EMRS: Thirty Or Turkey By Christmas?

EMRS Lib 26 (+1) ALP 23 (-1) ON 21 (+2) Grn 16 (+2) IND 14 (-2) others 1 (-1)

Seat estimate (new boundaries) Lib 10 ALP 9 ON 8 Green 5 IND 3

The latest EMRS poll is out as released on the Poll Position pod and the changes on state voting intentions are all within the in-theory margin of error from the previous edition.  There would be mild relief for the government that the dramatic slide in its numbers in the past few polls has stopped.  But political tragics could only have eyes for the ALP figure after leader Josh Willie was reported to have set a target of polling a 30% primary by Christmas.  (Something I somehow, in a feat tantamount to winning the Last Man contest for not knowing the Superbowl winner, managed to not find out about for a month!)  

This figure has led to predictable jibes from opponents that not only will Labor not attain this heroic goal (oh for the days when they were forever stuck around 30) but that Willie will instead be gone as leader by Christmas. Although Labor's primary has actually held up better to the One Nation surge than might have been expected, there seems to be an unusual level of Labor leadership chatter about.  It's hard to tell if it is serious or if a lot of people are just and quite rightly bored.  Some of it is connected to the issue that continues to dog (pun intended) the Opposition, with a Thick Of It style nadir when Willie was photographed defending Labor's unfashionable support for greyhound racing in front of a Dogs Prohibited sign. 

Ella Haddad has made it as clear as she's allowed to that she would put support for racing in the bin. Some may see her as the leader most likely to put the Rockliff Government in the bin via a mid-term transfer on the floor as well.   Willie took the leadership from Winter by threatening to force a member ballot where being from the left to Winter's right would have greatly advantaged him.  If there were to be a challenge from the left the same logic does not apply.

On the right of the party Shane Broad has been talkative well beyond his podcast and was thought to be flying a flag of some sort with noises about another unfashionable policy, the long-proposed kunanyi/Mt Wellington (see what I did there) cable car.   Frankly if it was possible to build the thing viably in a way that is sensitive to the aesthetics of the mountain as judged by the residents of South Hobart amd surrounds this would have happened in, oh, 1933, so he's clearly not playing to the New People's Republic of Clark.  The EMRS poll included a leadership question asking respondents to pick their preferred Labor leader out of Willie, Dean Winter, Haddad, Broad, Janie Finlay and none of the above, but I suspect that one's been held back for the next round of Poll Position commentary.

In view of Willie's 30% target, it's amusing that he himself has jumped from 25 to 30 as Preferred Premier, with Jeremy Rockliff down seven points to 37.  37% is in fact Rockliff's lowest Preferred Premier score so far, though seven points is far from his smallest lead.  (He even trailed Rebecca White in one poll).   That really doesn't fit a picture in which voters see Willie's leadership as circling the drain and observers who were committed to that prior will probably now take it as saying something very bad about Rockliff.  But what if anything in particular, and why now?  Net favourability polls just suggest voters have seen too much at the moment of the lot of them, with Rockliff down 11 points to net -7, Willie down 10 to net -14 and Rosalie Woodruff down 11 to net -16.  The poll also asked about voter impressions of the competence of Liberal ministers Felix Ellis, Bridget Archer, Eric Abetz and Guy Barnett; these numbers too have yet to surface.

Seat projection

In my article on the recent DemosAU poll I attempted to project that poll to possible seat results under the now finalised new electorate boundaries.  The boundaries will be official for the next federal election after 9 October this year, but their use at state level is approved by special legislation that is likely to take well into next year.  So depending on when the parliament next collapses, we don't know for sure that the new boundaries will be in play, but I'm assuming for the moment that they will be.

By adjusting my modelling for the DemosAU poll I came up with this as a possible projection, noting that the Others vote of 1% in this EMRS is likely to be unrealistically low:

The only difference from my read of the Demos poll is that I now have the final seat in new Clark as in doubt between the Greens and the second Independent.  This is on account of Peter George intending to serve only one term though it's not impossible that much of his movement's support might transfer to a new independent.  Plus polls tend to overestimate Independent support anyway.  At the moment I suspect some of this independent vote would scatter and that the Greens' statewide strength in this poll is strong enough for two seats in neo-Clark, though they would have to work on the profile of their second candidate.  I'm also still not totally sold on One Nation being good for a seat in neo-Clark; if they run a bunch of no-names they could easily be destroyed by within-ticket leakage.

My Clark projections continue to highlight Labor's problem - they have Willie, Haddad and Winter all with territorial claims on the new Clark but it is such an awful seat for them at recent state level that they are at big risk of only returning one member.   This may well change by election time but it does add extra value to the leadership.  The other thing this model continues to point to is the potential for a combined if narrow Liberal/One Nation seat majority.  By the next election things may be extremely different but that is an interesting live prospect.  

Federal

The poll also released federal figures (last seen in February).  There was similarly not a lot of movement, with Labor on 29 (down 1), Liberals 17 (down 1), One Nation 26, Greens 15 and independents 13.  Respondent-allocated two-party matchups had Labor leading the Liberals 57-43 and One Nation 63-37.  81% of respondent One Nation preferences flowed to the Liberals vs Labor but only 53% of Liberal preferences flowed to One Nation vs Labor.  The latter is likely to be an underestimate because of Liberal how to vote cards (assuming the Liberals do preference One Nation), but in Tasmania this isn't as big a factor as elsewhere.  The podcast reports that ALP vs One Nation is neck and neck in Braddon and that both it and the classic 2PP are very close in Bass; my eyeballing of these statewide primaries is that One Nation could win both if it managed to keep its primary vote this high and found good candidates.  On the Senate front anything like these numbers would be a straight 2 Labor 1 Liberal 2 One Nation 1 Greens.