2PP Average For Year 53.9 to ALP (+3.0)
At the end of each year I release a review of the year in federal polling. See the 2024 article here (ah, those days when the Coalition still seemed competitive!) and/or click on the "annual poll review" tab for all articles back to 2014. As usual if any late polls arrive I will edit this article to update the relevant statistics.
2025 was overall not a good year for the Australian polling industry. The year started well with good results for the two main pollsters in the field for the WA state election but this was followed by a rather serious federal election miss, where pollsters had the right winner but were almost as far off on the 2PP and primary votes as when they were wrong in 2019. As they had the right winner anyway there was little media interest in the error; there has also been precious little visible introspection from the industry since and none of the attempts at review processes we saw following 2019. The Tasmanian state election was then a mixed bag with a good result for local firm EMRS but a big shocker for YouGov. On the plus side, we are seeing more diversity in Australian polling, but still the average transparency level is ordinary.
This year welcomed one new entrant, Spectre, and also saw a low-key and unsuccessful return from Ipsos and a great step up in activity from DemosAU.
The state of play
As I write, 2PP polling since the election has shown on average a very slow decline, including months where it was arguably static, since a post-election peak which I have at 57.2 in mid-July. I currently estimate Labor's 2PP at 55.1% (dipping marginally below their election result finally for the first time after seven months) though it would be still at 55.8% but for my decision to treat Morgan's recent split sample as two readings. I made this decision because the second portion of the sample still had a sample size well over 1000 and aggregates function better using fresh data collected over relatively short windows than they do with data pooled over a month.
Morgan attributed the poor numbers for Labor in the second part of their sample to the travel rorts saga which was the first scandal of such a sort for this government. It is very common for Morgan to attribute any movement in their polls to anything whatsoever that is in the current news cycle; that said my perception is also that Morgan respondents for whatever reason actually do respond more to news cycle events than respondents to other polls. With all this being overshadowed by the Bondi terror attack days later we will never know whether the travel rorts matter was really a serious dampener of Labor support. And given the time of year at which it broke (when polls tend to be scarce on the ground around Christmas) we would probably not have known anyway. The previous poll to Morgan, a Redbridge/Accent poll that came out of field on Dec 12 with a 56-44 2PP for Labor, had not been visibly affected. The other polls out in December were an Essential that by last-election preferences was a term worst for Labor (I converted it at 54.3 though Essential's "2PP+" comes out at only 52.1) and a Resolve that was quite strong for Labor (55-45 but I converted it at 56.1). Newspoll was last out in late November and has never in 40 years appeared as late in December as now (but I'll keep an eye out on Sunday night just in case).
However the 2PP picture has been a relative sideshow in recent polling to the rise and rise of One Nation, lately running at about 17% primary and with no sign yet that their rise is stopping. Depending on swing models this suggests they would win from a few to several seats in an election "held now"; a DemosAU MRP had them on for as many as twelve though a difficulty here is in gauging Coalition to One Nation preference flows. In Hunter, Coalition preferences flowed 83% to One Nation but that was with Stuart Bonds as One Nation candidate, and Bonds is a local figure whose appeal is broader than the party (as can be seen from the Reps/Senate primary vote differences). In the DemosAU figures Calare would be retained by an independents, and at least Riverina and Lyne also look to me like Coalition wins on Labor preferences; other seats such as Groom and Hinkler might fall to indies before One Nation won them (there is a strong effect of indies on the One Nation vote in such seats). So I think for the moment something like seven seats is more plausible. But if they keep going upwards, that can increase rapidly.
We still don't know whether this is heading for a realignment on the conservative side (in which the Liberals are supplanted by a new main force on the right), a fracturing of the conservative forces into multiple parties or if this One Nation surge is mostly a protest bubble that can deflate if the Coalition starts looking more competitive and probably changes leader.
How many polls?
This year I counted about 106 federal voting intention polls, some of which doubled as MRP models. Despite this being an election year, this is down 25 on 2024, which partly reflects some polls becoming less frequent after the election (in some cases retreating to lick their wounds, in others a deliberate decision). My count includes 27 Morgans (18 before the election and nine after, one of which was split into two subsamples), 15 Newspolls (9 before 6 after), 13 Essentials (9 and 4), 12 Redbridges or Redbridge/Accents (5 and 6), 12 YouGovs (11 and 1), 11 Resolves (5 and 6), seven Freshwaters (6 and 1), six DemosAUs (4 and 2), three Spectres (1 and 2) and a lone Ipsos just before the election.
2PP Voting Intention
All of the polls in my records released a 2PP voting intention figure of some sort this year. Prior to the election Labor scored 40 2PP wins, six ties and 21 losses. They lost every released 2PP in January, lost nine of thirteen in February but had only three losses from 16 polls in March and won all 29 in April and May. Since the election Labor has won all 37 2PPs.
However a lot of the story of Labor's losses early in the year came from preferencing methods and assumptions that did not end up matching what happened at the election. Using the released primaries and 2022 election preferences, Labor lost only nine 2PPs if rounded to the nearest whole number, or fifteen if rounded to one decimal place.
I have said plenty before about the pattern before the election in which Labor's 2PP support, which had been declining gradually through 2024, turned the corner early in 2025, at almost exactly the same time that Donald Trump was inaugurated for a second term as US President. There was a steep climb in support for Labor from early March which national pollsters stopped finding evidence of in the final weeks. Had their finding been that it was continuing through this time, the final polls would have been far more accurate.
Prior to the election Labor's worst released 2PP was a 45 by respondent preferences from Resolve in February and their best was a 55.5 from Morgan in the third week of April (the only poll to even marginally exceed the actual result). By my last-election conversions the February Resolve was still the worst, but at a less outlying 47.3.
Since the election Labor's best released 2PPs were 59s from DemosAU in early July and Resolve in August. The lowest was a 52.1 (after conversion) from Essential in early December, but it's not possible for me to take Essential's mostly respondent preferences (or for that matter their perennially very low Ind/other primaries) seriously. By my last-election preferences the DemosAU at 58.9 was Labor's highest while the lowest was a 52.9 from the last week subsample of the December Morgan. Excluding that, the next lowest was a 53.9 from Redbridge/Accent in late Sept-early Oct.
As an average 2PP aggregated figure for the year (by last-election preferences) I have Labor on 53.9 for those times my aggregates were active but this is divided into 50.7 before the election and 56.3 since it. The 53.9, up three points on 2024, is not as high as the 54.9 that Labor averaged for 2022 or the 54.8 for 2023.
Leaderships
This year Anthony Albanese averaged a net -7.5 personal satisfaction rating in Newspoll, but this was divided into -11.7 pre-election and -1.3 after it, with a low of net -21 in February and a high of +3 in August. Both Coalition leaders fared badly with Peter Dutton averageing net -18.1 and Sussan Ley -19.1; Albanese led Dutton as Better PM by an average 10.6 points (increasing as the election approached) and Ley by 22.7. It is not unusual for new Opposition Leaders to struggle on such scores, especially when their party is losing the 2PP; indeed historically Ley could easily be behind by more and my sense is that the house advantage to incumbents on Preferred Leader scores has reduced.
Dutton actually beat Albanese as preferred PM in two Resolves early in the year and tied him in one Freshwater. Dutton was also ahead on net approval in all polls by anyone in January or February, but only in 3 out of 32 readings thereafter. Ley had a better rating than Albanese for "performance in recent weeks" in all Resolves through to October, but no-one else except the lone Freshwater. Resolve's latest net rating for her of +2 was a massive house-effect-fuelled outlier compared to Redbridge (-20), Essential (-19) and Newspoll (-29).
The path ahead
2026 will be a big year in terms of where Australian politics and polling is going. If the One Nation surge stabilises or recedes then we may get back to something more normal. If the Coalition then doesn't get supplanted or collapsed then Labor's already very long re-election honeymoon is likely to end sometime, perhaps sooner rather than later. On the other hand the upheavals on the right could overshadow the normal ups and downs of an incumbent government.
There is much speculation about the impact of the Bondi attack but a lot of it is coming from the same sources that believed antisemitism would be a very salient issue at the election (outside a few of the non-classic contests in Melbourne - and not even all of those either - it basically didn't register). Obviously the reality is far more significant than the speculation but will voters blame the government and shift their voting intentions or will they take the view that these things are not so easily avoided and that the government is trying to respond constructively? My own view is that whatever one thinks of the government's response so far, anything is better than US-style "thoughts and prayers".