Beware of Spin Doctors and Spin Professors

Written By: - Date published: 9:01 am, November 13th, 2016 - 97 comments
Categories: making shit up, Media, the praiseworthy and the pitiful, twitter, you couldn't make this shit up - Tags:

According to this article that appeared in the Herald this week Labour may as well give up now and not even bother contesting the 2017 election. National has won and John Key will be Prime Minister again so we have to suck it up and bear it.

The timing is interesting. On a day when the richest Western Democracy’s pollsters were shown to have no idea of what was happening this academic not only thinks that the next election is a foregone conclusion and it is too late for Labour to do anything. And her understanding of the MMP environment is poor in that she uses only the difference between National and Labour in the polls as the determining factor. The Green Party may as well not exist.

Interestingly the writer relies on historical polling data. This worked really well in the US of A where all of the pollsters got the result wrong. Nate Silver’s 538 was one of the best. But even it thought that the chances of a Clinton win were pretty likely.  It stands out because it did not think that a Clinton win was almost inevitable.

Here is 538’s graph from June 8 to two days ago.  This is the most pessimistic estimate of Clinton’s chances I can find.

538-forecasters

You have to question the quality of the analysis. It is not as if Spin Professor has no form. She has been anti Labour for as long as I can recall. It is worrying that a public servant such as a media personality like Mike Hosking is at least superficially biased. But an academic who is meant to weigh up the evidence dispassionately?

Some twitter traffic adds context …

Her tweet invited inevitable responses.

97 comments on “Beware of Spin Doctors and Spin Professors ”

  • Comments are now closed

  1. Bearded Git 1
    13 November 2016 at 9:18 am

    Real Clear Politics predicted Clinton 272-266 and was closer than 538. It only needed tiny shifts (0.6 % Clinton to Trump) in 3 states to upset the Real Clear prediction.

    • Lanthanide 1.1
      13 November 2016 at 10:23 am

      You misunderstand what the Real Clear Politics projection means.

      They gave Hillary a 100% chance of winning 272 electoral college votes, ie, 100% chance of winning the election.

      They were more wrong than 538, who gave her only a 71% chance of winning the election, and for the weeks prior, had repeatedly stressed that there was a real possibility Trump would win the EC despite Hillary winning the popular vote, because she was not very strong in the swing state polling.

      • Bearded Git 1.1.1
        13 November 2016 at 10:54 am

        @Lanthanide

        Disagree Lanth….you may be talking in pure statistical terms but anyone who understands the state by state US system would have looked at the real clear numbers and said this could go either way.

        When you see 70% Clinton on 538 that, to me, says she is very likely to win.

        • Lanthanide 1.1.1.1
          13 November 2016 at 11:12 am

          Like I said, 538 for weeks ahead, took pains to repeatedly say, 70% is not “a sure thing” and that Trump still had a very good chance of winning, and the reason is because Hillary’s showing in the swing state polls was not good.

          Also, you’re overstating what RCP was really showing. Their methodology is to take the raw poll numbers and simply average them, whereas 538’s model is much more complex. What happened on Monday the day before the election, is that a poll came out giving Trump +4 in Florida (he won by 1.3%) which nudged RCP’s model to give Florida to Trump, and hence Hillary 272 votes.

          Up until Monday though, RCP had Hillary winning Florida (and Trump New Hampshire) and her ending up on 297 electoral college votes.

          538 on the other hand, with their more complex model, re-scored that +4 Trump result as a +1 Trump result, and since the actual result ended up being 1.3 for Trump, 538’s model for that particular poll was better than RCP’s approach.

          If that poll had reported a number of +2.7 instead of +4 for Trump, RCP would have called Florida for Hillary and projected that she would have won 301 EC votes instead of the 272 they ended up predicting.

          So it’s only a stroke of luck (and bad polling data showing Trump at +4) that ended up with RCP giving Hillary 272 votes and not 301.

          538 give a probabilistic measure, and said Hillary had a 55% chance of winning Florida: http://projects.fivethirtyeight.com/2016-election-forecast/florida/

          Here’s the RCP polling for Florida, you can see it JUST squeeks ahead for Trump on the day before the election, resulting in the 29 point swing towards Trump.
          http://www.realclearpolitics.com/epolls/2016/president/fl/florida_trump_vs_clinton_vs_johnson_vs_stein-5963.html

          • Bearded Git 1.1.1.1.1
            13 November 2016 at 1:31 pm

            @Lanth mmm I stand corrected-you are right under that analysis.

            I think my problem is I don’t like the 70% thing and prefer the 272-266 thing.

            • Lanthanide 1.1.1.1.1.1
              13 November 2016 at 1:38 pm

              Oh yeah, I can completely understand that. It’s much harder to get your head around a probabilistic projection for a single binary event, and what it actually means in practice.

              It’s something that 538 have struggled with a lot, especially this time (but they’ve also proven many of their naysayers wrong).

              I think the takeaway point is that all of the aggregators were relying on the same underlying information, it’s just what methods they then applied to that information – and particularly in the case of RCP how they presented their results – that gave the difference in their projections.

              RCP had a very simply methodology, and a very simple way of presenting their results, that incorrectly gives the impression that they were more accurate than 538.

              Everyone got it wrong, but 538 was least wrong, and their commentary for the weeks ahead warned of this exact outcome.

  2. xanthe 2
    13 November 2016 at 9:20 am

    My gut feeling is that “polls” at this time are not intended to reflect public opinion but are rather tools to lead and manipulate opinion.

    i dont claim to have “evidence” of this thats just what i think is happening.

    anyone care to explain to me evidentially that polls are actually objective reflections of reality rather than a campaigning tool?

    • pat 2.1
      13 November 2016 at 9:38 am

      an observation I’m inclined to agree with.

    • Draco T Bastard 2.2
      13 November 2016 at 10:19 am

      Can’t do that but there’s research showing that polls do influence election outcomes:

      During elections, and major public policy events, much of the media coverage focuses on the “horse race,” or fluctuations in support for a candidate or policy. Reporting on public opinion not only affects support, but levels of engagement: donations, volunteering and turnout. These bandwagon effects can make polls self-fulfilling prophecies; the predictions of the polls come to pass because the polls not only measure public opinion but also influence public opinion and engagement.

      We need to ban the publishing of polls.

      • Colonial Viper 2.2.1
        13 November 2016 at 10:31 am

        A 7 day ban on polls before election day. It’s a very good idea.

        • Craig H 2.2.1.1
          13 November 2016 at 10:54 am

          More than 7 to avoid influencing advance voting, but I agree with the principle.

          • Colonial Viper 2.2.1.1.1
            13 November 2016 at 11:09 am

            I picked 7 days to balance it out with citizens right to free speech, particularly as you can have political advertising right up to election day (all of which influences advance voting).

            But yes, another time period could be chosen.

        • Draco T Bastard 2.2.1.2
          13 November 2016 at 11:01 am

          No, a ban on publishing the results needs to happen but not a ban on the polls themselves. After all, we do need our researchers to be able to do research. That’ll leave it so that political parties can do their own polling as well but it’s unlikely that newspapers will do them as they’ll no longer have then as clickbait for advertisers which means to say that the influence of the polling should drop and it’s the influence that we want to stop.

          And that’s a permanent ban that I’m talking about there.

          • Colonial Viper 2.2.1.2.1
            13 November 2016 at 11:08 am

            Yes a short term ban on public polls/publishing polls. Not a ban on internal polling.

            • Incognito 2.2.1.2.1.1
              13 November 2016 at 9:35 pm

              I do have huge reservations about political polls and the ways they are used and abused – there are four kinds of lies: lies, damned lies, statistics, and political polls. However, I don’t think that banning polling or publishing polls is the way to go for a number of reasons.

              First, there is no way to avoid accidental or deliberate ‘leaking’ of internal polls into the public domain.

              Second, how could a ban be enforced? Digi polls in MSM and also those ‘surveys’ on social media would need to be properly included.

              Third, the public is always hungry for information such as polls, especially during election time.

              Fourth, polls and surveys cross over into other areas such as policy setting and marketing.

              There must be better ways to deal with poll results. For example, better explanation of the limitations and meaning of these snapshots and the inherent problems with extrapolation and future predictions. Unfortunately, too many parties have ‘skin in the game’, e.g. political parties, polling companies, MSM.

        • Whispering Kate 2.2.1.3
          13 November 2016 at 11:12 am

          Pollsters are like Real Estate Agents, always in your face and persuading like there is no tomorrow. The flag referendum was another irritating in your face encounter. The flags had been displayed – what more needed to be done, visually it was enough for us to make up our own minds without celebs pushing their choices on to us and one particular pollie pushing his agenda Why can’t people be left to make up their own minds without this constant intrusion into their lives. Politicians give us enough information with policy announcements, their stupid and offensive behaviour and daily chit chats with the media to give most reasonably intelligent citizens enough knowledge to judge pollies and policies on their merits.

          I am with Draco here – they should be banned and Real Estate Agents should be banned from our telephones – good god if one wants to sell their home they don’t need constant stalking to do so. It is manipulation in its purest form and subtle – thank the lord there are some of us that can see through it all and tell them to bugger off.

          • Colonial Viper 2.2.1.3.1
            13 November 2016 at 11:19 am

            As I mentioned, I do think the publishing of political polls for 7 days before election day is a good idea. I wouldn’t go too much longer than 7 days for freedom of speech and informed electorate reasons.

            • Chuck 2.2.1.3.1.1
              13 November 2016 at 11:32 am

              “As I mentioned, I do think the publishing of political polls for 7 days before election day is a good idea.”

              Agree, that makes sense to me.

              Another unintended consequence in banning polls for public release would be a select few would have the inside running…while the public are kept in the dark (pretty much what CV also said).

              • Draco T Bastard …
                13 November 2016 at 12:14 pm

                Another unintended consequence in banning polls for public release would be a select few would have the inside running

                The inside running on what?

                • Chuck …
                  13 November 2016 at 12:34 pm

                  On pretty much everything.

                  The smaller parties will be all trying to say they are at or over the 5% threshold, in the hope they attract more voters…the larger parties will say they have private polls showing massive support to form the next government.

                  All this will go unchecked if the public are not allowed to see polling, that at least keeps most political parties from making outrageous statements on their true support.

                  The idea of banning public polls would hurt the left more than the right in NZ. As the left is cluttered with players…there is more chance of wasting votes on the left than the right.

                  • Draco T Bastard …
                    13 November 2016 at 12:53 pm

                    The smaller parties will be all trying to say they are at or over the 5% threshold, in the hope they attract more voters…the larger parties will say they have private polls showing massive support to form the next government.

                    They’re banned from doing so.

                    The idea of banning public polls would hurt the left more than the right in NZ. As the left is cluttered with players…there is more chance of wasting votes on the left than the right.

                    The problem there isn’t the existence or non-existence of publicly published polling but that we have a system that actually wastes votes.

                    • Lanthanide …
                      13 November 2016 at 12:57 pm

                      “They’re banned from doing so.”

                      Good luck with that.

                      “The problem there isn’t the existence or non-existence of publicly published polling but that we have a system that actually wastes votes.”

                      Unless you’re proposing a parliament of 1000 seats, then there will always be wasted votes. The alternative to award any party that gets 0.2% of the vote with a seat in Parliament, in which case you’re favouring “not wasting votes” over fair proportional representation.

                    • Draco T Bastard …
                      13 November 2016 at 2:25 pm

                      Unless you’re proposing a parliament of 1000 seats, then there will always be wasted votes.

                      Nope.

                      The alternative to award any party that gets 0.2% of the vote with a seat in Parliament, in which case you’re favouring “not wasting votes” over fair proportional representation.

                      Wrong.

                      The correct answer is: 0.8% of votes for a seat and get rid of electorates. Threshold set at 0.8%

                      That should minimise wasted votes.

                    • Chuck …
                      13 November 2016 at 5:57 pm

                      “They’re banned from doing so.”

                      I remember Colin Craig (2011 election) using a second tier polling company that did a private poll for Craig saying he was on track to win the Rodney seat. Craig used that poll to claim he was on track…

                      The public polls had Craig no-where.

                      Craig ended up with 2.65% of the party vote, and some 12,000 votes behind the National MP for the Rodney seat.

                      Imagine leaving political parties to themselves for private polls, without some kind of check the public currently have. Which is being able to see public polls.

                    • Incognito …
                      13 November 2016 at 9:54 pm

                      The Netherlands, for example, has got no legal election threshold, which means, in practice, that 0.67% of the vote is enough to get a seat in the Lower House (150 seats).

                      https://en.wikipedia.org/wiki/Election_threshold

                    • Draco T Bastard …
                      13 November 2016 at 11:36 pm

                      I remember Colin Craig (2011 election) using a second tier polling company that did a private poll for Craig saying he was on track to win the Rodney seat. Craig used that poll to claim he was on track…

                      Which bit in not being allowed to publish didn’t you understand?

                      Imagine leaving political parties to themselves for private polls, without some kind of check the public currently have. Which is being able to see public polls.

                      There aren’t any public polls. They’re all private. Some are published in the MSM. We generally don’t hear about the internal ones that political parties do.

                      I’m saying that we shouldn’t hear about any of them as it influences people.

            • Draco T Bastard 2.2.1.3.1.2
              13 November 2016 at 12:49 pm

              Polls don’t inform – they influence and it’s that that requires it to be a permanent ban.

              People need to be left to make up their own minds on the information that they have.

              • Colonial Viper …
                13 November 2016 at 1:03 pm

                Sorry, I think that position goes far too far against our democratic right to free speech.

                A poll free period 3 days, 7 days or maybe even 14 days before election day would probably be fairly balanced against that right.

                • Pasupial …
                  13 November 2016 at 1:22 pm

                  3days would be pointless, even 14 days seems a bit short. A month from polling day might be better, considering how much early voting there was in the last election. An advantage would be that this would force the media to focus on policy rather than the horse-race.

                  • Whispering Kate …
                    13 November 2016 at 1:50 pm

                    I agree with Draco – polls are manipulation of the minds of voters – people are not stupid and in fact I think the US punished the pollsters by just telling lies about who they were voting for. Nothing gets up the noses of people more than manipulators trying to meddle with our choices. There is enough in the media and on blog sites for us to find information and make our own decisions.

                    Dirty Politics is enough to contend with and its starting again already and giving our so called democratic system a very bad smell – just leave us all in peace to make our own decisions please.

                    • Pasupial …
                      13 November 2016 at 2:15 pm

                      While polls can certainly be used to manipulate through selective reporting, they are not bad in themselves. Your contention that; “the US punished the pollsters by just telling lies about who they were voting for”, is undercut by the fact that the polls were actually correct within their Margin of Error.

                      The problem I see with not ever reporting polls, is that they are still going to be conducted. So if they are entirely banned we would end up not knowing the basis on which the parties are developing policy. However in the immediate leadup to an election this is no longer so much of a factor.

                    • Draco T Bastard …
                      13 November 2016 at 2:18 pm

                      So if they are entirely banned we would end up not knowing the basis on which the parties are developing policy.

                      Shouldn’t the political parties be developing policies on what their members want?

                    • Pasupial …
                      13 November 2016 at 2:23 pm

                      DtB
                      Well, yes. But do you honestly believe; “political parties be developing policies on what their members want”. Maybe the Greens, but even their participatory democracy in creating policy is very much guided by the exec these days.

                • Draco T Bastard …
                  13 November 2016 at 2:16 pm

                  Sorry, I think that position goes far too far against our democratic right to free speech.

                  What you’re arguing for goes against peoples right not to be affected by others without their permission.

                  And now you’re going to say that people don’t have to read the polls but that’s a load of bollocks. I don’t read the polls but I get to hear about and be influenced by them anyway.

                  • Colonial Viper …
                    13 November 2016 at 5:33 pm

                    What you’re arguing for goes against peoples right not to be affected by others without their permission.

                    There’s no such right. It doesn’t exist.

                    • Draco T Bastard …
                      13 November 2016 at 11:41 pm

                      Are you sure about that?

                      It may not be written down but I’m sure that if I tried to build a skyscraper in the local street (ignoring local zoning) there’d be a lot of argument against it including the effect that I’d have on other people and they’d probably win in court. There’s been several wind farms that have been stopped by such arguments.

                      A lot of our laws are pretty much based on such a principle. We don’t have laws against speeding because you may kill yourself but because you may kill someone else.

            • Incognito 2.2.1.3.1.3
              13 November 2016 at 9:44 pm

              Could you elaborate on what you mean by “freedom of speech” and “informed electorate reasons”?

        • james 2.2.1.4
          13 November 2016 at 1:04 pm

          A 7 day ban on the FBI announcing investigations into any of the presidential contenders would prob make more of a difference.

        • Tiger Mountain 2.2.1.5
          14 November 2016 at 8:38 am

          and an even longer ban on partisan public servants stirring up the mud just prior to an election

      • KJT 2.2.2
        14 November 2016 at 12:30 pm

        How about simply a requirement for politicians to be truthful. Or be prosecuted for fraud.

        • KJT 2.2.2.1
          14 November 2016 at 12:43 pm

          And show a cost/benefit analysis for the policies.

          Like the one Wayne is unable or unwilling to provide for the China FTA he repeatedly touts.

    • mikesh 2.3
      13 November 2016 at 10:24 am

      This would assume that polls were being doctored in some way, but this seems unlikely. It is more likely that there is something wrong with methodology; perhaps people answer polls idealistically but vote pragmatically, or not at all.

      • Colonial Viper 2.3.1
        13 November 2016 at 10:30 am

        Firstly, Clinton supporters demonized Trump supporters to the extent that they stopped telling people who they actually supported.

        Secondly it was more than the polls. It was the entire corporate MSM who had decided that Trump was not going to win.

        MSNBC’s Morning Joe gets it:
        – the MSM decided that their Clinton supporting friends in Manhattan was all they needed to know about America.
        – MSM certainty about a Clinton win, variously put at 90%, 92%, 93% in the last month of the election, damaged Hillary’s turnout.
        – The New York Times totally sold out to become Clinton cheerleaders.

        • Lanthanide 2.3.1.1
          13 November 2016 at 11:17 am

          I find this article far more revealing about what was really going on:
          http://www.bloomberg.com/news/articles/2016-11-10/trump-s-data-team-saw-a-different-america-and-they-were-right

          Trump’s campaign only gave him a 30% chance to win. They think the Comey letter made a big difference, even while 538 and most MSM reporters eventually said they didn’t think it would matter because everyone had made up their mind.

          Nate’s pointed out that in the swing states that Clinton lost, Trump appears to have won late deciders at a 3/2 ratio, and given the late deciders made up as much as 12% of voters this time, they by themselves were enough to put Trump ahead by 2-3%.

          • Colonial Viper 2.3.1.1.1
            13 November 2016 at 11:28 am

            Thanks for that Lanth. I saw that article a bit before, and it’s a good one.

            An interesting tidbit – I understand that the Clinton campaign hired the same top data and targetting team that Obama used. (We all heard how good her “ground game” was and that it was inherited from Obama).

            And her HQ thought they were well on the way to winning until 9pm or even later, especially based on good exit polling that they saw on the day.

            One other detail. I just heard Michael Moore say that Hillary Clinton lost Michigan by just 11,000 votes. (I think the current figure is around 13,000 votes).

            Then he said this: 90,000 Michiganers completed both sides of their ballot papers BUT LEFT THE PRESIDENTIAL VOTE INCOMPLETE.

            They knew that they could not support a guy like Donald Trump. But neither would they support Hillary Clinton.

            Wow.

            • Lanthanide 2.3.1.1.1.1
              13 November 2016 at 1:05 pm

              Yip.

              I think all of this stuff about blaming the media in their little bubbles in cities is trite, and I think for a fair number of people saying it (now) it isn’t even their own idea, they’re just jumping on the bandwagon to try and ‘explain’ the election. Texas went MORE democratic in this election than it has been in a long time, and that was driven by votes in the cities. Nate is suggesting that by 2024 or 2028, Texas could be heading towards swing state status due to demographic change – that’s gotta scare the Republicans (and of course they should adapt their policies to suit).

              The only people worth believing who are explaining the election, are the people who said these types of things before the election. It’s very easy to come up with justifications in hindsight and say “see, told ya so”.

              You CV were among the people picking Trump would win due to dissatisfaction with Clinton, but I think the point that it seems to be the Comey letter that made a big impact, wasn’t even on your radar either.

              • Colonial Viper …
                13 November 2016 at 5:32 pm

                You CV were among the people picking Trump would win due to dissatisfaction with Clinton, but I think the point that it seems to be the Comey letter that made a big impact, wasn’t even on your radar either.

                It may have had an impact and I am certainly not ruling it out. More research will need to be done.

                However Trump also had many unexpected blows against his campaign. So if we are going to note ‘black swans’ affecting their campaigns both candidates had their share.

                • Lanthanide …
                  13 November 2016 at 7:33 pm

                  If you say the access hollywood tape was a “black swan” for Trump, it wasn’t really, because that’s the sort of jackass he actually is, so it’s not surprising something would be found that revealed his true thoughts since he’d been in the media for so long.

                  The Comey email, on the other hand, was someone at the FBI acting in a way that all their superiors advised against, and was against the department’s own procedures, and the letter didn’t actually contain any content. That was truly a black swan, because the whole investigation was supposed to have been closed months prior.

                  • Colonial Viper …
                    13 November 2016 at 7:52 pm

                    It will be very interesting to find out, probably in many years time, what the hell has been happening in the FBI and the DOJ over this last year.

        • Robertina 2.3.1.2
          13 November 2016 at 11:28 am

          The bleating about the MSM just doesn’t work when you consider the Trump coverage of, say, Mother Jones, the news site whose expose of private prisons brought about a federal-level ban.
          Oh, but then strong and independent journalism isn’t going to sit too well with the corporate shills propping up the transition team (plus Mother Jones has reported that private prison stocks shot up after Trump victory):

          ”The behind-the-scenes transition operation is being run by Ron Nichol, a senior partner at The Boston Group, a management consulting firm where 2012 Republican presidential nominee Mitt Romney launched his business career.

          Ken Blackwell…senior fellow at the Family Research Council…Veteran agribusiness lobbyist Michael Torrey…Energy industry lobbyist Mike McKenna…David Bernhardt…represents mining companies seeking to use resources on federal lands…Lobbyist Steven Hart, who focuses on tax and employee benefits, is leading the transition team for the Labor Department.

          Cindy Hayden…top lobbyist for Altria, the parent company of cigarette-maker Philip Morris…Homeland Security Department. Jeff Eisenach, a consultant and former lobbyist…Federal Communications Commission….Michael Korbey…former lobbyist who led President George W. Bush’s effort to privatize America’s retirement system….Shirley Ybarra…champion of “public-private partnerships” to build toll roads and bridges….Myron Ebell…man-made global warming is a hoax…David Malpass…Bear Stearns’ chief economist…Dan DiMicco…former chief executive of steel company NUCOR and a board member at Duke Energy…Former Rep. Mike Rogers…serves on boards for consulting firms IronNet Cybersecurity and Next Century Corp.

          Kevin O’Connor…partner at the law firm of close Trump adviser Rudy Giuliani…Jim Carafano…Heritage Foundation’s vice president for Foreign and Defense Policy Studies…retired Lt. Gen. Keith Kellogg…chief operating officer for Coalition Provisional Authority in Iraq…Mira Ricardel…vice president of business development for Boeing Strategic Missile & Defense Systems.

          Buckle up. This is going to be a rough ride.”
          http://www.motherjones.com/kevin-drum/2016/11/daily-trump-shitshow-about-begin

          http://bigstory.ap.org/7f2605f079334fddb0dfb341010b68ea
          http://www.motherjones.com/politics/2016/11/private-prison-stocks-are-soaring

  3. dv 3
    13 November 2016 at 9:22 am

    Margin of error??????

    • Pasupial 3.1
      13 November 2016 at 1:14 pm

      That’s the way I see it. It’s not that the polls were any more statistically inaccurate than usual, it’s that they were reported as if they were certainties. It’s not in this article, but just before the election I recall Silver mentioning a caveat about the polls because of “herding” that he’d seen in past elections (that polls that were seen to be rogue would go unpublished so possibly biasing the results).

      people mistake having a large volume of polling data for eliminating uncertainty. It doesn’t work that way. Yes, having more polls helps to a degree, by reducing sampling error and by providing for a mix of reasonable methodologies. Therefore, it’s better to be ahead in two polls than ahead in one poll, and in 10 polls than in two polls. Before long, however, you start to encounter diminishing returns…

      In fact, the error in national polls wasn’t any worse than usual. Clinton was ahead by 3 to 4 percentage points in the final national polls. She already leads in the popular vote, and that lead will expand as mail ballots are counted from California and Washington, probably until she leads in the popular vote by 1 to 2 percentage points overall. That will mean only about a 2-point miss for the national polls. They may easily wind up being more accurate than in 2012, when they missed by 2.7 percentage points…

      It’s one thing to criticize pollsters — or polling-based forecasts — if your personal prediction came closer to getting the outcome right. But I’d assert that most mainstream journalists would have given Trump much lower odds than the 30 percent chance that FiveThirtyEight gave him, and that most campaign coverage was premised on the idea that Clinton was all but certain to become the next president.

      http://fivethirtyeight.com/features/why-fivethirtyeight-gave-trump-a-better-chance-than-almost-anyone-else/

  4. save nz 4
    13 November 2016 at 10:31 am

    In my view anything could happen in the election. Natz have nothing! Nothing! But not sure Labour and Greens have put together a clear vision yet. Some sort of technical point to point complicated policy vision is probably not going to get people out of bed on election day.

    So it is up to Labour and Greens to articulate they can do a better job for NZ against the inevitable smears and media beats ups that have unfortunately become part of NZ political landscape under the Natz.

    John Key wil have probably lost the US interest for a start as they have plenty of their own problems now. Neoliberalism has been dealt a well deserved blow by public vote and Key as a big champion in the IDU has probably lost a lot of international mana. Is his neoliberal ideology working, nope the wheels are coming off.

    • Bearded Git 4.1
      13 November 2016 at 11:03 am

      @savenz agreed….surely the electorate are getting tired of Key/Bennett/McCully/English/Smith/Parata/Brownlee/Collins/Bridges and their policies.

      Collins was terrible on RadioNZ thsi morning re prison policy on rehabilitation.
      Earlier this week they have wrecked the RMA (and the environment) to help their developer friends in the face of criticism from the Law Society, Geoffrey Palmer and everyone else except fascist ACT and the quisling MP (may they rot in hell).

      30+12+8=Key playing a lot of golf in Hawaii (or 32+11+9 or 33+11+8…)

    • Draco T Bastard 4.2
      13 November 2016 at 11:06 am

      But not sure Labour and Greens have put together a clear vision yet.

      Yep, Labour and the Greens need to get together and paint a vision for the future that’s better for everyone and not more of the same failed neo-liberalism.

      • Gabby 4.2.1
        13 November 2016 at 1:18 pm

        They really need to avoid ‘ohforfucksake’ policies.

        • Draco T Bastard 4.2.1.1
          13 November 2016 at 2:28 pm

          ???

          • save nz 4.2.1.1.1
            13 November 2016 at 3:17 pm

            It’s Labour and Greens election to lose. All I can keep pointing out that the biggest demographic group is the Pakeha homeowners. In my view they feel neglected or targeted for extra taxes by Labour and Greens. If Labour and Greens decide early on not to increase taxes to resident Kiwis that would be a relief to that group, because interest rates and cost of living seem to be on the rise – obviously wages are stagnant!

            The main feeling I think people are feeling is that they want a fair go. Things are not fair in NZ anymore. Also inclusion – stronger together – including a stronger together political force of Labour and Greens to be strong and fair.

            People have to believe it though, not all rowing the boat apart or some complicated plan like the original TPPA 5 points. Just channel Trump like Natz channel Trump. (Tax cuts, smaller government.) (Fair go for all, The real clean Green pure NZ).

            One other thing they could do that would be exciting is if they have a referendum on the UBI – Gareth Morgan has a plan to introduce a UBI and a flat tax which means that you don’t pay any tax on first 40k. I haven’t looked at it, but the idea of a real safety net like Superannuation for all with the UBI could be a ‘big ticket’ idea for Labour like social welfare was when labour first bought it in, while appealing to low wage workers.

            A referendum would mean that voters could choose it or vote it out.

            Maybe rather than Gareth having his own party someone could look and see if he could come into Labour? Just a thought.

            • Chuck 4.2.1.1.1.1
              13 November 2016 at 3:43 pm

              “It’s Labour and Greens election to lose.”

              You are forgetting about NZF save nz, at least Bearded Git includes NZF in his assumptions.

              There is no creditable pathway for a Lab/Green combo to govern without the support of NZF.

              And to rely on Winston Peters is not a good position to be in.

              • save nz …
                13 November 2016 at 5:35 pm

                I think NZ First will go with Labour and Greens. It maybe some complicated independent arrangement but I don’t think they like the Natz. All NZ First policy is at logger heads with the Natz ideology and policy.

                If Labour and Greens can get some thing exciting on policy like a referendum on a UBI then they might get the missing million to vote and therefore not need NZ First.

                I think the best ideas are a combination of all three parties anyway so I am pro a 3 party government with Labour, Green and NZ First.

                • Colonial Viper …
                  13 November 2016 at 6:12 pm

                  NZ First will go with Labour and Greens, if Labour and Greens perform strongly enough. ~41%, 42%, 43%, or so.

              • Matthew Whitehead …
                15 November 2016 at 1:47 am

                What you mean is that current polling suggests NZ First is a required element to a winning coalition.

                It requires a 3% or so shift towards Labour and the Greens to open up a second option where they can flex between the small parties and NZ First.

                It requires an even smaller shift to National.

                It doesn’t matter where those votes come from, (ie. whether it’s new voters, or stealing between the left and right, or between one of the wings and the centre, ie. NZF) but it’s still absolutely within the bounds of probability (hell, it’s within the margin of error) that New Zealand First won’t end up being the kingmakers.

                Remember, with Trump the polls were out by 5%, and their polling in the USA is much more reliable. The polling is better than nothing, and normally reliable, but it’s not always going to pick the winner, especially if there’s a significant lag like there is with NZ polling.

            • infused 4.2.1.1.1.2
              13 November 2016 at 9:34 pm

              It’s also businesses. A repeal of 90 days is going to win 0 votes. Plus the other taxes they wish to impose.

              • Matthew Whitehead …
                15 November 2016 at 1:52 am

                Labour doesn’t necessarily have to steal any voters off National. They just have to get people who didn’t vote in the last three elections to vote for them.

                It’s entirely possible some of those people WOULD be motivated by a speedy repeal of the 90 day trial law- the issue is whether Labour and the Greens look like they have big ideas as well as the important tinkering details, and are unified and ready to govern.

          • Gabby 4.2.1.1.2
            13 November 2016 at 4:24 pm

            Like promising to raise the age of superannuation, thereby kicking manual workers in the nuts.

            • KJT 4.2.1.1.2.1
              14 November 2016 at 12:38 pm

              Exactly.
              CGT was actually popular amongst those who used to vote, Labour.

              Raising the super age was not. For obvious reasons. All those that support raising the super age are desk workers or finance industry shills, who gain from privatising super.

  5. Guerilla Surgeon 5
    13 November 2016 at 10:42 am

    “Natz have nothing! Nothing! But not sure Labour and Greens have put together a clear vision yet.”

    Not sure? You sound like someone who couldn’t make up their mind between Trump and Clinton. Of course they haven’t put together a clear vision. If they had they’d be ahead in the polls.

  6. rhinocrates 6
    13 November 2016 at 10:48 am

    Had the misfortune of working under Robinson a while back. Fucking moron.

    We were in a staff meeting and a memo from Student Health was read out, warning that disproportionate numbers from our college were showing up complaining of stress, anxiety, depression.

    Her response was to shrug and say “Not our problem.”

    A while after, one I know of committed suicide. I was forced by circumstance to counsel one who’d attempted twice already due to their stress over their work (they’re OK today, thankfully).

    To put it mildly, she is not someone who understands people very well. As an academic, her methodology is shit. Her institution has long had a problem getting rid of its dead wood and lawyer magnets.

  7. Keith 7
    13 November 2016 at 10:57 am

    Twyfords tweet rebuke of the media last week has seen a seething bitchy article from Stacey Kirk today. Twyord “cant be trusted”, “undemines” re Shearer and “has designs on the leadership”. Dun dun dun…… A sad sad bit of story writing for John and the team.

    The political media hate the left, so why no give up on them and turn against them. Trump proved those dickheads were pointless Be the underdog, leave the media establishment to continue jerking off National!

    • Lanthanide 7.1
      13 November 2016 at 11:19 am

      I replied to that article saying that it was misleading in the same way Andrea Vance’s report was. My comment wasn’t published, even though I didn’t really attack the author.

      I think the moderation at stuff is quite uneven, and has gotten worse in recent months. It seems like if you want to quibble with anything written in the article (and readers around here will know I do that a lot), your comment won’t be let through moderation.

      • Wensleydale 7.1.1
        13 November 2016 at 11:53 am

        Must not derail prevailing narrative! Must not derail prevailing narrative!

        I’m sure they have a siren and a flashing red light that goes off in the media room when some pertinent comment starts to get traction.

    • Gangnam Style 7.2
      13 November 2016 at 1:32 pm

      “Be the underdog” ! Yes!

    • Bearded Git 7.3
      13 November 2016 at 1:34 pm

      @ Keith agreed, as i said last week Standardistas should spend more time on policies and less time worrying and whinging about what the media says.

    • Matthew Whitehead 7.4
      15 November 2016 at 1:56 am

      Yeah, I think there’s actually a lot to be gained by going on the attack on the media. I’m just not sure Labour have the nous or instinct to pull it off properly.

  8. RedBaronCV 8
    13 November 2016 at 11:50 am

    Don’t you get the feeling that the MSM has been suckered by the NACT’s on this one and are just doing their dirty work for them as usual.

    Labour released a policy that targets a 17% ? unemployment rate among younger people. The costing may or may not be +/- $25m.

    Surely the main talking points Labour should stress are:
    Why is this rate so large and what has NACT done policy wise about this large group of young people except nothing. Why is NACT failing there better future and why do you (the MSM) not put the heat on NACT about that?

    Why are you (the MSM) obsessed with a possible $25m (about as much as the flag referendum or 2 Saudi sheep farms. Do you want a high unemployment rate for those kids?

    You know that meme of the Nacts – ” all labour’s fault”
    well the left could prefix everything from hereonin with “NAct are leaving us with a lot of expensive problems to deal with”

  9. The lost sheep 9
    13 November 2016 at 1:23 pm

    It is not as if Spin Professor has no form. She has been anti Labour for as long as I can recall.

    Even when she produced a report suggesting the Media had a structural bias towards National leading up to the 2014 Election?
    Can’t remember anyone here dismissing that on the basis she was anti-Labour….

  10. Andrew Geddis 10
    13 November 2016 at 4:55 pm

    We’ve been here before. In 2011, Claire made the same prediction of a National win more than a year out from the election. That caused me (and others) to sputter that she couldn’t possibly know that and that her assumptions were all wrong (see here: http://www.pundit.co.nz/content/it-was-different-then-and-thats-all-in-the-past).

    Then, when it turned out her prediction was in fact accurate, I had to follow up with a mouthful of crow (see here: http://pundit.co.nz/content/a-healthy-dose-of-humble-pie).

    So, y’know, is she being “anti-Labour”, or is she as an academic showing some impressive application of research-led theory to the real world?

    • Chuck 10.1
      13 November 2016 at 5:23 pm

      If her model showed a potential Labour win, I am sure it would of been greeted very warmly here.

      Being an academic Claire knows her work will be reviewed (as you did) and no academic would risk basing her/his work on dodgy assumptions (or personal preferences that do not align with the results). Its not good for future work prospects.

      • Stuart Munro 10.1.1
        13 November 2016 at 5:56 pm

        And yet history is full of examples of academics who have done exactly that.

      • Incognito 10.1.2
        13 November 2016 at 10:03 pm

        I believe the learned Professor Robinson has a tenured position so she could argue that the Earth is flat and still keep her cosy job.

        • rhinocrates 10.1.2.1
          13 November 2016 at 10:30 pm

          Talking to HR in her institution, you’re likely to be sprayed with spittle when you enquire about several senior academic staff at her college (or so my union rep told me). Lots of money has been spent on disciplinary action, lawyers and settlements for that place. They’re only kept on because the one-off costs of getting rid of them are unacceptable (even though the long term costs are higher).

          • Incognito 10.1.2.1.1
            13 November 2016 at 10:51 pm

            Doesn’t surprise me; dealing with HR usually is a Kafkaesk experience.

            What exactly do you mean by “her institution”; can you be more specific?

            There’s the dollar cost of keeping certain people employed, or getting rid of them to make way for fresh blood, and then there is the collateral damage that this causes.

            • rhinocrates 10.1.2.1.1.1
              13 November 2016 at 11:07 pm

              I suppose I could, but I have show some sense of legal delicacy (maybe I signed an agreement…). Google will help you.

              BTW, Kafka was an insurance clerk, which is near enough to HR.

      • Siobhan 10.1.3
        14 November 2016 at 11:21 am

        “Intellectuals Are the Shoeshine Boys of the Ruling Elite”

    • North 10.2
      13 November 2016 at 10:08 pm

      My recollection of Robinson is that she generally affects a tired somewhat bored tone about those who aren’t squarely into the status quo. The powers that be. Had the occasion to email her a few years ago, quite tidily questioning her academic objectivity.

      The charming response from Robinson – three words –
      “Go fuck yourself !” What an august professor !

      As to someone’s thought that she was employed by the National Party at some point – I don’t know but I do recall some tie-up around Shipley.

    • Anne 10.3
      13 November 2016 at 10:12 pm

      She was lucky to strike the jackpot. After all she had a 50/50 chance of being correct.

      I picked up on her demeanour/body language (whatever you want to call it) over the years when she used to be on TV1’s Q&A panel. She’s a Nat from the top of her head to the tip of her toes.

    • Incognito 10.4
      13 November 2016 at 10:38 pm

      Nice comment, thank you.

      As you mention in your links under MMP it is all about coalitions.

      My take on Prof. Robinson’s piece in the NZ Herald is that it is an opinion piece written by her as a “political commentator. Her Twitter account name suggests that she doesn’t take herself too seriously and thus we shouldn’t do that either, should we?

      If we are indeed to take the learned Professor’s predictions seriously than we need to know about the assumptions, the limitations, the bias in the data and what was done to counter this, etc. Ideally, we would get a reference to her publication in a peer-reviewed scientific journal.

      You did notice that Prof. Robinson does include at least one major proviso, which is the state of the economy? With Trump being President-Elect all bets are off, as if they were ever on, but uncertainty, market volatility, and possible geo-political instability are definitely on the cards [no pun].

      In other words, Prof. Robinson may or may not correctly predict the outcome of the 2017-elections but she will probably stick with ‘her model’ regardless. This is not the hallmark of good science …

      I predict that it in three years’ time we will get to read a date-adjusted version of the same article in the NZ Herald, but I hope that Prof. Robinson will prove me wrong.

    • Matthew Whitehead 10.5
      15 November 2016 at 2:07 am

      Except it’s lightweight stuff to call an election on a single fact even if it is a reliable correlation, as it’s difficult to tease out correlation and causation.

      Our history under MMP is too short for her to be making calls like this on something as flimsy as “whoever polls highest the previous July.”

      Firstly, at some point, it’s inevitably going to happen that Labour will win an election with less of the Party vote than National, purely because they have more powerful coalition options than National does. It may be in 2017, or it may be later. I’m happy to go on record for that, and I expect it will happen multiple times in my lifetime, so you may go ahead and laugh at me if it’s 2050 and every single government has still been formed by the largest party in Parliament.

      She may be right that looking at early voter intention is a reasonably reliable predictor of final electoral results to a degree.

      But she’s entirely disregarding the influence of the centre parties, coalition politics, and electorate wins, which can absolutely shift the government. They did back in the days of Bolger and Shipley, even though the winner coincidentally coincided with her metric of early decision voters.

      Several people here (myself included) were too optimistic about the chances last election. That’s a fair call. But Claire wasn’t right because she did good work. Her model is way too simplistic to call the entire election based on it a year in advance. She merely guessed right that time.

      Like any overly simplistic model, it has a perfect record on correlation- until you find out that whoops, it’s just correlated well in the past because it’s a contributing factor to who wins rather than an all-powerful determinator.

  11. Xanthe 11
    13 November 2016 at 9:38 pm

    As well as “spin-doctors” and “spin-professors” we need to beware of “spin-scientists”!

  12. swordfish 12
    14 November 2016 at 3:09 am

    Yep, she cynically regurgitates this shtick every Election.

    See my brief critique of her pre-2014 rendition of the same argument here … https://thestandard.org.nz/itching-scratches-on-nine-to-noon/#comment-1237358

The server will be getting hardware changes this evening starting at 10pm NZDT.
The site will be off line for some hours.