The New Pendulums
Posted by Possum Comitatus on January 7, 2008
The AEC has returned the writs, so we now have the complete results for the election.
First up, we can have a squiz at the national pendulum as well as the respective state based pendulums. They’re thumbnails so just click on them to blow them up:
What they show is the number of seats that would fall for a given uniform swing to either major party. To use it, pick a swing on the vertical axis, trace it across until it hits the black line, then trace it up or down to the horizontal axis to see how many seats that swing would deliver to the major party. The national pendulum also has the swings for seats blown up in the top left and bottom right quadrants for swings of less than 7% to make it easier to read in terms of the seats that would change hands for the usual smaller swings.
The other States and Territories are best given by a table:
With the Victorian (and national) pendulum, there’s a bit of a quirk involved with the seat of Melbourne where Adam Bandt from the Greens came second in TPP terms behind the ALP. As a result, Melbourne is sitting on a margin of only 4.71% for the ALP. However, if there were any swing to the Libs of a percent and a bit or more in Melbourne, the Libs would probably end up running second in TPP terms, forcing most of the Green vote to flow to the ALP via preferences and blowing the ALP margin out to over 15 points as result. So it’s best to keep that in mind. These charts will be archived and accessible from the top menu via the 2008 Swings for Seats button.
An interesting thing happens when we compare how many seats each party would gain from a given uniform swing. If you click the thumbnail below, it blows up into a chart that shows the number of seats each party would expect to gain from a given uniform swing.
What this means is that the Coalition gains less seats than the ALP from a given swing. For instance, the Coalition would need a uniform 1% swing to gain 5 seats whereas the ALP only requires a uniform swing of 0.21% to gain 5 seats. If we look at the swing needed for each party to gain 5, 10, 20 and 30 seats respectively we get:
|Seats Gained||ALP Swing%||Coalition Swing %|
There’s an electoral redistribution due before the next election so this may change a bit, but if the current national pendulum holds it rough shape, the Coalition will find itself in a rather difficult situation, having less fat in the margins of their 30 closest held seats than the ALP does in theirs. One would think that the next election campaign would be an old style battle for the marginals, with the ALP being in the prime position not only because of the pork barreling that government can supply, not only because the ALP will have more to spend as a result of them being in power everywhere, but also because they face the biggest seat bang for the swing buck.
To make it worse for the Coalition, there are 3 seats that the ALP could gain in Qld (the new centre of the government) with a swing of only 0.21%, and in WA where the Coalition hold all but 4 seats, 3 seats could fall to the ALP with a swing of 1.71%. If I were the Coalition, I’d be extremely worried about any potential government consolidation in Qld and a rebalancing of the seat numbers in WA. The ALP can gain seats by basically standing still, having only very small uniform gains in some states delivering the goods. The Coalition on the other hand, as you can see from the chart, find themselves in a much poorer position.