There’s a poll of sorts, but it would be a bit of a stretch to give it its own headline:
Roy Morgan has targeted a micro-sample of 200 voters in the crucial Victorian seat of McEwen, which could provide Labor with a desperately needed gain to offset losses in Queensland and New South Wales. Certainly that’s what the poll’s headline figure shows, with Labor leading 55.5-44.5, but the margin of error is approaching 7 per cent.
Peter Tucker at Tasmanian Politics has full results from the weekend’s EMRS poll from Tasmania, which pointed to a statewide 4 per cent two-party swing to Labor from primary votes of 43 per cent for Labor (unchanged on 2007), the Liberals on 34 per cent (down four) and the Greens on 20 per cent (up six). The sample on the poll is about 1000, with a margin-of-error or about 3 per cent. As usual, 200-sample breakdowns of each of the state’s five electorates are provided, and for what they’re worth they show Labor enjoying the full force of the swing in marginal Bass and Braddon.
Laura Tingle of the Financial Review wrote yesterday that more seasoned sections of the Labor camp believe they are just ahead and will fall over the line. This confidence was partly inspired by a conviction the party would be better placed to sway late undecided voters in the wake of a Labor launch which, Tingle accurately predicted, would seek to maximise the government’s apparent economic conservatism as it launches TV ads that portray Abbott as too big a risk to the economy with the world economy still shaky.
Peter Kerr of the Financial Review reported yesterday that Labor insiders in Western Australia were growing confident they were ahead in up to three (WA) marginals – Liberal-held Canning as well as Swan and Hasluck. The result in each was thought likely to come down to between 500 and 600 votes. The report also noted the significance of John Howard holding a fund-raiser for Canning MP Don Randall this week.
Simon Jackman in The Australian discusses the potential for the election to follow 1990 and 1998 in denying victory to the party with the greater share of the two-party vote. He also observes the disconnect between bookmakers’ odds on the overall result, which point to a clear Labor win, and individual seats, which point to Labor falling one seat short of an absolute majority.
UPDATE: Not sure if it’s already been linked to, but Essential Research has published state breakdowns from a combined three weeks of polling. The results are in line with other polling with the striking exception of NSW, where the swing is said to be 6.7 per cent. However, notwithstanding that Essential says more detailed weighting has produced slightly different estimates than simply averaging the published weekly results, I’m finding the state results hard to square with the reported national swing of 1.7 per cent. Weighting the averages for population puts it 1 per cent higher.
UPDATE 2: Essential Research have found an error in their state breakdowns, and revised NSW, WA and SA 1 per cent in Labor’s favour. So it’s now 5.7 per cent and 3.4 per cent against Labor in NSW and Queensland, and 0.7 per cent, 1.6 per cent and 0.3 per cent to Labor in Victoria, SA and WA.