September 29, 2014
September 5, 2014
These charts correct for poll bias. First the large parties:
And the rest:
So two weeks out I predict:
- Epsom won’t bother to vote for David Seymour and ACT will be gone from Parliament.
- Internet/Mana will win one or two electorate seats and either two or three MPs.
- The Conservatives will probably cross the 5% threshold.
- New Zealand First will be comfortably above the threshold.
- If the Conservatives make it then National might be able to form a government without New Zealand First in it. Which I think they’d rather do. Something tells me Winston Peters would quite like to be Justice Minister.
July 29, 2014
A look at the tracking polls for small parties helps explain what’s been happening over the last few days (larger interactive version here);
The Conservatives are only slightly higher than they were last election suggesting they’ll get 2.6% max. Probably less than that now that a seat has been ruled out and Conservative voters who saw their vote wasted back in 2011 might want to see their vote count this time around.
Craig’s strategy this year has been to try and target New Zealand First voters. That’s what National wanted him to do, and maybe there was some understanding that if he succeeded they try and help him out. I dunno. Anyway, it was a terrible strategy: most New Zealand First voters are soft Labour voters and the consequence for Craig attempting to target this unsympathetic demographic are that he’s toast.
I feel ambivalent about that. Sixty-thousand odd people voted for him in 2011, he has more of a mandate than an awful lot of sitting MPs – but the only reason he reached those people was because he happens to be rich. He’s talking about winning East Coast Bays in 2017, but I don’t know how much longer he can go on sinking hundreds of thousands of dollars a year into his political party. (That line from Citizen Kane comes to mind: ‘I did lose a million dollars last year. I expect to lose a million dollars this year. I expect to lose a million dollars *next* year. You know, Mr. Thatcher, at the rate of a million dollars a year, I’ll have to close this place in… 60 years.’ But I don’t think Colin Craig is Charlie Kane rich.)
Anyway, National’s refusal to help out Craig is why we’ve seen a sudden flurry of race-baiting from ACT and New Zealand First. They’re all competing for the same small pool of conservative elderly half-witted red-necks. Hopefully those votes will stick with the Conservatives and be wasted but they might give ACT enough votes to get Whyte into Parliament.
Winston Peters is well ahead of where he was at this point in 2011, and he made it over the 5% threshold very comfortably back then. He’ll be back again. Who will he go into coalition with?
A couple of weeks ago Colin Craig and Jamie Whyte showed up on one of the weekend politics shows, and both of them announced they probably wouldn’t be seeking Cabinet positions in a National government. Now, maybe its a coincidence that both of these political masterminds both had this same odd idea on the same day, but my guess is that it was an order on high from National’s leaders clearing the way for a nice, simple National-New Zealand First coalition. I still hear people saying that Peters is ‘unpredictable’ or that he ‘wants revenge on John Key’. I think the chances of Peters flipping a coin or picking revenge and going with a Labour, Greens, Internet/Mana et al are zero. He’ll be a senior Minister in the 3rd term National government. We might see Peters go even higher during the campaign as older Labour voters decide that Peters will ‘keep National honest’ and switch to him.
Finally, Internet/Mana really seems to have momentum and I think we can assume they’ll win at least one electorate seat, if not two. But it’s really hard to guess how they’ll perform on election day. Are the polls under-sampling all those cellphone only digital natives? Or will most Internet/Mana respondents fail to actually vote? I have no idea.
I do know that I’ve talked to a couple of students who are first-time voters who are ‘voting for Kim Dotcom’. He’s their big asset which is why he’s the headline act at all the Internet/Mana ‘Party Party’ functions. But he’s also, potentially, their biggest liability. Dotcom has promised that he’ll drop a ‘bombshell” on John Key five days before the election but National seems to have a few bombshells of their own that they plan to drop on Dotcom. Internet/Mana know that this is coming and their prominent use of Dotcom indicates they feel they can manage it when it does. We’ll see. It makes their result impossible to predict.
July 23, 2014
Updated the tracking poll. For variety’s sake this one goes all the way back to the start of 2005. It doesn’t correct for bias and the large circles intersecting the vertical lines are election results : Josie Pagani wrote a ‘what should Labour do now’ thing on Pundit that I partly agree with. (Although all such posts are predicated on the idea that Labour wants to win the election, and like I’ve said before, one of Cunliffe’s major challenges is a caucus filled with MPs who hate him and don’t want Labour to win the election while he’s leading it. They want to lose the election then assume control of the party and win in 2017, against a presumably weaker National-New Zealand First coalition. That’s not something Cunliffe can fix by ‘staying on message’ or ‘focusing on what matters’ or anything like that. It’s just a horrible reality that dooms him.)
Anyway, one of Pagani’s suggestions is:
Distance Labour from Dotcom. One reason for Labour’s poor polling is people just want to get rid of Dotcom and somehow he has become Labour’s problem now. Only because he is an enemy of our enemy. Labour should only ever say of Dotcom, “he shouldn’t be in the country and National should not have let him in. We want him and his party nowhere near government.”
About a month ago I wrote:
Now, Internet/Mana might make good on their promise to turn out loads of young low decile voters, which might not show up in the polls but will have a huge impact on the actual election – and good on them if they do. But at this point it looks like the Harawira/Harre/Dotcom contribution to the election will be to cannibalise the left vote and scare the centre vote over to National, possibly gifting them an historic victory.
If the trend continues Labour will have to take a risk (for once) and rule out Internet/Mana from a Labour government.
So why haven’t Labour done this? I think there are a couple of reasons. The friendship between Matt McCarten, Labour’s Chief of Staff, and Laila Harre has got to be a factor here. Also, if Labour rules out Internet/Mana they’re kind of ruling out hope. There is a chance that Internet/Mana will be a ‘game-changer’ that will swing the election. I think that chance is very tiny, but it is there, and that gives some purpose to the campaign that’s absent if you’re just fighting to get Labour back into opposition with slightly more MPs. Finally there’s the culture of the contemporary Labour Party which is very conservative and risk averse. No one would want to take ownership of a call like that in case it backfired somehow.
And the chance of a ‘game-changer’ (this has come to mean ‘event that generates enormous media coverage but changes nothing’, but I use it in the original context) increased slightly with Dotcom’s revelation that Glen Greenwald will be at his town hall meeting just before the election. Now, Greenwald might just show up and talk about the 5-eyes network. But he does have full access to the Snowden documents which includes Australian and possibly New Zealand intelligence material. John Key’s staff have, apparently, searched every piece of documentation they could think of to try and find a link between Key and Dotcom prior to the police raid and they’ve come up empty, which is why Key’s been so confident about dismissing Dotcom’s claims. But they can’t be sure Greenwald doesn’t have, say, a secret Australian intelligence report discussing Key being briefed on Dotcom.
That would be an actual genuine bombshell. But would it be a game-changer? Would a significant number of people change their vote because of it? I’m guessing no.
June 17, 2014
Updated the tracking poll to include the Herald digi-poll and last Roy Morgan poll, both of which are ‘bad for the left’. Bias corrected version below. Non bias corrected over the break.
What jumps out at me here is the comparison with the last election. Labour are trending down, just like last time – but now their votes are (mostly) going to National, not the Greens. Which makes sense to me: we have no idea what National plans to do in its third term, but that lack of vision is still preferable to being governed by a collection of left-wing parties who all hate each other but want to run the country together.
May 26, 2014
I’ve updated the tracking poll. The bias corrected poll is below. Non bias-corrected here.
- Things aren’t great for Labour. But they could be worse! According to John Armstrong, Shane Jones’ exit had Labour polling at ‘a mindblowing low of just 23 per cent.’ That was in National’s internal poll though. Weird how none of the public polls ever replicated that.
- Speaking of weird, there are going to be some interesting people a ways down the New Zealand First list who will probably be MPs in six months time.
- Not pictured but ACT is still on less than 1%, and they have a tough decision ahead of them: stick with running as a ‘classical liberal’ party or play the white supremacy card?
- As always, when the sun goes down and the night darkens all the rivers, cups the peaks and folds the final shore in, and nobody, nobody knows what’s going to happen to anybody besides the forlorn rags of growing old I think of David Shearer and look at Labour’s trajectory over this term. I think of David Shearer.
May 8, 2014
Via Andrea Vance’s bottom-feeding troll feed:
This attitude towards the Morgan poll seems to be a thing around Parliament rather than just Joyce spinning a bad result. Lots of journalists and staffers have told me that they don’t trust the Morgan poll, that it’s ‘always bouncing around’ etc. And the Morgan poll is always ‘bouncing around’, because the Morgan poll is roughly every two weeks, while Fairfax, say, only poll three or four times a year. Polls are noisy datasets and they jump around a bit but if you poll frequently you can generally see trends in the data while if you have very long gaps between your polls any changes you see might just be random noise.
David Winter and Gavin White both looked at how the polls performed when predicting the actual outcome of the 2011 election. I aggregated Gavin White’s data for the different parties and graphed them to show overall predictive accuracy for each pollster. (A zero would mean your poll result was totally accurate predictor – so the larger the bar the worse you did).
The Morgan poll is about middle of the pack. Their latest one might not be accurate though. It’s only one poll. After all, the gallery and political pundits all warned us that the departure of universally beloved blue-collar bloke Shane Jones meant the utter annihilation of the Labour Party, and this Morgan Poll shows them gaining tens of thousands of center voters, even before the Williamson resignation and Collins melt-down – so maybe it is a rogue. I guess the other outlets are waiting until after the budget to conduct their polls, so we might have to wait a while before we get additional data.
(NB: As Pete points out in the comments there’s an element of luck here. You might have a really accurate poll that happens to get one bad result just before the election and so you’d do poorly on this analysis.)
April 10, 2014
The Herald carries the details of the Greens attempt to form a more formal coalition with Labour:
Labour yesterday rebuffed a proposal by the Green Party to present both parties as a coalition Government in waiting during in the run-up to the September 20 election.
Labour leader David Cunliffe indicated that such a pre-election arrangement could have posed problems with post-election negotiations with other parties, such as New Zealand First.
Which might not seem like a big deal. But if you look at what’s happening in the polls over the last few months you see a pretty consistent pattern. Labour are losing votes to National and New Zealand First. But they aren’t losing any votes to the Greens.
Now there could be a bunch of reasons for that but my guess is that most Labour voters who are sympathetic to the Greens – and according to the Colmar Brunton poll that’s about 70% of Labour voters – don’t feel the need to switch because ever since the NZPower launch there’s been a kind-of-consensus that a vote for Labour is a vote for a Labour-Greens coalition. Labour’s announcement that this isn’t the case and that a vote for Labour could also be a vote for a Labour-New Zealand First coalition seems like a big risk. I can see why they took it: they want to win back those votes from National and think its going to be tough to do when they’re in a formal alliance with the Green Party. But I’d also note that Labour’s high-point during this electoral cycle came after the NZPower Labour-Greens co-announcement back in 2013, suggesting that center voters are less repulsed by the idea of a coalition than Labour’s caucus are.
March 31, 2014
Bias corrected aggregated poll of polls below. Non-bias corrected graph here.
Safe to say that Cunliffe isn’t working out as Labour leader. He’s losing voters to National and he’s also trending down in the preferred Prime Minister rating.
Not shown, but ACT are on 0.5%. Their new leader Jamie Whyte was on Q & A this morning debating climate change with Russel Norman. We’ve heard a lot (mostly from the New Zealand Herald) about how Whyte is an intellectual giant who will rebuild ACT and restore it to its future glory. Based on his performance during the live debate I predict that ACT under his leadership will not reach 1% and he will not be elected as an MP.
National has a tricky decision to make regarding the Conservative Party. The bias-corrected poll has them on 2.9%. That’s three or four MPs IF National throws them an electorate seat. But if they do that then they might lose some voters to the Conservatives and a whole lot of center-voters might panic and switch to Labour, New Zealand First or the Greens.
I wonder what David Shearer thinks when he looks at the gap between National and Labour since the election? At the time I thought ditching Shearer was the right thing to do, but its starting to look like it was a horrible, horrible mistake.
February 2, 2014
- National – 44.5 percent, down 1.9 percent
- Labour – 33.5 percent, up 1.3 percent
- Greens – 12.4 percent, up 2 percent
- NZ First – 5.7 percent, up 1.5 percent
- Conservative – 2.1 percent, down 0.7 percent
- Maori Party – 1 percent, down 0.2 percent
- Mana – 0.3 percent, down 1 percent
- ACT – 0 percent, down 0.8 percent
- United Future – 0 percent, down 0.1 percent
I know, I know – it’s only one poll. But it feels right, right? Sure, there might be a run of polls next week that completely contradict it but you can see the logic behind the movements: Key endorsed Winston Peters so a bunch of National voters switched their votes to New Zealand First. I can see that. So let’s assume the poll results are broadly accurate and the movements aren’t just random noise.
- It tells us that voters (or at least voters in this demographic) are listening and responding to signals from politicians. Key only made his announcement about Peters a couple of weeks ago.
- It tells us that Conservative voters and New Zealand First voters don’t seem to overlap as much as National seemed to think they did. Key and Joyce’s grand plan, remember, is for New Zealand First and The Conservatives to both get just under the 5% threshold, wasting their votes and thus putting National into power, somehow.
- No votes for the Internet Party. (Unless you think all those Kim Dotcom lovin’ young urban National voters aren’t being sampled because cellphones.)
- Voters aren’t responding to National’s ‘rock-star economy’ marketing campaign. Or maybe they are? Perhaps some voters feel more comfortable changing governments during a time of economic growth?