Resolve Strategic is publishing monthly Victorian state poll results for The Age ahead of the election in November, padding out the sample in its regular national polling and inquiring about state as well as federal voting intention. The latest result finds Labor and the Coalition remaining tied on the primary vote, with both up a point by 27%, while One Nation are down two to 22% and the Greens are steady on 12%. Jess Wilson’s lead over Jacinta Allan on preferred premier shifts from 39-20 to 43-24. The poll was conducted last Tuesday to Saturday from a sample of 1000.
There is also a report in the Herald Sun on a 6500 sample poll conducted for Trades Hall by RedBridge Group which has One Nation on 27%, Labor and the Coalition each on 26% and the Greens on 12%. The polling “includes low, median and high scenarios for seats likely to be obtained by each party, and Labor is understood to be able to scrape across the line with its number of seats in the mid-40s on a ‘high’ sensitivity scenario – but could fall into the 20s under the ‘low’ scenario”.
There was a major development over the weekend with reports the government will shortly legislate to abolish group voting tickets for the Legislative Council, which it has taken a confoundingly long time to get around to despite the urgings of a parliamentary committee inquiry in December. The reports are scarce on details, beyond that fact that the reforms follow other jurisdictions in allowing above-the-line voters to number multiple preferences, rather than voting for a single party and accepting its full nominated preference order. The existing system helped two parties win seats from less than 1% of the vote in 2018, and has recently encouraged a rash of registration applications by parties plainly devised with the intention of harvesting preferences. Antony Green noted a fortnight ago that the Victorian Electoral Commission has advised a bill would need to be passed at the next sitting of parliament later this month to take effect in time for the election.
UPDATE: Further results from the poll show 70% support for a royal commission into corruption in the building industry, which Jacinta Allan is resisting, with 6% opposed.
Trent – I thought I’d include Redbridge’s commentary again with the percentages as it does seem to accurately describe how either major party can win with a low primary vote.
I’d also be interested to see how the primaries from the recent DemosAU and Freshwater polls come out in your model.
As expected with the lowest ever primary vote Labor have recorded, the DemosAU one (ALP 21, LNP 30, GRN 15, ONP 23) results in an absolute bloodbath for Labor. The model gives me:
27 ALP
43 LNP
6 GRN
7 ONP
5 Tossups
That would likely be a very narrow Coalition majority, and the 15% GRN primary vote in that poll puts Northcote & Footscray in the Greens column.
Using the Freshwater poll (23 ALP, 27 LNP, 14 GRN, 25 ONP) it gives me this:
32 ALP
34 LNP
5 GRN
10 ONP
7 Tossups
That almost guarantees a Coalition-ONP minority regardless of how the tossups land in that scenario. Again of course noting that’s just what that very blunt model spits out with all its flaws, not an actual prediction.
On those numbers with Greens -1 and Labor +2 compared to the DemosAU poll, the Greens still gain Northcote but Footscray returns to an ALP/GRN tossup.
I think DemosAU was an outlier though and I don’t see Labor getting anywhere near a 21% primary vote. I think when all is said & done, the Labor campaign machine kicks in, and the prospect of a Coalition-ONP minority starts scaring some moderate or multicultural ALP/LIB swing voters whose vote is currently polling with the Libs, Labor will end up with a primary vote somewhere in the 27-30% range.
North-East Metro in particular is a region where the Redbridge breakdown is pointing at a HUGE swing away from Labor, but there are some very multicultural communities in that region that traditionally leaned Liberal, swung hard to Labor over the 2018-22 cycles, have probably returned to the Liberals in these polls, but would probably be among the most easily frightened back into Labor’s arms as the prospect of One Nation being part of a Liberal government becomes more real as the election nears (if polling still points to that) and a lot of campaigns focus on that.
Thanks, Trent. Very interesting to see how the full range of polls come out through your model and shows how damaging a high One Nation primary can be to both majors, including to the Coalition in DemosAU despite their dominant primary vote in this poll.
It will definitely be interesting to see how the campaign plays out, including in N-E Metro, where the Liberals will need to maintain their current polling to offset being excluded from ALP-ONP contests elsewhere in Melbourne.
Interestingly, further details from the most recent Redbridge poll have been “leaked” and are now reported in the AFR, presumably with the intention of driving voters away from One Nation. The poll is an MRP and gives the following seat predictions as median (range):
– ALP – 35 seats (25-44)
– L/NP – 39 seats (28-50)
– ONP – 12 seats (7-19)
– GRN – 2 seats (0-5)
Of particular interest:
“The Trades Hall-commissioned poll showed Labor was likely to finish in the top two on election night in 72 of the 88 seats; One Nation and the Coalition were likely to finish top two in 49 seats.”
This is consistent with Redbridge’s previous commentary and shows how the Coalition is benefiting from preferences, but still needs to finish in the top two to win enough seats. Compare:
– ALP – 35 seats from 72 contests
– L/NP – 39 seats from 49 contests
– ONP – 12 seats from 49 contests
Preferences are clearly damaging both Labor and One Nation when they finish in the top two.
The Coalition will be clearly disappointed to only be finishing in the top two in 49 seats, though also reassured that they can efficiently translate these contests into seats (relative to the other parties).
Very interesting reading.
I think a lot of the pro-Coalition media (including The Age) are starting to worry that One Nation will be a significant disrupter to the Coalition winning government in November.
While One Nation certainly damage both major parties pretty significantly, their disruption overall benefits Labor much more than if the swing against them was translating straight into Coalition support, because:
a) There are likely to ALP v ONP contests that ALP retain, but would otherwise lose vs LIB due to preference flows likely being weaker from LIB to ONP than from ONP to LIB;
b) With Labor really only needing to improve their primary vote by 3-4% (based on the most recent couple of polls) to be seriously competitive at retaining government, the more these type of numbers that indicate a Coalition-One Nation minority get published and circulated, the more likely it is that swinging ALP/LIB voters (say 2010-14 LIB / 2018-22 ALP / currently leaning LIB for ’26) will return to Labor out of fear of ONP which could also save Labor in some tight ALP v LIB contests.
It’d be tricky for those sort of media outlets to use these sorts of numbers, like the AFR has, to drive voters away from ONP by demonstrating how it can hurt the Coalition’s chances, without also simultaneously reinforcing the narrative that Labor themselves are developing a scare campaign around, effectively giving Labor’s scare campaign ammunition, data & quotes.
Yeah. Clearly both major parties would rather not have One Nation polling this strongly.
It could be argued One Nation’s presence can both help and harm the Coalition, depending on the seat. If One Nation is in third place in a given seat, it could help strengthen the flow of preferences to the Coalition. However, if One Nation performs too strongly and knocks the Coalition out of the top two in enough seats, the Coalition could be deprived of a majority.
Labor can use this MRP to say “don’t vote for an L/NP-ONP government”.
The Coalition can use this MRP to say “only a vote for us can remove Labor”.
And One Nation would probably rather ignore it altogether.
And more details of specific seats from the Redbridge MRP, also courtesy of the AFR:
“Premier Jacinta Allan and deputy Ben Carroll would be turfed out of the Victorian parliament and Labor would lose a swag of Melbourne seats near the Suburban Rail Loop project if the state election were held today.”
“One Nation would pick up eight seats from Labor and four from the Coalition, and the Allan government would lose 15 seats to the Coalition”
“Under the median case scenario, Labor would pick up the seats of Prahran and Melbourne from the Greens.”
“However, it would lose Bentleigh, Box Hill, Ringwood, Glen Waverley, Monbulk, Bendigo West, Bayswater, Ashwood, Niddrie, Bellarine, Eltham, South Barwon, Bundoora and Essendon to the Liberals.”
“Labor would lose the seats of Ripon, Hastings, Bendigo East, Bass, Pakenham, Melton, Sunbury, Yan Yean and Eureka to One Nation. The right-wing minor party was also tipped to win the seat of Berwick from the Liberals and Morwell, Murray Plains and Euroa from the Nationals.”
Interestingly with the model I’ve been using, 3 Labor seats have consistently been clear ONP gains regardless of which poll’s numbers I used, even the ones that are weakest for ON and strongest for the Libs like DemosAU:
Melton
Yan Yean
Pakenham
So the MRP is consisent with that.
Seats like Hastings and Bass were always Labor losses but whether they were ON or LIB gains depended on the poll used.
Interestingly a couple of my results had Labor losing Bundoora and Eltham too, results I found a bit odd, but all my different scenarios had Labor retaining Bentleigh & Essendon, other than the DemosAU one which had Bentleigh as a tossup.
There’s no way Labor are winning Melbourne or Prahran while they’re at such a low point. Maybe Melbourne could be a very, very long shot if the Liberals put the Greens last and there’s an increased right-wing vote adhering to that, but in Prahran there’s just no way Labor are finishing above the Greens to make the 2CP if they couldn’t even do that in 2018 and 2022.
Every scenario I’ve done has had the Greens retaining Melbourne and comfortably retaining/regaining Prahran, with Labor in a pretty distant third there (low 20s primary).
Melton, Yan Yean, Pakenham, Hastings and Bass are all seats Labor had apparently already “written off”. The other four seats are Ashwood, Bayswater, Glen Waverley and Sunbury. All nine are losses in the MRP.
Agree Essendon is especially surprising. I would have thought Labor would lose other seats before losing Essendon.
Also surprising with the Greens losses. Labor didn’t even stand in the Prahran by-election last year.
The MRP definitely suggests One Nation is doing better in Labor seats in regional areas.
The Liberals would be happy to only have lost one seat to One Nation in this modelling, though they would be disappointed that it’s Brad Battin’s.
I find it funny that LNP will gladly accept ONP support if it results in a Coalition government…But if ONP received 30 or 40 seats there’s no way the LNP would support a One Nation government . In that instance LNP would side with Labor. Pauline Hanson is a fool.
I don’t think the Victorian Coalition would “gladly” accept One Nation support, at least in the lower house. They’d be terrified.
And One Nation can’t get close to 30-40 seats based on this MRP. Despite leading with a primary of 27%, they’re only getting a median of 12 seats (range 7-19), thanks to the distribution of their vote and the effect of preference flows.
In other news, Moira Deeming has been dumped by the Liberals (again).
https://kevinbonham.blogspot.com/2026/07/victoria-2026-election-where-sorts-of.html
Victoria 2026: The Election Where All Sorts Of Things Can Happen!
That’s a great read Kevin, thanks for that.
I do have a feeling that Coalition preferences will probably flow quite a bit weaker to ON than the other way around, and than in elections in other states too.
I think firstly, with ON polling so high, a lot of the more conservative or populist leaning Coalition voters would already have switched their primary vote, leaving the remaining Coalition vote being far more comprised of “moderates” than in most elections.
Victoria as well is a state where I think the Coalition base is does lean more moderate than conservative as well.
And then the Coalition polling probably includes quite a lot of ALP/LIB swing voters who voted Liberal in 2010-14, then Labor in 2018-22, and are currently back in the Liberal camp. Or at least that type/profile of voter. Many of them would be likely to put Labor above ON, in my view.
So given the more conservative cohort of Coalition voters (especially after Battin was replaced by Wilson) have probably already jumped ship to ON, it’s plausible that the remaining Coalition vote is probably much more strongly comprised of moderates, ALP/LIB swing voters and teal types than their vote would usually be.
ON are very divisive so I can imagine that when their primary vote is so high, preference flows to them could be quite weak even from centre-right voters.
Labor needs to improve primary vote by 4 and can still win. I don’t see it happening with JA. There is still a chance One Nation may fizzle out. Certainly Hanson’s rating have taken a hit.
Agree Gorks.
Jacinta Allan is very actively dragging the vote down. A neutral leader would probably translate to a 2-3% boost immediately. Not because changing leader would be a positive, not because the new leader would even need to be a net positive either, I just think she probably has a solid 4-5% personal drag on their vote and half of that would be recovered even factoring in the negative impact of a late leadership change.
That alone gets Labor back around 28-29%, but here’s the most important factor:
Labor are a great campaign machine, but people have tuned out from Allan. So with her still as leader, I even think Labor lose the edge they have as better campaigners, because the message won’t get across if no one is listening.
That’s what the biggest benefit – along with somewhat neutralising her personal drag on their vote – of changing leader would be.
It gives them the best chance run a winning campaign. People are more likely to listen to a relatively unknown commodity who they’re curious about, than a leader that they have simply tuned out from.
Yeah a neutral unknown leader is much better than -40 approval JA thats been polling disastrously for 2 years.
Dont know what’s going on in the party , do they wanna win or not ? I certainly dont want austerity Jess and One Nation weirdos in charge.
It has to be a dignified exit by JA though, cant have contested messy transition.
I agree. Ideal scenario is Jacinta Allan steps down and decides not to recontest Bendigo East in November.
She doesn’t seem to want to do that but perhaps the likelihood of losing her seat anyway would sway that decision..
While I think winning in 2026 might be a poisoned chalice for either side, 4 years is a long time. I never think losing to make your chances better in 4 years time is a good idea because so much could change, and also I would absolutely not want a Coalition-One Nation alliance to have the reins for 4 years.
Gorks
Whoever wins will be doing austerity, that is why whoever wins will probably lose in 2030. The Allan Government has started laying the groundwork with the Silver Report.
That MEP poll makes little sense.
The Greens hold 3 seats, there are 4 Labor vs Green seats under 5% and Prahran is Liberal vs Green, but the poll has the Greens only winning 2 seats.
Euroa a ON gain, Benambra stays Liberal, but Euroa’s MP is well liked and Benambra has no incumbent.
ON polling 18% in Northcote has to be poor sampling, and is one of the 4 marginal Labor vs Green seats.
I can’t see the Greens not winning at least 5 seats (Melbourne, Richmond, Brunswick, Prahran, Northcote) in November.
Footscray looks very competitive for them too.
Liberal preferences are an issue for the Greens; some seats they would win without them but Liberal preferences were a major reason why they did so well in the lower house last time. It is possible that if Labor does really badly Labor could be out in third in some of the Green target seats rendering Liberal preferences irrelevant.
MRP models don’t really take account of personal vote effects; it’s one of their weaknesses.
Melbourne is a seat where if Labor do really badly I can see them dropping below the Libs. And ON preferences going to LIB could help them leapfrog Labor.
Richmond is the main seat where I think Labor could have a shot at a gain. That’s more likely to remain ALP v GRN because everything north of Bridge Rd is very hosand there’s a significant enough LIB vote around the Burnley, Cremorne and southern Richmond that their preferences flow could be decisive.
I tip Northcote to be a Greens gain, but you’re right that if the Libs put Labor above the Greens, it could offset the swing against Labor and help them retain it.
In Footscray, especially after the federal election results I think there will be a big swing to Victorian Socialists which will benefit the Greens. VS could even finish above the Liberals in some booths. That’s going to be an interesting seat to watch I think because while it’s still inner city, it’s more like the type of working class, multicultural area that’s gaining Greens (and VS) voters unlike the more gentrified inner north/South where they seem to be declining a little bit, but it still has some more affluent & gentrified pockets with an established Greens vote too.
The Libs may direct preferences to the ALP over the Greens but it would seem a high possibility that the Libs voters might not necessarily follow the ticket. At the federal election 25% of Lib votes in Wills and 35% in Grayndler leaked to the Greens. In the main Greens seats, it is highly likely that there will bo no Libs staffing lots of booths. Even if preferences are directed to Labor, Liberal may have a greater loathing for the Allan Government despite what the HTV says.
I agree with Trent’s list of Greens wins above. Footscray would seem a good chance and Albert Park should not be ruled out if the Labor vote really tanks. The problem for the Greens is that they seem to have so little traction at the Victorian state level. They almost seem to relish their invisibility.
That’s a good point about the lack of Lib volunteers in some of those seats too.
Even in one of the 2022 elections (can’t remember if it was state or federal), my polling place in St Kilda didn’t even have Liberal volunteers. And their vote is usually around the mid-teens in St Kilda, which is in seats where they also make the the 2CP.
So I can’t imagine they would bother resourcing booths in suburbs like Fitzroy, Collingwood, Northcote, Footscray, Carlton, etc where they typically get single-digits, and are in seats where they finish a distant third.
On the topic of Richmond and Melbourne, an issue that will help the Greens retain is Labor’s plan to demolish the public housing towers. More than half of Victoria’s commission towers are in those two seats alone (Prahran has the third most with about 7). Those housing estates are traditionally the Labor strongholds, just look at the Labor vote in North Richmond, but the Greens have been campaigning hard to save the towers. Labor will lose a lot of votes among those residents.
Ah ..the public housing towers – no policy process, no justifiable need, social dislocation, constant government attempts at trying to hide something, enourmous cost, questionable benefits. Says it all about government and governance in Victoria. There is no doubt that a lot of the public housing stock in Victoria was very tired but it was mainly the low rise buildings. I often wonder where the tenants of the Walker Street estate in Northcote have gone now that is has been replaced by a very upmarket apartment complex. Trent, good point, it could have an impact on voting behaviour. There might be some fallout in Albert Park as well, there must be hundreds of voters in Park Towers and some of other South Melbourne and Port Melbourne estates.
Yep 100% agree.
Independent reviews have calculated that it would have been far cheaper and less disruptive to retrofit the towers (Atherton Gardens in Fitzroy had only recently had a big overhaul), and the fact that housing stock is only increasing by ~10% makes it not worth the huge cost.
The low-rise ones were generally very bad and needed replacement.
All they should have done is refurbish the towers, and replace the low-rise with new (but still 100% public) medium density buildings. That would have increased the public housing stock and upgraded the existing stock, at least cost with less disruption.
Yep the seat of Albert Park has two towers in South Melbourne and one in Albert Park itself. Park Tower especially is probably the tallest individual HCV tower.
I think a Port Melbourne estate (not high rise though) has already been replaced with a more upscale development too.
But I can see the Labor vote really declining in the booths near those larger public housing estates in North Richmond, Fitzroy, Prahran & South Yarra, Flemington etc where there are 4-5 towers grouped together.
Love looking at the bludgertrack for the vic state election – but last time I checked today, the 2PP pie chart, and swing dial are still stuck at 50% and -5.1% swing.
Are the graphics stuck? I swear we have had a couple, maybe even a season of polls where it’s more like 53% 2PP to LIB (or at least The Right), and yet it isn’t showing.
Other modellers have their own decisions, but something like aeforecasts show a 52.8% 2PP to Libs, vs the dead heat here.
Are the polls closer than the media and other aggregators see it? Is there some other reason?
The Herald Sun is reporting today that “Jacinta Allan faces a showdown on Tuesday” linked to a JWS Research Victorian poll commissioned by Australian Energy Producers. Primary votes have the Coalition on 29%, One Nation on 24% and Labor on 23%. A 2pp of 56-44 (L/NP-ALP) is also provided.
“Multiple sources said there was the prospect of a delegation visiting Ms Allan to advise her that her time was up early next week, but key factions still seem unsure whether Deputy Premier Ben Carroll – the only person they say would be a viable candidate – wants the job.”
The Victorian Coalition would be happy with this primary lead, which is closer to the recent DemosAU and Freshwater polls, and suggests it could secure a majority, unlike the recent Redbridge and Resolve polls.
The 56-44 2pp looks very good for the Coalition. However, we have seen that a 54-46 2pp lead does not necessarily translate to a majority in Redbridge’s modelling, because of One Nation knocking the Coalition out of the top two in many seats (the Coalition only ended up in the final count in 49 seats in Redbridge’s MRP).
She really just needs to go.
Her unpopularity is completely unrecoverable and would be actively dragging the Labor vote down by at least 3-4%.
As I said in a previous comment, people have totally tuned out. Labor are a campaign machine, it’s their strength, but even the best messaging and policy announcements won’t matter if nobody is listening to the messenger anymore.
@Trent, I agree as the current vibes is just anti-Jacinta and that makes it easy for ‘Fake Moderate’ Jess Wilson to hide to the median voter that her policy promises would actually make it the most Right-wing Government in Victoria’s history (and that doesn’t even take into account a potential One Nation Coalition). Once much the negative baggage is gone, then it could be easier to expose.
It looks almost like Pierre Poilievre for the Conservative Party of Canada where anti-Trudeau sentiment made it easy to hide to voters that he was planning to govern hard-right but once the PM changed to Carney and Trump’s trade war, it made it much easier to expose.