Wednesday, August 5, 2015

Poll Roundup: Bad Bishop Off The Board

2PP Aggregate: 52.8 to ALP (+0.2 since last week, highest for 14 weeks)
Labor would win election "held now" with small to modest majority

It's been a week for obvious chess analogies.  In the wake of the richly deserved end of a dreadful appointment, commenters have been scrambling to decide whether Bronwyn Bishop was "captured", "sacrificed" or "gambited".  However, I'm here to do more justice to this theme.  Chess players are well aware of the concept of a Bad Bishop - a piece that is unable to develop and interact with the opposition properly, stuck behind its own lines mingling with its own pieces all game.  Heck. I've had some that have spent the first 28 moves chatting to their king and queen in the party room, and now, every time this happens to me again, I'll be unable to look at the damn thing stuck there on c8 without thinking "Bronwyn".   Some bishops are even so bad that we just call them big pawns.

There could also have been comments about whether Bishops Bronwyn and Julie were "bishops of opposite colour" (one moves on light squares, one on dark, and never the twain shall meet).  Bishops working together are normally a bit better than knights, but when a couple of knights really get going in a tangled position they can whip up quite a storm.  So much so that in obscure Hobart chess lingo, really damaging knight pairs (like, say, Fairfax and Murdoch in tandem) were known as "the whirling helicopters of doom".

Anyway, enough chess talk.  On the sidebar I've put up a Not-A-Poll where readers can vote for their choice as Speaker.  Assuming that it will be a Coalition MHR (to make the poll less predictable) I've added everyone who Sportsbet had at less than $80 then chucked in a few obvious joke candidates (some of whom are polling far too well).   There are almost no wrong votes here: most of those listed, even most of the silly ones, even Wyatt Roy who for whatever reason is currently leading, would probably be an improvement.  (Not much love for the fancied contenders, by the way.  I really need to spam more on Liberal-oriented sites).

This week's polls

There was no Newspoll, presumably because it is being held off for the resumption of parliament, which has frequently led to three-week spells between polls in the past.  For about fifty minor Ellis clones tweeting on the #Newspoll tag, this was further evidence of a Murdoch conspiracy to protect Abbott's government from bad ratings. Never mind that the Murdoch tabloids had spent much of the last two and a half weeks chasing the Speaker in a way highly conducive to terrible polling, or that the Bishop scandal started with a report by Annika Smethurst in the Herald Sun.  Heck, in these days of social media, people so often report being polled that it is hardly as if a pollster could get away with sitting on its data if it wanted to.

There was, however, a ReachTEL late last week which came out at 53:47.  We also have 53:47s from Essential this week and last, and a Morgan which was 53.5 to Labor by last-election preferences, or 54:46 respondent-allocated.

After processing the primaries and adjusting the Morgan for house effect I counted the ReachTEL as 52.8 to Labor, the Morgan as 52.3, last week's Essential as 52.7 (not that it is included anymore) and this week's as 53.2.  That's pushed my aggregate out to 52.8 to ALP, the highest reading since late April.  Depending on the behaviour of next week's Newspoll, I may remove the "global house effect" from my model and back-apply the correction, but that would only take the Labor score down 0.1 of a point if so.

The smoothed tracking graph (below) shows just a hint of blowout, with Bill Shorten's TURC appearance barely a blip on the line anymore.  However, it's nothing to get carried away about yet, especially not given that the major cause of it has been addressed.


Two polls had new leadership figures.  ReachTEL had its usual leadership figures, showing a trivial loss in support for Tony Abbott (net rating down 1.2 to -26.2) and no real net rating change for Bill Shorten (net -26.3).  However, Shorten's "Very Good" score dropped from 8.1% to a parlous 6.8%.  Also, Shorten's Better Prime Minister score was down 1.2 points to 55.1 on ReachTEL's distinctive scale (which unlike those of other pollsters, tends to closely follow the 2PP).

Essential had a smorgasboard poll for preferred Liberal and Labor leaders with quite different ratings to Morgan's from last week.  Essential's polls have much higher "Someone else" and "Don't know" ratings than Morgan's, most likely because Essential has to offer these upfront as options whereas Morgan, in its phone poll, doesn't.  That said, the published result that 187% of Greens voters don't have a preferred Liberal leader might be a slight exaggeration! (It's clearly meant to be 18).

Essential's results are quite a bit cooler on Malcolm Turnbull than Morgan's, with 23% of decided Coalition respondents, 54% of Labor and 54% of Greens preferring him, compared with 33%, 64% and 67% in Morgan.  I'm unsure how much of that's down to the options list and how much to Essential's more direct way of asking the question.  Overall, Turnbull (24) had a meaningless lead over Abbott (18) and the Bishop who wasn't forced to resign this week (17).

On the Labor side, just four options were offered and the results were apathetic to say the least, with Shorten, Plibersek, Albanese and Bowen appearing in the same order and with proportionally similar results as they did in Morgan.  Essential differed from Morgan in finding Shorten most preferred by some margin among Labor voters (Morgan had Plibersek preferred), but also in finding an even spread for the four candidates among Coalition voters (whereas Morgan had Shorten very low).

Other polling

The two Bronwyn Bishop polls are a footnote in history now but their use in establishing the lack of support for the Speaker's position shouldn't be forgotten; ReachTEL's is at question 5 and here's Essential's.  The polls are rather different in design with Essential's use of a temporary stand-down option making it difficult to compare them, but overall the results seem fairly similar.  An oddity was that the ReachTEL found the youngest age group (18-34) both more likely to say that Bishop shouldn't resign, and less likely to support "a federal anti-corruption commission to oversee political donations, allowances and entitlements" (overall this proposal had 73.4% support with most of the rest undecided).

Essential found perceptions of asylum seeker policy were intriguingly little different from when the question was asked in June 2013, while Julia Gillard was still PM, with the main movement being unhelpfully from "None of them" to "don't know".  A comparison they didn't give is to the late July 2013 poll in which Labor's standing on the issue was briefly more competitive during the Ruddstoration - a time when Labor's 2PP was considerably lower than now.  There's no sign of Labor's change having impressed anybody here, with Labor supporters especially unlikely to think their own party has the best policy on the issue.

Essential also showed that Coalition voters were surprisingly positive about Labor's renewable energy targets even when told it was an ALP policy.

Do Daleks Rort Their Entitlements Too?

Lastly, Essential asked a cracker of a question which was undone by some rather dubious (albeit comical) analysis.  Asked whether there was likely to be intelligent life somewhere else in the universe, 60% said yes to 26% no, with party breakdowns being 67-19 for Labor voters, 71-18 for Greens voters, 55-31 for Others voters and 54-35 for Coalition voters.

Out-there Green ideas?  Closed-minded conservatives?  In my view, the real cause for these breakdowns is religion.  A US Survata survey found that atheist respondents were much more likely to believe in extra-terrestrial life than Christians, and that's exactly what I would expect.  For many atheists, it's a bit hard to explain why a cosmically unremarkable planet would happen to be the only one where life ever rocked up, especially with "earthlike" planets being found orbiting distant stars frequently.  For many Christians, there's no real problem with believing that God only wanted life to have one place to misuse its travel expenses in.  For the more literalist types there can be problems with fitting in the idea that God created humanity in Her image and then made some other more intelligent morally aware bunch somewhere else (but didn't mention them in scripture).

Of course, there are atheists who believe we are alone and are incredibly lucky, and Christians who are confident we're not.  But I suspect the tendency of very religious voters to prefer the Coalition and the Christian right parties while non-believers tend to lean left is a big part of the explanation here.

No comments:

Post a Comment