We’ve had two polls come out in the last few hours showing remarkably divergent results.
Galaxy has brought out their second poll of the weekend. After a poll conducted on Thursday and Friday gave the ALP a lead of 52-48 on a two-party preferred basis, a poll conducted on Saturday night for the Nine Network has put the two-party preferred vote at an even 50% for each party.
In contrast, Newspoll, in its first poll in three weeks, has the ALP up from 53% to 55%.
Galaxy has a primary vote split of 44-38-12 in favour of the Coalition, while Newspoll has a split of 42-38-12 in favour of Labor.
Consider this an open thread for discussion of the federal election.
The only logical thing to do here is to do what I did during our election in the UK, average out the polls.
July 18th 2010: Labour 51% Coalition 49% (Labour lead of 2%)
Labour had a lead of 6% at the last election, so there has been a swing of 2% to the Coalition. If repeated on a national uniform swing (something that does not usually happen in the UK, but I cannot say if it does in Australia), that means Labour would lose the following electorates: Herbert, Robertson, Solomon, Swan, Macquarie, Gilmore, Macarthur, Dickson, Hasluck, Corangamite, Bass, Bennelong, Deakin and Longman. This would put Labour on 74 seats and the Coalition on 74 as well (with the two Independents holding the balance).
Presumably they are allocating their preferences differently?
I’m a little out of the loop as I live in Washington DC, but didn’t Green second preferences desert Labor in the Penrith by-election compared to the 2007 flow?
Assuming 3% error margin, if the real position was 52-48 (say), then both polls could be “right”. It would also fit in with the 52-48 position given by ACN and the previous Galaxy, and the last Morgan (53-47 I think?)
This will be a low enthusiasm election, swings will be highly variable and sitting MPs votes will be significant. I had said it before but it will 1984 all over again with the difference that the Lib surge has come earlier.
Does anyone know how the polls are going on a state by state basis?
My gut feel is:
NSW – ALP being tarnished by unpopular state gov but overall swing to be small.
VIC – ALP doing very well. May pick up seats.
QLD – Expect swings to the LNP.
WA – Very strong state for Libs who will take 12 out of 15 seats.
SA – Overall swing small but volatile between seats.
TAS – Expect swing to Libs which could put Bass and Braddon in doubt.
NT – CLP to pick up Solomon.
ACT – ALP to hold seats.
Is there any polling to back this up? Or to refute it?
SMc – my info is (and I can’t tell you where):
NSW – LNP will pick up Macquarie and Robertson. ALP will pick up Greenway. Macarthur and Hughes in doubt but trending ALP. LNP should retain Gilmore. Eden-Monaro and Lindsay both ALP. Eden-Monaro especially so.
QLD – most interesting. In some seats the Libs can’t make any inroads. In others, they look good. Herbert to ALP for sure. Dickson to ALP (but close). Bowman to be held by LNP. Longman, Flynn, Dawson, Forde and Petrie all look LNP. Brisbane, Ryan and Bonner in doubt. ALP doing better in Leichhardt for some reason (but close). If Rudd doesn’t stand Labor will lose Griffith.
VIC – agree with you to some degree. Corangamite to be held by ALP. McEwan looks like ALP. Don’t agree on LaTrobe – Libs will hold this one I think.
WA – not much movement. Stirling could go ALP. Swan is very close. Hasluck looking like Lib. Brand seems close in some polling, but I expect Labor to hold based on trend. I can’t see the Libs winning 12/15 seats there at this point which would mean the ALP only have Fremantly, Perth and Brand. The ALP should probably get one more.
SA – little change. ALP doing well as in VIC. Only possibly Hindmarsh in doubt. ALP will hold Kingston comfortably.
TAS – ALP will win all 5 seats. More likely for the Libs to pick up Braddon than Bass. I don’t know why.
NT – agree that Solomon will go to the Libs on a big swing.
I think it will be little changed unless Gillard collapses. Note, 2 months ago I would have said the Libs were heading for victory so the polling can change quickly. The Libs might pick up a net 3 seats from the 83/64 at the moment. It is less likely that the ALP will increase their majority but the trend is back to the ALP. But the new preferencing deal between ALP/Greens might change everything above and make it a larger margin.
Harry Hatfield – uniform swings don’t happen anywhere – it is a purely rational concept designed to give an indication of the NUMBER of seats that could be lost based on the swing rather than the actual seats. Seats are influenced by local factors and demographics that negate the possibility effect of a “uniform swing” as it assumes that seats are uniform, which they are not.
Isaac Levido – you are indeed absolutely correct, upwards of 60% of Greens votes exhausted in that by-election, and this was part of the reason for the lopsided swing to the Liberals. You may or may not have heard but Labor and the Greens have just announced a preference deal, and of course, preferencing is compulsory at the federal level.
SMC – I essentially agree with your assessment with the exception of NSW and Victoria – regarding NSW, think there will in fact be a swing to Labor, as voters in that state are registering high approval for federal Labor and may be scared into voting Labor in larger proportions due to the impression being given by the media and the Liberals that the election could be lost due to NSW. This factor may also play out in Queensland, but polling so far does indeed back you up in Queensland. In regards to Victoria, Labor is already pretty much at its high water mark in terms of seats in this state and further gains seem unlikely.
Comments are closed.