THERE have been many howls of outrage following the publication of the Sunday Examiner EMRS poll suggesting the Labor vote is strengthening in the final weeks of the campaign.

The Greens and Liberals have derided the poll as inconsistent with the feedback they have received on the ground. Well, surprise, surprise.

Not only do they have to say that (because some voters will vote Labor only if they think Labor can win outright), but political candidates in general are remarkably deluded creatures. Many of them, especially those of the more idealistic variety, just don’t realise that by the end of the campaign they are mainly hearing nice things about themselves because most voters are either giving them lip service or else saying nothing and waiting for them to go away.

An even stranger contribution came from Bruce Montgomery on this site (Rat?). According to Montgomery, the Sunday Examiner “merely told us what the likely split of seats was in each electorate, gave us some figures for Lyons and Denison but all we know about the other three is that the Greens ‘are well behind’”. That’s odd because my copy of the paper in question includes, in the reports for each electorate, seat by seat breakdowns for the three parties in Franklin (47-33-20), Bass (44-38-18) and partially Braddon (48-??-14 –since the “other” vote is negligible in the poll as a whole the ?? is 38 or just below).

Effectively almost the full breakdown has been released (the breakdowns for Denison and Lyons appear to be 49-23-25 and 54-22-23 respectively). Not only does Montgomery’s conspiracy theory therefore have no basis in fact, but even if the figures were omitted his argument would still be groundless — the 7-8% swing from the Libs back to Labor in Lyons is not suspicious but on the edges of the margin of error for a 200 vote sample even assuming no change at all in voter intention.

Comparing the seat-by-seat breakdown given by EMRS with the combined TasPoll/EMRS figures I provided previously (Comparing polls and papers) some remarkably clearcut patterns are apparent.

Labor is up in every single electorate, by between four and seven points. The Liberals are down in every single electorate. While it is always possible that the EMRS sample is a rogue poll and its figures are a load of total baloney, there is a far more obvious explanation for the differences between this and previous recent EMRS polls. The proportion of undecided voters has dropped to 10% (it was 20% in May and August last year, 17% in November and 23% this February). Most likely the Government has picked up a boost as voters who have been unsure of their votes have decided, as they usually do, to stick with what they know. There may have also been a small drift back among voters who had previously formed a view, possibly as voters either realised the health issue was far from straightforward or became concerned about just how unstable another hung parliament might be.

(As an aside I don’t agree with Wayne Crawford’s point about stability of governance being a non-issue for voters. Voters probably just don’t think of it as an “election issue” when the question is put to them, even if it nonetheless influences their vote. Furthermore while it is true that a hung parliament is not an issue if the Liberals put principles ahead of power, nobody really expects a politician to do that.)

The latest EMRS poll shows a statewide swing of +6.2% to Labor, -3.8% to the Liberals and –2.4% to the Greens since the combined EMRS/TasPoll results I calculated in this article (Comparing polls and papers). If this swing was applied uniformly to the seat-by-seat breakdowns it would result in only two Labor seats being at risk — Braddon (which they would probably retain over the Greens) and Franklin (which would be very close with the Liberals).

The following is my current electorate-by-electorate view of this election (with ferret jammed firmly in neutral in case any polls in the final days produce any further surprises.)

Bass

Bass is a no-brainer and always has been since Kathryn Hay decided not to recontest. It will split 2-2-1 and the only issue of interest is whether either Peter Gutwein or Sue Napier might be beaten by David Fry. I think this is relatively unlikely although Fry should poll strongly. Gutwein’s prospects are very relevant because he was treated extremely harshly by Rene Hidding early in the last parliament and may be a potential rat should Labor win 12 seats and be looking for an independent supporter.

Braddon

The Sunday Examiner reports a breakdown of 48-38(see above)-14 and on this basis claims Green Paul O’Halloran has an “outside chance”. This is understating how close this seat actually is. For instance a move of two points from Labor back to the Liberals — a totally meaningless move in the context of a 200-vote sample, would very likely give the seat to the Greens. Even 48-38-14 is not a certain Labor win since leakage from Bryan Green’s likely surplus could harm them. The outcome will be either 3-2-0 or 2-2-1 (with either Kons or Best, probably the latter, at risk) but will depend on exactly how close to three quotas Labor can manage to get. I still think this seat is waiting to be claimed by the Greens one day as soon as they can get a favourable distribution between the other parties.

Lyons

Even within the limits of the small seat-by-seat samples, Labor continues to poll astonishingly well in this electorate. The problem for the Liberals is their leader. While the Liberals’ new candidates, Page and Howlett especially, will probably poll rather well, preferences of minor Liberal candidates will be burned up on trying to get Hidding over the line. Hidding only just retained his seat last time. One of the few things that can be gleaned from EMRS’ candidate samples with any reliability is that he is still not a popular man in the electorate despite his profile. If any of the new Libs do get up it will most likely be at their leader’s expense (Hidding losing his seat has not been flagged anywhere but stranger things have happened) and I expect this electorate to most likely stay 3-1-1 with 2-2-1 the only other likely result.

Denison

Labor’s revival in this electorate has been remarkable — from lagging around the two-quota mark it is now pushing towards three — an increase that is outside the margin of error for the EMRS polls and is therefore probably meaningful. The Greens have also crashed by the greater portion of a quota, but the 36% they were polling in a previous EMRS poll was never believable anyway. Most likely their real support level in Denison is in the mid to high 20s. For reasons discussed in my previous articles in this column, that is probably not enough, especially not if Labor’s vote is in the mid 40s but only one of their candidates (probably Sturges) gets over the line quickly. I now think it is most likely Denison will remain 3-1-1. 2-2-1 and 2-1-2 are also possible but looking less likely by the day.

Franklin

The EMRS poll distribution (47-33-20) is flagged by EMRS as a likely 2-2-1 result. Again, at that sample size it is impossible to say much. Furthermore Labor can easily beat the Liberals in Franklin from slightly behind on the basis of Green preferences and reduced exposure to leakage (virtually all Labor’s votes will go to Lennon, Wriedt and Giddings but the Liberal vote after Will Hodgman will split messily between several candidates making leakage a serious issue.) This seat will be 3-1-1 or 2-2-1 but I still have no firm view as to which.

Overall

Throughout the campaign I have thought that whatever the polls said otherwise, Labor would most probably lose either one seat or two, hence either just retaining their majority or just missing out on doing so. This remains my view and the latest EMRS poll is consistent with it. I am currently expecting 12-13 Labor, 7-9 Liberal, 4-5 Green, and think Labor has a very serious chance of retaining its majority. I may add more comments and hopefully a predicted exact breakdown using the comments section for this article between now and Saturday.

Kevin Bonham is an ecological consultant, election scrutineer and psephelogical hack. He is looking forward to Saturday night in the tally room, based on his experience last time in which he was pointedly and offendedly ignored by a staffer for a certain Liberal Senator and winked at for no apparent reason by a former Premier!