Poll Roundup: What Do We Do With A Split Coalition ... Or A Rampant One Nation?

2PP Aggregate 53.5-46.5 to ALP (-1.3 since end of 2025, +0.6 in two weeks)

Shadow-2PP Trend Estimate 54.3-45.7 to ALP vs One Nation.  One Nation has made accelerating gains.

Labor would easily win an election "held now", probably losing several seats

The last few weeks have been remarkably messy ones for Australian politics- and poll-watchers.  Following the Bondi massacre the Opposition pressured the Government to recall parliament as soon as possible to pass measures in response, only to end up wedging itself when it got what it asked for, with the result that the National Party has again broken away.  So should analysts and pollsters still treat these parties as the same entity, and if we don't, what especially do we do in the case of Queensland where Liberal and National party room members run in separate seats under the Liberal National banner?  In the meantime, One Nation has exceeded the ex-Coalition's total in three of the ten polls released so far this year and tied it in two others.  An election right now would make Queensland 1998 look somewhat orderly, with all manner of messy multi-way seats and probably One Nation making fifty or more 2CPs with perhaps something like twenty wins - though this stuff is very hard to model.  So is it time for a Labor vs One Nation "two party preferred" figure as well?

What do we even call a Labor vs One Nation national figure?  It's not a two-candidate-preferred because that is a term that refers to a single seat, while the term "two-party preferred" is reserved for Labor vs the major conservative force of the time (well, by the next election that might be clearly One Nation, but we're not there yet).  I've decided for now to call it a "shadow 2PP".

I decided that it was time to try to model an indicative national Labor vs One Nation figure.  I previously wrote about issues with modelling the ALP vs ON shadow-2PP here. I've used that work as the basis for a preference formula in which, all else being equal, One Nation does slightly worse on excluded party preferences vs Labor than the Coalition do (although they do do better on some groupings, such as non-Independent Others).  (Note also Antony Green's figures for Queensland 2017 and 2020 here, with flows to One Nation in the high 70s - as these would be among One Nation's strongest seats it again figures the overall flow would be somewhat lower).

That is, however, on the basis that preference flows stay the same as what I think they were in 2025 - that might not happen, in either direction.  At this stage, I don't have the time (or to be honest the motivation after how pleased I was to have got the normal aggregation design finished!) to do a full aggregation of Labor vs One Nation but it did occur to me that I could run conversions for all the polls released so far, and then of course that I could graph them.

The result (above) startled me for its similarity to the projection graphs I did a few years back for the Voice referendum, in that Labor's shadow-2PP vs One Nation at this stage shows much the same accelerating decline pattern as the Voice Yes vote did.  Now I am not (on current evidence!) saying that Labor are going to be smashed 60-40 by One Nation at the next election but that graph should be at least a little bit of a pause for thought for ALP triumphalists crowing about the split state of the conservative parties.  While the combined Liberal/National/One Nation primary vote is very badly split, it has also been quietly going up, and it is now averaging 45.8% in polls this year compared to just 38.2% at the election - something that seems to not have been generally noticed.  Some of One Nation's poll gains - perhaps as much as three points - are clearly coming from Labor.  

If anyone thinks that accelerating shadow-2PP poll gains for One Nation vs Labor are going to suddenly stop, or even that the more or less linear increase in the ON primary vote will not continue, well that could be true but I would ask them on what basis (apart from a Liberal leadership change - maybe!) is now the time to be so lucky.  This is especially as this sort of accelerating pattern can (and in the Voice's case I believe did) result from a sort of organically spreading social permission.  People who are resistant to an idea because they fear that it is socially unpalatable may become more open to it once they know that someone who they respect holds that same idea.  Voting no to the Voice and preferring One Nation to Labor are in a similar boat here because both carry the taint of perceived racism.  

It might be argued that with so many seats that would be a mess off the current voting intentions neither a conventional 2PP nor a shadow-2PP are going to be much use, but I don't agree.  For sure they should be treated with a great deal of caution for seat predictions but if Labor remains well ahead on both scores then the likelihood is that Labor wins a majority, probably a smaller one than normal because of cases where one conservative force has strong local support and can beat it.  If Labor is much worse placed versus one of the opponents than the other, that's probably because the better placed opponent is well ahead of the other on primaries and is more or less the serious opposition, so that measure is likely to be a good predictor of Labor's fortunes.  If Labor is 50-50 or so vs both, that would probably result in enough combined wins for the conservative parties to push Labor into minority. Below that some kind of combined right majority becomes likely.  

Pollster responses so far have varied.  In their most recent offerings now:

* DemosAU and apparently Essential have ditched two-party preferred.  DemosAU will instead be issuing seat total projections based off its quarterly MRPs.  These are distinctly on the bullish side for One Nation (even after adjusting their Coalition flow downwards) with the most recent finding the National Party could be reduced to only a few seats.  No wonder the Nationals are terrified.

* YouGov and Fox&Hedgehog have offered both conventional 2PP and Labor vs One Nation

* The most recent polls from Resolve, Morgan, Freshwater and Newspoll have continued traditional 2PP and have not offered an alternative figure (but I understand at least one of these to be working on one)

Morgan is (as it has long done, albeit often with a lag in publishing results) splitting off the standalone Nationals in its primary figures.  YouGov is splitting off both the standalone Nationals and the Queensland LNP.  At the moment I am simply adding together all parts of the former Coalition for aggregation purposes, and maintaining the usual approach to 2PP in which whichever is the strongest of the Liberals and Nationals is the 2PP candidate in each seat.  Things will get messy if we get to an election where the Liberals and Nationals are running as distinct parties and against each other in Queensland (and more than usual elsewhere), but that is a long way off if it happens at all.  

Ye Olde 2PP

This article will be long enough without going into particular polls in detail but here is an overall summary of where things stand on the Labor vs Coalition front.  Only one poll, the December Resolve, was released between the Bondi terror attack and the end of 2025.  Probably Labor was already below 54 2PP at that time, but because of the lack of data it wasn't obvious.  This was quickly corrected once polls resumed early in January with 53-47s from DemosAU Fox&Hedgehog (my conversions 52.6 and 52.8) and a 52 from Morgan (51.8).  A Newspoll 55-45 (55.4) cancelled out a very bad Resolve 51-49 (51.2).  Newspoll looked a bit of an outlier and has on average been a point better for Labor than my aggregate in this term, but the general conventional-2PP run of polls since has been mostly pretty strong for Labor:  Mogan 53.5/53 (respondent/last-election) (53.0) and 56.5/54.5 (54.1), DemosAU no official 2PP (54.2), YouGov 55 (54.6), Freshwater 53 (52.6) and Essential no 2PP (51.6) as the weakest.

All up my aggregate suggests Labor's position vs the "Coalition" may if anything have improved slowly over the last few weeks - but this is mainly because the "Coalition" vote is still crashing.

Leaderships

I also don't see the need here to repeat figures readily available at Poll Bludger but it's obvious Anthony Albanese has taken a modest personal hit with his net satisfaction ratings down into the negative teens (still far from awful historically, indeed every PM after Gorton has at some stage been polling worse).  There is a lot of variation in results for Sussan Ley who has terrible net ratings in Newspoll (-28) and YouGov (-31) but is beating Albanese on net rating in Freshwater (-5) and Resolve (-7).  Both these polls displayed similar behaviour at times in the last term with Peter Dutton.  Nothing much is happening on Better Prime Minister.

There has been more interest (and more complaining) re the entry of Pauline Hanson into these figures.  While Hanson has often been polled in the past, it has usually been in a net likeability context not a performance context.  Some confusion has arisen this year from some people (and a certain errant twitter AI account) comparing Hanson's net likeability scores with the approval ratings of Albanese and Ley.  DemosAU did ask a question to provide a net rating on the same footing for all three leaders but it was "What is your opinion of?" and not a specific question about performance; this may explain why Ley had a massive neutral rating.  (Albanese 27 positive 41 negative, Ley 15-33 and Hanson 35-40).

DemosAU's addition of Hanson to the preferred PM race (Albanese 39 Hanson 26 Ley 16) drew some complaints along the lines that Hanson is a Senator and not a putative Prime Minister and therefore shouldn't be included in such a poll.  There's a general view that Senators can't really be Prime Ministers even though they technically can - practicalities around Question Time and the media focus on the House of Reps meant that Gorton immediately relocated and no other Senator has been seriously considered as a potential PM since.  However, this is One Nation we are dealing with here, they really do not care about conventions or how people expect parliament to operate.  They'd probably get a kick out of making Question Time unworkable or irrelevant and show me anyone who would even claim to care who has not spent at least fifteen years saying that Question Time is terrible.  If they did somehow get into government, and Hanson was still a Senator, and their will was to make her PM, then I'm sure they would find a way.  This debut for Hanson in these questions is very unflattering for Ley but it's not so long ago that Nick Xenophon led in a three-way preferred Premier for South Australia and wound up winning zero seats.  

The air of unreality

For some observers what is going on at the moment is quite hard to comprehend.  Walk down a proverbial average town street in Australia and, if the polls are right, sometime in the last eight months one in every six voters in that street has been Pauline Hanson-pilled.  Which one is it? Can this and soon more be true? 

One Nation?  That joke of a party famous for having around 80% of its MPs through its history leave or be kicked out or disqualified within a term, which has never won a House of Reps seat or any single seat electorate outside Queensland in its own right?  That lot would presume to govern?  For those who are used to Australian politics in isolation this is hard to comprehend.  It's also worth remembering that at state level One Nation has a long history of threatening to win a raft of seats and usually (except to a degree for Queensland 1998) tanking as they get in sight of voting day.  Polls often pick up the tanking but even final polls have sometimes overestimated the party (eg Queensland 2020).  

One reason One Nation bubbles have tended to burst in the past is that they are very prone to candidate malfunctions when someone is discovered to be beyond the pale for even them.  However if One Nation were to run into a federal election with consistent polling in the twenties or higher it is likely they would attract a more prominent calibre of candidate, including several more ex-Coalition defectors.  Perhaps this would also head off the sorts of candidates who have run as independents and dented the ON vote in certain seats in the past.  

The UK is an obvious comparison with the rise of Reform UK to the polling lead, but Reform got there off the back of a far more prolonged and agonising Tory implosion coupled with the new Labour government being completely on the nose too - both sides of establishment politics being terrible at the same time in a nation with more locally severe socioeconomic issues.  While Reform still lead for now, their polling has dropped off by about five points in the last few months and the Tories seem to be recovering a little, so even there there may have been a ceiling.  Australia has in the past seemed unusually immune to the widespread erosion of the establishment right by nativist insurgents but perhaps our time is finally here.  

Is a return to normal for the Liberals and Nationals only a leadership change away, or is it already too late?  We may find out the answer to this very soon.  The problem is that the Liberals shouldn't have made Ley leader in the first place, not because she was too moderate but because she wasn't up to it.  (I was giving her some credit for exceeding my low expectations on that front til she followed Sharri Markson down the Joy Division beatup rabbit hole).  But now if they remove her it looks like it will be a boys' club rolling the party's first female leader at the behest of the National Party because the Nationals are curled up in the corner by One Nation.  Some interesting weeks ahead.