Australia - 7 September 2013

[Filuwaúrdjan]

Liberal candidate for Charlton 'withdraws' under something of a cloud.

[Filuwaúrdjan]

Both regional breakdowns of national polls and constituency polls are inherently and notoriously unreliable; it shouldn't be that surprising if they don't seem to link up that well. Fairly clear that the election is probably lost though.

[Filuwaúrdjan]

And 2016 is hardly unwinnable for the ALP, unless the party collectively decides that it is (and acts accordingly). There's a first time for everything - and both Howard and Rudd/Gillard only barely survived their first re-election attempt.

[Filuwaúrdjan]

In the absence of Rudd, Griffith would not be a safe seat at all (he has a substantial personal vote for whatever reason), but I think this is one of those things you have to put into the noted and very important category of 'I'll believe it when I see it'. Now, a by-election following hastily on from a Rudd 'resignation' five seconds after an election defeat...

[Filuwaúrdjan]

Regarding divisions in areas deemed likely to swing heavily, be sure always to check what the swing was last time round. If it was large and against the ALP, then they'll be less room for a large swing ('large' defined as relative to the rest of the wider area in question) this time round. And vice versa, etc.

[Filuwaúrdjan]

One issue for Labor in Victoria historically was its weakness outside Melbourne (in the Whitlam victories, for example, the only Labor seat not in Melbourne was Corio) and the tendency for its support within Melbourne to be even more geographically concentrated than other large cities.

[Filuwaúrdjan]

In what direction(s?) are that curious Queensland double act directing their respective preferences?

[Filuwaúrdjan]

, but KAP and ALP have done deals for the QLD Senate and for a few crucial seats in FNQ.

oooo...

[Filuwaúrdjan]

It's worth noting that the one Labor seat left in 1949 (Wilmot; since renamed as Lyons) was actually a gain in 1946. And with a further swing in Labor's favour. Also, the division gained in 1958 (Braddon) had not been Labor since King O'Malley's defeat in 1917, but was then held - and comfortably, even in 1966, until the electoral debacle that followed the dismissal of the Whitlam government.

[Filuwaúrdjan]

Denison in the post-war era is evidence that even utter idiots can have huge personal votes if they've got themselves a sufficiently shiny war record.

[Filuwaúrdjan]

Though much of that swing was already in evidence in the last federal election.

[Filuwaúrdjan]

I think by this point he's doing some kind of complex salchow like thing...

[Filuwaúrdjan]

More constituency polling: Coalition wave in Western Sydney, per Newspoll. We'll have to see what's right: constituency or national.

Is 'Western Sydney' just being just as a catch-all term for 'potentially endangered Labor seats in Sydney' these days? Though - and to take the pedantry in a different direction - that sort of poll is not constituency or national; it's called a marginals poll.

[Filuwaúrdjan]

Nothing says community of interest like Page and Kingsford-Smith!

[Filuwaúrdjan]

Primaries in Brand last time round were 41 ALP, 39 Lib. The poll, in other words, shows no swing from last time.

[Filuwaúrdjan]

About to start posting some predictions. Will be doing this state-by-state and probably in separate posts. Throughout I'm assuming a national 2PP swing of about 3pts with less regional variation than was the case in 2011 (though still with quite a bit). I accept that I am very likely to be wrong about this, but whateversky.

[Filuwaúrdjan]

New South Wales - Labor seats

Grayndler - even if the Greens manage to stay ahead of the Liberals, the change to Liberal preferencing policy pretty much completely rules out the prospect of an upset that wasn't looking likely anyway.

Sydney - a very safe seat and will doubtless remain so, with even the vague prospect of a Green upset (which was never very likely) removed. I do have vague worries that demographic changes will eventually catch up with Labor here, but perhaps the current sweet spot can be maintained indefinitely.

Cunningham - Labor stronghold based around Wollongong and only ever lost in a freakish by-election during the Crean leadership (and even then not to the forces of darkness).

Shortland - safe seat in the Hunter.

Charlton - another safe seat in the Hunter, though an open seat. But with a disgraced Liberal candidate.

Hunter - another safe seat in the Hunter, this one includes a lot of natural Nationals territory that's more than counterbalanced by the inclusion of Cessnock.

Newcastle - this is, shockingly enough, yet another safe seat in the Hunter, though one that has occasionally produced a semi-wobbly margin in recent decades. It's an open seat, but defeat is basically unthinkable.

Chifley - as solidly Labor as it is solidly proletarian and will remain so even if there is another exaggerated swing in Sydney. Which, given its name, is exactly as things should be.

Blaxland - a rather ironically named division (Gregory Blaxland was a wealthy landowner and this is one of the poorest parts of Sydney) that was once the electoral fief of a man of fine cultural tastes and a charmingly foul wit. Labor hold.

Throsby - this is actually safer than its margin implies because it's rather polarised (Port Kembla and Bowral et al do not vote the same way, shockingly enough).

Watson - successive boundary and demographic changes have all been rather kind to Labor here, something not dented by the large swing last time round.

Fowler - grim banlieue upon grim banlieue and solidly Labor even in the 2011 state election nightmare. A huge swing (nearly 14pts) last time was mostly down to bizarre selection drama that reflected poorly on the ALP.

McMahon - this ought to be a safe Labor seat and will probably continue to be (and consider that my prediction), but if you were to pick a safe Sydney seat to be lost in an upset this would - given particularly grim 2011 results and the comparatively (but only comparatively!) low swing last time - perhaps be topping your list.

Richmond - a new age dawned nearly a decade ago in the most New Age of divisions. Labor hold.

Barton - ought to be secure enough. Labor hold.

Werriwa - this has been floated around as a possible upset, but it's worth noting that the large swing here last time was to a large extent due to the same absurdity that caused an even larger swing in Fowler. Notably, Labor won't have been far behind here even in the catastrophic state election. Ought to be a hold, then.

Kingsford Smith - demographic changes and a large swing last time means that this looks at least halfway vulnerable for the first time since more or less ever. Still think Labor will hold.

Dobell - circumstances are beyond atrocious and a hold seems extremely unlikely. Of course (and mentioned just because) this was also true of neighbouring Robertson last time.

Parramatta - I would give a very slight edge towards a hold here, but would not be surprised at either outcome.

Eden-Monaro - can rural incumbent strength cancel out a general swing? Again? I think so, maybe. Just about.

Page - as Eden-Monaro, but with even less confidence.

Reid - the name of Reid has traditionally (ironically) gone to safe Labor seats, but the last redistribution changed that. Still, the swing was so large last time and the division so neatly polarised that it might not be as horribly vulnerable as it looks? Nonetheless, I suspect a loss, but a hold wouldn't come as a total shock.

Banks - looks horrifically vulnerable, but is perhaps a little less so than it looks: is there really much more room for swing? The Liberals won everything here in the Great Electoral Fiasco of 2011 (of course) but not by massive margins, something that ought to cause at least a little pause. Genuinely unsure about this one.

Lindsay - too much has been written about Lindsay over the years, so I'll just predict defe...

Robertson - asking for lightening to strike twice is probably asking for a little much.

Greenway - Liberal candidate woes have turned a probable gain into something quite messy and unpredictable. Sufficiently so that I shall not predict.

[Filuwaúrdjan]

New South Wales - Coalition & Other

Note that I'm not bothering with Coalition seats with margins over 6% without very good reason.

Macquarie - named for a man who loved to name things after himself, this strange jumble of unalike places at the point at which raw physical geography halts the western sprawl of Sydney ought to be pretty safe given the general dynamics of the election (as it was during the Howard years), yet has been mentioned as a potential against-the-grain gain. I tend to think that's unlikely, but this is certainly worth keeping half an eye on just in case.

Macarthur - is, amusingly enough, named for one of Macquaries enemies. It's a big slice of exurban horror and will record a large swing to the Coalition.

Bennelong - see Lindsay, but here I expect the incumbent to ho

Hughes - will presumably see a large swing to the Coalition.

Gilmore - the Liberal margin here is greatly inflated by the (retiring) incumbent's personal vote. Were it not for the general atmosphere of the election I'd be calling a Labor gain here (it was that rarest of things last time: a Coalition seat where Labor polled more votes in the Senate elections), as it is a Liberal hold with a 'surprisingly' unimpressive swing seems likely.

Paterson - a comfortable Liberal hold is a dead cert.

Lyne - a straightforward gain for the Nationals.

New England - see Lyne. Note that until Windsor's victory in 2001, New England had always voted for the Nationals/Country Party/Whatever since the emergence of that straw-chewing political movement in 1919.

[Filuwaúrdjan]

Victoria - Labor seats

Batman - the only question is whether Labor win by a large margin (if the Greens come second) or an utterly mountainous one (if the Liberals do).

Gellibrand - absurdly safe seat in western Melbourne.

Gorton - another absurdly safe seat in western Melbourne, ironically so given who its named for.

Wills - Bob Hawke's old seat and named for a fatally incompetent explorer. It actually contains one of the Greens strongest districts in Melbourne (Brunswick) but unlike Batman they missed out on second last time. Not that it matters.

Lalor - this ever growing (and ever cut back on at redistribution time) patch of western Melbourne suburbia was Julia Gillard's seat. The loss of her personal vote will have some kind of impact on the swing, but the seat will remain monstrously safe.

Scullin - solid block of solidly proletarian northern Melbourne suburbia.

Calwell -  appropriately named not just because Calwell was Labor, but because of his role in boosting the number of immigrants from non-British countries (provided they were white countries, of course).

Maribyrnong - the DLP might have re-grouped (ha, ha, ha) but they have no baleful electoral impact here these days.

Hotham - Crean's retirement and the subsequent messy preselection drama could perhaps do things with the swing, but this is still very safe.

Holt - someone should make sure that this division has a coastline. Poor taste? Anyway, a safe seat these days, but a former Liberal MP for Holt was once a Conservative MP for the old Wrekin constituency in Shropshire (Bill Yates).

Corio - the first non-Melbourne division (it's based around Geelong), though still rather safe.

Ballarat - the transformation of this seat into a safe one has been a strange thing to behold, but safe it is for now.

Jagajaga - safe seat that tends not to swing so much anyway.

Isaacs - this was a marginal not so long ago at all, but looks utterly secure at present.

Bendigo - is an open seat and as such may be unusually prone to a fairly large swing. Still, Labor's continued victories in the area at state level seem to argue against the sort monstrous swing needed to actually threaten Labor's hold here.

McEwen - boundary changes have turned this rapidly growing lump of exurbia into a safe Labor seat, at least on paper. Reality could be less friendly, but this shouldn't be close.

Melbourne Ports - one of the lowest swinging seats in Australia for various structural reasons, thus another solid (worst case 'solid enough') Labor hold is basically certain.

Bruce - there's a bit more room for a swing here, but it includes Dandenong so...

Chisholm - this really ought to be held, but if the likely Coalition victory is at the upper end of the scale of possibilities, might - just - fall. In other words, it's high on the list of possibilities for 'cold sick' losses, if it's that sort of election.

La Trobe - given that this is a natural Liberal seat gained when Labor had a Victorian leader, defeat really does look very, very likely. Of course if Rudd somehow clings to power, then a failure-to-gain here would be a 'cold sick' result for Liberal supporters...

Deakin - looks pretty much gone.

Corangamite - again...

[Filuwaúrdjan]

Polling in Tasmania is not very reliable anyway. This is because Tasmanians are good people.

[Filuwaúrdjan]

As for LaTrobe, it's in my handful of seats to keep watching before I lock in my final prediction.  It's borderline, I think.

It's the kind of seat that 'ought' to be lost (so a hold would be an 'upset') but might not be, after all.

[Filuwaúrdjan]

If they aren't named, the journalist probably made it up.

[Filuwaúrdjan]

Let's... call a halt to this line of discussion, yes?

[Filuwaúrdjan]

Victoria - Coalition & Other

Melbourne - basically this depends on how loyal Liberal voters are to their cards. So it's basically not possible to call it with any safety, though a Labor gain from the Greens is more likely on a balance of probabilities basis. Either way, an anti-Coalition seat.

Aston - this seat is prone to absurdly large swings (back in 2004 it was the 'safest' Liberal seat in Melbourne IIRC) and, given the general tone of the election, will presumably deliver one in the general direction of the Coalition.

Dunkley - the follies of one of its state MPs can hardly be welcome for the incumbent given the tight margin here, and if there is an against-the-grain result in Victoria it would presumably be here... but I don't think such a result is likely.

McMillan - should be safe enough, given the election.

Higgins - no chance.

Indi - a safe Liberal seat under threat from a Natdependent (as part of a baroque response/revenge for the Liberals contesting Mallee?). I don't know enough to make a serious prediction, but I'll call Independent gain for the sake of hilarity (Mirabella being an objectively awful human being and all that). The result will have no impact on the wider result.

Mallee - in what the Nationals presumably regard as a dick move, the Liberals are contesting this open Nationals seat. In what the Nationals presumably regard as an act of malicious trolling, Labor are directing preferences to the Liberal. The last time Mallee was open the same thing happened and the Nationals nearly lost the seat. Will tentatively call a Nationals hold regardless, as they have a very strong candidate.

[Filuwaúrdjan]

Queensland - Labor seats

Basically Queensland is confusing the hell out of me. My general thinking is that the swing (no matter what direction) is probably better for Labor outside Greater Brisbane, but that's purely because that was the pattern in the Great Electoral Armageddon of 2012. Anyway...

Griffith - Rudd's electoral fief in south central Brisbane is, on paper at least, the safest Labor seat in Queensland. There have been a few ominous/dodgy local polls here, but Rudd ought to be more than fine provided that this isn't one of those random Queensland Wipeout Years that the ALP occasionally suffers from. In the absence of Rudd - who, somewhat remarkably given his personality, has a large personal vote - this seat would not be safe at all, though would still lean Labor.

Oxley - a rather odd division that includes a random assortment of southern Brisbane suburbs - including Inla, one of only two Brisbane seats to stay Labor in the Great Electoral Cataclysm of 2012 - and the eastern suburbs of Ipswich, and is a safe ALP seat by nature (it used to include all of Ipswich and was very safe, except for 1996 - and the less said about that the better). Ought to be a Labor hold, no matter what.

Rankin - another fairly safe Labor seat in the southern suburbs of Brisbane, with strong Labor support in Logan (part of which even stayed Labor in 2012) finding itself somewhat diluted by more wishy-washy territory inside Brisbane proper. This is an open seat (which may be less than great news), but ought to be held if statewide polling tells the truth.

Blair - an uneasy mix of strongly Labor Ipswich and a sizeable swathe of rural territory. This was a comfortable Labor hold in 2010 and ought to be held again if the polls are correct. Queensland, though, is Queensland, so, you know.

Capricornia - a big slice of regional Queensland that has tended to remain pretty loyal to Labor while other divisions in that part of the world have become (at best) patchy in recent decades; the reason for the difference being the fact that the Labor vote was (is) based around the coal industry. It is occasionally lost, however, and the chances of that happening are elevated this year by the retirement of Kirsten Livermore. All things considered I'll call a Labor hold here, but this is definitely one to keep an eye on.

Lilley - there is a very real prospect of Swan losing as he did in 1996 (though the boundaries are a tad better for Labor now than then, if I remember rightly. Which I might not) and whether he does or not mostly depends on what version of political reality in Queensland is the real one. Very much in Swan's favour is the fact that Lilley is (unusually for Queensland) a low-swinging seat, and my general suspicion is that Swan will hang on through sheer bile.

Petrie - the reality problem strikes even harder here: if Labor aren't doing so badly in Queensland (even if only because the swing happened in 2010), then there's a very good chance that this seat will be held (meaning that it will he held by an opposition member for the first time in a quarter of a century). If that is not the case, then Petrie is horribly vulnerable. Genuinely undecided.

Moreton - the same problem as with Petrie, but on a significantly narrower margin. As such, it logically follows that I should predict a Coalition gain, albeit with some vague hesitancy. Can see anything from another narrow Labor hold (if the optimistic scenario is correct) to a not-that-close-at-all-actually loss.