Antony has the new pendulum for 2010 over on his site. What I thought was interesting about it was the chunkiness of the swings for seats for each party around the center of the pendulum – say, a 6% uniform swing each way. If we take Labor first and chart the seats they would gain with a given uniform swing (and note the seat number this would be for them in a new parliament) we get:
The 99th ALP seat would become Hinkler from a uniform swing to Labor of 1.5%. But to get the 100th seat – Boothby – the uniform swing required is 2.9%, nearly twice the swing of the 99th seat. That’s a lot of uniform swing for no action between 1.5 and 2.9.
For the Coalition we have:
For a very small swing, the Coalition picks up relatively large numbers of seats. Some of these are seats currently held by the Coalition anyway, but have been theoretically moved into the Labor column because of the redistributions. A 1% uniform swing would deliver them 70 seats in Parliament. To gain government outright, a swing of 2.3% is needed.
What’s worth mentioning about this, is that the broader electoral system still slightly favours the Coalition after the redistributions – with the Coalition needing 49.6% of the two party preferred to govern in their own right, 49% to govern with the support of the independents (assuming uniform swings).
You’ll also notice that Bowman is in each chart – that’s because nominally Bowman is pretty much a tie at the moment where any swing of a few tenths of a percent brings it nominally into that party’s column.
UPDATE:
Here’s the new pendulum as an image, with the seat name, the State it’s in, the margin (red for ALP, blue for Coalition) and the nth seat that would be for the ALP in any new Parliament assuming a uniform swing. You’ll notice that only 147 seats are listed – this is based on the assumption that the three independents will win regardless of what happens elsewhere.
31 Comments
This is more like it.
If an election were held today (or at the time of the last poll) what would the result be?
And is anyone else bemused by seats like Batman and Grayndler – ultra safe. I always get the feeling looking at these pendulums that even if Labor lost every other seat, they would always have those two.
Ah… looks like someone asked the same question on Antony’s site (and Antony answered it) (link):
Alexander,
I’ve just updated the post with an image of the new pendulum. Based on the current phone poll average of 57.5% to Labor, that’s a 4.8% uniform swing which would theoretically deliver 109 seats.
Excellent as usual. Tried to print it, but no go, any possibility of being able to have a printout.
Thanks
Geez a 99-48 win for the ALP seems eminently achievable…
Hey Poss, given the redistribution and current polling, do you see the LNP diverting resources to firewall their own safe seats so as to not put themselves too far behind for the 2013 election?
Gravel@5:
Have you tried just copying the graph, pasting it into a document, and printing that?
Sounds like the news.ltd strategy…
Very nice Possum (as usual).
In your view, is this usual, unusual, or just a quirk of the numbers? (I am particularly interested in this since if one were to do a similar exercise for the NSW state pendulum, there are very few seats that would change hands, even on 4% swing)
It would be a fitting end to Malcolm Turnbull’s political career if he was to follow John Howard’s lead and be run out of his own electorate while at the head of the party. I used to live in Wentworth though and I couldn’t imagine it ever flipping.
Don@8 Thanks, it worked.
deconst
I live in Wentworth and have to agree. However, in my view, Labor needs to think a bit strategically about the candidate to be selected, who could potentially appeal to the electorate here – economically conservative, socially liberal – and probably have more than one election cycle to have a go. I thought George Newhouse was a shocker last time, even against a formidable Liberal candidate in Turnbull; poor media skills and the campaign turned into a circus…
I am not sure that parachuting a “celebrity” candidate would do the trick either (some talk of Dr Phelps). Suspect someone with profile, willing and able to put in hard yards for quite a while, a post 2010 redistribution might improve Labor chances.
But having said all that, real hard to see it shifting, so absent some absolute Liberal collapse, or a “campaign of distraction” to eat up Liberal resources, may not be key target for Labor for a multi-electoral-cycle commitment
Laocoon,
It’s just a quirk of the numbers. Sometimes you’ll end up with a dozen seats changing hands if 1% of voters wake up on election morning with a bad mood, sometimes – like NSW – you need big swings before a significant number of seats start falling.
I’ve often wondered what a hypothetical election campaign would look like if every seat in the House sat on a margin of less than 5%.
Thepro went:
I’m not brave enough to hazard a guess! The Libs at the moment aren’t even doing the basic 1 percenter type stuff – those really little things that in isolation don’t achieve diddly squat, but collectively can be the difference between winning or losing a seat or two.
For instance, why is Fran Bailey (the Member for McEwen) still sitting in one of the very few seats behind the Opposition dispatch box in Question Time that gets those members regular media exposure, when she is retiring?
What will be interesting at the next campaign though is how much dough the Libs will be willing to pump into Sturt to try and save Christopher Pyne.
Thanks Possum
ANd will they talk him up as much as they have tried to talk up Peter Dutton…
Interesting stuff, but let’s not get carried away with how good this is with the accuracy of an assumed uniform swing.
Good for illustration of a state of play and is probably enough to put seats generally into the ‘go either way’ or ‘win/loss’ column at present.
But the ‘go either way’ seats will be quite large. We can’t really look at this on a seat by seat basis, but rather assume that the acutal non-uniform swing will even out somewhat.
Basically, we know generally labor will win and pick up more seats than they lose unless things start to change really quickly. We have a reasonable idea of what seats to watch… but functionally it means very little in Australian politics. Labor will get past 50% of the seats in the lower house. It’s all about the senate, which we know much less about.
This is not a critisism- I’ve mentioned this before and it seems Poss is getting near the best out of the polling numbers we have without fancy (perhaps dodgy) assumptions.
Still for I’d love to see you try!
poss @ 14 – I’d love to see safe seats on both sides eliminated and all seats on less than 10% margin
we might then get better representation
Possum
How would you rate Labors chances in Canning? How likely would it be won on a lower national swing to Labor but higher in WA as labor did so badly last time? Labor would need 51.1 in WA to win Canning as u prob already know. Labor clearly has a stronger candidate this time and did well in 2001. But is that enough or can you see any other bolters based on state statistics.
DR,
A friend of mine who runs corporate end Geographical Information Systems education and development training programs recently used the 2007 election results at the booth level in Qld as part of his course. One of the assignments he created was to build 25 hypothetical electorates in Qld where the two party preferred vote was between 47 and 53 for each party – although this is a US company operating in the US,so for the people undertaking the course it was an exercise in shape fitting to a given set of population parameters that nearly all of them wouldn’t have had a clue about.
Apparently the shapes of the electorates you had to come up with to fullfill the brief were pretty quirky! It made the old gerrymanders in Qld, and the US Congressional redistricting farce look like an exercise in ‘fair and balanced’!
Scotty,
It depends on what the WA vote as a whole is doing as we approach the election proper.
The state level vote estimates in the polling of federal politics is the best predictor of how any given seat will go at the election, and WA appears to be swinging to Labor at the moment.
If the ALP keeps that up, then Canning will be in serious danger of falling – but at the moment, that’s still a big if.
One of the things in Canning that’s probably running against the Libs is the that Don Randall is a product of his geography *combined* with his generation and Canning has been having a bit of population turnover among relatively younger age groups that could take away from his generic vote.
But WA has also been a very conservative centric state at the federal level for a decades, so if there is a movement away from the Coalition at the next election at a level that the polls are currently suggesting, it would go down as a bit of a political realignment moment.
So I’m a bit wary of the polling figures in WA at the moment – not in terms of whether they are accurate or not (because good polls are generally accurate – people at the moment *are* saying they will vote for Labor in large numbers), but in terms of whether there will be some significant gap that might occur between the stated (now) and revealed (at election) preference of voters in WA.
But if the polling in WA stays the same as it is for the next few quarters, it’s hard to see how Don Randall in Canning will hang on.
Election calculator updated
http://www.abc.net.au/elections/federal/2010/calculator/
Cheers possum, quite Insightful. it shall be interesting to see what happens WA.
Possum: In the past you have stated that Higgins may become a marginal seat. I believe you thought the Libs might see a weakening of their hold at the next election and a possible win by Labor in 2014. Was their thinking correct not to contest the next election?
And, thanks to the extraordinarily, well, extraordinarily unknown figure, Clive Who? Whom the Greens, in their infinite wisdom, have decided to run, namely one Clive Hamilton. Makes this despairing voter ask you: Couldn’t this choice just increase the vote to the Libs? For if there is anything which can be said about the citizens of Higgins, the words environmentally angst-ridden do not spring to mind.
In other words, could Labor have effed up their chances for the next fifty years, in the House of Higgins?
Poss, is that pendulum based on country wide swing, or adjusted for known state swings?
Would it look much different if it was the latter?
That pendulum is based on the known results of the 2007 election. As for what happens at the next election, as yet there are no known swings, national or otherwise.
I know it’s not an original point (see Antony’s excellent analysis over at his blog for instance) but the next election really has to be all about ‘renewal’ for the Liberals. Amongst the now notionally Labor, very marginal and at risk seats are far too many of the younger Liberal party members with any hope of becoming the future of the party. At the same time the has beens like Bishop and Ruddock are standing firm, goodness knows why, in the blue ribbon seats that are safe as houses even on the current extraordinary polling figures.
As much as I’m shedding no tears at the plight of the Coalition as such, we need an effective and constructive opposition and denuding the Liberals of what little talent they have left is not going to help make one appear anytime soon.
The Ruddocks and Tuckeys and Heffernans should be politely asked to leave, and if that fails – disendorsed, and the rising stars facing the axe moved in to their seats instead.
The Libs need to prune the deadwood and nurture new buds.
2% swings, 4% swings, bah! I live in the seat of Mallee which has a 21% margin. I wonder what sort of magical candidate could be put forward by Labour or the Greens to create some change in this part of the world. I always enjoy the election night coverage except that is always one of the first seats Antony gives to the Nationals.
Jesus himself might do the job Dino, but he’d still have to go to preferences in Mallee
I live in Berowra, and this area will be represented by a Liberal forever. Labor got something like 42% 2PP in 2007 – that’s about the best they can achieve here, there are too much rusted on Liberal voters and over 50 year olds(who are naturally conservative). And although I despise Ruddock, we could get a Liberal candidate far more hard right than him.