At long last, I present my assessment of the state of play in the upper house. I always reserve the right to change my mind up until the close of polls, and that goes double here because there are no doubt subtleties that I have missed. If anyone thinks they can spot any, please raise them in comments.
Northern Metropolitan: The key test here is likely to be the combined Labor and Greens vote, which will produce three seats for Labor and one for the Greens if it adds up to more than two-thirds of the total. Stephen Mayne of People Power has constructed an optimistic alternative scenario in which their candidate Barbara Biggs gets ahead of Family First with independent and Democrats preferences, but it would seem more likely that the latter bloc will be outweighed by the combined DLP and Family First vote, so that Biggs will be eliminated and the Labor surplus will instead go to the Greens. The more likely threat to an outcome of three Labor, one Liberal and one Greens would be an unheralded 10 per cent dive in the Labor vote which failed to transfer to the Greens, in which case the Liberals might win a second seat at the expense of the Greens. This would have to be rated as very unlikely.
ASSESSMENT: Labor 3; Liberal 1; Greens 1.
Eastern Metropolitan: With Liberal and Labor sure to bag two seats each, the final place looms as a three-way struggle between Liberal, Labor and the Greens. If the combined Liberal and Family First vote is over 50 per cent, the seat will go to Liberals. Given that the Liberals scored 44.3 per cent in this area in 2002, and the Family First 2004 Senate vote here was 2.2 per cent, the Liberals will be looking good if they are picking up a swing in excess of 3 per cent in the eastern suburbs. Otherwise, the key test will be whether the combined vote for the Greens, Democrats and People Power is greater than Labor’s surplus over the 33.3 per cent they will need to win their second seat. However, Antony Green raises the prospect of a wild card outcome, noting that the contest "gets interesting" if a) the Liberal plus Family First vote is below 50 per cent, b) the Labor plus Greens vote is below 50 per cent, and c) the combined People Power plus Democrats vote is higher than the combined Family First and DLP vote. One of the first two scenarios is plausible, but the third looks very unlikely given that the combined Family First and DLP vote in this area was 3.8 per cent at the 2004 federal election. Stephen Mayne’s scenario for a win for People Power candidate Karin Orpen puts the People Power vote at a very optimistic 2.9 per cent.
ASSESSMENT: Liberal 3; Labor 2.
South-Eastern Metropolitan: This is the Greens’ weakest metropolitan region, so it seems very unlikely that they will poll higher than Labor’s surplus over 33.3 per cent, even after the addition of preferences from the Democrats and People Power (and especially in light of the Liberals decision to put them last on preferences, depriving them of any Liberal surplus). On the basis of the 54.2 per cent Labor scored in this area at the 2002 election (as calculated by Antony Green) it seems equally unlikely that Labor could fail to reach the 50 per cent they need for a third seat after they receive the Greens vote as preferences. For the final seat to go to either Liberal or the Greens, Labor would need to suffer a double-digit swing.
ASSESSMENT: Labor 3; Liberal 2.
Southern Metropolitan: Southern Metropolitan is Labor’s weakest metropolitan region there is little prospect of them winning a third seat, and a slight risk that they won’t win a second. If Labor does win a second seat, the final position looms as a tussle between the Liberals and the Greens. The Liberals polled 44.2 per cent here in 2002 and will receive preferences from Family First, the DLP and an independent candidate, which the 2004 Senate figures suggest will add up to around 3 per cent. If that adds up to more than 50 per cent, the Liberals will win the seat. The Greens vote in this region was 15.4 per cent at the 2002 state election and 12.8 per cent at the 2004 Senate election; the only boosts they will receive beyond their primary vote will be preferences from People Power and the Democrats, along with Labor’s surplus over 33.3 per cent. If that all adds up to more than 16.7 per cent, the Greens will win the seat. The alternative scenario is that Labor’s vote (which was 37.9 per cent in 2002) falls well below 33.3 per cent, giving the Greens a slim chance of overtaking Labor’s second candidate (one Evan Thornley) and winning his seat, for a final result of three Liberal, one Labor and one Greens.
ASSESSMENT: Liberal 2; Labor 2; Greens 1.
Western Metropolitan: The Greens would need a big improvement on their recent performances to win a seat here, following the Liberal decision to put the Greens last on preferences. There would need to be a substantial boost in their vote at the expense of Labor, and little improvement in the Liberal vote. If that does not occur, the Liberals could win a second seat if their vote reached 33.3 per cent after the addition of Family First preferences. Otherwise, Labor would win a fourth seat and be well on their way to maintaining their upper house majority. The Liberal vote in 2002 was 25.6 per cent, while the Family First 2004 Senate vote was 1.5 per cent.
ASSESSMENT: Labor 4; Liberal 1.
Eastern Victoria: The Coalition parties have done Labor a huge favour here by preferencing the Greens last, giving them some hope of winning a third seat. Past experience suggests the Greens vote will be in the 8 per cent ballpark, which can now be added directly to a Labor vote that was 40.6 per cent at the 2002 state election and 30.4 per cent at the 2004 Senate election. If the surplus of this sum over 33.3 per cent is greater than the equivalent figure for the combined Coalition vote (47.2 per cent in 2002, 50.8 per cent for the 2004 Senate), Labor will win the seat. However, the odds are stacked against this outcome and it is more likely that the final seat will go to either the Liberals or Nationals. The determining factor here will be whether the Nationals vote (10.2 per cent on 2002 figures) exceeds the Liberal surplus over 33.3 per cent (3.7 per cent in 2002), after the Nationals receive preferences from Family First, People Power and the DLP and the Liberals receive preferences from the Country Alliance. As such, the most likely outcome would look to be two Labor, two Liberal and one Nationals. Wild card possibilities include the dim prospect of the People Power getting ahead of the Greens with preferences from the DLP, Family First and the Labor surprlus, and then coasting to victory on Greens preferences.
ASSESSMENT: Labor 2; Liberal 2; Nationals 1.
Western Victoria: Labor and the Liberals are each assured of two seats; the final seat could go to Labor, Liberal or the Greens. Labor’s vote here in 2002 was 47.3 per cent; if this falls by less than 5 per cent, they are very likely to win the final seat. The Greens have a narrow window of opportunity: they need for the combined Labor and Greens vote to be near or above 50 per cent, and for their own share of it to be greater than Labor’s surplus over 33.3 per cent. On 2002 figures, Labor and the Greens respectively polled 47.3 per cent (a 14.0 per cent surplus) and 8.3 per cent. Alternatively, the Liberals might reach the 50 per cent needed for a third seat after receiving preferences from the Nationals, Family First and the DLP. The Coalition vote in 2002 was 42.3 per cent, while preferences should boost them a further 3 to 4 per cent. The Nationals vote in this region is too low to give them a serious shot of squeezing out the Liberals. There are a couple of wild card scenarios here, one of which involves the DLP snowballing ahead of the third Liberal with preferences from Family First, the Country Alliance and People Power.
ASSESSMENT: Liberal 3; Labor 2.
Northern Victoria: The Coalition will surely improve upon its combined 49.1 per cent vote in 2002, and are thus assured of three seats. Barring a Nationals collapse, they are all but certain to win one of these three seats. The question is whether Labor can manage a second. They polled 37.4 per cent, and will only receive preferences from the Greens and half those of the Country Alliance. Even so, their vote would need to fall by about 10 per cent if they were to fall short. A possible wild card is celebrity chef and independent candidate Stefano di Pieri. If his vote breaks the 5 per cent barrier, he might get ahead of the Greens and pull off a miracle.
ASSESSMENT: Labor 2; Liberal 2; Nationals 1.
Add that all together and you get a final result of Labor 20, Liberal 16, Nationals two and two for the Greens. Look back later though and you may find I have changed my mind.
UPDATE: Be it noted that my original assessment of Liberal 3, Labor 2 in Southern Metropolitan has been withdrawn.