I use poll midpoint dates,
- their samples sizes are small, and
- only the national numbers are available, so they do not impact differences between regions.

Projection as of the latest national poll (midpoint: August 28)
LIB - 162.4 (34.9%)
CON - 139.5 (35.9%)
NDP - 18.0 (12.2%)
BQ - 13.5 (4.5%)
GRN - 3.6 (9.3%)

PPC - 0.5
IND - 0.5

If you're new to my 2019 projections, view key interpretive information here.

No seat lead changed hands. Note that the polling average is now even (actually the LIBs lead by 0.01%): the CON advantage in the projection is entirely due to the turnout adjustment.

Sunday, September 1, 2019

Projection Update: EKOS 8/27-29

The following poll has been added to the model:
EKOS' 8/27-29 national poll (current weight among national polls: 15%)
For a full list of included polls, see previous Projection posts.

The major news in this poll would normally be the LIBs' big lead in ON, but we've known for a while about EKOS' numbers there. Note that since this poll overlaps with EKOS' previous ON release, I no longer include that release in the polling average (using this poll gives EKOS a higher weight than using that one).

The other big news is the NDP's collapse not just in ON, but also in QC and even BC. This makes them fifth in the QC polling average (they're still third in BC). The EKOS numbers do not align with other pollsters', though that may partly be because the other national IVR pollsters (see the recent online vs. IVR post), Mainstreet and Forum, have not released any national or regional polls conducted in August.

Also, the BQ's number in this poll is very good. EKOS tends to produce high numbers for the BQ, but this is still the best they've had in months.

Projection as of the latest national poll (midpoint: August 28)
LIB - 161.5 (34.8%)
CON - 140.5 (36.0%)
NDP - 18.0 (12.2%)
BQ - 13.5 (4.5%)
GRN - 3.6 (9.3%)

PPC - 0.5
IND - 0.5

If you're new to my 2019 projections, view key interpretive information here.

The LIBs are now above 160 for the first time, and the BQ is also at its highest since I started projections for 2019 about a month ago. Conversely, the CONs, NDP and GRNs are all at their lowest seat-wise (though not vote-wise for the CONs and NDP) since I started.

Seats changing hands since the last projection:
- LIBs regain Egmont from CONs.
- LIBs regain Richmond Hill, Burlington and Oakville North--Burlington from CONs.
- CONs take Burnaby South from the NDP.

Saturday, August 31, 2019

Trends 2019: First Instalment

Update Sept. 1: The trends now reflect the EKOS poll released Sept. 1, which has the same field dates as the Léger poll released Aug. 31.

Below are the projection trends since the first projection issued in early August. The plots show how the projection would have evolved with the current model, had the polls been released in the order in which they were conducted. This means that the seat counts are recalculated when an older poll is published after a newer one, or when I tweak the methodology. The result is not quite the same as the retroactive best guess at what the situation was at the time (since I don't use the subsequent data to "backcast" how things were), but it's closer than just plotting projections as they're issued, as most other sites do.

Details for how these trends were generated are given at the end of the post.


The trends show the following evolution in late July and August:

- The CONs lost their seat lead in late July, and were remarkably flat around 145 seats and 36% support during most of August. The latest EKOS ON numbers and Léger national polls, however, ended this period of stability. It will be interesting to see whether this sticks, like the drop in late July.

- The LIBs have generally been in the 33-34% range and in the low-to-mid 150s in terms of seats. The Dion report had no discernible effect. The latest EKOS ON numbers and Léger polls pushed them above 34% and to nearly 160 seats for the first time. Again, it will be interesting to see if this rise is confirmed or is just a blip.

- The NDP has been flat around 20 seats and 13% support. But don't let the flat trend fool you: there is wide disagreement among pollsters regarding where the NDP is. Update: Also, although the monthly trend is flat, it will be interesting to see if the drop in the past two weeks continues.

- The BQ has quietly made some gains, passing the critical threshold of 12 seats. However, whether it would, in fact, win 12 or more seats remains essentially a coin flip. Many BQ races, both projected wins and losses, are very tight, so the uncertainty is very large relative to the seat count.

- The GRNs have been quietly fading in terms of vote share, for example falling to fourth place in BC. Remember that in 2015, the NDP was riding high during the summer thanks in part to Rachel Notley's win in AB, only to be sorely disappointed in the fall. Does the same fate await the Greens after encouraging results in recent provincial elections? Or do they manage to leverage Elizabeth May's popularity and eat into the Liberal/NDP vote, thereby earning the balance of power (or causing a Conservative majority)?

Going forward, I plan to update trends every week or two.

Here are some methodological details for these trends:
- The dates correspond to the midpoint of the most recent poll in the projection. Polls with midpoint dates within a day or two may be grouped. For example, there are Abacus and DART polls with a midpoint date of Aug. 16.5, and Léger and Ipsos polls with a midpoint date of Aug. 17.5. All of them are first included at the Aug. 17 point.
- Methodological updates (e.g. change in turnout adjustment) and adjustments to ridings identified as special cases lead to recalculations (even if those adjustments informed by subsequent polls, as long as those polls likely mainly reflect things baked in before the first projection). Poll-based adjustments to other ridings or to sub-regions do not lead to recalculations, as they may reflect actual evolution of the situation.
- Regional, sub-regional and riding-level polls are included in the numbers at the same time as the nearest national poll (except, as mentioned above, for certain polls in special-case ridings). For example, the Corbett ON poll with a midpoint date of August 12 begins to be included at the August 11 point, as the nearest subsequent national poll has a midpoint date of August 16.5.

Projection Update: Léger 8/27-29, Mainstreet Riding Poll

The following polls have been added to the model:
Léger's 8/27-29 national poll (current weight among national polls: 28%)
Mainstreet's 8/27-28 Vancouver Granville poll
For a full list of included polls, see previous Projection posts.

The Léger poll is the first national poll in 4 weeks (since Mainstreet's late July poll) showing the Liberals ahead, on the basis of strong Atlantic and BC results. The numbers in all other regions are very close to current polling averages. The poll oversampled QC (1,001 respondents), and provided a very useful breakdown of the province, which I used to update sub-provincial adjustments.

The Mainstreet poll is very close to what the model was expecting - a bit better for the LIBs, and a bit worse for the CONs. Therefore, in Vancouver Granville, the adjustment for the LIBs is changed from -10% to -9%, and the adjustment for CONs is changed from -10% to -11%. (Update Sept. 1 and 5: The adjustments have changed slightly after contemporaneous EKOS and Abacus BC numbers subsequently come out.)

Projection as of the latest national poll (midpoint: August 28)
LIB - 158.8 (34.3%)
CON - 141.9 (35.9%)
NDP - 19.6 (12.7%)
BQ - 13.2 (4.4%)
GRN - 3.6 (9.5%)

PPC - 0.5
IND - 0.5

If you're new to my 2019 projections, view key interpretive information here.

The LIBs have gained 5 seats in the projection in the past week, mostly from the CONs. This is not at all a big change (the CON-LIB gap shrank by less than 1 point), but still the largest we've seen in August. I will likely write a post on projection trends (with past projections retroactively computed to net out the methodological tweaks I've been making) this weekend, so stay tuned!

Seats changing hands since the last projection:
- LIBs regain Charlottetown from CONs.
- BQ takes Laurentides--Labelle and Laurier--Sainte-Marie from LIBs.
- LIBs take Vancouver Granville from Wilson-Raybould.
- LIBs take Victoria from GRNs.

Please enjoy the updated images below:

Polling-Related Adjustments 2019, Round 2

This post reviews polls of smaller areas than the six standard polling regions (BC, AB, SK/MB, ON, QC, Atlantic), and polls providing breakdowns of smaller areas. Resulting adjustments to the model are indicated in bold. You can view the initial round of polling-related adjustments here.

Update Sept. 5: Justason's Vancouver Granville and Vancouver polls added These polls were sponsored by the Wilson-Raybould campaign, and have therefore been removed.

Update Sept. 8: Abacus' ON breakdown added.

Update Sept. 10: MQO's summer polls added, methodology for Atlantic adjustments slightly revised.

Update Sept. 11: Forum's QC poll and 4 Mainstreet riding polls in QC added.

Update Sept. 12: QC breakdown from Léger's 9/6-9 national poll added.

Update Sept. 18: Added comment about regional adjustment applied to Portneuf--Jacques-Cartier.

I. Regional Polls

Atlantic Canada
MQO Summer 2019 polls (7/25-31 NL7/31-8/6 PEI8/6-14 NB, 8/20-27 NS)
Narrative's 7/31-8/22 poll
These polls largely confirm the initial round of adjustments. The biggest tweak is an increase for the CONs in PEI, which is useful for them as PEI is currently unusually competitive. The updated adjustments are (note: may be re-updated when the MQO NS numbers come out):
NL: CON +9.5, NDP -1.5, LIB -2.5, GRN -5.5
NS: CON -3, NDP +3, LIB -1.5, GRN +1.5
NB: CON -3.5, NDP -2, LIB +4.5, GRN +1
PE: CON +4, NDP -3, LIB -5, GRN +4
The final set of numbers from MQO shows a widening LIB lead in NS, to a size similar to what Narrative shows. This caused a slight change in NS adjustments. At the same time, I have reviewed the adjustments methodology to put more weights on recent polls and better account for differences in the size of electorate between ridings - this caused changes favouring the CONs in NL and hurting the LIBs in PE:
NL: CON +11, NDP -2.5, LIB -3.5, GRN -5
NS: CON -3.5, NDP +3, LIB -1, GRN +1.5
PE: CON +4, NDP -1.5, LIB -7, GRN +4.5

Québec
Forum's 8/26-29 poll
Léger's 8/27-29 poll
These polls large confirm the first round of adjustments: the CONs are gaining disproportionately outside the Quebec City and Montreal CMAs, and are likely not gaining at all in Quebec City. As the first round was cautious, I have significantly increased the magnitude of these adjustments, which are now:
Quebec City CMA: CON -7 -9, NDP +0.5, LIB +3 (1.5 was typo), GRN -1, BQ +0.5
Montreal CMA: CON -1 (excluding far suburbs), NDP +0.5 (excluding far suburbs), LIB +1, GRN +0.5 (Island of Montreal only), BQ -0.5
Rest of Quebec North: CON +3 (3.5 was typo), NDP -0.5 -1, LIB -2 (-2.5 was typo), BQ +0.5
Rest of Quebec South: CON +3.5 (3 was typo) +4, NDP -0.5, LIB -2.5 (-2 was typo), GRN -0.5, BQ +0.5
(Salaberry--Suroît is mostly outside of Montreal CMA but shares some characteristics with it. Since the Montreal CMA and Rest of QC S adjustments go in opposite directions, I have left this riding unadjusted. Moreover, due to the large difference between Quebec City and Rest of QC North adjustments, I am computing the adjustment in Beauport--Côte-de-Beaupré--Île-d'Orléans--Charlevoix and Portneuf--Jacques-Cartier according to their rough population distribution: 55% and 58% respectively in Quebec City CMA, 45% and 42% outside.)
This can help the Tories make gains outside the Quebec City area, but the two Liberal seats in Quebec City are now unlikely to flip. In fact, the projection for Québec is now within 2 points for all parties of the riding poll published by Mainstreet in July. I had refrained from making an adjustment based on the riding poll because I through the regional adjustment might end up doing the job, and that's exactly what happened.

Léger's 9/6-9 poll
Further tweaks to the adjustment in bold:
Quebec City CMA: CON -10, NDP +0.5, LIB +3.5, BQ +0.5
Montreal CMA: CON -1 (excluding far suburbs), NDP +1.5 (Island of Montreal), +1 (near suburbs), LIB +1 (Island of Montreal), +0.5 (near suburbs), GRN +0.5 (Island of Montreal only), BQ -0.5
Rest of Quebec North: CON +3.5, NDP -1.5, LIB -1.5, BQ +0.5
Rest of Quebec South: CON +4, NDP -1, LIB -2, GRN -0.5, BQ +0.5

Ontario
Corbett Communications' 8/11-13 poll
Update Sept. 8: Abacus' 8/23-29 ON breakdown
Nothing out of the ordinary here - no resulting sub-provincial adjustment. There is no huge deviation from expectations based on provincial averages. However, now that we have two pollsters providing a breakdown, I am comfortable making adjustments based on moderate deviations where there is some agreement, namely that the NDP does better than expected in the Toronto suburbs (but, per Corbett's finer breakdown, not in Hamilton-Niagara) and Eastern ON outside Ottawa, but worse than expected in the North and Southwest:
Toronto CMA outside 416: CON -0.5, NDP +2, LIB -0.5, GRN -1
Southwest: CON +1.5, NDP -2, GRN +0.5
North: NDP -1.5, GRN +1.5
East outside Ottawa: CON -3, NDP +3, LIB -0.5, GRN +0.5
I have also made a minor change to the adjustment for the 7 central Toronto ridings to CON +2, NDP -5, LIB +2.5, GRN +0.5 (previously +2/-5/+3/0).
All these adjustment are fairly cautious - very roughly half the size suggested by the polls - to account for the possibility that some of the apparent effect might be due to random variation. Because they are broadly consistent with the Corbett poll, for the purposes of retroactive computations to produce trends, these adjustments will be deemed to have been made in response to that poll.

Manitoba/Saskatchewan
Campaign Research's 8/9-13 poll
EKOS' 8/20-22 poll
Update Sept. 1: EKOS' 8/27-29 poll
Both All of these suggest that the NDP is falling a lot more in SK than in MB, and that, conversely, the CONs and GRNs* are rising much more in SK than in MB. However, due to the very small sample sizes (there were part of national polls) and the ongoing provincial election in MB, I am not making any adjustments for now.
*GRNs was a typo - the numbers didn't show that, even before the update.

Vancouver
Justason Market Intelligence's 9/3 poll
This poll is broadly in line with the current projection for the City of Vancouver (unlike the April poll by Justason, which had some very strange numbers and a very small sample). The Tories are a bit lower than expected, and the GRNs a bit higher. Justason overestimated the BC Greens in the last provincial election, so I'm going to hold off on any adjustments.


II. Riding Polls (by Mainstreet unless otherwise noted)

Note: A riding poll will result in a model adjustment ONLY IF it shows a substantial deviation from regional swing and/or it is in a riding with unusual factors. Most riding polls will therefore be ignored: they generally did not help much in 2015.

Victoria, 8/9-10
This poll did not come as a huge surprise to the model because it takes into account that, in 2015, the Green vote was inflated and the Liberal vote was depressed due to the Liberal candidate dropping out (but too late to be removed from the ballot). In fact, there are several ways to make the adjustment, and I picked one of the relatively cautious ones; the poll suggests that my caution was unwarranted. Therefore, rather than making this a poll-based adjustment, I amended the baseline adjustment (as previously noted in the model post) by CON +1.3 (thus cancelling the CON adjustment), NDP -2.8 (to -6.5), LIB +6.5 (to +22.8), GRN -5 (to -16.3). This shifts all numbers toward (but not quite all the way to) the poll.

Windsor West, 8/21
Because this poll essentially guides the Sandra Pupatello adjustment rather than provides information about a differential evolution of the riding per se, it leads to a baseline adjustment. Details were provided in the model post.

Vancouver Granville, Mainstreet 8/27-28 and Justason 8/28-30
The Mainstreet poll isn't very different from the May Vancouver Granville poll by Mainstreet, if you compare the party figures from the May poll (which are the ones I used) to the candidate figures in this one (which are more relevant now). Similarly, Justason continues to see a moderate lead for Wilson-Raybould, like in its April poll. The adjustment is amended from CON -10, NDP -8, LIB -10, GRN -2 to CON -11, NDP -8, LIB -9 -9.5, GRN -2 -1.5 (update due to contemporaneous Léger and EKOS polls subsequently published) CON -13, NDP -8, LIB -10.5, GRN -1.

Beloeil--Chambly, 9/3
This is almost exactly as the model expects!

Trois-Rivières, 9/3
LIBs and BQ a bit higher than the model expected, and other parties a bit lower, but not enough to warrant an adjustment.

Sherbrooke, 9/3
LIB much higher than expected. A cautious adjustment for now: CON -4, NDP -2, LIB +6.

Louis-Hébert, 9/3
Almost exactly as the model expects for LIB, GRN, NDP. CON a bit lower and BQ a bit higher than expected, but not enough to warrant an adjustment.

Thursday, August 29, 2019

Projection Update: Angus Reid's 8/21-26 Poll and EKOS' 8/25-28 ON Poll

The following polls have been added to the model:
Angus Reid's 8/21-26 national poll (current weight among national polls: 24%)
Update: EKOS' 8/25-28 ON poll (current weight among ON polls: 4%)
For a full list of included polls, see previous Projection posts.

The Angus Reid poll is good news for the Tories in the Atlantic and ON, where they're marginally ahead. And it's excellent for the Liberals in QC.

Update: The EKOS poll is similar to the ON portion of EKOS' recent national release. In the past, I have been reluctant to include regional portions of a national poll released without national numbers by EKOS. This time is a bit different though: we do have a recent national poll by EKOS, which means that the weight on these numbers would be low (as you can see above - and it's effectively even lower if you consider that the weight on the previous EKOS numbers goes down). So this is effectively somewhat like a national poll with an ON oversample, similar to polls by Campaign Research.

Projection as of the latest national poll* (midpoint: August 23.5)
LIB - 155.0 156.5 (33.9 34.1%)
CON - 146.1 145.0 (36.3 36.2%)
NDP - 20.2 19.7 (12.9 12.7%)
BQ - 12.1 12.0 (4.3%)
GRN - 3.7 (9.4%)
IND - 0.5

PPC - 0.5
If you're new to my 2019 projections, view key interpretive information here.
*For now, the EKOS ON poll is treated as if it also has a midpoint date of August 23.5.
(Very minor changes outside ON because I fixed a data entry error for auxiliary information about the Angus Reid poll.)

With the Angus Reid poll, despite the good national headline figures for the Tories, it's the Liberals that gained about an expected seat from the Bloc. Why?
- The weighting formula now puts some negative weight on the July Angus Reid poll to reduce the pollster house effect. That poll (like many previous Angus Reid polls) was terrible for the Liberals.
- This poll undersampled ON, so the CONs' good ON number didn't help them as much as one might expect. By contrast, the QC sample, which was in proportion with the national sample size, really helped the LIBs.
- Additional Tory vote gains in SK/MB aren't very useful in terms of seats.

Update: With the EKOS poll, unsurprisingly, the LIBs gain at the expense of the CONs and NDP in ON.

Seats whose projected lead changes relative to the last projection:
- CONs gain Charlottetown from LIBs.
- LIBs gain Rimouski-Neigette--Témiscouata--Les Basques from the NDP.
- LIBs take Salaberry--Suroît, Laurier--Sainte-Marie and Laurentides--Labelle from the BQ.
- CONs regain Cambridge and Vaughan--Woodbridge from the LIBs. (This no longer happens after the EKOS poll.)