2019 Result Map

Here is a map of the results of the 2019 General Election, as well as a potential seating plan (before the election of the Speaker). Over the next few weeks, I will add a series of analysis posts, so check back from time to time!

Projection by Region

Here are some details of the final projection by region. I will pick one riding per region where either I am skeptical of the projection, or that will be interesting to watch tonight.

Atlantic Canada (32)
LIB - 18.8 (19 projected ahead), 38.5%
CON - 9.0 (9), 28.6%
NDP - 4.1 (4), 19.7%
GRN - 0.1 (0), 10.2%
Riding where I don't believe the projection: Acadie--Bathurst. The NDP's severe organizational problems in NB are not reflected in the model, as there was no breakout of NB in any poll since August.

Québec (78)
BQ - 35.3 (39), 31.3%
LIB - 31.6 (30), 32.1%
CON - 8.7 (7), 15.7%
NDP - 1.8 (1), 12.5%
GRN - 0, 4.1%
Riding where I don't believe the projection: Saint-Maurice--Champlain. Minister François-Philippe Champagne has a high profile in QC and has spent a lot of time with the people of his riding. I expect him to survive.

Ontario (121)
LIB - 64.9 (68), 38.5%
CON - 44.3 (43), 33.9%
NDP - 11.7 (10), 18.7%
GRN - 0, 5.8%
Riding to watch: I have trouble picking since there are so many...

Manitoba/Saskatchewan (28)
CON - 19.9 (20), 48.4%
LIB - 5.0 (6), 21.9%
NDP - 3.1 (2), 3.1%
GRN - 0, 3.7%
Riding to watch: Regina--Wascana. Long-time MP and Liberal Minister Ralph Goodale could be threatened. He survived in 2011 when things were much worse for the Liberals overall, but this year, there is anger in SK, and the Conservative leader is from a neighbouring riding. How much will the NDP vote rally around him?

Alberta (34)
CON - 32.2 (33), 62.4%
NDP - 1.4 (1), 15.5%
LIB - 0.4 (0), 16.4%
GRN - 0, 2.6%
Riding to watch: Edmonton Strathcona. This is the one projected NDP seat, though I think the model is probably too sanguine in terms of margin of victory. A Mainstreet poll this week was publicly described as showing a very tight race.

British Columbia (42)
CON - 18.5 (18), 32.8%
NDP - 11.8 (13), 25.0%
LIB - 9.9 (9), 27.1%
GRN - 1.1 (1), 11.3%
Riding to watch: Nanaimo--Ladysmith. This isn't just a two-way NDP/Green race, but actually a three-way NDP/Green/Tory race. If the NDP and Greens split the left-wing vote evenly, the Conservative candidate could squeak by. This is also where the limitations of my shortcut for estimating probabilities of victory are apparent: Green MP Paul Manly obviously has a better than 10% chance of holding this riding, but his probability of victory is being underestimated due to being in third place.

FINAL 2019 Projection: \o/

The following polls have been added to the model:
Mainstreet's 10/19-20 national poll (current weight among national polls: 30%)
Research Co's 10/18-20 national poll (current weight among national polls: 9%)
EKOS' 10/17-20 national poll (current weight among national polls: 11%)
Forum's 10/18 national poll (current weight among national polls: 3%)
Nanos' 10/19 and 10/20 national polls (national results only; current weight among national polls: 6% and 18% respectively)
I have also updated ON regional adjustments using the breakdown from Campaign Research's 10/16-20 national poll, as well as Atlantic Canada and BC regional adjustments using the breakdown from Innovative Research's 10/15-17 national polls.
For a full list of included polls, see previous Projection posts.

And I'm finally done! This projection update includes polls that are favourable to the Liberals (EKOS, Forum) and polls that are negative for them (Mainstreet, Nanos). Because the latter have more weight thanks to their recency and sample sizes, the projection moves against the Liberals.

As you can see, the final projection gives an insignificant edge to the Liberals, which is flipped from the "quickie" projection I posted on Twitter just after the final poll was released. The Liberals gained roughly 1 seat, at the expense of the NDP. What accounts for the difference?
- In order to avoid Nanos and Mainstreet being the only polls with a midpoint date after Oct. 19, I used the Oct. 19-20 roll-up from the EKOS poll, and inferred Oct. 17-18 results, which were included as a separate poll. This slightly increases the weight on EKOS, which helps the Liberals and hurts the Tories. (As it turns out, the inferred results were given a weight below 0.5% by the model, so it doesn't matter if they're slightly off.)
- Updates to regional adjustments for BC, ON and Atlantic Canada help the Tories and Liberals and hurt the NDP and Greens.

Projection as of the latest national poll (midpoint: October 20)
LIB - 133.2 (32.1%)
CON - 132.7 (33.6%)
BQ - 35.3 (7.1%)
NDP - 34.5 (18.1%)
GRN - 1.2 (5.9%)

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

Without the turnout adjustment, the projection would be:
LIB - 134.2 (31.6%)
CON - 126.0 (32.1%)
NDP - 39.1 (19.1%)
BQ - 36.1 (7.1%)
GRN - 1.4 (6.9%)

IND - 0.7
PPC - 0.6

I very crudely estimate the confidence intervals of the projection as follows:
80% confidence intervals
LIB 95-175
CON 95-170
BQ 20-50
NDP 15-50

95% confidence intervals
LIB 85-190
CON 85-185
BQ 15-55
NDP 12-55

Seats changing hands since the last projection:
- In NL, LIBs get St. John's South--Mount Pearl back from the NDP.
- In NL, CONs regain Avalon from LIBs.
- In PE, CONs regain Egmont from LIBs.
- In QC, the BQ regains Beauport--Côte-de-Beaupré--Île d'Orléans--Charlevoix from CONs.

- In QC, the BQ regains Saint-Maurice--Champlain and Argenteuil--La Petite-Nation from LIBs.

- In ON, CONs regain Peterborough--Kawartha from LIBs.

- In ON, the NDP regains Oshawa from CONs.

- In MB, the NDP gets Elmwood--Transcona back from CONs.

- In BC, CONs regain Cloverdale--Langley City from LIBs.

- In BC, the NDP retakes Nanaimo--Ladysmith from GRNs.

Below is a map showing the seat-by-seat leading party. Please keep in mind that my projection uses riding polls only if they are not behind a paywall - i.e. very few riding polls outside QC are used. Moreover, I only take into account factors described in my methodology description. So I am tying one hand behind my back - no "the NDP isn't winning Acadie--Bathurst because of their organizational problems in NB and because this year's candidate does not share a last name with long-time MP Yvon Godin."

As you can see, the Liberals have a 5-seat lead in ridings projected ahead. The sums of probabilities are roughly equal because the Liberals win more tight races. Below are the 75 ridings where the gap between the top two parties is less than 5%:

A gap of less than 5% means that there is at least a 30% chance that the projected winner won't actually win. I expect to get almost half of these ridings (say, 30-35 out of 75) wrong.

Next, here are the 55 ridings where the gap is between 5% and 10%:

In "standard" two-way races, these projections should be correct 70-85% of the time. However, in three-way races, the probability of error significantly increases. This is also the case in special situations, such as the Territories, for which we don't have any polling. Therefore, I would expect to get 15-20 of these wrong.

Of the remaining 208 ridings, the standings are:
91 CON
82 LIB
20 BQ
14 NDP
  1 GRN
Of these:
- 56 have a gap between 10 and 16 points (85-95% chance of being correct in a standard two-way race),
- 58 have a gap between 16 and 22 points (95-99% chance of being correct), and
- 94 have a gap above 22 points.
Therefore, I would expect to get 10-15 of these wrong. Adding up the expected numbers of mistakes, there should be 55-70 of them; in other words, I expect 79-84% accuracy on an average night. If the polling averages are right, accuracy will likely be slightly higher, while if there is a significant polling error, then accuracy could be slightly lower.

Here's a potential pre-Speaker seating chart of the House of Commons:

Trends 2019: Seventh and Final Instalment

For new readers: these trends are different - and I would say more meaningful - than 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. My post on comparing models highlights other advantages of my approach.

As I work on my final projection (considering adjustments based on some sub-regional breakdowns from Campaign Research and IRG), here is the final trends post. As usual, all numbers in this post reflect my turnout adjustment of CON +1.5, LIB +0.5, NDP -1, GRN -1.

Methodological note: The data points for September 26, October 3 and October 11 are not plotted: each of those dates was the midpoint of only one public national poll (Nanos tracker), making those projections less reliable given that discounting has become very aggressive.

1. The major story of the final week of the campaign was a steep further rise of the NDP, peaking at 40 seats and 19% national support on Monday/Tuesday, followed by an equally steep pull back, to 31-32 seats and under 17% on Friday/Saturday. However, the final NDP number is back up to the mid-30s and 18.2%: the last two polls (by median date) showed the NDP either not dropping from its mid-week levels (Mainstreet) or actually surging on the final day (Nanos). If the momentum that Nanos picked up is real and continues on Election Day, the NDP could approach, or even surpass its 2015 seat count of 44. However, if the final polls just happened to see noise on Sunday and a few tight races don't go the NDP's way, it could easily end up in the 20s.

Either way, it'd be a much better result than envisaged at the start of the campaign, when the NDP was hovering around 15 seats. Will Jagmeet Singh continue to perform well once the election is over, or will he prove as ineffective as he was before the campaign?

2. The Liberals appeared to have turned things around at Thanksgiving. They hit rock bottom a day or two before Thanksgiving, and have been on an upward creep since, except for the final day polls. Were strategic considerations making their way into voters' minds? If the final day downtick was noise and/or their turnout matches the Tories', the Liberals should eke out a win. If both are true, they should win comfortably. And if in addition, the IVR pollsters are right about Ontario so that the Liberals win it by ~10 points instead of 5, they may start thinking about a majority. If, on the other hand, the Tories beat the polls significantly like in 2011, the NDP gains more Liberal voters, and the Bloc's supporters, who tend to be older, turn out in droves - i.e. the perfect storm against Liberals - the Grits could fall dangerously close to 100.

If the Liberals win the most seats, it will probably be because some soft voters, having not seen any other plausible option (those in ON possibly being turned off the Tories by Doug Ford), reluctantly came home. Given the country's relatively good economic performance in the past few years, it seems likely that Trudeau has been a net drag on his party's fortunes this year.

3. The Tories fell through Thanksgiving weekend, though there appears to have been a modest recovery in the last few days. The general trend of the past month has been ugly for them. One cannot help but feel that, with just an average-quality leader, the Tories would have capitalized on Trudeau's baggage and be on their way to at least a strong minority - strong enough that Trudeau would probably feel obliged to resign. Indeed, that would be the case if the Tories just held their constant August level of 36%. Unlike for the Liberals, it is hard to see the Tories getting a majority because they are so far behind in QC. Even if they somehow tie the Liberals in both ON and Atlantic Canada and get above 20 seats in BC, they'd probably still need 20 seats in QC, which is just not happening. On the other hand, it is not difficult to imagine the Tories failing to make substantial gains: one just needs the IVR pollsters to be right in ON and the Tories dropping some tight races in the Atlantic and BC.

4. The Bloc's explosive rise appeared to stop just before Thanksgiving, and its support has held steady this week. It's looking like a seat count in the 30s for them if the tossups don't all fall one way or another. The main risk for the Bloc is that a significant fraction of its supporters actually approve of the Liberal government. If these voters decide to "play it safe" after all, the Liberals could hold up surprisingly well and avoid significant losses in QC, which would limit the Bloc to picking up NDP seats and getting in the 20s. On the other hand, Quebecers like to jump on bandwagons; if the Bloc capitalizes on that, it could well crack 40.

5. As I've commented in every (or almost every) trends post since the first one in late August, the Greens are on a downward track. Last week, I ventured that the Greens could fall below their 2008 popular vote of 6.78%, and it looks like that will be really close. At this point, it's all about saving Nanaimo--Ladysmith for them - gains are still possible, but they would be gravy. Just like Trudeau and Scheer, May needs to go.


*Changes made on or after Sept. 25 are only reflected back to the start of the campaign (Sept. 11). These include the re-weighting of standalone regional polls (i.e. regional numbers not part of a national poll), the Guilbeault baseline adjustment in Laurier--Sainte-Marie and fixing an old spreadsheet error. These only make a barely visible (if at all) difference on the graphs.

Mapped Projection Update: Campaign Research 10/16-20, Abacus 10/17-19, Ipsos 10/17-19

The following polls have been added to the model:
Campaign Research's 10/16-20 national poll (current weight among national polls: 35%)
Abacus' 10/17-19 national poll (current weight among national polls: 13%)
Global News Ipsos' 10/17-19 national poll (national and BQ results from Global News article only; current weight among national polls: 17%)
I have also considered the following polls for a potential update to ON regional adjustments, but no change was made:
Nanos' 10/7-13 ON breakdown
Léger's 10/10-14 GTA poll
For a full list of included polls, see previous Projection posts.

Taken together, these polls show a fairly static race between the Tories and Liberals. Abacus and Ipsos show the NDP losing some ground. Abacus has the Bloc up significantly since its last poll, but since that poll is over a week old and other polls have shown a Bloc increase since then, this does not shift the projection much. Abacus also has the Liberals jumping out to a lead in BC, but Ipsos shows them falling further behind, so there is still no clarity there (though Ipsos BC numbers cannot be included in the projection).

Accordingly, the NDP is slightly down in the projection, which is mostly stable.

Projection as of the latest national poll (midpoint: October 18)
LIB - 137.8 (32.4%)
CON - 133.0 (33.7%)
BQ - 34.0 (7.1%)
NDP - 30.4 (16.9%)
GRN - 1.8 (6.7%)

IND - 0.6
PPC - 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 get Beauport--Côte-de-Beaupré--Île d'Orléans--Charlevoix back from the BQ.
- In ON, LIBs get Cambridge and Peterborough--Kawartha back from CONs.
- In ON, CONs get Oshawa back from the NDP.
- In MB, CONs regain Elmwood--Transcona from the NDP.
- In BC, GRNs retake Nanaimo--Ladysmith from the NDP.

Here is, belatedly, the map that I promised for last night. Of course, the map will be updated again when I post the final projection tonight!

Projection Update: Léger 10/17-18 and Nanos 10/17-19

The following polls have been added to the model:
Léger's 10/17-18 national poll (current weight among national polls: 14%)
Nanos' 10/17-19 national poll (national results only; current weight among national polls: 30%)
For a full list of included polls, see previous Projection posts.

Update Oct. 20: Added the Nanos regional breakdown for yesterday (with imputed values for MB/SK and AB) that were publicly released. (Earlier version of this post had just the QC and BC numbers discussed on CTV.)

Even though its headline is a national tie, the last Léger poll is rather bad news for the Liberals:
- Reduced lead in ON
- Bloc still increasing in QC (contrary to what most other polls show)
- Lost lead to Tories in BC
Even worse, this Léger poll's QC breakdown shows a reversal from the Léger megapoll of QC in that the Liberal vote seems to be increasingly concentrated in Montreal again. It's possible that the Liberals dropped more in the Montreal area to the NDP last week, but is getting some of those voters back. Or this could be noise. In any case, the QC adjustments have been revised. (I still need to work on the ON ones, though I will probably wait for the final Campaign Research poll for that.)

The Nanos poll is also bad news for the Liberals, who relinquish their national lead from yesterday. Nanos should have another release tonight with today's data though, so stay tuned.

All this bad news for the Liberals sends the projection back to a virtual tie.

Projection as of the latest national poll (midpoint: October 18)
LIB - 134.6 134.9 (32.1%)
CON - 133.1 132.8 (33.5%)
BQ - 34.0 (6.9%)
NDP - 33.4 (17.5%)
GRN - 1.7 (6.8%)

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

Seats changing hands since the last projection:
- In QC, the BQ takes Richmond--Arthabaska, Trois-Rivières, Beauport--Limoilou and Beauport--Côte-de-Beaupré--Île d'Orléans--Charlevoix from CONs.
- In QC, LIBs retake Outremont from the NDP.
- In ON, CONs retake Peterborough--Kawartha and Cambridge from LIBs.

Projection Update: Campaign Research 10/16-19

The following poll has been added to the model:
Campaign Research's 10/16-19 national poll (current weight among national polls: 48%)
For a full list of included polls, see previous Projection posts.

Today's Campaign Research update adds about 1,500 respondents to the 2,000-respondent poll published yesterday. This update did not change the numbers much: the Tories and Liberals are up marginally, while the NDP and Greens are down marginally. Accordingly, the two main parties make small gains in the projection at the expense of the NDP and Bloc.

Projection as of the latest national poll (midpoint: October 17.5)
LIB - 139.1 (32.2%)
CON - 130.9 (33.0%)
NDP - 33.0 (18.0%)
BQ - 32.2 (6.9%)
GRN - 1.7 (6.6%)

IND - 0.6
PPC - 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 Trois-Rivières, Beauport--Limoilou and Richmond--Arthabaska from the BQ.
- In SK, CONs regain Saskatoon West from the NDP.

Projection Update: Mainstreet 10/16-18 and Riding Polls, QC Regional Adjustments

The following polls have been added to the model:
Mainstreet's 10/16-18 national poll (national results only; current weight among national polls: 19%)
Mainstreet's 10/? Québec poll
Mainstreet's 10/15 Vancouver Granville poll
Mainstreet's 10/16 Laurier--Sainte-Marie poll
Moreover, within-QC adjustments were updated using Léger's 10/13-15 QC poll. (Nanos' 10/4-13 QC breakdown and Forum's 10/11 QC poll were also considered, but were much less useful.)
For a full list of included polls, see previous Projection posts.

First, the main lesson from Léger's megapoll of QC is that the Liberal vote is slightly less concentrated in Montreal than was the case earlier during the campaign. This could be because Liberal losses to the NDP came disproportionately in Montreal. The Liberals are surprisingly competitive in Eastern QC, which is good news for someone like Minister of National Revenue Diane Lebouthillier. These regional adjustments push the Liberals up 0.5-1 seat in the projection, at the expense of the BQ and Tories.

The Mainstreet tracker shows a small Liberal increase at the expense of the NDP, Greens and BQ. In the projection, however, it is the NDP that profits: Mainstreet's anti-NDP lean means that just by staying at a high level (even with a small drop), the NDP impresses the model. At the same time, the modest Liberal increase in the poll was mostly priced in. The Bloc, Greens and Tories are down very slightly (under 1 seat each).

All three riding polls resulted in adjustments, though the ones for Vancouver Granville and Laurier--Sainte-Marie are inconsequential, as the difference between the top two candidates is almost unaffected. However, for Québec, the adjustment flips the riding, pushing Minister Duclos ahead by a few points.

Net effect of all this? Liberals and NDP gain about one seat each, Tories and Bloc lose about one seat each. Underwhelming, I know...

Projection as of the latest national poll (midpoint: October 17)
LIB - 137.1 (32.2%)
CON - 129.2 (33.0%)
NDP - 35.3 (18.0%)
BQ - 33.7 (6.9%)
GRN - 1.6 (6.6%)

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

I'm mostly caught up now, except for considering the following information for within-ON adjustments: Campaign Research's ON breakdown, Nanos' ON breakdown and the Léger GTA poll. (I will not be adding the Lethbridge College AB poll: due to its age, it would only have a ~1% weight in the AB average, not shifting any party's number by more than 0.1%.) This will be done tonight, at which point I will post a mapped projection update.

Seats changing hands since the last projection:
- In QC, the BQ takes Richmond--Arthabaska, Beauport--Limoilou and Trois-Rivières from CONs.
- In QC, LIBs get Avignon--La Mitis--Matane--Matapédia, Gaspésie--Les Îles-de-la-Madeleine, Québec and Saint-Maurice--Champlain back from the BQ.
- In QC, the BQ retakes Laurier--Sainte-Marie and La Prairie from LIBs.
- In QC, the NDP retakes Outremont from LIBs.
- In SK, the NDP gets Saskatoon West back from CONs.

Projection Update: Nanos 10/16-18, Insights West 10/13-16 BC Poll

The following polls have been added to the model:
Nanos' 10/16-18 national poll (national results only; current weight among national polls: 19%)
Insights West's 10/13-16 BC poll
Mainstreet's 10/9 Jonquière poll
For a full list of included polls, see previous Projection posts.

Nanos has some good news for the Liberals this morning, with the Grits gaining a point, the Tories losing a point, and the NDP also down slightly. The only issue for the Liberals is that the Bloc also jumped: Nanos now finally has it above 30%, even as other pollsters have it down from a few days ago.

This update also incorporates regional adjustments based on Insights West's BC poll. Because that poll does not give decided+leaning numbers, it does not figure in the projection average (manually distributing them is unreliable due to weighting issues). However, it is used to update sub-BC adjustments, which resulted in the Tories gaining about 0.5 seats at the expense of the NDP. The MB/SK breakdown from yesterday's Campaign Research poll was also used to update the MB/SK adjustments. This was also marginally positive for the Tories and negative for the NDP.

Finally, I've finally gotten around to adding Mainstreet's Jonquière poll, which resulted in a riding adjustment making a Bloc win only marginally less certain.

I'm now finally working on incorporating the following information from this week: Léger fine sub-QC breakdown, Campaign Research's ON breakdown, Nanos QC and ON breakdowns, Léger GTA poll and Forum QC breakdown. Today's public riding polls will also be added later.

Somewhat surprisingly, the projection did not move much. It appears that the good news for the Liberals was mostly pre-empted by yesterday's Campaign Research poll, which had the same field dates.

Projection as of the latest national poll (midpoint: October 17)
LIB - 136.2 (32.1%)
CON - 130.2 (33.0%)
BQ - 34.7 (6.9%)
NDP - 33.9 (17.5%)
GRN - 1.8 (7.1%)

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

Seats changing hands since the last projection:
- In ON, LIBs get Peterborough--Kawartha back from CONs.
- In SK, CONs regain Saskatoon West from the NDP.

Projection Update: Innovative 10/15-17, Campaign Research 10/16-18

The following polls have been added to the model:
Innovative's 10/15-17 national polls (current weight among national polls: 13% and 8%)
Campaign Research's 10/16-18 national poll (national results only; current weight among national polls: 36%)
For a full list of included polls, see previous Projection posts.

Note #1: This post will be updated later with the ridings where the lead changed. The regional breakdown on the left will also be updated at that time.

Note #2: I have not yet incorporated Campaign Research's sub-regional breakdowns into the projection, and I am aware of Insights West's BC poll. I'm also behind on adding the following information released earlier in the week: Léger fine sub-QC breakdown, Nanos QC and ON breakdowns, Mainstreet Jonquière poll, Léger GTA poll and Forum QC breakdown. I hope to have an update tonight where I catch up on all this...

Overall, the IRG polls were fairly neutral for the LIB/CON race: both parties lost equal ground relative to previous IRG polls. The Campaign Research poll helps the Liberals, who pull into a national tie and a 4-point lead in ON. These polls are not great news for the NDP: the smaller IRG poll has a very low NDP number, while Campaign Research has it down 1 point from last week.

As a result, the Liberals are up in the projection, while the NDP is down. After both flirting with 40 seats, both the NDP and Bloc are back down in the mid-30s. The Liberals move back to where they were on October 8, and are once again ahead of the NDP in MB/SK, AB and BC.

Projection as of the latest national poll (midpoint: October 17)
LIB - 135.9 (32.2%)
CON - 130.2 (33.2%)
NDP - 34.8 (17.7%)
BQ - 34.2 (6.9%)
GRN - 1.8 (6.9%)

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

Seats changing hands since the last projection:
- In QC, the PPC retakes Beauce from CONs.
- In QC, the BQ retakes Longueuil--Charles-LeMoyne from LIBs.
- In QC, LIBs retake Outremont from the NDP.
- In ON, LIBs get Davenport, Toronto--Danforth, Parkdale--High Park and Nickel Belt back from the NDP.
- In ON, LIBs get Burlington, Vaughan--Woodbridge and Cambridge back from CONs.
- In BC, CONs retake South Okanagan--West Kootenay from the NDP.