Inefficiency and the Opposition: Both the Conservatives and NDP Made Gains in All the Wrong Places in 2019

The 2019 election produced an interesting result.  Despite winning a plurality of the popular vote, the Conservatives ended up with just 121 seats compared to the Liberals’ 157.  The NDP also had a disappointing result, finishing a couple of points below the 18 percent that polls had projected for them on the eve of the election.  With just 24 seats, the party won fewer seats than it had in any election since 2004.  Both the Conservatives and NDP had trouble winning seats because they did particularly poorly in competitive ridings.  While the Liberals lost vote share in most ridings, they were able to avoid suffering particularly large losses in swing ridings.

In this post I look at the different changes in vote share across ridings in which parties were either safe, competitive or well behind.  To determine competitiveness I estimated parties’ vote shares in ridings using on a universal swing model based on regional polling averages from the CBC poll tracker from the day before the election.  I use the polls and not the actual results in order to get a sense of what parties could have expected based on the knowledge that would have been available to them prior to the election.  I count any riding in which a party is within 10 points of either winning or losing as competitive.  Any riding in which the party is up by 10 is counted as a safe riding and any riding in which the party was predicted to be down by 10 was considered to be a lost cause riding.  While my model was not perfect in predicting which ridings were safe for parties, it did pretty well.  The Conservatives won all but one of their 84 predicted safe ridings (99%), the Liberals won every one of their 89 predicted safe ridings, the NDP won 18 of their 21 safe ridings (86%), and the Bloc Quebecois won 16 of their 18 predicted safe ridings (89%).

The graph below shows the change in vote share since 2019 for different types of ridings for each party, with the dots showing the average and the lines showing the ranges for a 95% confidence level.  It is striking that most of the Conservatives’ gains came in ridings that were safe for them, while the party made only marginal gains in competitive seats.  The party averaged vote shares almost 7 percentage points above their 2015 result in safe seats but only less than a point above their 2015 result in competitive seats.  This meant that the Conservatives could only win competitive seats where Liberal or NDP losses were sufficiently large to make up the difference between the Conservatives and their strongest competitor in 2015.  Almost all of the Conservative growth since 2015 came in ridings that were safe, and therefore had only a limited impact on the number of seats that the party won.  Because both the Liberals and NDP averaged losses in ridings that were competitive for them, the Conservatives were able to pick up competitive seats without substantially growing their vote share.  They would have been able to make more gains, however, if they had managed to increase their vote share in competitive ridings.

Change Since 2015 By Seat Competitiveness

The distribution of gains and losses is even worse for the NDP.  The party averaged losses in safe, competitive, and lost cause ridings, but its worst losses came in competitive ridings.  In such ridings, the NDP average vote share was over 7 points below the party’s 2015 results.  Compared to this, the party managed to do relatively well in safe seats, losing less than a couple of percentage points.  The NDP’s failure to do well in competitive ridings was particularly notable in Ottawa and Toronto.  The party lost Ottawa Centre, Parkdale High Park, and Toronto Danforth by over 14 percentage points even though these were all ridings that the party had done well in prior to 2015 and which the party was likely targeting.

The Liberals, meanwhile, managed to do slightly better in competitive ridings than in safe or lost cause ridings.  The party managed to keep its losses in competitive ridings to just over 5 percentage points compared to almost 7 percentage points in safe ridings and almost 8 in lost cause ridings.  As a result, the Liberal vote in 2019 was more efficient with respect to turning votes into seats than the vote for any other party.  Had the Liberals not been able to limit their losses in competitive ridings, they may not have been able to hold on to a minority government despite not winning a plurality of the popular vote.

It is finally worth noting that the Bloc Quebecois managed large gains across Quebec, but also that those gains were strongest in safe and competitive ridings for the party. It is not surprising that the Bloc Quebecois made significantly smaller gains in lost cause ridings.  These ridings are disproportionately Anglophone and so would be less likely to be affected by the surge in Bloc Quebecois support that occurred over the course of the campaign.

Both the Conservatives and the NDP can take lessons from the way that their vote was distributed in this past election.  For the Conservatives the lesson must be that they need to do more to reach out to voters beyond their base.  This is particularly true in the GTA, where the Conservatives failed to get close to winning many of the ridings that Harper picked up when he won a majority in 2011.  Playing up concerns over Western alienation is only likely to make their vote more inefficient and increase the likelihood that the Liberals will be able to win a plurality of seats without a plurality of the vote.  The worst thing that the Conservatives could do, at least in so far as their future electoral success is concerned, is to spend the coming minority parliament focused on issues that are most important in the Interior BC, Alberta, and Saskatchewan ridings that the party already wins by large margins.  It may be more fruitful for the party to focus on issues relating to affordability, particularly with respect to housing, then to attack the Liberals over their handling of the Trans Mountain pipeline or their decision to implement a carbon tax.

The NDP, meanwhile, needs to think carefully about what it needs to do to win the urban ridings that it is competition with the Liberals for.  It is notable that the Liberals won 17 of the 18 ridings that my model was showing as Liberal vs. NDP races on the eve of the election, and that the other one (Kenora) was won by the Conservatives.  In addition to the aforementioned Ottawa and Toronto ridings, the party failed to compete with the Liberals in Halifax, lost the Burnaby North Seymour riding they had been hoping to pick up, and failed to make inroads in either the Brampton area ridings that Singh had represented provincially or the Newton and Surrey ridings that the party won in 2011.  The NDP’s struggles may be a result of strategic voting, or they may reflect the Liberal party’s ability to appeal to urban social progressives.  The NDP will have a difficult task ahead of them in this minority parliament.  They will have to exert enough influence over the government that urban voters start to see the party as one that can influence policy (and thus feel less of a need to strategic vote).  At the same time, they will have to find ways to oppose the Liberals on issues popular with the urban social progressive voters that live in the swing ridings that they lost.

The Liberals’ ability to hold on to government after the 2019 election came about in part because of both major federal opposition parties’ inability to make gains in swing ridings.  As the Conservatives and NDP move forward, they will have to think carefully about how they can move beyond their regional bases of support to appeal to the voters that live in the ridings that regularly determine the outcomes of federal elections.

Standard

Missing the Centre: The Conservatives Fail to Make Inroads in Canada’s Largest Provinces

The 2019 federal election will leave Canadians with much to talk about.  It has led to a Liberal minority government, but one with enough seats that it will be able to work with either the Bloc Quebecois or the NDP.  This should make for an interesting couple of years of parliamentary politics.  The NDP did better than one might have expected a year ago, but worse than one would have expected based on the last week of the campaign.  For the first time in a couple of elections the Bloc Quebecois is back and will be an interesting voice for Quebecois nationalists that the other parties will have to contend with.  Finally, the election produced a “wrong-winner” with the Conservative winning the plurality of the popular vote and the Liberals winning a plurality of the seats.  Importantly much of the swing against the Liberals came outside of Ontario and Quebec.  The Conservatives did strikingly poorly in the suburban Toronto and Greater Toronto Area (GTA) ridings that were central to Stephen Harper’s majority government.  The party will have to find ways to win these voters if it wants to compete in future elections.

Despite dropping around 6 points in the national vote, the Liberals managed to hold on to most of their vote share in Ontario and Quebec.  In Ontario the Liberals lost just 3 points of the popular vote and ended up with just one fewer seat than they had in 2015.  Quebec was friendlier to the Liberals with regards to popular vote, the Liberals lost just over a percentage point, but less friendlier with regards to seats, the Liberals ended up with 6 fewer seats in the province than they had in 2015.  The greater decline in seat share in Quebec was due to the consolidation of the non-Liberal vote in the province.  Where the Liberals’ strongest competitor in Quebec in 2015, the NDP, won just 25% of the vote, but their strongest competitor in 2019, the Bloc Quebecois, won almost 33% (less than a couple of percentage points fewer than the Liberals).

The Conservatives meanwhile saw their vote stagnate in Quebec.  The party’s 16% of the vote leaves it more or less where it was in 2015, though the greater Bloc Quebecois vote share means that 16% was worth 2 fewer seats in 2019 than it was in 2015.  Quebec may have been the first missed opportunity for the party.  It is too much to expect the Conservatives to win the most seats in Quebec, but the party can do better than 16%.  It managed 24% in 2006 and almost 22% in 2008.  Given the splits between the non-Conservative parties in Quebec were greater this election than in 2006 or 2008, a vote share of over 20% would have likely yielded more than 10 seats for the party.  This is not dominance in the province, but had it been coupled with gain in Ontario, such growth could have given the party enough support to win a plurality of seats nationally.

There is also reason to believe that the Conservatives can run a campaign in Quebec that gets them into the 20% range.  The party’s support for provincial autonomy should play well in the province, and in the most recent provincial election, the Coalition Avenir Quebec (CAQ) demonstrated that a centre-right platform can win votes in Quebec (though the Conservatives cannot be expected to do as well as the CAQ).  A successful Quebec Conservative campaign, however, requires that the party moderate its position on pipelines and ensure that its social conservative wing is kept quiet.  Neither issues play well for the party in Quebec, and while it was not necessarily the fault of the leadership that the party had deal with attacks for its social conservatism, the party failed to keep either issue from being prominent in the election.  Attacks on the Liberals over their handling of the SNC Lavalin scandal also did not help the Conservatives as both the Bloc Quebecois and Liberals decided to frame those attacks as threats to jobs in Quebec.  Too many of the issues that were important to this election were issues on which Quebecers and the Conservatives tend to disagree.  That prevented the party from using its support of provincial autonomy to make gains with centre-right Quebecers.

It is remarkable that the Conservatives managed a worse result in Ontario in this election than in 2015.  In 2015, the Conservatives had the fact that they had been in government for almost 10 years, Harper’s personal unpopularity, and a slight downturn in the economy all working against them.  Andrew Scheer did not have to deal with the desire to replace the Conservatives in government and should have not have been as unpopular as Harper was in 2015 (he did not have to deal with the scandals that inevitably develop when a party spends a long time in government).  Yet, the Conservatives won 35% of the vote in the province in 2015 and only 33% in this election.  To make matters worse, the party did particularly poorly in the swing ridings in the GTA that were essential to Harper’s 2011 majority.  In 2011, the Conservatives won every seat in the Don Valley, 2/3 seats in Etobicoke, and every seat in Brampton, Mississauga, and Oakville.  Not only did the Conservatives lose every seat in those regions in this election, they were not even close to winning them.  In every seat in the Don Valley, Etobicoke, Brampton, Mississauga, and Oakville the Liberals won with at least 45% of the vote.  In no seat did the Conservatives manage over 40%.  Of the 20 seats in the areas mentioned above, the Liberals won over 50% of the vote in 12.  The Conservatives do not need to win all of these to win elections, but they must win some, and they cannot do as poorly as they did in this election.

It is harder to say how the Conservatives might have approached these ridings differently in order to produce a different result.  A message focused on affordability has worked for them in the area in the past, though that message may have gotten drowned out by other issues in this election.  The decision to run such a negative campaign may have hurt.  These are all ridings that made the switch from the Harper Conservatives to the Trudeau Liberals in 2015.  It is unlikely that Trudeau is disliked as much in these ridings as he is in much of the West.  The Conservatives also benefited in 2011 from the work that Jason Kenney did to reach out to immigrants and minority communities throughout the 2000s.  While the Conservatives have likely not abandoned attempts to win voters from such communities, it is clear from this election and from 2015 that their current efforts are not working.  The party may have to rethink the way it appeals to an ethnically diverse electorate.  Given the ethnic and cultural diversity in swing ridings in Canada, particularly in Toronto and the GTA, the Conservatives will not win elections if immigrants and visible minorities consistently disproportionately vote for the Liberals.

Because the Conservatives win so many votes on the prairies, there is always a danger than their vote will be inefficient.  The party has most seats in the region, so an increase in their prairie vote share does little to help them win seats.  The party’s campaign may have only worsened this problem.  While the Conservatives made significant gains in every Western province, their vote stagnated in Ontario and Quebec, two of the provinces that they needed to make gains in if they were going to compete with the Liberals to form a government.  The party needs to carefully rethink its strategy if they want to make gains in future elections.  Strategies that play up regional tensions are only likely to hurt the party in the Ontario and Quebec ridings that it needs to win if it wants to form government.

Standard

Final Competitiveness Estimates: The GTA and Toronto

This coming Monday the 2019 election campaign will come to close.  In anticipation of the election I have spent this past week taking a last look at which parties are competitive in which ridings.  In previous posts I have discussed most regions in the country.  In my last post I will look at the Greater Toronto Area (GTA) and Toronto itself.  In the past two elections this has been a key region.  Gains in the GTA and Toronto were essential to the Harper Conservatives’ majority in 2011 and to the resurgence of the Liberals in 2015.  Despite Liberal losses in Ontario, the Conservatives may struggle to make many gains in the Toronto area in 2019.  The party is competitive in a number of seats but has not made inroads beyond the GTA and into the suburban Toronto ridings that it won in Etobicoke and Scarborough in 2011.

To estimate competitiveness, I calculate a universal swing based on the difference between parties’ vote shares in the previous election and regional polling averages from the CBC poll tracker.  These estimates come from averages from October 18.  I have been updating my model throughout the week to reflect changes in the polls so each of my posts has used a slightly different polling average.  I count a party as competitive in a riding if it is either within 10 points of losing or within 10 points of winning.  I count a seat as safe for a party if it is up by at least 10 points.  I do not account for the effect that retiring incumbents will have on parties’ chances in some ridings, though I will discuss ridings where I think incumbency is throwing off my estimates throughout the piece.

For context, I have included the graph that I had in my last piece looking at the changes in support for the different parties between now and the 2015 election.  Because I do not have separate polling estimates for Toronto and the rest of Ontario, I use the same differences between party’s support in 2015 and party’s support today to estimate competitiveness in both the ridings I looked at in my last post and the ridings I am looking at in this post.  I have updated the graph to reflect changes in the polls that occurred between October 17 and October 18, but these changes were minimal.  The Liberals are still significantly below their 2015 vote share.  The Conservatives have also lost ground since 2015, though they have lost less than the Liberals.  The NDP has gone up, but not by as much as one would expect given the Conservatives’ and Liberals’ declines.  This suggests that a lot of these Toronto area ridings will be decided by which of the two largest parties loses less support.

Change from 2015 (Ontario) Oct 28

The distribution of competitive seats in the GTA and Toronto is relatively straight forward.  My model suggests that the Liberals have safe leads in 28 of 50 seats, slightly more than half.  The Conservatives have only one safe seat but are competing with the Liberals in 14 and with the NDP in 1.  There are also 5 ridings that my model is showing as Liberal vs. NDP races.  As a general rule, as one moves from the outer GTA towards the centre of Toronto the Conservatives get weaker and the Liberals and then the NDP get stronger.

Competitive Seats (GTA and Toronto)

The strongest region in the Toronto area for the Conservatives is the outer part of the GTA.  The Liberals have just two safe seats, Ajax and Markham Thornhill, in the region.  This region also has the only Conservative safe seat in the Toronto area, Peter Kent’s Thornhill riding, which the party picked up in 2008 and held with over 50% of the vote in 2015.  Most of the ridings in the outer portion of the GTA are Conservative vs. Liberal races.  Three of the 6 Conservative vs. Liberal races are ridings that the Conservatives would have won with more than 50% of the vote in 2011 had that election used the 2015 riding boundaries.  These ridings include Aurora Oak Ridges Richmond Hill, King Vaughan, and Vaughan Woodbridge.  My model, however, shows all these races as very close.  The Conservatives are going to be in tough fights in this election to try to win back ridings that they won easily in 2011.  Aurora Oak Ridges Richmond Hill will be particularly interesting because the Conservatives are running Leona Alleslev in it.  Alleslev won the riding in 2015 as a Liberal but crossed the floor to join the Conservatives.  My model has the Conservatives and Liberals within one point of each other in this riding.  If Alleslev is able to keep even a small number of the voters that supported her when she was a Liberal, she may be able to tip a very close riding in the Conservatives’ favour.

Outer GTA Table

Two ridings in the outer GTA are competitive for candidates who are not Conservatives or Liberals.  The most interesting of these ridings is Markham Stouffville where former Liberal Jane Philpott is running as an independent.  Philpott resigned from cabinet in support of Jody Wilson-Raybould during the SNC Lavalin scandal and was later expelled from the party along with Wilson-Raybould.  Prior to her resignation she was seen as a very effective cabinet minister.  The affair and her support of Wilson-Raybould will have raised her national profile and give her a chance of holding on to the riding.  My model has no way of accounting for Philpott running as an independent and has it as a close Conservative vs. Liberal race, with the Liberals up by 3 points.  Even if Philpott does not win the riding, it may be hard for the Liberals to win it with Philpott running because of the votes that she will draw from the Liberals.  My model also has a very close race in Oshawa between the Conservatives and NDP.  The NDP has not won the riding since 1993 and lost by just under 7 points in 2015.  The combination of the decline of the Conservatives and the growth in support for the NDP relative to 2015, however, should make the riding a much closer race in this election.

The Brampton, Mississauga, Oakville area is much friendlier to the Liberals.  My model has the Liberals with a safe lead in every riding in Brampton and half of the ridings in Mississauga.  It is notable that, had the 2011 election used the 2015 boundaries, the Liberals would had lost every single riding that my model is showing as safe for them in this election.  The NDP would have won Brampton East and the Conservatives would have won the rest.  Prior to 2011, though, most of these ridings have elected Liberals most of the time.  The fact that the Conservatives are not more competitive in Brampton and Mississauga highlights the extent to which the party has failed to make the inroads it did in 2011.  It is also notable that most of these ridings are highly culturally diverse.  The Conservatives’ success in the region in 2011 and failures in 2015 and this election likely reflects significant changes in their support with ethnic minority voters.  The Conservatives do have a number of ridings in Mississauga and Oakville where they are competitive and where they did well in 2011.  The Conservatives will need to win these ridings if they want to be successful.

Brampton etc. Table

Historically the NDP has struggled in Brampton but they have been campaigning hard in the city in the final weeks of the election.  Singh held a major rally in the riding this past Thursday.  When Singh was a member of the Ontario legislature he represented Bramlea Gore Malton, which included what are now parts of the Brampton East, Brampton Centre, Brampton North, and Mississauga Malton ridings.  Singh’s history as a Member of Provincial Parliament may be able to give the NDP a boost in these ridings, though they would have to outperform my model by a significant margin to be competitive.  The NDP did well in what is now Brampton East in 2011 and there is no incumbent in the riding.  This makes it the riding in which the NDP are most likely to make a breakthrough, though it is still a long shot for the party.

Suburban Toronto is the best region in the Toronto area (and in all of Ontario) for the Liberals.  My model estimates that every riding in this region is a Liberal safe seat.  Many of these ridings have been longstanding Liberal strongholds.  The Conservatives managed to win a few, Don Valley East, Etobicoke Centre, and Etobicoke Lakeshore in 2011 (and would have won Don Valley North had it existed).  The Conservatives need to be very popular in Ontario for these ridings to be competitive for the party.  With the Conservatives at just over 30% in Ontario it is unlikely they can compete in any of these ridings this election.  It is worth noting that Renata Ford, the wife of late Toronto Mayor Rob Ford, is the People’s Party’s candidate in Etobicoke North.  To the extent that the Fords are popular in Etobicoke this may give the People’s Party a long shot chance at this riding.  The Liberal incumbent, however, is cabinet minister Kirsty Duncan, who managed to win the riding by 10 points in 2011, when the Liberals were at their weakest in Canadian history.  It would be surprising to see the Liberals lose Etobicoke North.

Suburban Toronto Table

Once one moves from suburban Toronto to Central Toronto the Liberals start to face more competition from the NDP.  My model has the Liberals safe in 5 ridings in this part of the city and in competitive races with the NDP in another 5.  The three best ridings for the NDP in Central Toronto are Davenport, Parkdale High Park, and Toronto Danforth.  All of those ridings have been represented by prominent NDP MPs in the past.  Andrew Cash won Davenport in 2011.  The fact that he is running again in this election after losing in 2015 should give the NDP a good shot at winning the riding.  Parkdale High Park was won by Peggy Nash in 2006 and 2011 (but not 2008), though she is not running again.  Liberal incumbent Arif Virani does not have a strong national profile, so NDP candidate Paul Taylor should have a decent shot at winning the riding.  Toronto Danforth was Jack Layton’s riding.  It was a riding that many were surprised to see the NDP lose in 2015.  If the NDP are able to grow their support in Central Toronto by even a couple of points they should have a good shot at winning back the riding.  In documentary filmmaker Min Sook Lee the NDP they have a candidate who comes into the election with some name recognition.  Still, Liberal candidate Julie Dabrusin is the incumbent and so should not be counted out.

Central Toronto Table

The Conservatives are competitive in one riding in Central Toronto, Eglinton Lawrence.  Joe Oliver won this riding for the Conservatives in 2011, but lost to Liberal Marco Mendicino in 2015.  This has historically been a Liberal riding, but Conservative attempts to win the support of the riding’s large Jewish community started to make it competitive in 2008.  It is unclear if the Conservatives can have the success they had in 2011 and be as competitive as they were in 2008 without Oliver running for them (Oliver was the losing Conservative candidate in 2008).  This election will be an interesting test of whether the Conservatives have turned Eglinton Lawrence into a competitive riding.  It is certainly the Central Toronto riding that the Conservatives have the best shot of winning.

To form majority governments the Conservatives need to win a substantial number of ridings in the Toronto area, particularly in the GTA.  My estimates suggest that the Conservatives have little to be optimistic about in the city.  The party is competitive in a number of GTA ridings, but is well back of the Liberals in many of the ridings that were important gains for the party in 2011.  As Harper demonstrated in 2006 and 2008, it is possible to for the Conservatives to form a minority government without winning many of the Toronto area ridings they picked up in 2011, but a majority would be very difficult without them.  The Liberals, by contrast, have a strong base of support in the city.  The decline in their support means that they are likely to lose a few seats, but they should be able to hold on to a large majority of the seats in Toronto and a good number in the GTA.  The NDP, who were shut out of Toronto in 2015, have a good chance of picking up at least couple of seats, though they have no safe seats so they will have to compete hard for every riding that they win.

Standard

Final Competitiveness Estimates: Ontario Outside of Toronto

With the election campaign finishing on Monday I am taking a last look at which parties are competitive in which ridings.  Throughout the week I have been going region by region, breaking down the different kinds of races in the West, Quebec, and Atlantic Canada.  In my last two posts I will look at Ontario.  Because Ontario has 121 ridings, I have broken seats in the province into two groups to make writing about them more manageable.  In this post I will look at ridings outside of the Greater Toronto Area and Toronto itself while my next post will focus on the GTA and Toronto.  With only three parties competitive in Ontario, party competition is much less complicated than in British Columbia or Quebec.  Most of the ridings that I will look at in this post are either safe seats or Conservative vs. Liberal races.  There are a handful of Conservative vs. NDP races or Liberal vs. NDP races and one three-way race.

To estimate competitiveness, I calculate a universal swing based on the difference between parties’ vote shares in the previous election and regional polling averages from the CBC poll tracker.  These estimates come from averages from October 17.  I have been updating my model throughout the week to reflect changes in the polls so each of my posts has used a slightly different polling average.  I count a party as competitive in a riding if it is either within 10 points of losing or within 10 points of winning.  I count a seat as safe for a party if it is up by at least 10 points.  I do not account for the effect that retiring incumbents will have on parties’ chances in some ridings, though I will discuss ridings where I think incumbency is throwing off my estimates throughout the piece.

The graph below shows the change in each party’s share of the vote in Ontario compared to 2015.  Because I do not have separate polling estimates for Toronto and the rest of the province, this graph includes the GTA and Toronto ridings.  It is noteworthy that both the Conservatives and the Liberals have lost ground in the province.  The Liberals have lost a little under 7 points and the Conservatives have lost just over 4.  The party that appears to have benefited most from these losses are the Greens, who are 6 points ahead of where they were in 2015.  The weakness of the Greens in most ridings in 2015 though, means that a 6-point gain is not enough to make them competitive in the province.  Meanwhile the NDP is a little under 3 points above where they were in 2015.  This is not insignificant, but it is notable that the NDP have not gained more given the amount of support both the Conservatives and Liberals lost.  To the extent the Conservatives and Liberals are competing with each other in a lot of ridings in Ontario, the outcome in many Ontario seats will be determined by which of the Conservatives and Liberals loses less support as opposed to who makes more gains.

Change from 2015 (Ontario)

These vote shifts result in a number of competitive Conservative vs. Liberal races in Ontario outside of Toronto.  Of the 71 seats in the region, there are 23 seats (32%) that are competitive between only those two parties.  There are 40 safe seats in this part of Ontario, 17 of which are safe for the Liberals, 15 of which are safe for the Conservatives, and 8 of which are safe for the NDP.  The NDP is does not have a lot of competitive seats that they can win in Ontario outside of Toronto, though they are competing with the Liberals for 5 seats, with the Conservatives for 2, and with both parties for an additional seat.

Competitive Seats (Ontario Outside Toronto)

Ottawa is probably the Liberals’ best region outside of Toronto in Ontario.  The party is competitive in every seat in the city and is safe in 5 of them.  My model has the Conservatives up by only about 5 points in Carleton, but Pierre Poilievre is the Conservative incumbent in the riding.  He is the party’s finance critic has a strong enough reputation that it would not be surprising to see him outperform the model.  The Liberals may still be competitive in the riding, they were within 4 points of winning it in 2015, but winning this riding is likely a long shot riding for them.  Kanata Carleton was a new riding in 2015 which Liberal Karen McCrimmon won.  Had the 2011 election been run using the 2015 boundaries though, the Conservatives would have won this riding by a large margin, so it will be interesting to see if the Conservatives can win the riding.  Ottawa Centre is a trickier riding.  Environment Minister Catherine McKenna is the incumbent, and her reputation will give her a very good shot of holding the riding.  At the same time, this was an NDP riding from 2004 until 2015.  My model has the NDP up by 5 in the riding, but that is likely because it is underestimating the strength of McKenna and the Liberals.  It is more likely that the Liberals are the favourites, but the NDP should not be counted out.  Regardless, it is very unlikely that the Conservatives will win the riding and so progressive voters should feel comfortable voting for their preferred party or candidate.

Ottawa Table

If Ottawa is the best region outside of Toronto for the Liberals, Eastern Ontario is the best region for the Conservatives.  They have 9 safe seats in the region and are competitive in all but 3.  The Liberals have 3 safe seats in this part of Ontario.  The NDP not only does not have any safe seats in Eastern Ontario, my model does not even have them competitive in any seats.  Progressive strategic voters in this part of the province are probably best served by voting Liberal.  Peterborough Kawartha is an interesting riding in the region in that in that the party that wins it often ends up being the party that forms government.  With cabinet minister Maryam Monsef the Liberal incumbent in the riding, the Liberals stand a reasonably good chance of holding it.  The fact that a cabinet minister in running in the riding, however, may make it less of a bellwether.

East Ontario Table

Probably the most competitive region in Ontario outside of Toronto is Southern Ontario.  All of the three largest parties have a substantial number of safe seats in this region.  My models shows the Conservatives and Liberals each with 5 each and the NDP with 6.  Both the Liberals and the NDP have a seat that my model shows as safe but which may actually be competitive.  After the Greens won Guelph in the most recent Ontario provincial election there has been a lot of speculation that the riding may be won by the Greens federally as well.  It helps the party that most riding boundaries in Ontario are the same for provincial and federal elections, so the federal Greens in Guelph will be going after the exact same voters that their provincial counter-parts did.  At the same time, the federal Liberals are a lot more popular than the Ontario provincial Liberals, so the Greens will be facing much stiffer competition than they did in the provincial election.  Windsor West may also be more competitive than my model suggests.  The Liberals are running former provincial cabinet minister Sandra Pupatello in this riding.  She has a well-established reputation in Ontario and has represented the riding provincially.  At the same time, Windsor West (and the city in general) has been a long time NDP stronghold in federal politics.  NDP candidate Brian Masse has represented the riding since winning a by-election in 2002 and has won 50% of the vote or more in the last 3 elections.  The choice of a high-profile candidate raises the likelihood that the Liberals can make this riding competitive and that they might win the riding.  At the same time, this has historically been one of the ridings in Ontario where the NDP have been strongest and thus will pose a significant challenge for the Liberals.

South Ontario Table

If a riding is competitive in Southern Ontario, there is a good chance that it is a Conservative vs. Liberal race.  14 of the 18 ridings that come up as competitive in my model are such races.  One of these, Milton, features an interesting race between Conservative incumbent Lisa Raitt and Liberal star candidate (and Olympic gold medalist) Adam van Koeverden.  Raitt is one of the most prominent and best performing members of the Conservatives’ shadow cabinet, so winning this riding will be difficult for the Liberals.  Van Koeverden may have the name recognition, though, to pose a significant challenge to Raitt.  Several of the Conservative vs. Liberal races in this region were ridings that the Conservatives won by substantial margins in 2011 (or would have if the 2011 election used the 2015 boundaries) but lost to the Liberals in 2015.  These include Burlington, Cambridge, Hastings Lennox and Addington, Kitchener South Hespeler, London West, and St. Catharines.  It will be interesting to see how many of these ridings the Liberals will be able to hold on to given that Conservative success in them was central to Harper’s ability to win a majority in 2011.  It is hard to see the Conservatives getting back to a majority government without winning back most or all of these ridings. Burlington will be particularly interesting to watch given that the Liberal candidate in the riding is cabinet minister Karina Gould.

There are two interesting ridings for the NDP in Southern Ontario.  One is Hamilton East Stoney Creek which the NDP held from 2006 until 2015 when they lost it to the Liberals.  Winning this riding would be an important step for an NDP trying to rebuild parts of the Layton coalition that allowed the party success in 2011.  With a Liberal incumbent in the riding, the party may be at a disadvantage compared to previous elections though.  Niagara Centre (formerly known as Welland) will also be an interesting riding for the NDP.  My model has this as a three-way race, with the Conservatives and Liberals also competitive in the riding.  The NDP candidate in the riding is Malcolm Allan, who held it for the party from 2008 until the 2015 election.  Allen was a popular MP, and like Jack Harris in St. John’s East, the fact that he has won the riding in the past should give the NDP a good shot at the riding.  Allen beat a Liberal incumbent in 2008 to win the riding, but the fact that the Liberals have an incumbent running for re-election means that they should not be counted out.  While my model has the Conservatives down by 8 points, that is close enough that a vote split between the Liberals and NDP could lead to a Conservative victory.  The popularity of Allen coupled with the fact that the Liberals are the incumbent makes it very hard to tell which of the Liberals or the NDP has the better chance of beating the Conservatives in the riding.  This is a difficult riding for progressive strategic voters.

The weakest region in Ontario outside of Toronto for the Conservatives is Northern Ontario.  They have no safe seats in the region and are only competitive in Parry Sound Muskoka.  With 4 safe seats in the region, the Liberals are probably the best positioned party in Northern Ontario.  3 of the 4 Liberal safe seats are seats that the NDP has won at some point since 2000 (Nipissing Timiskaming is the exception), but it is not clear that the NDP is in a position to challenge the Liberals in these ridings this time around.  Notably, former NDP and Green MP Bruce Hyer is running for the Greens again in Thunder Bay Superior North (Hyer left the NDP after being disciplined for voting in favour of ending the long-gun registry), but he is facing Liberal cabinet minister Patty Hajdu and it would be very surprising if he was competitive in the riding.

North Ontario Table

Nickel Belt and Thunder Bay Rainey River were both ridings that the Liberals took from the NDP in 2015 and which the NDP will be looking to take back.  The NDP lost Nickel Belt by about 5 points in 2015, but would have won over 50% of the vote in the riding in 2011 had 2011 used the 2015 riding boundaries.  It is likely the easier of the two ridings for the NDP win, though both should be interesting Liberal vs. NDP races.  Kenora is a riding that the Liberals took from the Conservatives in 2015, but which the NDP almost won.  Former Ontario provincial leader Howard Hampton came within 2 points of winning Kenora for the NDP in 2015.  Normally I would suggest that the presence of a prominent NDP candidate in 2015 is leading my model to over-estimate the strength of the NDP, but the NDP have again have a candidate that is well-known locally running for them.  Grassy Narrows Chief Rudy Turtle is the NDP candidate in this election and has a good chance of winning, though one should not count out Liberal incumbent Bob Nault.  It is also worth noting that this is the riding where Jagmeet Singh challenged a reporter for asking him about the cost of building a water treatment centre in Grassy Narrows, arguing persuasively that the cost of providing clean drinking water would not be a question if one where talking about a major Canadian city such as Toronto or Vancouver.  The fact that the NDP have highlighted Grassy Narrows as an example of a community that the government needs to invest in if it is going to meet its commitments to reconciliation likely helps to make a persuasive case for the party in Kenora.

Most of the interesting races in Ontario that are outside of Toronto are in the Southern portion of the province.  The ability of the Conservatives or Liberals to form a government will likely hinge on how successful they are that part of the country.  There are, however, also some interesting races to watch in Ottawa and in Northern Ontario, especially for the NDP.  The number of seats in Ontario means that it always important and usually interesting in Canadian election campaigns.  This election campaign will be no different, and the interesting races will not be limited to Toronto and the surrounding area.

Standard

Final Week Competitiveness Estimates: Quebec

With the 2019 election campaign coming to an end I am taking a look at which parties are competitive in which ridings.  In previous posts I have looked at ridings across Western Canada and in the Atlantic.  In this post I am looking at Quebec.  The surge of the Bloc Quebecois in the province has created several interesting races.  It also does significant harm to the Liberals’ chances of success in the election as any Liberal path to a majority requires them to win a large number of seats in Quebec.

To estimate competitiveness, I calculate a universal swing based on the difference between parties’ vote shares in the previous election and regional polling averages from the CBC poll tracker.  The estimates for this post are based on averages from October 16.  Later posts looking at Ontario will use polling averages from later dates as I try to ensure that my estimates keep up with changes in the polls.  I count a party as competitive in a riding if it is either within 10 points of losing or within 10 points of winning.  I count a seat as safe for a party if it is up by at least 10 points.  I do not account for the effect that retiring incumbents will have on parties’ chances in some ridings, though I will discuss ridings where I think that retirements are likely throwing off my estimates.

My model includes a very rough estimate for the People’s Party’s support.  I calculate what I call a “Bernier score” by multiplying the share of Maxime Bernier’s vote in the Conservative leadership race by the Conservative party’s share of the vote in 2015.  I then calculate the difference between each ridings’ score and the average Bernier score.  To finally get a predicted vote share for the People’s Party in each riding I added the percentage that they were polling at in the region to that difference.  So if a riding has a Bernier score that was the exact average for the country, and the People’s Party is polling at 2.5% in the region that the riding is in, I would estimate a vote share for the People’s Party of 2.5%.  I have done this for my analysis of other regions, but none of the People’s Party’s estimates in those regions are high enough to affect my conclusions.  I do have a riding in Quebec where the People’s Party estimate matters.  Given that this is a very rough estimate of the People’s Party’s vote share, one should be careful reading too much into it.

The most striking vote shifts when one compares the CBC poll tracker’s regional averages to the 2015 election result are the growth of the Bloc Quebecois and the decline of the NDP.  The NDP decline is expected given the party’s struggles in Quebec since 2015.  The growth in the Bloc Quebecois is more recent, but may reflect a failure of the four other federalist parties to do a good of job of speaking to issues that are important in the province.  Notably, the Liberals are also down in Quebec, having lost over 4 points.  This makes the party’s success in the province precarious.  Even though they won a large number of seats in the province in 2015, they only won 36% of the vote.  I expect that the 31% that they are at now is lower than the vote share they need in order to win enough seats in the province to get them to a majority.  Notably it is not clear that the Conservatives have lost any ground in Quebec.  Part of this because their vote share in 2015 was already reasonably low, at just under 17%.  Scheer’s poor performance seems to have cost the Conservatives the opportunity to expand their vote share in the province, but it does not seem to have led to a decline for the party.

Change from 2015 (Quebec)

While the Liberals and the Bloc Quebecois are within the margin of error of each other with respect to the vote in Quebec, the Liberals have slightly more safe seats with 21 compared to 19.  The Conservatives are unlikely to be shut out of the province, and have 6 safe seats, while the NDP have none.  There are a wide range of different permutations of party competition in the province.  The plurality of competitive seats (15) are races between the Bloc Quebecois and the Liberals.  Unlike the rest of the country, there is only one Conservative vs. Liberal race in the province.  There are also a few three and even four-way races.  It is worth noting that the rise of the Bloc Quebecois means that my model has a number of Bloc Quebecois safe seats in ridings that the Bloc Quebecois lost by considerable margins in the last election.  A 10 point jump party coupled with a 13 point drop for the NDP will produce such outcomes.  Some incumbents might be able to use their reputation to keep the Bloc Quebecois from picking up seats that I see as safe gains for the party, but I expect that there will be a number of new seats that the Bloc Quebecois will win by large margins.

Comeptitive Seats (Quebec)

North Central Quebec is one of the stronger regions in the province for the Bloc Quebecois.  My model has them with a safe lead in 8 ridings in the region.  The Conservatives have safe leads in two ridings, both of which they won in 2015 and both of which are ridings in which their incumbent MPs are running for re-election.  The only safe seats for the Liberals are the three in the Gatineau area.  It is unsurprising that the Liberals would have safe seats in this area given its proximity to Ottawa and the larger share of federal government employees that live in those three ridings compared to others.  Notably, my model has the Bloc Quebecois competitive in every non-safe riding in the region.  They are in two-way races with the Liberals for 6 ridings, the Conservatives for 2, and the NDP for one.  They are also in three-way races with the Liberals and NDP in two ridings and in a four way race in Trois Rivières.

North Central Quebec Table

The race between the Bloc Quebecois and NDP in Berthier Maskinongé is particularly notable as that is Ruth Ellen Brosseau’s riding.  Brosseau made headlines by winning for the NDP in 2011 despite not campaigning or living in the riding but has since built up a strong reputation for constituency service and House of Commons performance.  This has made her one of the NDP MPs that many speculate might be able to hold on to her seat in the face of the sharp decline in support for the party.  This election will be an interesting test of how much Brosseau’s personal popularity can protect her against her party’s decline in support.  The two Abitibi ridings are also ridings where the NDP did well in 2011 and 2015, but both NDP incumbents that won the ridings in the past have retired.  The Bloc Quebecois and Liberals may be able to take advantage of this, though they will be in close competition with each other and my model suggests that the NDP should not be completely counted out in those ridings either.  Chicoutimi Le Fjord is a final riding in the region worth touching on.  This is a riding that former QMJHL hockey coach Richard Martel picked up for the Conservatives in a by-election.  The Conservatives did not do well in the riding in 2015 so my model does not have them as competitive in it, but the by-election and Martel’s name recognition could change that.  This riding may actually be a three-way race between the Bloc Quebecois, Conservatives, and Liberals.

Quebec City is probably the best region for the Conservatives in Quebec.  They have two safe seats in the city and my model suggests they are competitive in all 5 of its ridings.  It has them in three way races with the Bloc Quebecois and Liberals in Louis Hébert and Québec (the riding that shares its name with both the city and province).  Louis Hébert has elected a candidate from all four of the largest parties since 2000, with the Liberals winning in 2015.  Québec elected a Bloc Quebecois MP in every election between 1993 and 2008 but switched to the NDP in 2011 and then to the Liberals in 2015.  The Liberal incumbent, Jean-Yves Duclos, is a cabinet minister, which may help the party in the riding.  The final riding of note in the city is Beauport Limoilou.  My model has this as a four-way race between the Bloc Quebecois, Conservatives, Liberals, and People’s Party.  This being said, my model has both the Liberals and People’s Party down by 9, so both are probably long shots to win.  This may, in reality, be a Bloc Quebecois vs. Conservative race.

Quebec City Table

The Bloc Quebecois also stands to make significant gains in ridings that are South of the St. Lawrence river.  Here my model has the Bloc Quebecois safe in 7 ridings including Beloeil Chambly where leader Yves-François Blanchet is running.  By comparison the Liberals have 3 safe seats and the Conservatives have 2.  To add to this, the Bloc Quebecois are in two-way races with the Liberals for a further 8 ridings in the region.  Three ridings stand out as particularly interesting.  In Beauce, the Conservatives are trying to keep People’s Party leader Maxime Bernier from holding on to his seat.  My model has this as a Conservative safe seat, but only because it is using the 2015 election when Maxime Bernier ran as a Conservative as a point of comparison for estimating the Conservatives’ vote share.  Given that Bernier likely took many of his voters in the riding with him, I felt it was appropriate to count the riding as competitive.  I am not, however, confident that Bernier took enough voters with him to make the riding a safe seat for the People’s Party.  The other interesting ridings are Rimouski Neigette Témiscouata Les Basques and Sherbrooke.  Rimouski is where Guy Caron is running for the NDP.  Like Brosseau, Caron is a candidate who may be popular enough on his own to survive a large swing against his party.  Still, it is hard to see the riding as safe given the magnitude of the swing against the party, and my model has both the Bloc Quebecois and Liberals competitive in it.  Sherbrooke is represented by Pierre-Lu Dusseault who won the riding in 2011 as a 19 year old.  My model has Dusseault behind both the Bloc Quebecois and the Liberals, but not by so much that it is inconceivable that he could hold on to the riding.

South St. Lawrence Table

Montreal is the least competitive of all of the regions in Quebec.  The vast majority of the seats are safe Liberal seats, reflecting the party’s high level of support with Anglophones on the West side of the city.  My model has 3 Bloc Quebecois safe seats in Montreal, though one of them is Laurier Sainte Marie.  My model may be over-estimating the Bloc Quebecois’ support in this riding because former leader Gilles Duceppe was running in 2015 and so it is using a vote share from 2015 that may be larger than it otherwise would have been.  The Liberals and NDP also have notable candidates in the riding.  The Liberals are running environmental activist Steven Guilbeault while the NDP are running activist Nimâ Machouf who is the wife of former Quebec Solidaire spokesperson (Quebec Solidaire calls their leaders spokespeople) Amir Khadir.  The reputations of both the Liberal and NDP candidates are something that my model cannot account for (though the NDP has a retiring incumbent that may also lead the model to over-estimate the party’s strength).  This riding could easily be a Bloc Quebecois vs. Liberal competitive race or even a three-way race with the NDP competitive as well.

Montreal Table

Three other ridings in Montreal are worth making note of.  My model is showing Outremont as a Liberal vs. NDP race but that is probably because Thomas Mulcair pushed up the NDP’s 2015 vote share there.  After Mulcair retired the Liberals won a by-election in the riding by a large margin, and it is likely that without Mulcair this is a safe Liberal seat.  Mount Royal is a seat that the Conservatives have been trying to win for a few elections.  Once a safe Liberal seat, repeated efforts by the Conservatives to court the riding’s Anglophone and Jewish community have made the riding more competitive.  That being said, the Conservatives’ best shot to win the riding was likely in 2015 when Irwin Cotler retired and the Liberals were without an incumbent.  With Anthony Housefather now running as a Liberal incumbent, this riding will be a much harder for the Conservatives to win.  Finally, Rosemont la Petite Patrie will be a riding that the NDP will be trying hard to hold on to.  This is the third riding in the province where the NDP have an incumbent, Alexandre Boulerice, who could conceivably hold on to the riding on the strength of his own popularity.  Indeed, Boulerice was named deputy leader of the party by Singh and is probably the most important person to the party’s Quebec campaign.  My model suggests that Boulerice’s main competition will come from the Bloc Quebecois.  Indeed, the Liberals have not won the riding since 1984.  If there is one riding that the NDP are going to hold on to in Quebec, Rosemont la Petite Patrie is probably it.

The significant shifts towards the NDP in 2011 and then the Liberals in 2015 have made Quebec interesting in successive elections.  A resurgent Bloc Quebecois promises to make Quebec interesting again in this election. The significant number of Liberal vs. Bloc Quebecois races in the province make success in this region central to both parties’ chances of overall success in the election (indeed it is the only region that matters to the Bloc Quebecois).  The Conservatives, NDP, and even the People’s Party have ridings to watch in Quebec as well.  In the past block voting in Quebec used to deliver a large number of safe seats to whichever party was favoured in the province.  A more divided Quebec electorate means that almost every party has a few seats that it can win in this election.

Standard

Final Week Competitiveness Estimates: British Columbia

With the federal election taking place on Monday I am revisiting my analysis of which parties are competitive in which ridings and updating it to reflect recent shifts in the polls.  Over the course of the week I have been looking at different regions, and in this post, I am looking at British Columbia.  BC has an interesting mix of competition.  Four parties have at least one safe seat and there are 6 different permutations of competition in the seats that are competitive.  Given that BC is the last region to report results, this should make for an interesting end to election night.

To estimate competitiveness, I calculate a universal swing based on the difference between parties’ vote shares in the previous election and regional polling averages from the CBC poll tracker.  These estimates come from averages from October 15.  Later posts looking at other regions will use polling averages from later dates as I try to ensure that my estimates keep up with changes in the polls.  I count a party as competitive in a riding if it is either within 10 points of losing or within 10 points of winning.  I count a seat as safe for a party if it is up by at least 10 points.  I do not account for the effect of incumbency on parties’ chances in some ridings, though I will discuss this if I think a retiring incumbent is throwing off my estimates.  I have grouped the riding in the Yukon in with British Columbia for lack of a better region to put it in.  Because the population of the Yukon is so small, it is impossible to do a separate analysis just for that riding (or even for Northern Canada as a whole.  Because it is likely that the Yukon is different from other ridings in BC, analysis of it should be taken with an extra bit of skepticism.

Like the Prairies and Atlantic Canada, the Liberals have lost a lot of ground in BC since 2015.  The CBC poll tracker regional average has them 8 points behind where they were in 2015.   Like in Atlantic Canada, this may simply be a return to normal for the Liberals.  The 35% they won in BC was significantly higher than in previous elections.  Prior to 2015, the last time the Liberals broke 30% in BC was in 1974 under Justin Trudeau’s father.  The 27% they are polling at now is consistent with their vote shares in the province in the early 2000s.  It is notable that there is no indication that Conservatives or the NDP have made significant gains since 2015.  Indeed, the NDP is almost 3 points below their 2015 vote share.  Instead, the Greens have made large gains and are 7 points ahead of where they were in 2015.  This sets up several interesting races.

Change from 2015 (BC)

The Conservatives have the largest number of safe seats in the province with 9.  12 other seats are either Liberal, NDP, or Green safe seats.  Almost as many seats are Conservative vs. Liberal races as are Conservative safe seats, though one of those Conservative vs. Liberal seats is Vancouver Granville and is really a Conservative vs. Liberal vs. Independent seat.  There are 5 races where the Conservatives are competing with the NDP and a couple of three-way races (though the four-way race that I had in Esquimalt Saanich Sooke in an earlier post has turned into a Green vs. NDP race).  This wide range of permutations of party competition makes it particularly important that progressive strategic voters know who is competitive in their riding.

Competitive Seats in BC

Despite its reputation as a Conservative stronghold, the Conservatives only have three safe seats in Interior BC.  The Liberals have one in the Yukon and the NDP have one in Skeena Bulkley Valley (though they have a popular incumbent, Nathan Cullen, retiring in that riding so it may be a bit more competitive than my model suggests).  Surprisingly the Liberals won Kelowna Lake Country in the last election even though the riding had only ever elected a Conservative, Canadian Alliance, or Reform MP before 2015.  The Conservatives should still be competitive in the riding and the Liberals have not completely collapsed in the province, so my model has this as a Conservative vs. Liberal race.  The Conservatives and NDP are competing over three ridings in the Interior.  Kootenay Columbia and South Okanagan West Kootenay were both seats that voted NDP in 2015.  Both have NDP incumbents running for re-election.  The combination of the Conservatives’ strength in the Interior, NDP incumbency, and a history of NDP success in the British Columbia Southern Interior riding that made up parts of both ridings prior to 2015 should make both ridings interesting races in this election.  Kamloops Thompson Cariboo is a long shot for the NDP.  The Conservatives have won that riding in every election since 2004.  The Liberals are also running former provincial MLA, Terry Lake, which may pull some votes from the NDP.

Bc Interior Table

The South Eastern portion of the Lower Mainland is likely the place in BC where the Conservatives stand to win the most seats.  They have 5 safe seats in this part of the province and are uncompetitive in only 2.  Most competitive races in the South Eastern Lower Mainland are Conservative vs. Liberal races.  A couple, such as Cloverdale Langley City and Delta, were surprise Liberal wins in 2015 that the party will have to fight hard to hold on to.  That cabinet minister Carla Qualtrough is the Liberal candidate in Delta may help them to hold that riding.  My model has the NDP as just barely competitive in Surrey Centre, putting the party a little under 9 points behind the Liberals.  As such, this riding is likely a long shot for the NDP, though they would have won it in 2011 had the 2011 election used the 2015 riding boundaries.  My model has one three-way race in the region, Pitt Meadows Maple Ridge.  This will be a difficult riding for progressive strategic voters as my model has the Liberals and NDP quite close to each other (and behind the Conservatives).  The Liberals might have a slight advantage because they have an incumbent in Dan Ruimy running for re-election in the riding.  He only narrowly won in 2015 and does not have much of a national profile, so incumbency may be less of advantage in this riding than in others and it would not be surprising to see it end up as a close three-way race.

SE Lower Mainland Table

If the South Eastern half of the Lower Mainland is friendly to the Conservatives, the North Western half is not.  The Liberals have 4 safe seats in the region, the NDP have 3, and the Conservatives have none.  On top of this, of the 5 seats in which the Conservatives are competitive in they have one in which they are running against NDP leader Jagmeet Singh (Burnaby South) and another in which they are running against popular independent Jody Wilson-Raybould (Vancouver Granville).  With Singh and the NDP surging in popularity as the election gets nearer it is hard to imagine Singh losing his seat.  Wilson-Raybould will also be incredibly tough to beat, though if the Liberals and Wilson-Raybould split the vote it is conceivable that the Conservatives could win that riding.  It may well be that Wilson-Raybould is the better bet for strategic voters though- though it is hard to tell for sure because riding polls are not that reliable.  In Vancouver South Defence Minister Harjit Sajjan gives the Liberals a good shot at out performing my model’s expectations, though Wai Young is also a former MP for the riding and ran unsuccessfully for mayor of Vancouver.  She likely has her own name recognition to compete with Sajjan’s.

North West Lower Mainland Table

Port Moody Coquitlam has a retiring NDP MP in Fin Donnelly, and my model’s failure to account for retiring incumbents may mean that it is overestimating the NDP in this riding.  At the same time, it is hard to see the NDP losing so much ground that they become uncompetitive.  The hardest choice for progressive strategic voters in the North Western Lower Mainland will be in Coquitlam Port Coquitlam where all three parties are competitive.  My model has the NDP in third and the Liberals have an incumbent running for re-election in the riding, so the Liberals may have the better shot at winning it but it is hard to know with any kind of certainty.  It is finally worth noting that the easiest choice for progressive strategic voters will be in Burnaby North Seymour.  This is where the Conservatives dropped their candidate after homophobic comments that she had made came to light.  Progressive voters in the riding can feel confident that they can vote for either Liberal incumbent Terry Beech or former New Democrat MP Sven Robinson without worrying that a vote split will lead to a Conservative winning.

If the chances of Conservative success in the North Western Lower Mainland are bad, the Liberals’ chances on Vancouver Island are abysmal.  My model does not have the Liberals competitive in any ridings on the Island.  It has one NDP safe seat, North Island Powell River, and one Green safe seat, Elizabeth May’s Saanich Gulf Islands.  The growth in Green party support means that my model is showing three Green vs. NDP races.  These are in Cowichan Malahat Langford, Esquimalt Saanich Sooke, and Victoria.  At 9 points down, my model has the Cowichan Malahat Langford as a long shot for the Greens.  Like in Fredericton, though, my model may not be capturing the level of concentration of the growth in Green support and so Cowichan Malahat Langford may be more competitive than my model suggests.  In contrast, Courtenay Alberni is a Conservative vs. NDP race, with NDP incumbent Gord Johns in a fairly good position to hold on to the riding.  There is one riding that will be tricky for progressive strategic voters, Nanaimo Ladysmith.  The NDP won this by almost 10 points in 2015, but then lost it in a by-election to Green Paul Manly.  Because by-elections sometimes yield unusual results it is hard to know whether the NDP’s success in 2015 or the Greens success in the by-election is more instructive with respect to which party is stronger.  While my model has the Conservatives in third, they are not so far behind that it is inconceivable that a Green/NDP vote split will lead to a Conservative win in the riding.  It is not clear whether a vote for the Greens or a vote for the NDP is the best way to block the Conservatives from winning the riding.

Vancouver Island Table

British Columbia will offer an interesting conclusion to election night.  On the mainland there is an interesting mix of Conservative vs. Liberal and Conservative vs. NDP races.  There are also some three-way races where it will be difficult to tell the winner until election night.  On Vancouver Island it will be interesting to see whether the Greens are able to hold on to Nanaimo Ladysmith, and whether they will be able to pick up seats in other ridings on the Southern half of the Island.  Given how close the national polls are, which parties do well in BC could make the difference over which party holds the most seats and who holds the balance of power in a minority situation.

Standard

Final Week Competitiveness Estimates: Atlantic Canada

As the 2019 election campaign comes to a close I am revisiting some of the analysis that I have done regarding which parties are competitive in which ridings.  This allows me to update it to account for the way that polls have shifted over the course of the campaign.  With little time left in the campaign, I am also able to do a more careful riding by riding analysis as there is less time left for a large shift in the polls to affect which ridings are competitive.  In order to further limit the chances a late election shift in the polls will throw my analysis off, I started with the regions that are least competitive and am working my way towards the more competitive regions over the course of the week.  In this post I look at Atlantic Canada.  In many ways Atlantic Canada is the opposite of the Alberta and the prairies.  Where both Western regions are Conservative strongholds that have seen the largest party hold on to its vote share, Atlantic Canada was swept by the Liberals in 2015 but has seen the party lose a large amount of support.

To estimate competitiveness, I calculate a universal swing based on the difference between parties’ vote shares in the previous election and regional polling averages from the CBC poll tracker.  These estimates come from averages from October 14.  Later posts looking at other regions will use polling averages from later dates as I try to ensure that my estimates keep up with polling changes.  I count a party as competitive in a riding if it is either within 10 points of losing or within 10 points of winning.  I count a seat as safe for a party if it is up by at least 10 points.  I do not account for the effect that retiring incumbents will have on parties’ chances in some ridings, though I will discuss a couple of ridings where I think that retirements are likely throwing off my estimates.  This will be particularly tricky to grapple with in Atlantic Canada because of the Liberal sweep in 2015.  It is unlikely that the Liberals will sweep the province again, but it is hard to know which incumbents have built up enough of a reputation to protect themselves against a Liberal decline.

A look at the change in vote shares shows that the Liberals have lost a lot of support since 2015.  This is not surprising.  The party won a massive 58.7% of the vote in the region along with every seat.  For a party that had not won more than 50% in an Atlantic province other than PEI since 1993, this was a surprising and unsustainably strong result.  Some regression back to the party’s normal levels of support wast to be expected.  Even with their support dropping by 18 points, the party is still at a very respectable 40% in the region and is well ahead of its competitors.  It is notable that the NDP has not been able to regain any of the ground it lost in 2015.  The party is sitting at 16% and is still in some danger of being knocked to 4th place by the Greens.  Instead, most of the Liberals’ losses have gone to either the Conservatives or the Greens.  The Conservative recovery puts them around 27%, which is probably still a bit lower than what one would expect the party based on its performance in Atlantic Canada over the 2000s.  The growth of the Green party, in contrast, makes them a serious competitor in the region for the first time.

Change from 2015 (Atlantic Canada)

Despite their decline, the Liberals still have a large number of safe seats in Atlantic Canada, with 17.  My model gives the Conservatives 4 safe seats and the NDP 1, but all but one of these safe seats has a Liberal incumbent running in it (the Liberal incumbent in Tobique Mactaquac is not running for re-election).  An incumbent that has built up a strong reputation may be able to protect themselves from a swing against their party, and so one should be careful not to be too confident that these seats are safe.  At the same time none of the Liberal incumbents in ridings that my model is showing as safe for other parties has much of a national profile, and all won for the first time in 2015, so it is hard to tell whether they have built up their reputation without being in the riding.  In addition to those safe seats, my model is showing 6 Conservative vs. Liberal races, 3 Liberal vs. NDP races, and one three-way race.

Competitive Seats in Atlantic Canada

The most interesting province with respect to competitiveness in the Atlantic is New Brunswick.  All 4 seats that my model is showing as safe for the Conservatives are in this province.  Fundy Royal, New Brunswick Southwest, and Tobique Mactaquac were all seats that the Conservatives won with over 50% of the vote in both 2011 and 2008.  Miramichi Grand Lake was created for the 2015 election, but a transposition showing how many votes parties would have won had the 2011 election been run with the 2015 boundaries show the Conservatives over 50% here as well.  The strength of the party in previous elections suggests that these are all seats that the Conservatives should be able to win should they return to respectability in Atlantic Canada.

New Brunswick Table

Fredericton, Moncton Riverview Dieppe, and Saint John Rothesay all come up in my model as Conservative vs. Liberal races.  There is not much reason to doubt this with respect to Saint John Rothesay, but Fredericton and Moncton Riverview Dieppe are trickier.  Fredericton is probably the Atlantic riding that the Greens are most likely to win.  My model has them 15 points down, but that does account for the fact that provincial Green leader David Coon won Fredericton South with over 50% of the vote in the most recent provincial election.  If the Green party’s growth in Atlantic Canada is more concentrated than model is able to account for, this could actually be a three-way race between the Conservatives, Greens, and Liberals.  Moncton Riverview Dieppe is tricky because Ginette Petitpas Taylor is the incumbent in that riding.  She is the Health Minister and may have a strong enough reputation to protect her against broader Liberal declines in the region.  Finally, it is worth noting that Acadie Bathurst is the riding my model shows as a Liberal vs. NDP race.  The model may be overestimating the NDP because it is not accounting for the limited attention that the party has paid to the province in the run up the election.  At the same time, the NDP held this riding from 1997 until 2015, though it is unclear how much of their prior success was due to the popularity of then MP Yvon Godin.  Regardless, if the NDP is to win a riding in New Brunswick it is likely that it will be Acadie Bathurst.

Despite the decline in support in the Atlantic, much of Nova Scotia still is safe Liberal territory.  4 seats fall outside of that group.  I have Central Nova and South Shore St. Margarets as Conservative vs. Liberal races.  Central Nova is Peter Mackay’s former riding, so it is not surprising to see the Conservatives competitive there.  South Shore St. Margarets was also a long held Conservative riding.  Including 2015, this riding has only been won by the Liberals twice since its creation in 1968 (the Conservatives or Progressive Conservatives have won it in every other election).  The Liberals’ strength in Nova Scotia in 2015 coupled with the fact that the Conservatives are still polling a little lower in Atlantic Canada than they usually do mean that it is possible that the Conservatives will be shut out of Nova Scotia.  If they are not though, it will likely be because they will have managed to take at least one of these two ridings.

Nova Scotia Table

My model has the NDP competitive in Halifax and Sackville Preston Chezzetcook.  Both are ridings where it may be tricky to separate the NDP’s success in past elections from the individuals who ran for the party.  Halifax was held by the NDP from 1997 until 2015 but by former leader Alexa McDonough and then by popular MP Megan Leslie.  Leslie’s popularity may have led the party to a higher vote share in 2015 and thus may be leading my model to over-estimate the NDP’s strength in the riding.  To add to this, notable Green candidate and former journalist Jo-Ann Roberts is running in Halifax, though it is difficult to tell whether that does more damage to the Liberals or the NDP.  If Roberts takes more support from the Liberals, the NDP may take back the seat.  However, if Roberts wins over more NDP voters than Liberal ones, a Green/NDP split may help the Liberals hold the riding.  My model has Sackville Preseton Chezzetcook as a three-way race, but with the Conservatives down by 9.5 points, it is more likely a Liberal vs. NDP race.  Peter Stoffer held the riding for the NDP from 1997 to 2015 and, like with Halifax, that may be leading the model to over-estimate the NDP in the riding.  All of this being said, with the Liberals dropping by 18 points it is reasonable to expect the NDP to win 1 or 2 seats in Nova Scotia, and these would be the two ridings that they could win.

There are only a couple of ridings between Newfoundland and Prince Edward Island that are competitive.  My model has Egmont, in Prince Edward Island, as competitive.  The Conservatives won this riding in 2008 and 2011, but Prince Edward Island is also very Liberal, so while it is reasonable to expect that the party could win this riding, they probably are not the favourites.

PEI Table

In Newfoundland I have the NDP competitive in St. John’s East and St. John’s South Mount Pearl.  In both cases the NDP incumbents pushed up the NDP’s vote share in 2015 and, in the case of St. John’s South Mount Pearl, that may be causing my model to over-estimate the NDP’s strength in the riding.  It probably also does not help the NDP that the candidate that won St. John’s South Mount Pearl for them, Ryan Cleary, had an on-again off-again relationship with the party.  To the extent that Cleary was able to pull votes to the NDP because of his local popularity, it may difficult for the party to replicate the level of support it had in the riding in either 2011 or 2015.  St. John’s East is a different story because the NDP candidate is Jack Harris, who won the riding for the party in 2008 and 2011 by large margins.  Harris’ popularity took the party from under 20% of the vote in 2006 to over 70% in both 2008 and 2011.  Harris lost by a narrow margin in 2015 and it is unlikely he wins the same large shares of the vote he had in 2008 and 2011.  At the same time, he is a popular former NDP MP running against a Liberal party that has seen a significant loss of support in the region.  Even if the model is underestimating the Liberals’ strength and this should be a competitive Liberal vs. NDP seat instead of a safe NDP seat, one has to think that Harris has the best shot of any New Democrat in Atlantic Canada of winning a seat.

Newfoundland Table

Despite the decline in Liberal vote share there are still a large number of safe seats for the Liberals in Atlantic Canada.  The Conservatives should win at least a few seats in New Brunswick and have a shot at a couple in Nova Scotia and maybe one in Prince Edward Island.  The NDP meanwhile has about 5 seats that they are competitive in, though two of those are long-shots for the party.  Progressive strategic voters do not have a lot of hard choices in the region.  Where the Conservatives are competitive, the Liberals are their closer competitors.  Where the NDP is competitive, the Conservatives are uncompetitive and so progressive voters in such Atlantic Canadian ridings can feel relatively safe that a vote for the NDP is not going to lead to the Conservatives winning a seat.  Even in Sackville Preston Chezzetcook, which my model has as a three-way race, the Conservatives are well behind the Liberals and NDP.  With poll numbers where they are, it is hard to see vote splits leading to Conservative gains in Atlantic Canada.

 

Standard