44

Prodigal sons and daughters


 

From our friend Paul Adams, who seems to be everywhere this autumn and who is blogging on these Ekos robo-polls for — well, for his employer, Ekos, I’ve brazenly swiped this damned interesting chart. In return for my larceny, I ask you to follow the link to Paul’s explanation. My own explanation comes below, after the chart…

Voter Retention

Reported Vote – 2006

Vote Intention – 2008

CPC

LPC

NDP

BQ

Green

Did not vote

Conservative

84

18

5

9

11

35

Liberal

6

62

13

5

13

18

NDP

5

11

74

11

6

30

Bloc Québécois

1

1

1

71

2

1

Green

4

8

7

4

68

16

This chart collates some of what pollsters call “cross-tabs” from the latest Ekos Cylon Terminator robo-poll. Stated party support from the 2006 election is cross-indexed with stated party preference in this election right here now. So if you look down from “CPC,” you see that 84% of people who seem to recall voting for the Harper Conservatives in 2006 are now planning to vote for the Harper Conservatives again. Similarly, the NDP seems likely to keep 74% of its 2006 voters, the Bloc 71% of their voters, the Greens 68%, and… the Dion Liberals are on track to keep 62% of those who voted for the Paul Martin Liberals.

Where are those leaking Liberals leaking to? One in five of them plan to vote for the Harper Conservatives. In contrast, only 6% of 2006 Harper Conservative voters are angry enough to plan to switch to the Liberals.

Anyway, Paul Adams has more.


 

Prodigal sons and daughters

  1. Am I reading this wrong, or is the Bloc keeping an enviable 71% of 2006 voters, with only 9% going to the Tories? Isn’t that — not what’s supposed to be happening?

  2. More accurately, Kady, it would seem that the Bloc has lost some 29% of their voters to a bunch of other parties, and the Conservatives, finishing a competitive second in a number of their ridings, don’t need the full 30% to win them this time.

  3. I was wondering the same thing, Kady, maybe Lib supporters in Que are moving to Cons more than BQ voters? BQ voters seem to be going NDP.

  4. As Adams was pointing out before I stole his graphic, the further you dig into the numbers the more shimmy you get from sample-size considerations. But even the more CPC-friendly Quebec polling shows swings of less than 10 points…

  5. What’s going on in Quebec then? Are lots of apolitical Quebecois so inspired by Harpermania that they’re voting for the first time in years, or will the Tories’ performance there underwhelm in a bit less than a month?

  6. curiouser and curiouser. I had no idea the LPC was bleeding as bad as it is hmmm.

  7. The Conservatives already took a large chunk of the right-of-centre Bloc vote last election. According to this poll they’re set to take a good chunk more, with almost no back-leakage.

    However this election the Conservatives are clearly targeting the right-of-centre Liberal vote in Quebec. They do that best, confusingly, by attacking the Bloc.

  8. Note that the off-diagonal numbers in the BQ row are all 1 or 2. The 29% is pure bleed, as they’re not poaching from anyone else.

  9. I dunno, something strikes me as really bizarre about those numbers. The people who voted for Paul Martin’s liberals, in the wake of Adscam and the NDP’s smear investigation of Ralph Goodale, are now moving to the Conservatives in the wake of Cadscam and In-And-Ou.. oh wait.. I think I get it now. Some people just like corruption in their politics.

  10. The important question is what the Liberals do about it. They could try to pull back the conservatives they’ve bled, or hit the “Harper majority/strategic voting” buttons to try to scoop up wavering NDPers and Greens.

    (Which actually makes things complicated in Quebec and the West, since there are seats there that the NDP could conceivably win that the Liberals probably have no shot at.)

  11. Just a small quibble with your post – it is not quite autumn. Those of us clinging to the last few days of summer would prefer a correction.

  12. So there’s a real niche potential for Bloc Guité.

  13. Very interesting. So it would seem Harper hasn’t disappointed those who have backed him all along, and the outrage is really just coming from those who never supported him in the first place?

    The most interesting number here to me was the fact that 35% of people who didn’t vote last time are going Tory. So Harper has won over more former non-voters than any other party – whilst the Grits are barely ahead of the Greens on that count.

    So how do the Liberals reverse this? Is it too late? They will have to do something huge to shake it up.

  14. You get some really bizarre results if you put these into the UBC ESM forecasting tool:

    CON: 186 seats (38.3%)
    LIB: 46! seats (24.4%)
    NDP: 30 seats (19.6%)
    BQ: 45 seats (8.4%)
    OTR: 1 seat (9.3%)

  15. Also, you get the Bloc winning 1% of the vote in ROC.

  16. On the bright side, should the Liberals manage to squeak in to power this time, they’ve got a strong incentive to push for proportional representation now.

  17. And the Blocs newfound concern with the NDP suddenly makes sense: that’s where most of their disaffected voters are going!

    This is far-and-away the most illuminating poll I’ve seen so far this election.

  18. ALW, small point: surely non-voters from last time are likeliest to not-vote next time?

  19. One wonders how many of the didn’t-vote-last-time crowd will become didn’t-vote-this-time-either on election day?

  20. Paul Wells,

    Are the parties generally comparable in their abilities to “get the vote out”, or is there evidence that some are better organized than others in this respect?

  21. Sean S., that sort of thing varies over time, but I think you’d get pretty broad agreement that Conservatives are lately good at GOTV, to the tune of a couple points’ advantage over rivals, and that the Bloc has consistently had difficulty translating vote intentions into votes.

  22. As for the didn’t-vote-last-time crowd, haven’t the Tories spent all kinds of money to try to identify who they are, whether they would lean their way, and how to get them to the polls on election day if they do? Like, in a way more sophisticated fashion than everyone else?

  23. Thank-you for this Link Wells, interesting.

    I still think we will have a minority government.

  24. Paul, it’s not clear from the chart whether the people in the last column are (a) people who didn’t vote last time but had a preference then or (b) people who didn’t vote last time and also had no preference. It seems logical that people with a preference would be more likely to vote than people with no preference.

  25. Paul,

    Do you have any insight into how much of an impact Martin’s rise, reign and relapse had on the on-the-ground organization of the Liberals? GOTV can be vital in swing ridings.

  26. Nick Nanos just out : Results of today’s CPAC-Nanos tracking poll show the Conservatives are still holding strong nationally at 38%, followed by the Liberals at 31%, the NDP at 17%, and the Green Party and the Bloc Québecois tied at 7%. On the CPAC-exclusive leadership index scorecard evaluating trust, competence and national vision, Stephen Harper leads with 108 points, followed by Jack Layton at 51 points, Stéphane Dion who has dropped to 32 points, Elizabeth May at 14 points and Gilles Duceppe at 13. Stephane Dion’s one day drop (from 48 to 32 points) coincided with Bob Rae’s intervention in the campaign yesterday attacking the Conservatives and the NDP. Even factoring the personal Dion drop, there was no impact on support for the Liberals.

  27. “Even factoring the personal Dion drop, there was no impact on support for the Liberals.”

    I think that is just the law of diminishing returns at play – there is virtually nobody who is voting Liberal because they like Dion, so they’re beating a dead horse.

    “The most interesting number here to me was the fact that 35% of people who didn’t vote last time are going Tory.”

    This was interesting given that the Tories are usually weaker among young voters (it does explain the NDP strength there). It may also be misleading – what percentage of the total fell into this category? If it is low, this subsample probably has a very high margin of error.

  28. Mike Moffat beat me to the UBC stats. I’ll do the provincial/territorial breakdown.

    Party Cons. Lib. NDP Bloc Other

    BC 24 4 8 0 0
    AB 28 0 0 0 0
    SK 13 1 0 0 0
    MB 10 1 3 0 0
    ON 71 20 15 0 0
    QC 16 13 0 45 1
    NB 9 0 1 0 0
    NS 5 4 2 0 0
    PE 3 1 0 0 0
    NL 6 1 0 0 0
    YT 0 1 0 0 0
    NT 0 0 1 0 0
    NU 1 0 0 0 0

    TL 186 46 30 45 1

    These predictions don’t take into account local events that may change the results. Check these results with DemocraticSpace’s 2008 predictions.

  29. Mike — I get even more outlandish numbers for the UBC matrix:

    CPC – 198 seats (39.7%)
    BQ – 42 seats (8.3%)
    LIB – 38 seats (24.7%)
    NDP – 30 seats (20.9%)
    Other (Greens) – 0 seats (6.4%)

    Which puts Gilles Duceppe in again as Leader of Her Majesty’s Disloyal Opposition.

    ***

    Not happening (the model has Goodale losing Wascana!), but it’s fun to look at…

  30. WRT Quebec, what we seem to be seeing is that, unless the Bloc vote falls to some “tipping point” (defined as “a magic thing we make up to make the results more interesting”), the Bloc shouldn’t be expected to fall much below 40 seats. They got 37, if I recall correctly, in 2000. They’d have to do really badly, by their standard across the last four elections, to fall below that.

  31. Oh, I see — I read down, you guys read across.

  32. I notice a few things. The Conservatives are keeping 84% of their voters while the Liberals only keeping 62%. The other parties are somewhere in between.

    There is very little transpostiton of votes between the NDP and Greens. Someone who voted Green in the last election is more likely to vote Conservative than NDP.

    The Conservatives have largely keep their supporters while the other parties are fighting for each others.

  33. Goodale may indeed lose Wascana.

  34. And Ignatieff losing Etobicoke-Lakeshore.

    These are possible outcomes — just not plausible.

  35. I’d be interested to see the missing row at the bottom. “Not voting this time.” Not very helpful for predictions, but useful for the parties to know. People who are disillusioned, but unwilling to move elsewhere are a good target to pull back into the fold.

  36. Based on the UBC Election Forecaster, here are the voting predictions for each party:

    Conservative 38.3%
    Liberal 24.4%
    NDP 19.6%
    Bloc 8.4%
    Other 9.0%

    For those of you who comment about there being no Green category or transfer of votes from Non-Voters, the UBC Election Forecaster does not have categories for these.

    The results above seem pretty close to the latest polls.

  37. Guys, you’re not being fair to Paul Adams. This discussion should take place beneath his blog post, not the other Paul Well’s. Damn good thread though, I have to admit.

    How about a joint Paul & Paul election blog? Paul A brings the tables, Paul W brings the friends. Sounds like a party.

    JK – sure we can still have a minority. This is a snapshot of the situation as of earlier this week. Still lots of time to go, unless people think election campaigns don’t change the way people vote anymore. And seeing the enthousiasm here, few people seem to believe that.

  38. I’m most surprised by the 68% retention rate of the Greens – at a season when their support is so high. Wasn’t aware that the voters perceived this year’s Green Party as so different from last election’s. I mean, I know there’s been a makeover, May has arrived, etc. but I’m surprised they’ve got 32% revolving door support (to coin a phrase).

  39. PW: the Bloc has consistently had difficulty translating vote intentions into votes.

    MYL: Paul, if I am not mistaken, responsible pollsters have, years ago, developed a “separatist in public to save face, not so separatist facing the secret ballot” fudge factor for Quebec politics, no?

  40. Well, a lot of pollsters allocate undecideds asymmetrically in Quebec, according to assorted formulas. But in most (but not all!) elections since I’ve been covering politics in or near Quebec, there’s still a ballot-box bonus for federalists even after the pollsters thought they had accounted for it.

  41. I recall a Québécois comedian who had a routine about voting in a referendum — not sure if it was numéro un or deux. He comedically agonized over the choice he would make. He selected OUI. The crowd roared in approval. Then he nervously stated how proud he was of his selection, so proud he put the oui ballot in his pocket and walked away. The crowd offered a very understanding laugh.

Sign in to comment.