17 Responses to “Minority Math is Hard on the Liberals”

1. Ron says:

   September 28, 2015 at 10:22 am

   Your analysis is a little like reviewing a movie based on viewing a few frames. The situation is fluid and the Libs have slo-mo momentum. Look for trends, not snapshots.

   Ivison is not a writer I agree with often but today he has some interesting words:
   “There may be seismic shifts taking place that are not yet being felt on the doorsteps.
   Trudeau’s message that he offers a new approach, compared to the “Harper-lite” program proposed by Mulcair, seems to be starting to register in some parts of the country.
   If that is the case, the high number of undecided voters suggests the impact will be tsunami-like. But for now, Quebec voters are still waiting for a wave.”

2. BC Voice of Reason says:

   September 28, 2015 at 12:21 pm

   “The Liberals have run a strong and energetic campaign thus far under leader Justin Trudeau, so it seems unfair how hard it will be for them to find enough seats to claim the head of a minority government.”

   Are you part of the Liberal campaign team?

3. Kevin Logan says:

   September 28, 2015 at 12:27 pm

   So you wrote all this negative spin on the Liberals to tell us that they will likely triple their seats, while the NDP loses seats? Should the headline not read “NDP dreams of third way government a mirage as it melts before their eyes in the face of a tripling of Liberal seats?”

4. Jack D says:

   September 28, 2015 at 12:33 pm

   Your analysis is way over simplifying the current situation, Alice.

   I agree with the commenter above; look for trends not snapshots. Its a bit misleading to base your hypothesis of Liberal fortunes on a current projections when there is so much fluidity. You’re assumption is based entirely in the hope that the NDP’s support base in Quebec is solidified enough for them to build on, when in fact, its quite volatile. It is also predicated on the notion that the NDP can remain as competitive nationally with e-day drawing closer and closer. Regional fluidity is a reality and with an apparent slide in the NDP numbers for Quebec and Ontario, it’s showing weakness in the NDP’s ability to be competitive in a huge chunk of their own ridings nationally.

   I think it was a pollster who said this, if either the Liberals or the NDP show any signs of weakness in the latter half of the election, expect to see a quick coalescence of the “anti-Harper” vote around the other party. The NDP is plunging in Quebec and barely registering in Ontario. Those Quebec seats are now being threatened. The NDP can’t fight a war on two fronts and expect to make gains everywhere.

   This analysis is drawing foregone conclusions that are based on projections that are shifting as we speak.

5. rumleyfips says:

   September 28, 2015 at 12:45 pm

   I did my calculation Saturday morning while walking the dogs.

   Seat projections seemed off to me . In some ridings the party with the most support is shown to have a lower percentage chance of winning than the second place party. Two assumptions are made by most pollsters, pundits and aggregators. There is a strong correlation with the last election, incumbency, and the CPC votes are more efficient because many are rural.

   Both assumptions seem questionable. Things have changed since the last election. Liberal support is way up . NDP support is a little down. CPC support is down. There are 30 new ridings with no track record to use and boundaries have changed in a lot of places. A large number of CPC ministers and members chose not to run this time. Riding history tells us less this election.

   In Atlantic Canada, the CPC seems to be losing at least 10 seats . Their support in Quebec is low and rural seats some of their rural seats may be at risk. They may lose some rural seats in Ontario , Manitoba and the North and will in BC . If they are at risk in 30 rural seats, their vote efficiency is gone.

   Riding level polls done by Environics show the weakness of polling algorithms this election.

6. Bluegreenblogger says:

   September 28, 2015 at 2:18 pm

   @ BC voice of reason: Don’t be silly, there is such a thing as an objective statement you know. And Alice is correct, the Liberals have run a strong disciplined campaign. The NDP has had a difficult job, but hey are running a good campaign too.

7. The Pundits' Guide says:

   September 28, 2015 at 3:09 pm

   So, I’m awake from my nap, have been through the day’s nomination updates from Elections Canada and my Inboxes, and have time to address some of these comments.

   THIS WAS NOT A SEAT PROJECTION.

   It WAS an exercise for me to try and get a handle on the new battlegrounds in light of the possibly changing / possibly not party system in our country.

   So, just my luck that it got published just as the Nanos weakened further for the NDP in Ontario, and Abacus showed further movement in Quebec. We will all have to incorporate what that means into our sense of what’s possible going forward.

   The Liberals can get to minority territory by winning all the contests in the Atlantic, Ontario, and their strongholds in Quebec.

   The NDP can lose many of their contests to both the Liberals and Conservatives. I don’t think either outcome was precluded by the above exercise.

   I’m not alone in pointing out some current Liberal challenges in BC: for example, see the research referred to in Adam Radwanski’s column of this morning: http://www.theglobeandmail.com/news/national/trudeau-struggling-to-win-over-bc-voters-despite-direct-tv-pitch/article26556323/

   Some people see the heavily Chinese seats in Richmond going Liberal, but I believe they stay Conservative based on pot, safe injection sites, and other socially conservative values. I also never accepted the seat projections of either Delta or South Surrey-White Rock as going Liberal either. So, a 5-7 seat ceiling for the Liberals in BC for the time being, unless things start to move there, because as the Liberals come up, they’re so far cutting into the Conservatives.

   As for the blue-orange seats in southwestern and northern Ontario, Stephen Harper visited a lot of those in the first two months of the campaign. A retooled Conservative ballot question may indeed be behind their growth in Ontario, at the NDP’s expense, which was behind most of the NDP’s hoped-for seat growth in that province.

   I’ll publish more of the assumptions behind the exercise in coming posts. Continue to pick away at them. It’s all good.

8. Shadow says:

   September 28, 2015 at 3:33 pm

   Hmm so we’re wrangling over who gets the PM coalition nod, eh ?

   Guy with most votes LPC, or guy with most seats NDP.

   With Mulcair running as Paul Martin and Justin Trudeau running as Jack Layton we’ve truly entered the narcissism of small differences territory.

   Bottom line is that Alice’s central point is correct – the math is much harder for the Liberals because of their inefficiently distributed support. That doesn’t mean she’s projecting anything. Just looking at how things might break down if things stay deadlocked.

   Of course if the Liberals have a big sweeping win they’ll get a big sweeping chunk of seats. 100% of the vote takes 100% of the seats every time. But putting aside the fevered dreams of partisans some of us are interested in how things may break down under more likely scenarios.

9. Ron says:

   September 29, 2015 at 12:03 am

   A very solid Munk debate from Mr Trudeau.
   I believe he is now perceived as “the agent of change” and will gain support across the country.

10. Ed says:

    September 30, 2015 at 1:52 pm

    I did this exercise with the electionprediction.org site. I have problems with their methodology, and their success rate at calling contests seems to be about 80%. I think they are underestimating Conservative and overestimating NDP strength. Still, the site is good enough to be a start and the comments give some sense of what is happening in the constituencies.

    As of a couple days ago, they had the NDP at 98 ridings, the Conservatives at 88, and the Liberals at 68, with other (BQ/ Green at 2, and 82 “too close to call.”

    I went through the “too close to call” predictions. I wound up giving 50 to the Conservatives, 22 to the Liberals (mostly in and around Toronto), 8 to the NDP, and 2 to other (BQ/ FD/ Gr/ Ind), which I think corrects for the bias in the underlying predictions. There are alot of ridings where the Conservatives had big margins in 2011 that are not going to go away, absent a 1993 style meltdown which isn’t happening. This implies a Conservative minority, which I’ve thought was the most likely outcome for some time. But yes, short a big upward rise in the polls, I don’t see how the Liberals get into majority territory, there are just not enough races they are competitive in. Its just hard to do that from their 2011 result in one election.

    The Liberals could have pulled it off with a sort of “Trudeaumania version 2”, but once Harper and Mulcair firmed up their support, the window of that happening closed. It probably wouldn’t have worked with a long campaign, which is one reason why we got a long campaign. What he could do is make a good impression and get a big increase in seats, maybe come in second, which is what seems to be happening.

    The Conservatives have a big reserve of rural ridings west of the Ottawa River where they will get at least 70 MPs without having to do much. By the way, the figures show that they could get wiped out completely in the Atlantic provinces and still get a majority if they just held what they notionally held everywhere else (this would have to include the 5 Quebec ridings, but they are up in Quebec). On the other hand, NDP support in Quebec has the composition of jello, but they can afford to lose some votes there, and still do OK as long as their percentages in the other provinces stay within 3% or so of the 2011 result.

    I calculate that 36% of the national popular vote will deliver a majority. At the end of the day the Conservatives are still the likeliest of the three to get the last minute uptick that will take them there. For the Liberals, it is just too much to go from about 20% to 36%in one election, and for the NDP, getting to 30% last time was so unprecedented that they will be lucky to keep most of that.

11. Ed says:

    September 30, 2015 at 1:55 pm

    I guess what I just said was that for the Liberals and the NDP the election was always really about who comes in second. There would be an opening for a majority with a Conservative meltdown, but its far enough into the campaign to see that this won’t happen.

    The second place finisher will be well positioned as the main alternative next time, and obviously if no one gets a majority, would be the lead partner in any coalition or agreement of the two “left” parties.

12. BC Voice of Reason says:

    September 30, 2015 at 8:01 pm

    @Bluegreenblogger

    You have very low standards for what you consider a great campaign. There have been no major issues that have excited the general population. A year ago according to the polls there were about 40% of Voters that were supporting Trudeau. His lack luster performance and continuous gaffes have driven down the people who wanted to vote for him.

    Mulcair has done no better. They have not provided a clear vision of their Canada.

    I really can’t see any of the polite/considerate women cohort who have been moving to the political front over the last decade considering to vote for Trudeau after his juvenile rude debate behavior. Maybe this is acceptable in the HoC but there is absolutely no place for that in the real world.

    The voter turnout is going to be an all time low. The 40% of Canadians that do not vote will not be inspired by either the personalities or the platforms presented by Trudeau or Mulcair. They are basically OK with the current Harper government.

    In 2004 65% of the voters voted. The increase was because the Liberals were exposed as near totally corrupt. ie the Electorate was worked up enough to kick the bums to the curb. There is no Shawinagate or Adscam equivalent to get the electorate worked up and no new policy that has grabbed the country’s imagination.

13. Ed says:

    September 30, 2015 at 9:26 pm

    I double checked. Since MacKenzie King retired, eleven of the nineteen Canadian elections has resulted in a party getting a majority in the House of Commons. The other eight resulted in parliaments with no overall majority. Canadians are more comfortable with parliaments with no overall majority than, say, the British (or more specifically the English) who will bandwagon for the party most likely to get a majority to make sure they get it. There have been only two UK parliaments elected with no overall majority during the same period.

    Two of the majorities, in 1984 and 1993, reversed majorities held by the other party in the earlier parliament. This is not very common. More often governments change by producing a parliament with no overall majority, then the alternative party wins a majority later.

    The NDP broke with precedent with getting from their usual third place finish to finishing second in 2011, thanks to a perfect storm of factors, but no party has gone from third to first. The NDP itself under Broadbent came closest to this in the 80s but obviously it didn’t happen.

    The Conservatives, with a bank of safe seats in rural areas where people speak English, a record of convincing 40% of the electorate to vote for them next time, and no compelling reason to eject them, plus a campaign pretty much rigged to be fought on Harper’s terms, have a reasonable chance of getting their percentage of the vote into the high 30s and a majority. There is still probably a better chance of the Conservatives losing their majority but still taking the most seats.

    So I will reiterate my usual point. I actually think the younger Trudeau has pretty much performed well and the Liberals will gain in this election and will be back in contention for a majority in the next election. They still have a good chance of a second place finish and minority government or getting back to being the main opposition party. A majority, though, was never in the cards short of Trudeaumania version two in a short campaign, which was deliberately checked by Harper by having a long campaign. Compared to other politicians, Harper is short on charisma but unusually good on strategy.

14. Ed says:

    October 1, 2015 at 9:58 am

    Sorry for the triple post. This should have been consolidated into one post.

    The party that gets 36% plus of the national popular vote should get a majority.

    I don’t think the distribution of Liberal support will keep them from a majority. The problem is getting to 36%. If the Liberals get 36% on election day, they will be pulling victories in ridings out of their rear. They will be winning ridings that were thought to be CPC-NDP contests, or safe for the other two parties. The ridings will materialize somewhere.

    The problem for them is getting to 36%. Its a big leap from five years ago, and they really needed one of those sudden surges of enthusiasm for their leader that sometimes happens with electorates, but is unlikely with a long campaign.

    The Conservatives could well get to 36%, they got more than that last time, and its really unlikely but possible for the NDP. I don’t think its more than theoretically possible for the Liberals.

15. Ed says:

    October 12, 2015 at 9:28 pm

    I posted here earlier that to get a majority, the Liberals would need about a 16% increase from their national popular vote in 2011. This seemed to be implausible. But polls now show that they are within reach of this, and I remembered that the Canadian public is prone to massive, often inexplicable electoral swings. So I went back and looked to see if there is a pattern from past federal elections when Canadian voters gave big percentage increases or decreases in the vote for particular parties.

    I found 30 instances where a Canadian political party either picked up or lost at least 7.5% of the popular vote between two federal elections. There is an additional instance of 2004, where the newly merged Conservative Party lost 8.6% of the vote from the combined Reform/ Progressive Conservative vote in 2000, but this is ambiguous, since it gained 4.2% from the 2000 Reform vote alone. The other thirty instances fall into four categories.

    A. 1993 election (5)
    First there is the split in the right, and the uniting of the left under the Liberals in 1993, which accounts for five of the thirty instances by itself:

    1993 Liberals +9.3% NDP -13.5% CPC -27% Reform +16.6% BQ +13.5%

    B. Bay of Fundy (8)
    These are cases where a party gains a big swing to itself in one election, and then gives it all back in the last election. If a party gets a big increase in its vote in an election, it is set up for big losses the next time. There are eight instances of this happening, including 1945-9 where the Liberals lost a big chunk of their vote, and then got it back after Mackenzie King retired :

    1874 Conservatives -8.6%,
    1878 Conservatives +12%

    1945 Liberals -11.4%
    1949 Liberals +9.4%

    1958 PC +14.6%
    1962 PC -16.8%

    1962 SC +9.2%
    1965 SC-8.3%

    C. Conservative fracturing 1920s/ 1930s (6)
    The Conservatives lost a lot of their voters to the Progressive movement in 1921, then got them back in 1925. In 1935 the problems with the Bennet government meant the Conservatives lost support to the new Reconstruction Party. The Tories eventually merged with most of the Progressives as well as with Reconstruction. This accounts for six instances:

    1921 Conservatives -27% Progressives +21%

    1925 Conservatives +16.2% Progressives -12.6

    1935 Conservatives -18.5% Reconstruction +8.7%

    D. Liberal meltdown (4)
    The Liberals lost 7.8% of the vote to the Diefenbacker Progressive Conservatives in 1957, which they never really got back, though they gained majorities in a half a dozen elections in the future, they never dominated as they did in the Mackenzie King-St. Laurent years. In 1984, the Liberals lost 16.3% of the vote, and the Mulroney Conservatives gained 17.6% of the vote. While the Mulroney coalition fractured in 1993, the post-Trudeau Liberals lost a step, they got over 40% of the vote in only two subsequent elections (1993 and 2000), and over 36% of the vote in another two (1997 and 2004):

    1957 Liberals -7.8% PC +7.8%

    1984 Liberals -16.3% PC +17.6%

    E. Other/ uncategorizable (6)
    That leaves six other instances. There are three big Liberal increases in the nineteenth century which I will leave to experts on nineteenth century Canadian political history to explain. There is a jump in Conservative support in the Great War election in 1935, and two big CCF/ NDP increases, one in 1935 and one in 2011:

    1872 Liberals +12%

    1887 Liberals +12%

    1900 Liberals +8.9%

    1917 Conservatives 8.4%

    1925 CCF +9.3%

    2011 NDP +12.4%

    Does the situation in 2015 fit any of these categories? I think a “Bay of Fundy” situation is a very real possibility, where the NDP gives back the 12.4% gain it achieved in 2011.

    However, the Liberal problem was not so much a big drop in support in 2011, but the air slowly leaking from the balloon after Chretien got 40% in 2000. Their support just kept dropping bit by bit until it reached a tipping point. I still don’t see any grounds for them to get anything like a 16% popular vote increase (a swing this size has happened only eight times in Canadian electoral history). Any NDP drop will more likely flow back to the Bloc, where they got a lot of their 2011 increase, and to the Greens as much as the Liberals. Nor are the conditions there for a big Conservative drop, there is no alternative party on the right to split their vote, and the Harper conservatives have built up their support a little at a time in successive elections, instead of one big swing.

    So yes, I suspect the polls are now greatly overstating Liberal support, and the likeliest outcome is a Conservative majority or very large minority.

16. Ed says:

    October 19, 2015 at 5:11 pm

    The last few polls showed a Liberal surge and the Liberals moving into majority government territory, so my posts here could turn well turn out to be completely wrong. We will find out in five hours.

    A number to keep in mind is 17.6%. That was the percentage gain in popular vote by the Progressive Conservatives under Mulroney in 1984. That holds the post-World War 2 record for an increase or decrease in a party’s national popular vote. There were some greater swings before World War 2, but they involved voters going back and forth between the Conservatives and Progressives/ Reconstruction, which dynamic doesn’t apply here.

    If the Liberal popular vote percentage gain is anywhere near that, and polls indicate this is, then this election would mostly closely resemble 1984, not 1993 as some people in the media claim. The party would still get under 40%, while Mulroney got a popular vote majority, so “landslide” would not be quite the right word, “stampede” might be appropriate. Incidentally, when Trudeau first became leader I thought this result was certainly possible, but I (along with Harper) thought a long campaign would eliminate the possibility of such a stampede.

17. Ed says:

    October 20, 2015 at 8:58 am

    The actual Liberal gain was 20.6% of the vote. This sets a record for port-World War 2 elections. It was also the second largest percentage increased in a party’s vote share in a Canadian federal election in Canadian history, behind the Conservative collapse in 1922. The CPC and the NDP also had significant percentage decreases, though in the NDP case this is explicable as they basically gave back their 2011 gain.

    This was a stampede election, literally in the sense that voters seem to have decided to vote Liberal in the last weekend of the campaign, and I think in the coming days people will blame pollsters too much for not picking it up earlier or the other two parties and their leaders for not preventing it. Polls taken during a parliament can’t capture voter decisions made at the last minute, and you can’t get in the way of a stampede.

    The NDP result is actually something they would have been thrilled with if they had gotten it in 2011. It was the second highest riding total and third highest popular vote percentage in their history. The CPC, at 99 seats and 31% of the vote, is exactly where it was when Harper took over in 2004, 99 seats and 30% of the vote.