The Courier-Mail has published a Galaxy poll of the Queensland marginals Bonner, Moreton, Longman and Herbert, which shows a collective Labor lead of just 51-49. Based on results in these seats in 2004, Adam Carr calculates a swing to Labor of 5.1 per cent – enough to reel in Bonner (0.5 per cent) and Moreton (2.8 per cent), but not Longman (6.7 per cent) or Herbert (6.2 per cent). Also just outside the range is Blair (5.7 per cent). The Courier-Mail refers to a 5.6 per cent swing in Bonner, but no seat-by-seat breakdown appears to be available online. This is the most encouraging poll to appear for the government since June, when Galaxy showed a Labor lead of 53-47 result nationally and Westpoll had the Liberals on track to gain a seat in Western Australia. The poll was conducted over the weekend, from a combined sample of 800, so it’s conceivable that Labor’s political difficulties of last week might have been a factor.
In other news, this blog will be moving to new premises reasonably shortly, which partly explains why my reaction to the election announcement has been less dynamic than it might have been.
78 Comments
New newspoll: http://www.theaustralian.news.com.au/files/newspoll-15-oct.jpg
Isn’t it slightly bizarre that the Oz would leave that sort of file just lieing around on its servers?
From the article, does it appear that they’re directing FF preferences back to the Coalition? It’s quite a confusing article.
The numbers are very weird. He’s miles ahead on PPM but still only 51/49?
‘Mr Rudd is still more popular than Mr Howard, with 54 per cent of those surveyed naming him as their preferred prime minister, compared with 38 per cent for Mr Howard.’
It seems they assume most of FF will go back to Coalition. The problem is this is a sample of 800 voters in 4 seats, so on average 200 per seat. That is a margin of error of about 7%!
Adam says it’s a 5.1 per cent swing, but the Mail says 5.6 per cent. Can anyone adjudicate here?
I don’t get what the big fuss is over QLD. In terms of the 15 most marginal liberal held seats, only two of these come from QLD.
Infact of the most marginal the 3rd queensland seat doesn’t arrive until number 22.
Sure, everyone expects Rudd to do better in some QLD seats because he’s from QLD, and I reckon this may well be true, however its quite plausible for him to win government only picking up 2 QLD seats.
SA is far more important with 3 Ultra marginals at under 1%. Of course QLD will likely be more interesting, since those 3 SA marginals are expected wins, whereas QLD has a few seats a bit further out that may move if Rudd goes well.
I’m sick of hearing those moronic lines in the MSM about Rudd needing some huge number of QLD seats to win.
Quite dissapointing really for the start of a campaign where the Govt might recover a little.
A reaction to McLelland’s stuff up and the Govt making it look as though they supported the Bali bombers? Or simply MOE.
If the Newspoll continually confirms a 55/56 then there must be some severe business going on elsewhere.
Williams, you may want to directly link to the article below, which has more of the actual polling details
http://www.news.com.au/couriermail/story/0,23739,22585658-3102,00.html
I agree LTEP, it is a dreadfully confusing piece and after finding out the details, I’m not sure whether the poll has much value except to reveal that the marginals are somewhat close (not that was a huge revelation).
I think Adam was just abstracting the 5.1% from the headline figure of 51/49. The 5.6% is probably their exact figure swing.
The huge differences between national polling and seat-by-seat polling are quite bewildering.
Kina, Newspoll was confirmed at 56/44
Once the labours ads start rolling the polls could move further Labours way.
Do you think so? I think adverts just keep people onside, I don’t think they really switch votes.
Doesn’t make a lot of sense against a TPP 51/49. Do they ask everyone all the questions?
How much did the minors get? Seems many must be preferencing Labor.
Thanks, KT.
While these are underwhelming numbers for the ALP in Queensland, I do agree with James @ 7 – the ALP really only have to win Bonner & Moreton to be on track. And there are other seats in Queensland which may prove to be of interest (Blair & Ryan for starters).
This ad could work well:
http://www.howardfacts.com/features/howardstwentytwopercent.php
I’m not convinced by this Galaxy Poll. There are too many doubts and questions. 200 per seat is very questionable. The other figures stated don’t match the voting intention result.
I’m also mindful of the role Galaxy likes to play in keeping John Howard in the race. Briggs is a big Howard supporter.
neither labor nor the unions have been running ads and theres been a blast of government ads, adding to that there was the death penalty bugga up and the labor camp has been fairly quiet with no big announcements, now it’ll be all stops out and Rudd can dole out his ammunition a bit at a time to get momentum, he’s in a much better position than Howard.
the Galaxy pollster must be getting a feel for whats happening for him to say Howard has only a 10% chance of winning– i’d take Rudd’s 90% chance any day.
Why would Briggs say “Howard has only a 10% chance of winning” when he publishes these figures and puts so much emphasis on Queensland as to Labor’s chances of winning?
Rudd-v-Howard
PPM
Newspoll: 48 – 39
Galaxy: 54 – 38
Satisfaction
Newspoll: 60 – 47
Galaxy: 71 – 49
A great many more of the Galaxy respondents prefer Rudd to Howard but are apparently masochists. Almost no contest!
Newspoll 56/44
Galaxy 51/49
Hmmm…I think the preferences should go 90% to Labor then.
One last thought before hitting the sack – where are both major parties’ primary votes? It may be sleep dep, but they do not appear to be there. Hmmmm…
HANG ON A SECOND.
Galaxy Polling took a poll of 4 Marginal Seats.
It found that COMBINED that Labor was polling 51% TPP to Libs 49%.
Then the author goes on to say that “this is a swing to Labor of 5.1%”.
BULLSHIT. This is an 8% swing to Labor from the 2004 Queensland results.
They then go to claim this: “enough to reel in Bonner (0.5 per cent) and Moreton (2.8 per cent), but not Longman (6.7 per cent) or Herbert (6.2 per cent).”
BULLSHIT. If you take a poll of marginal seats, and then find Labor gets 51% of the TPP, how can you then go and say there is not enough “swing” to pick up the seats you JUST POLLED?!
It’s wrong on the swing. It’s wrong on the analysis. It’s wrong in it’s assumptions.
Yeah, this Galaxy poll just does not seem right. The PPM and Satisfaction numbers are streets ahead of the 2PP. What’s going on there?
one of the 4 in Longman
the Mal Brough personal vote affecting it?
dodgy all round
The Galaxy has to be taken to mean that Qld MAY not be a happy hunting ground for the ALP this year with only 5.?% swing. So what?? If they get 5% swings in NSW, Vic and SA they are well into government. Don’t think Qld is any more important that SA in this election, possibly less so. Anyhow, I think it will be won and lost in NSW / Vic.
Calm down, LaborVoter. The 5.1 per cent figure came from paid-up ALP member (correct me if I’m being presumptuous here) Adam Carr, who I presume added together the results from the four seats in question. Whereas I gather you are comparing it with the overall Queensland result (correct me if I’m wrong), which I don’t think you should be. The four seats have margins of 0.5 per cent, 2.8 per cent, 6.2 per cent and 6.7 per cent. It is entirely reasonable to conclude that the Coalition would still hang on to the latter two with an even 51-49 result.
Okay for everyone here that is “dismayed” at only 51% in Queensland LIBERAL marginal seats, have a read of this:
http://www.gatago.org/aus/politics/49384416.html
51% in Queensland is landslide territory, despite the spin from the Author.
This Galaxy seems to be saying that even when Queenslanders know someone else is better and they prefer them they still wont change sides.
Well, the MOE is about 3.5% on this poll, so for all we know it could be at the bottom of the range. Very intriguing pair of polls to start the election!
LaborVoter, it’s not a poll of Queensland, it’s a poll of Bonner, Moreton, Longman and Herbert. The average margin in those electorates on the new boundaries is just over 4 per cent. Therefore, a combined result of 51-49 is a swing of 5 per cent, not 8 per cent.
Okay but how does he then come to the conclusion that getting 51% of TPP from 4 seats they polled shows Labor NOT winning all 4 seats?
Without knowing how many people were polled from which electorate, and without seeing the results from each electorate do you agree that the author is making the assumption that the vote was increased by exact equal amounts in all 4 seats?
LaborVoter, by “the author”, I believe you mean me. Unless you are referring to the writer of the Courier-Mail article. Yes, of course the author/the Mail is making an assumption that the vote increased by an equal amount in all four seats. What other assumption would you suggest? That it happens to be more heavily concentrated where Labor would like it to be?
#32
If Bonner polled 55 ALP as the Courier Mail stated, the others could be 52,49 and 46. This averages as 51. Still only two of the four seats changing hands. This still fits with the swings being larger in some of the notionally safe seats.
Good question.
If it is 800 for all 4 seats, and assuming an average of 200 per seat, then that makes the margin of error 7% _per seat_, which is almost to big to draw any meaningful conclusions from.
If the ALP primary vote is 45, and Greens are on 7, then the ALP 2pp should be _at least_ 51 (45 + 6 = 51), but probably even more because the Liberals wouldn’t be getting 100% of the Family First preferences.
Even if the Liberals are getting 75% from Family First, that would be another 1 or 1.5% for Labor (5% X 0.25 = 1.25%). So I think they are on 52 or 53 rather than 51.
Ooops… should be 55,52,49,48.
All psephed out for today.
ShowsOn, the points you make are sound, except for the “good question” bit. LaborVoter’s gripes have nothing to do with the MoE or the preference calculation.
This will all be academic anyway. I’m tipping Antony to call the election by 6.55pm anyway. I Think it’s GST and monthly mortgage repayments. Does anyone (dis)agree?
I am wondering if they ask all the questions of everyone or include answers where the respondent won’t answer all the questions.
ie some people revealed who they liked but not who they would vote for.
The satisfaction ratings and PPM if it were a Newspoll [consistent correlation all year] would indicate a TPP in excess of 56/44. In fact Newspoll has never had approval and PPM ratings this high.
http://www.theaustralian.news.com.au/files/newspoll-15-oct.jpg
The only way to explain it is if there are a very high proportion of minor party voters preferring Rudd. I feel this particular poll must surely be 51+ with a bullet.
I’ll call the election before any mainland results come in if the swings to Labor in Bass and Braddon are in double figures.
Tasmanian results only come in early if the election is held is in October or March, when Tasmania is the only state on daylight saving.
William,
As a fellow Sandgroper, what’s your thoughts of Howard’s donation to Telethon and how he was trumped by the State Govt’s donation ?
And likewise the AC Neilsen – there has been a consistent correlation between PPM, Satisfaction rating and TPP. Not that they are predictive just generally synonymous. At least not counter-intuitive.
http://au.acnielsen.com/news/documents/NielsenPoll2007.pdf
The Galaxy is the opposite of the Newspoll and AC Neilsen. I don’t have other Galaxies to look at unfortunately.
I would be interested to see if the poll results were significantly different in Herbert. It’s a North Queensland seat so there would be a lot of opposition to council amalgamations. The sitting member, Peter Lindsay is also the parliamentary secretary for defence and Townsville has a high military population.
It woud be interesting to compare with Leichhardt. Particularly as Warren Entsch is retiring.
Same goes for Longman. Mal Brough is the man of the hour. Is Howard really this clever? Methinks that St Kevin is going to pick up a few roughies in QLD….might be worth a visit to sports bet…Johnny is fast running out of fingers to plug his dyke with.
Interesting to keep track of this one from time to time though it is not representative of course – the Courier Mail on-line poll results.
http://www.news.com.au/couriermail/poll/display/1,23815,5031238-952-1,00.html
Fear not Laborites, this poll is nowhere near the reality up here in sunny QLD, but I would not be putting any money on Mal Brough losing Longman.
The reality, even in those four that got our attention, is much closer to the 56/44 Newspoll published on the National scale.
Bonner, Moreton, Herbert and Blair [4] are not affected by the amalgamations issue, whereas Longman, Petrie, Flynn and Dickson are.
That leaves us with Bowman, Hinkler and Ryan.
Bowman has two factors going in Labors favour: it is the 28th highest mortgagee seat in the country (33.9%) and there will be some image damage to the sitting member (Dr. Andrew Laming) who was caught up DPP investigations for misappropriation of printing funds and a ‘ghost staffing’ investigation, not prosecuted due a lack of evidence.
Ryan is a high middle income earner mortgage belt seat dominated by wet liberal professionals and large families with a keen interest in the RBAs next meeting on Nov 6th.
If mortgage rates go up in November I would put both Bowman and Ryan in the possible surprises basket, although Ryan is nothing like an outer suburban mortgage belt seat where you would normally look for the above average swing if there is one on in a given State. And I wouldn’t be writing off Hinkler altogether.
IMO Labor have some hope of doing much better than the 2 seat gain this article and poll would want us to beleive will be the outcome in Queensland.
Looking at these seats from the 2004 election, we have ALP 2PP votes of:
Bonner 49.5
Herbert 43.8
Longman 42.3
Moreton 45.8
Average 45.4
As already noted, a 51-49 in these seats represents a 5.6% swing overall. That’s a fairly big swing. 51-49 margin is not “stratospheric”, as in 57-43 or similar, but the swing is quite large, considering just these seats.
My question to Galaxy is this: why weren’t the following seats polled?
Bowman 40.9
Dickson 42.2
Hinkler 45.2
Petrie 42.1
Average 42.6
Bowman has a large swing required (9.1%) but was an ALP seat in 1998 and 2001, and thus is very volatile and susceptible to going back to the ALP. Hinkler has a margin of less than 5% and the other two are about the same as Longman.
An interesting poll, but it could have gone further.
In Hinkler, where I have been doing street stalls for Gary Parr, the internal polling shows that the ALP vote is solid and the coalition vote has collapsed. It is showing however a 30% undecided, up from 10% on previous elections this close to an election.
If Gary Parr can get 20% of those undecided, would give him a primary vote of about 45% and a winning lead.
Once again the phenomenon of Labor’s vote in marginal seats far less than the national polls would suggest it should be. And this in the home state of Rudd and Swan. It has to be a worry for Labor.
“Galaxy poll shows Rudd in trouble at home”
What an absurd headline. Labor needs to win 2 seats from QLD, it’s in the lead in 2 seats. Sure it’s a lot less than other polls are predicting but it hardly amounts to ‘trouble’.
Phil Robbins 50
I suspect that the inclusion of Longman in the Galaxy poll skewed the results somewhat. Without knowing local politics there, I suspect Mal Brough is fairly popular (hey, he’s one of the few Colation ministers I actually like, and I think he’s a future leader in the making) and will hold his seat.
Bowman, on the other hand, with the one-term Andrew Laming as the sitting member, is much more vulnerable. Hinkler as well, as noted in post 49.
I’d be worried, for sure, but I still think the ALP will win Bonner, Herbet, Moreton, and Bowman. I think they could also win either Petrie or Dickson, maybe both.
That’s 4-6 seats, and a third of what they need for government. Not a bad result.
Whoops, I missed Hinkler. So make that 4-7 seats. Even better.
This is a nothing result by the inclusion of Longman. The Coalition were always going to win the seat thanks to the higher public profile of Mal Brough and his intervention. Even hard nosed ALP supporters suggested that they were pushing it uphill to gain the seat.
This skews the result lower for Labor than it otherwise should be. A better idea would have been to include Blair or Petrie which are more in doubt.
Sorry, Flynn or Petrie…
NSW is joining Tasmania as of March next year. Iemma has a bill in parliament right now to legalize it. It didn’t get going in time for the start of DST in Tassie earlier this month but we in NSW will end it with Tasmania next April. Then going forward in the future, NSW will be on line with Tasmania. Suspect Victoria, ACT and SA won’t be far behind
Two polls have been released with differing results, but they’re not mutually exclusive.
The national Newspoll reports the same result it’s been reporting for a long time 48/39 primary 56/44 TPP, ~ 9% swing.
The Galaxy poll in 4 QLD marginal seats shows 5%. Newspaper polls in SA show similar swings.
Assume the 5% figure is actually all the ALP will get in the marginals, then we can conclude two things:
1. that’s sufficient to deliver 18 seats by itself with the hypothetical uniform swing.
2. there are some very large swings going on elsewhere in the nation delivering seats that no-one thought would fall before. If 40 marginals (L/NP and ALP) are delivering a 5% swing to the ALP, then the other 110 seats will be delivering over a 10% swing.
Definately in line with my prediction of a 5-7 seat majority.
I can imagine Labor won’t find it too hard to get together the following:
SA: 3 seats
Qld: 2 seats
Tas: 2 seats
NSW: 3 seats
Total: 10 seats
However, I imagine them losing Cowan in WA is a real possibility. This gives them a net gain of 9 seats. Can they scrounge up another 5 seats to gain minority government? Definately not out of the question but I wouldn’t be betting the farm on it.
Why dont we run a lie detector over this election campaing here is number 1:
Prime Minister John Howard has promised to run a clean election campaign, denying knowledge of a Coalition dirt unit.
http://www.smh.com.au/news/national/well-keep-it-clean-pm/2007/10/15/1192300637422.html
Re Bluebottle at #47…
Herbert is affected by the amalgamations issue because the Townsville and Thuringowa councils are being amalgamated and both mayors have expressed opposition.
http://www.thuringowa.qld.gov.au/council/message/index
These results do not surprise me at all, but as James at # 7 and some others have said, the ALP does not necessarily need to gain more than two seats in Queensland to win the election.
In any case I think that the ALP will gain Bonner and Moreton and Blair besides, due to a larger than average swing there.
Bluebottle @ 47: 56/44
That would certainly make more sense than everyone indicating they prefer Rudd and want him as PM but say they will vote LNP. I am suspecting that 51/49 is being very generous to the LNP.
Labor’s internal polling will no doubt have its finger on the pulse.
Well may Mr Howard say that everything depends on the economy [BTW, a position not inconsistent with Marxism]
What he appears not to realise is that, in turn, the economy itself rests on three crucial foundations, all of which he has sorely neglected.
They are:
The environment,
Education training and research and
Investment in public infrastructure.
He clearly has been a negligent economic manager, riding on the reform coat-tails of Keating and a global boom.
He lamely boasts of surpluses. But they have been gleaned from the neglect of the above crucial areas and by scrimping on payments to the states.
Playing with Anthony Greens calculator on the ABC site is interesting. 5% swing in NSW, Vic and QLD, followed by 3% everywhere else (which is pretty conservative given the current polling i think!!) and the ALP romp it in with an 8 seat majority.
Ok, not quite the bloodletting that some are hoping for, but it will do the job. Also, i really think that the Liberals need to lose big time at this one so that the “liberals” can go about reclaiming the party from the “conservatives” in the aftermath.
I also think Antony will call the election by 6.55pm, but if it’s anything like the Victorian election, he’ll call it for Family First and the Greens!
you can line up this poll with Newspoll.
some ’safe’ seats will fall and some ‘marginals’ will not but seats will fall nevertheless. quite a few in fact.
Mark 44: “I would be interested to see if the poll results were significantly different in Herbert. It’s a North Queensland seat so there would be a lot of opposition to council amalgamations.”
You are deadwrong on this mate. I live in Townsville(Herbert Seat) and most people here support the council amalgamation, so if anything it’s a plus for Labor.
Most of the “Thuringowa” area is now swallowed up in the city that is Townsville, and people in Thuringowa see themselves as being an area OF Townsville… not a seperate city.
I will be glad if Howard pulls his council amalgamation plebescite stunt here in Townsville on election day, because it’s going to backfire badly on him and we may see Peter Lindsay down at the centrelink queue the week after the election.
The importance of Queensland
Those who don’t hail from the Sunshine state might be forgiven for wondering what the fuss is about here. Why is it so important to this (and most) Federal elections?
1. It is the most “bipolar” of states. Voters here can landslide Labor at State level (2006) right after delivering 23/29 Federal seats to a coalition.
2. It has the highest growth in population. This significantly changes a) the voter mix, b) the electoral boundaries and c) the issues mix (local/state issues can blur more readily).
3. It has a lot of people. Sure Sydney and Melbourne as gravitational centres outstrip Brisbane but are more politically and demographically stable in terms of population. However, South East Queensland, from Coolangatta to Noosa and West to Toowoomba, with over 2 million residents, is the biggest pool of swinging voters in the country.
It is not “fair” that Queensland and Tasmania have disproportionate influence on our Federal politics but influence it they do. If you want to win this election, you need to win South East Queensland.
Erratum with post #67
This should read ” 23/29 Federal seats to conservatives, including 22 Coalition and one independent (Bob Katter) in Kennedy”
Sorry Bob
Peterg @ 62
I agree.
This is clearly evident in the graphs (1 and 2) contained within the coalition’s own ‘vision’ statement released on Friday, here:
http://www.liberal.org.au/info/docs/documents/AustraliaStrongProsperousAndSecure.pdf
Pretty clear trends starting around the Hawke/Keating governments.
John Howard has certainly been good at not spending money, even if he’s promised to spend it.
This Galaxy Poll is a bit of trickery by David Briggs – a Howard supporter – He deliberately includes Longman and Herbet, which for reasons mentioned by others may not reflect the general swing in the rest of the state. Why did he exclude Blair which is more marginal than both of these? You can be certain that on election night, across the country, some govt seats with 5-7 percent margins will be retained while others requiring a 10%+ swing will be lost. For this reason polls of selected marginals should be treated with caution.
GO at 67, Labor don’t need to win Qld as such, they just need to do better. Even if the swing in Qld is less than 4% (& indeed even if this is accompanied by a swing of less than 3% in WA, SA, Vic, Tas & the NT) Labor can still gain sufficient seats to form government if they get a swing above 4% in NSW.
This election will be very close, as per my post pointing out that the opinion polls are reflecting expectation and not voting intent. I expect the Coalition to fall over the line 76-72-2.
http://ker-plunk.blogspot.com/2007/10/prediction-coalition-to-win-november-24.html
Sorry Jack, are you arguing that opinion polls are people saying who they think will win rather than who they will vote for? What basis do you have to believe that?
Not saying I disagree with your count, I’d probably be a little more generous for the Coalition than you, but I’m interested to see why you think the opinion polls reflect ‘expectation and not voting intent’.
Jack, the polls have been incredibly consistent with a swing of 8%+ for 9 months. You don’t get polls like that unless something is going on out there. Please provide some evidence to your assertions rather than wishful thinking…
The Narrowing can be bit of a myth – whilst possible I wouldn’t be putting my house on it and there can always be the ‘widening’.
Possums did an analysis of it here:
http://possumcomitatus.wordpress.com/2007/09/27/the-narrowing/
Summary
“So The Narrowing is 2 from 5 at best, 1 from 5 that can be called certain, and where both of those occasions can be explained by the longer term trends and movement occurring in the polling series over the previous months.
While The Narrowing makes for a simplistic bed time polling story and is to be expected to be used by political parties as a weapon for their own electoral self-interest, when it comes to providing actual explanations of electoral behaviour to a broader public, the dogmatic representation of “The Narrowing†as some divinely inspired inevitability is a poor substitute for polling analysis based on the evidence contained within the polling data.”
Possums ominous comments about the two possible narrowings -
“….both of those occasions can be explained by the longer term trends and movement occurring in the polling series over the previous months.”
This is what has not happened to Labor’s polling, it has staid steady if there has been a recent trend in recent times it has seemed to be in the opposite direction.
Anything is of course possible.
If I was the ALP I would be fine with these poll numbers.
Now I could be wrong but I feel Longman is a bit like Deakin always marginal but never ALP.
We are seeing from the polls 5% swings in seats like Bennenlong, these four Queensland seats and Stuart / Boothsy.
Possum has reported the swing is said to be bigger in safer Liberal seats and I would suggest apart from 2004 the ALP have in recent timers improved there polling numbers.
The sample size is only 200 per seat and I would imagine that all booths in these seats have more voters than the sample size.
Jack @ 72,
Where do you come up with such hilarious statements? You don’t work for the Government Gazette or for the Liberal party do you?
I’ve had a look at your blog and I’ve found this very interesting comment:
“While the PM has been looking calm and relaxed over recent weeks, Kevin Rudd has been looking decidedly stressed and it’s my view that it’s because he knows what internal party polling is telling him regarding Labor’s election chances”.
Rudd looks pretty relaxed to me, it’s the Howard who looks tense – a shadow of the PM he was in 2004.
Oh and here’s another beauty from your blog:
“Both Labor and Liberal polling suggests that Victoria will remain with the status quo and while Deakin and La Trobe will be closely fought battles I expect them to stay in Liberal hands”.
How do you know what internal polling for BOTH parties is saying? I thought such polls were confidential to important party members only?
I don’t believe your statements, but hey everyone’s entitled to their opinion – especially when they’re good for a laugh. It’s a pity that Glen and Nostradamus et al are not posting on this thread tonight – they’ll be delighted to meet you.