Whose preferences do the Australian Greens need the most? Friday, March 5, 2010

The other day I looked at which party would benefit most from Green preferences in the senate at the next election. The result was that of the three battleground states simulated, Green preferences only really matter in Western Australia, where they look to play a decisive role in whether the Australian Labor Party (ALP) or the Liberals/Coalition pick up a third seat.

Today I am going to reverse the situation and ask the question: Whose preferences do the Australian Greens need the most?



Results
From the simulations run, the results are as follows: in New South Wales (NSW) the Greens would be better off with ALP preferences, in Queensland they would benefit more from Liberal preferences, and in Victoria (VIC) and Western Australia (WA) preferences from the major parties make no material difference to whether a Green senator gets elected or not.

Methodology
Using the 2007 election as a base template for my simulations, I update the votes received for each party based on the latest aggregated state-by-state newspoll surveys. With this basic 2010 scenario in place, I run 10,000 simulations per state with random preference allocations. It is my theory that since no single candidate has control over how all other candidates allocate their preferences, for most intents and purposes it can be said that preferences are allocated randomly at a senate election. This is backed up by the results that show randomly allocating preferences provides a fairly accurate simulation of existing election results.

In today's exercise, once I have the results of a senate election with random preferences, I run three new batches of 10,000 simulations. One where the ALP preferences the Greens, one where the Liberals preference the Greens and finally one, for comparison, where the ALP and the Liberals preference each other.

New South Wales

With the Greens sitting on a 2.6% swing of the vote in NSW, they win a seat in 3,554 simulations (35.54% of simulations). The state splits 3/3 with majors in 62.92% of simulations. However, with ALP preferences the result swings markedly towards the Greens, with the Greens picking up a seat in 51.1% of simulations. This is in start contrast to the when the Liberals preference the Greens, with the result essentially the same as random preferences. In that scenario, the Greens win in 35.05% of simulations. When the majors preference each other, the Greens fall to winning only 21% of simulations, while the majors splitting the six seats rise to 78.2%. ALP preferences in NSW would almost certainly guarantee the Green's lead senate candidate, Lee Rhiannon, a seat.

Victoria

In Victoria, the Greens are sitting on a very solid 3.7% swing which, if it holds through to election day, will virtually guarantee them a seat. In the random preference allocation, they already win a seat in 98.6% of simulations. With ALP or Liberal preferences, this rises to over 99%. Even if the majors try to freeze out the Greens, on this level of vote candidarte Di Natale still wins the seat in 98.15% of simulations!

Queensland

In Queensland, it is Liberal preferences that help the Greens candidate Larissa Waters the most. In the random preference, ALP and majors-swap scenarios, the Greens win in about 72% of scenarios. Only in the Liberal scenario do we see a significant change, with the Greens winning a seat in nearly 85% of simulations.

Western Australia
Western Australia is shaping up as a really interesting battle this election, with the Liberals in danger of losing their traditional third seat to the ALP. With a 3.5% swing towards them since last election, the Greens are sitting pretty with a near total lock on a seat for their candidate Rachel Siewert (93.66%) ... indeed, their preferences may decide the final seat for the other parties. When the ALP preference the Greens they win in 94.84% of simulations, with Liberal preferences, 94.66%. When the majors preference against the Greens, they still win in 90.85% of simulations.

Conclusions
The Greens do not really need preferences from the majors in Victoria or Western Australia. They will probably coast in on minor party preferences. However in Queensland, the Greens would really benefit from Coalition preferences more than ALP preferences. In New South Wales, they would benefit more from ALP preferences than ones from the Coalition. In Victoria and Western Australia, this suggests that the Greens could focus on beneficial preference deals in the Lower House at the expense of their senate tickets in those states without costing them those two seats.