An early preview of how the seats might fall in this year’s federal election:
Of course, no psephologist makes serious predictions until a few days
before an election; too much can happen to change things. If we can, we
leave it till the day itself (or better still, a few days after!). All
the same, the electorate is not as volatile as most commentators make
out; big swings in the last few months are very much the exception
rather than the rule, so it should be possible to say something
sensible at this stage.
So here’s a first attempt at previewing the way seats will fall in this
year’s federal election. I’m giving three possibilities; not, of
course, the only three (a government victory is still very much a
possibility, although looking fairly unlikely), but enough to get some
sense of how things look.
First possibility: a 2% swing to Labor.
This is about the minimum required to be confident of a change of
government (I think 1.5% would probably do it, but it’d be much less
certain). From the pendulum, 2% uniform swing would gain Labor 11
seats. My prediction is it would actually gain it the following:
Solomon, Canning, Adelaide, Hindmarsh, Parramatta, Herbert,
Eden-Monaro, Hinkler, McMillan, Bowman, La Trobe and Makin. Total 12
seats, or a narrow but workable majority for Labor. The last 5 are all
above the 2% line, but seem to me particularly vulnerable: McMillan
& Bowman have Labor sitting members, La Trobe a retiring government
member, Hinkler a sugar seat, and Makin a member in exceptional
Second possibility: a 4% swing to Labor.
This about where I think things are at the moment; the polls are a
little better for Labor, but some allowance should be made for late
movement towards the incumbent. A uniform 4% would be a gain of 22
seats, but I call it 21 – the previous 12, plus Dobell, Deakin,
Richmond, Longman, Moreton, Gippsland, Page, Lindsay & Dickson (a
Labor majority of about 20 seats). The first 3 of those actually have
margins below 2%, but I think are a bit safer than that: Dobell because
of a new sitting member (sophomore surge), Deakin because Latham seems
to play less well in Victoria, and Richmond just from gut feel. The
last 2 have margins above 4%, but Lindsay is vulnerable because it’s
Latham heartland territory, & Dickson because the 2001 result was
exaggerated by the Kernot factor.
Third possibility: a 6% swing to Labor.
This is roughly what the polls are showing now, and although I think
that exaggerates things slightly it’s well within the bounds of
possibility. On a uniform swing it’s a gain of 29 seats, which I’m
calling 30: the previous 21, plus Paterson, McEwen, Petrie, Kalgoorlie,
Corangamite, Leichhardt, Macarthur, Robertson & Boothby. The first
three seem safer than their raw margins due to sitting member factors;
the last four seem less safe (Leichhardt for sugar, Macarthur and
Robertson for Latham/NSW, & Boothby for Mitsubishi). Net result a
Labor majority in the high 30s, which could reasonably be described as