Party vote predictions

In response the the Newshub poll on Thursday Rob Jones (@cossackstomper) predicted (or stirred):

Greens will be 10 percent on Election day. Labour 28percent. Winston 10 percent

I think that trying to predict a specific party vote, especially over three months out from the election, is poke in the dark pointless.

Elections in the US and UK have shown how much actual vote counts can move from polls.

I think that there’s a wide range of possibilities for our election, probably within ranges like this:

  • National 40-48% (2008 – 44.93%, 2011 – 47.31%, 2014 – 47.04%
    Very unlikely they will get much if anything more than the last three elections. They could slump but I think something drastic would have to happen for them to drop below 40%.
  • Labour 20-35% (2008 – 33.99%, 2011 – 27.38%, 2014 – 25.13%)
    The biggest unknown – they could get a bit of a boost during the campaign if Andrew Little somehow finds a way to connect but more than 35 looks like miracle territory, but they could just as easily slump.
  • Greens 10-15% (2008 – 6.72% , 2011 – 11.06%, 2014 – 10.7%)
    The should be able to hold up their vote but will struggle to grow it much unless they benefit from a Labour collapse.
  • NZ First 5-15% (2008 – 4.07% , 2011 – 6.59% , 2014 – 8.66%)
    They look likely to improve on their last result if nothing major goes wrong with Peters and Shane Jones doesn’t turn potential voters off, and they also benefit from a lack of a credible alternative to National, but how much they grow will depend a lot on Labour’s gains or losses.
  • TOP 1-5%
    I think they are good for at least a percent or two but will struggle to get over 4%. It really depends on whether they can get some high profile candidates and a committed and capable leader that the media takes seriously.
  • Maori Party 1-4% (2008 – 2.39%, 2011 – 1.43%, 2014 – 1.32%)
    The trend doesn’t look good and if Labour succeeds in the Maori vote battle they may lose their one list MP, but they could benefit if Labour falls over and gain a seat or two.
  • ACT 0.5-2%  (2008 – 3.65%, 2011 – 1.07%, 2014 – 0.69)
    There support fell off through major upheavals. David Seymour has done a good job of steadying and re-establishing  so they have a good chance of improving, especially if National sheds support, but probably not by a lot. They should be very pleased if they manage to get another MP or two.
  • Mana Party 0.5-1% (2011 – 1.08%, 2014 – 1.42% with Internet Party)
    They should have some core support still but it’s hard to see much more than a fraction of a percent.
  • United Future 0.1-0.5% (2008 – 0.87%, 2011 – 0.6%, 2014 – 0.22)
    The trend looks likely to continue unless Damian Light manages to be taken seriously by media and the look like more than one person.
  • Conservative Party 0.1-1% (2011 – 2.65%, 2014 – 3.97)
    There will be some core votes but Colin Craig has ensured they won’t get a lot.

There are two key and related things – whether National can maintain their support or shed votes, and whether Labour recovers some support or slumps even further.

The small parties will as usual pick up the crumbs, although there could be some sizeable crumbs available.

UPDATE: Coincidentally this update today:


Election predictions

Polls are generally snapshots of opinion in the past, but they are often used to try to predict future election results. This can be interesting but accuracy is obviously not guaranteed.

An interesting approach here to New Zealand general election forecasts

This page provides experimental probabilistic predictions for the 2017 New Zealand General Election. It draws on multiple opinion polls, but goes a step beyond a straightforward poll aggregator in that the estimated voting intention from successive polls is used to forecast the chances of each party to actually win seats on election day, taking into account uncertainty. Polling results are also adjusted to take into account different polling firms’ past performance in predicting different parties’ results.

Every election can have it’s own unique pattern of lead-up polling so past trends won’t always predict future trends correctly, but this is as good as predictions can get.

It has some flaws but I expect some of these to be dealt with.

Oddly that doesn’t include a National coalition with NZ First.

What I like about this site is that it shows ranges of probabilities, showing that there is always margin for ‘error’ or inaccuracy.

This shows that a range of outcomes are possible for all parties. I presume it will be refined as we get closer to the election – it will be interesting to compare these predictions with later ones.

The poll results and trends show how volatile the polls have been for Labour, Greens and NZ First.

Perhaps ominously for National their results have been tightening in a downward trend since John Key stepped down, but it’s too soon for poll responses to the budget.

For the term to 2014 National had been trending down but that turned around, but well in advance of the election. They will be hoping for a late upswing this year but there is no sign of that yet.

NZ First have had late and significant upswings for previous elections after dipping mid term, but this term their poll support has held up. It’s difficult to predict whether this pattern will repeat this year or not.

Poll volatility

The latest Roy Morgan poll continues a ‘trend’ of volatility for National and Greens this year and a lack of progress for Labour.

Roy Morgan polls so far this year:

National 43.5 47.0 48.0 48.5 45.5 43.0 48.5 42.5
Labour 33.5 33.0 30.0 30.5 31.5 32.0 28.5 31.0
Greens 12.5 11.0 12.0 10.5 14.0 13.0 11.5 14.5
NZ First 4.0 4.5 5.5 4.5 3.5 5.5 5.5 6.0
Conservative 2.5 1.5 1.0 2.5 1.5 2.5 2.0 0.5
Internet Party 0.5 1.0 1.5
Maori Party 2.0 1.5 0.5 1.5 2.0 1.5 1.0 1.0
Mana Party ^ 1.0 1.0 0.5 ^ 0.5 1.0 1.0
ACT ^ ^ 1.0 1.0 0.5 0.5 0.5 0.5
United Future 0.5 ^ 0.5 0.5 0.5 0.5 0.0 0.5

Rounded to 0.5
^ is less than 0.5

Party summaries with predicted election support range based on current performance:


This is the lowest they have been this year but dropped below this six times last year, with a lowest of 40.5 (April) and a peak of 51 (July).

This volatility suggests a significant number of voters are soft supporters of National, sometimes. National have benefited from a variable and often disappointing performance by Labour but their fold on power looks precarious.

Predicted election range 40-50


Labour have bounced back from a worrying 28.5 in the last poll but at 31 are still failing to impress. Last year’s low was 29 (July) and they peaked at 37 for two consecutive polls in September when Cunliffe took over leadership. Since then they have dropped to mostly low thirties.

While National (and Greens) fluctuate Labour have been more consistent but this is not where they want to be in the polls, totally reliant on Greens,  probably also reliant on NZ First and possibly also needing any mix of of Mana, Internet Party and Maori party.

Predicted election range 25-35


Greens are back to their peak, last reached in November last year but they got to 14 in March. They should be happy with this and look well prepared for the election. Their low last year was 10 in July.

They are benefiting in polls from Labour’s weaknesses but they are relying on Labour to do well enough to give them their first shot at being in Government.

Predicted election range 10-15

NZ First

NZ First are looking good for making the threshold this election. Last year they ranged from 3 (several times) to 6.5 (August) but in the past have done better in elections than polls. They are benefiting from Labour’s weaknesses and National’s missteps.

Predicted election range 5-10

Internet Party

The Internet Party keep nudging up in the third poll and will be satisfied with this progress. The big questions are how much they will climb to and what effect any arrangement with Mana may have. It’s difficult to predict how much big money and a big presence of someone who can’t stand for election will end up impacting.

Predicted election range 0-5

Maori Party

Last year’s range was 0.5 (January) to 2.5 (February and March), this year they have been 0.5 to 2.

The Maori Party has never needed to worry about party vote in the past, they have always got more seats than their proportional allocation by winning electorates. This will be much harder for them this year with some predicting difficulty winning any of them. At least one electorate looks likely but this may bring their party vote into play for the first time. Will they seek party votes?

Predicted election range 1-3

Mana Party

Mana have mostly been 0.5 or 1 this year and last but got to 1.5 in July and dipped below 0.25 this March.

Hone Harawira may or may not be challenged strongly by Kelvin Davis in Te Tai Tokerau. If he retains his electorate they are in.  otherwise everything is up in the air depending on what arrangement they may come to with the Internet Party.

Predicted election range – depends on any arrangement they may make with the Internet Party.


ACT are virtually a new party this year. They have struggled between 0 and 1 last year and so far this year. They have intellectual rather than charismatic leadership so may have to rely on retaining Epsom with a new and relatively unknown candidate. They will have to find something different to impress beyond that.

Predicted election range 0.5-1


United Future have polled 0 to 0.5 last year and so far this year apart from one blip to 1 in March last year. Most of their election efforts will go in to retaining Ohariu which Peter Dunne might manage to pull off but he has had a difficult term. Party support doesn’t look like changing unless they can change dramatically and look like something other than Dunne.

Predicted election range 0.5-1

Conservative Party

Conservatives should be worried about the last poll dropping to 0.5 although they have been there before, in January and September last year. They have reached 2.5 several times including two polls ago.

Colin Craig’s dithering over what electorate to contest, his apparent concession that they need to get 5% and the lack of any connection with National makes their election looking likely to be another expensive folly. Money may pull something out of the hat but there’s no sign of magic yet.

Predicted election range 0.5-3

Source (PDF)

Election prediction 2014 #1

My current election predictions:

  • National 50-59
  • Labour 31-40
  • Greens 10-18
  • NZ First 0-7
  • Maori Party 1-3
  • Mana Party 1-2
  • United Future 0-1
  • Act Party 0-2
  • Conservative Party 0-1

This is based on what is know at the start of the year, with a long time until the election and much unknown.

If Labour and Greens move it’s likely to be in opposite directions as vote tends to move from one to the other.

NZ First is make or break on making the 5% threshold.

Act need to come up with a credible Epsom candidate to make 1 and a credible list to rise above that.

On current polling the Conservative Party don’t look like making 5% but there are indications and rumours National may help then get an electorate seat.

The Civilian Party is currently an unknown but could add some interest.

Unsuccessful predictions for the end of the world

Documented predictions for the end of the world go back as far as 634 BC. Since then there have been many obviously unsuccessful guesses/premonitions/attention seekings.

Wikipedia lists over two dozen (some listed as ‘various’) that predicted an end this millenium up until now, including Isaac Newton.

Apart from tha Mayan madness there were two other predictions for this year, four for last year etc etc.

I guess someone may eventually get it right (if they can see it coming).

The next one (until some other attention seeking nutter pops up) is:

2018–2028 F. Kenton Beshore Beshore bases his prediction on Hal Lindsey‘s failed 1988 prediction that Jesus would return within one biblical generation of the founding of Israel in 1948. Lindsey stated a biblical generation was 40 years. Beshore argues that Lindsey’s prediction was correct, but that his definition of a biblical generation was incorrect, which is actually 70–80 years, placing the Second Coming of Jesus between 2018 and 2028.

But why the hell would some interpreters of old Mayan writings be any more accurate? It simply makes no sense that unless you can see something like a large asteroid motoring towards you it’s something that isn’t able to be predicted.

The end is nigh megaphone

And more from

bullet Will the world end in the near future?
bullet The end of the Millennium: What people thought would happen: The Y2K Problems: Social, Economic, and Spiritual Aspects
bullet Past end-of-the-world scenarios which have failed
bullet Predictions that the end of the world will happen between now and 2010
bullet Predictions for 2012-DEC-21 or so
bullet Predictions for the 2010s
bullet Premillennialism and other competing Christian theories about the end of the world.

Successful predictions for the end of the world