The Herald Sun reports a DemosAU/PremierNational poll of Victorian state voting intention has the Coalition’s two-party lead over Labor narrowing from 55-45 to 52-48, from primary votes of Labor 25% (up two), Coalition 30% (down two), One Nation 20% (down two) and Greens 14% (up one). Ben Carroll’s personal ratings are up two for positive to 23%, down two for neutral to 40% and steady on 37% for negative; Jess Wilson’s are down two on positive to 33%, up one on neutral to 42% and up one on negative to 25%; and Warren Pickering’s ratings have responded only moderately to the recent elevation to his public profile, up three on positive to 22%, down five on neutral to 38% and up two on negative to 40%. The poll was conducted September 10 to 20 from a sample of 1020. A full report featuring detailed breakdowns should be on the pollster’s website later today.
UPDATE (1/10): The Herald-Sun today has a Freshwater Strategy poll that gives the Coalition a 53-47 lead on two-party preferred, from primary votes of Labor 24%, Coalition 29%, One Nation 22% and Greens 14%. A three-way preferred premier question has Jess Wilson at 37%, Ben Carroll at 30% and Warren Pickering at 13%. The poll was conducted Thursday to Monday from a sample of 1030.
I am also pleased to announce that the Poll Bludger’s Victorian election guide is now online, with all the customary features: tables, charts, interactive maps and analysis of all 88 lower house seats and eight upper house regions, plus a general overview of the electoral terrain and state politics generally over the past parliamentary term.
No less exciting is a new addition to the BludgerTrack poll aggregate that aims to make sense of the new electoral terrain by projecting outcomes for all 88 seats, based on modelling and the latest reading of the poll aggregate. To cut a long story short, the model is based largely on the relationship between federal and state election results in South Australia, the only case study so far of how voting patterns have changed since One Nation’s apparent transformation from minor to major party, together with Indigenous Voice referendum results (a pointer to strength for One Nation), data on outright home ownership (a pointer to the Liberals) and whether the seat has a Coalition incumbent (which appeared to be a relevant factor in the seats where the Liberals held out against One Nation in South Australia). If this approach proves its worth, I may end up applying it to the federal context.
The seat projection currently suggests Labor will be copping it at both ends, losing exactly enough seats to the Coalition to give the latter a bare majority of 45 seats while also bearing almost the entire brunt of One Nation’s 15 seat haul, in contrast to South Australia where Labor’s primary vote was 37.5% as compared with BludgerTrack Victoria’s current estimate of 25.2%. The former includes what would undoubtedly be some surprise Liberal gains in Melbourne, including Albert Park, Eltham and not least the Premier’s seat of Niddrie, with the very considerable qualification that the model makes little allowance for candidates’ personal votes. One Nation is credited with a new empire extending from Melbourne’s outer north and west through the Ballarat seats to Ripon, along with scattered gains elsewhere. UPDATE: A correction/revision to the model has made its projection less rosy for the Coalition, who are now at only 38 seats.
I like the model;
I’m glad William Bowe worked out that voice vote is a strong predictor of someone being socially conservative, and someone being socially conservative is a strong predictor of them considering flipping to one nation.
Yeah nah Arange, there were plenty from all over the political spectrum who voted no to the Voice
Greens to win 5 to 6 seats imo, the model looks interesting but is a bit out is my guess
If won’t take much more tightening for this to turn into a cliff hanger election
Although I don’t like the forecast, terrific work William. Thank you.
Hard Being Green
I didn’t realise the Progressive No campaign was such a widespread and successful one.
Do you agree with Boerwar that the Greens helped kill the voice?
The Greens and the Voice?
Six months of openly opposing the Yes, the internicine battles over whether to oppose it or support it which resulted in the exit of the Greens’ Indigenous senator, and then only half-heartedly campaigning for it helped to kill the Voice.
Shoebridge is doing the same utterly Greens destructive arsehattery with Australia’s possible seat in the Security Council.
Carroll was at a luncheon at the Royal Melbourne Show yesterday.
He didn’t get any pushback when walking through the crowd.
Which was interesting.
https://www.abc.net.au/news/2026-09-23/ben-carroll-one-nation-victorian-election-jess-wilson-liberals/107186180
I was very disheartened by the amount of people in my sphere who voted no on the voice.
Thankfully my immediate family were all in the yes column.
But it didn’t make me feel any better for it.
The model doesn’t seem to be picking up on the Greens in seats where they should be a factor, e.g. Northcote is a Labor-Greens marginal but the Greens aren’t showing in the 3CP. (And the apparently surprising model for Albert Park – which Labor didn’t lose even in 1992 – has One Nation rather than the Greens in the 3CP, which would surprise me if it happened).
At this point, I would be happy with a minority Labor Green govt.
The notion of a coalition and one nation govt is giving me conniptions.
Albenese was a passionate supporter of The Voice, but he could only persuade a handful of labor held electorates to vote yes.
Sooner or later the explanation for the no vote needs to be laid at the feet of the concept itself. It was a bogus idea that was fundamentally anti-democratic.
The no vote wasn’t due to posturing by any single political party, what happened was the majority of voters across the spectrum realised it was a dilution of the democratic principle of one person/one vote and rejected it accordingly.
Personally, I don’t see much correlation from The Voice to the surge in one nation…unless it is an indirect connection to the sense that LNP/Lab have been asking for citizens to vote for them but then centring their plans for economy and community around next years migrants….which is also an erosion of democracy
Mr Squigglesays:
Tuesday, September 29, 2026 at 9:04 am
“Personally, I don’t see much correlation from The Voice to the surge in one nation”
So it’s just a coincidence that in the SA election, the biggest swings to one nation were in the areas with the highest no votes?
Ben Carroll is doing a good job getting on the front foot, setting the agenda and removing barnacles. Too bad he is not very articulate or an orator, hopefully he warms up to give a good contest.
At the very least there is now a plan rather than drift.
William’s model makes for a grim reading but small shifts to ALP could save large number of seats.
Arange, I havent seen that data, do you have a link?
I’m very interested in the SA situation, partly because the state government legislated a voice, despite the state voting no.
Thank for your hard work, William. Very impressive.
Mr Squigglesays:
Tuesday, September 29, 2026 at 9:33 am
“Arange, I havent seen that data, do you have a link?”
No, but you can just look up the corresponding state electorates that cover the federal electorate of Spence.
Like the state electorates of Light or Taylor, for example.
Nice work William
It’s traditional at this stage of the electoral cycle for the side that’s losing to claim the polls will tighten once the campaign starts. It’s also traditional that such tightening (almost) never happens.
With that in mind, the Labor government looks doomed. The only possible question mark is whether a significant number of prospective ON voters get cold feet at the last minute and deliver the Coalition a comfortable majority.
Regarding the link between the No Vote to the Voice and One Nation Support, there is a correlation in the SA Election with polling booths having the ONP (in TPP) do roughly on average 20% worse than the No Vote for the Voice with the largest deviation being immigrant heavy areas where it ONP TPP did 30-40% worse than the No Vote to the Voice. Farrer by election ONP TPP did 17% worse than the No Vote and Secret Harbour’s ONP TPP is around 12-13% worse than the Voice. I can say the areas where it is a heavily White, high Labor Vote and High No Vote happens to be areas with a large ‘Red One Nation’ Vote base.
It was only a social media post but Mildura MP Jade Bedham did a post on how Labor could be returned, the thing that caught my attention was she only gave ON 5 seats, is it possible she was hinting at the internal polling showing the public polling overestimating ON.
FWIW, Ch 7 news in Victoria quotes figures from a company called SEC Newgate:
– 26% say state is heading in the right direction (59% after the last election)
– 19% positive rating of government’s performance (49% after the last election)
– 60% rate performance as poor or very poor
https://www.youtube.com/watch?v=KS_Ex8tyjVU
For me, the YouTube algorithm dishes up a heap of Ch 7, News24 (aka SAD) and Falun Gong anti-China stuff. Presumably that’s punishment for my sins of watching websites critical of Trump.
Because there’s lots of evidence public polling has overestimated the One Nation vote in recent elections, right, Landlord?
I think the Nationals are just trying to sell themselves as the best vote to get rid of the ALP.
The Wombat
ON is clearly polling strongly in the 20s but does that translate to seats, it might but it’s just as possible ON falls short after preferences.
Yeah, I agree. A significant proportion of the traditional conservative vote just aren’t going to ever preference One Nation. I look at my own parents, who absolutely hate the ALP with every fibre of their body, bit they will preference the ALP before One Nation.
My gut feeling is One Nation might underperform the general public’s expectations in Victoria, the Coalition will fairly comfortably win, and this will lead to a bit of a swing back towards the Coalition nationwide. I’m convinced part of what is driving the swing to One Nation is frustration with the Coalition’s performance in the last federal election.
@ Ante Meridian, a recent poll found the core supporters of One Nation in Victoria is 18% which is actually barely lower than this Demos Poll of ONP support so 20% already around the floor of ONP support in Victoria so for the Coalition (let alone Labor) to even convince Core ONP Voters to switch to them would probably require them to oppose multiculturalism with culture war rhetoric and adopt Conspiracy Theories spouted by Pauline Hanson and the far-right base.
I’m a bit wary of talk about ‘core’ ON support.
Unlike voters of Labor, the Coalition, and the Greens, ON supporters haven’t had decades to become welded on. They might say they’re core, but anyone who’s just jumped on a bandwagon can be fickle.
Landlord of the Yearsays:
Tuesday, September 29, 2026 at 11:20 am
“ON is clearly polling strongly in the 20s but does that translate to seats, it might but it’s just as possible ON falls short after preferences.”
Am I the only one who remembers people saying this before the SA election, a Labor landslide where one nation nearly won 14% of Labor’s seats, and did win about 30% of the Lib’s seats.
Arange
Anything is possible in Victoria, it’s shaping up to be one of the most chaotic election results in ages.
For the combined vote of ALP and Greens’ to be in the high thirties is not looking good. This is in what is touted by its residents as Australia’s most progressive state. The state branded as Australia’s least progressive, delivered a better result for the ALP at an election characterised by very similar issues, hostile media and momentum. Perhaps it’s time for the Vic premier to wheel out the 50c tram fares?
I think your wrong with the seat of Williamstown. I live in the area and the Libs are making the biggest effort i have ever seen. They think they can take it. Also the west Party is in with a chance.
I would think Mr Carroll’s pothole blitz is much better targeted than 50c fares.
I hope WB model is wrong for the sake of Victorians and Victoria Labor losses aren’t many.
I hope Victorian public polling is wrong and Carroll somehow pulls rabbit out of the hat.
Anyone telling you with certainty what will happen in Victoria has their hand on it.
This election will have surprises all over the place. That is the only thing that is guaranteed.
I cannot ever remember an election in Victoria where the libs did not declare it was wharfie time before the actual vote. They were right in 1995/6 and in 2010(by one seat). So let’s all calm down just a tad.
Personally I think the ALP are in a spot of bother, but nothing would surprise me. Nothing.
It’s an acknowledged limitation of the model that it produces 3CP estimates for Labor, Coalition and One Nation. I have crudely accommodated the Greens by replacing Labor with them in the four seats they won in 2022. So the model doesn’t allow for the possibility of Labor winning those four seats, or the Greens winning anywhere else.
Heh heh. ‘Wharfie Time’.
Freo’s never going to live that down. It’s part of the national lingo already.
Hi William,
A fantastic resource and excellent work all round. The visual information makes the data much more accessible and it has helped me show other people who usually get bogged down with numbers how better understand some of the polling. I hope it lives up to your expectations for election night!
“I didn’t realise the Progressive No campaign was such a widespread and successful one.”
The Greens progressive no almost made me abstain. In the end I reconsidered & vote yes after thinking on it, but at the same time it made me feel the referendum itself was doomed because it was already on the back foot with the LNP’s reversal after changing to no so they could use it for their culture war as a replacement for competent policy.
If the Greens couldn’t even decide to support it wholeheartedly, there was zero chance the electorate was going to either.
BW: “Six months of openly opposing the Yes, the internicine battles over whether to oppose it or support it which resulted in the exit of the Greens’ Indigenous senator, and then only half-heartedly campaigning for it helped to kill the Voice.”
—
Amazing that Labor is yet again trying to blame their own failures on the Greens.
The Greens were the first party to support the Uluru Statement from the Heart.
The Greens also warned Labor that Truth and Treaty needed to come first before the Voice. Labor did not listen and rushed straight to the Voice in search of short term political gain, only to set back First Nations rights by decades with their failure.
As for Lidia Thorpe, she is to be commended for continuing to represent the Greens platform that she was elected on, despite no longer being a member of the party. She has my utmost respect and thanks for the way she has conducted herself. Lidia hasn’t sold out her values.
In Senator Thorpe’s own words recently when she was speaking about the retiring Senator Whish-Wilson: “So, you know that you’ve just left such an impact on me personally as a sovereign black woman who, you know, also supports the Greens, even though I’m not one anymore.”
I wonder if they ever got around to having that J? lol
“The Greens also warned Labor that Truth and Treaty needed to come first before the Voice”
That’s not what the statement asked for.
Labor will be happy with their primary up 2 compared to the August DemosAU poll, though these primaries would likely still result in a Coalition majority. This is also the weakest One Nation primary since before the Labor leadership change, which should help the Coalition reach the top two in more seats. The full regional breakdown will be interesting to see when it’s released.
Resolve, which was one of Labor’s stronger series in August, has moved towards the Coalition (L/NP +1) in September.
Whereas Redbridge and DemosAU, which were particularly strong for the Coalition in August, have moved towards Labor (ALP +1 and +2 respectively) in September.
The new projection model is very interesting – thanks for this addition. The current projected Coalition bare majority of 45 seats is right in the middle of the range of outcomes suggested by recent polls.
citizen – The SEC Newgate survey is also reported in the AFR:
“Victorians are the most pessimistic in the country about their state’s government and its future, according to a survey showing Premier Ben Carroll has been unable to arrest a decline in Labor’s electoral fortunes.”
“SEC Newgate’s latest Mood of the Nation report, which surveys more than 1200 people every two months about their views on issues dominating the national agenda, has laid bare the massive challenge facing the Carroll government as it tries to clinch Labor a historic fourth term on November 28.”
“Only a quarter of Victorians believe the state is headed in the right direction, down from 51 per cent at the 2022 state election. Sixty per cent of respondents rated the government’s performance as poor or very poor – substantially higher than other states.”
“The online survey of 1659 Australians, including 756 Victorians, was conducted between September 8 and 14 for non-partisan strategic communications firm SEC Newgate.”
“Cost of living remains the highest issue for Victorians, like the rest of Australia, but a third of Victorians identified crime as the second most important issue heading into the election. This is double the remaining country.”
https://www.afr.com/politics/victorians-the-nation-s-most-pessimistic-people-20260928-p61119
Vote one nation and the state will move in the right direction… sure.
Re One Nation’s ‘Core’ support – amuses me because Reform UK had a ‘core’ until Restore Britain came along, and in a FPTP system that could be fatal. Here it is hard to know – literally break-away factions should be adjacent to what they broke away from, but the history of the DLP hating Labor suggests this may not be the case.
I still wonder if too much of ON is in the rural seats to be useful to them.
Congrats on getting this up, William. I feel informed. I was interested in Pakenham as it was feeling like an interesting three way contest between ALP, LIB and ON but with ON putting up their leader in the seat and Labor not really looking like they are fighting for it, it does look stronger for ON unfortunately. Never a sure thing… Libs could easily come up the middle.
I do think we could get a echo of the recent SA result where the Libs and ON split the vote and the ALP squeak (not thunder) back in across the state… although Jess Wilson is carrying the Libs hard.
Hmmm, the more I think about it the more I come to the conclusion that anyone who confidently says what is going to happen in November is full of it!
Arangesays:
Tuesday, September 29, 2026 at 12:14 pm
Landlord of the Yearsays:
Tuesday, September 29, 2026 at 11:20 am
“ON is clearly polling strongly in the 20s but does that translate to seats, it might but it’s just as possible ON falls short after preferences.”
Am I the only one who remembers people saying this before the SA election, a Labor landslide where one nation nearly won 14% of Labor’s seats, and did win about 30% of the Lib’s seats.
————————————————————
Not sure if I’ve read this right. Correct me if I haven’t. ON didn’t win any Labor seats. They got close in a few outer suburban seats, but weren’t close in others. ON v Liberals, ON got a 4% higher primary vote than the Libs but still got one less seat and only equalled the Independent’s numbers. Didn’t think they’d preference well.
I am sure that the Greens have conceptually integrated their erratic behaviour during the Voice referendum to their own satisfaction. A plain reading of the facts is that the Greens contributed to the defeat of the Referendum. That is all I am claiming. IMO, the Greens’ behaviour was not decisive. It was merely a minor contribution.
Once the Liberals and the Nationals came out for the NO it was all over red rover. That referendum did not have a chance in hell. The NO is therefore owned by the Coalition.
IMO the NO has had three large subsequent impacts.
The first is that it dudded Dutton and his claque of advisors that the federal election could be won easily.
The second is that it created a polarization and a bandwagon effect that are both still energizing ON supporters.
The third impact is that the decisive nature of the NO seems to have destroyed a generation of Aboriginal national leaders. IMO, the resulting leadership vacuum has yet to be filled.
Preference roulette roolz.
Truth, Treaty and Voice, in whatever order, are off the agenda for the foreseeable future. The window of opportunity was briefly open, as it was for the Republic nearly 30 years ago. Either could have got up if the Yes supporters had been able to: agree on the details, which would have required give and take, not ideological purity; had anticipated the opposition tactics and had a plan to combat these; and especially in the case of the Voice had been able to mount a coherent campaign to sell the proposal.
has closed and won’t reopen any time soon, if ever.
@Steve777, even with details and work which would enable Coalition supporting, it would have to already contend with opposition from One Nation, Gina Rinehart, half of Murdoch commentators and Maverick Coalition MP’s (including Jacinta Price). We might never have a successful referendum again as having Greens and One Nation in the mix.
One Nation didn’t win any Labor seats in SA, although they got close in some. Of the four seats they did win, two were Liberal seats where the Liberals dropped to third and One Nation and Labor were the final pair, and two (both of which they won very narrowly) were seats where sitting Liberal members – both of whom were facing criminal charges – were disendorsed and recontested as independents.