Polling update

Because the reaction to the May Budget appears to have been both quick and significant, I have loosened the day-to-day innovation term in my Gaussian Random Walk model from 0.1 to 0.15 percentage points. This term is the standard deviation of the day-to-day change in underlying voting intention: a larger value lets the modelled series move more rapidly in response to genuine shifts. It also means the model is less able to filter out noise, and the median lines will not be as smooth as they were previously. 

Primary votes

The aggregation tells a story of a fracturing major-party vote. Labor's primary has fallen almost continuously across the year, sliding from the high 30s to a current median of 27.9 percent, well below its 34.6 percent result at the 2025 election and the lowest point in the series. The Coalition has fared no better, drifting down from around 30 percent to 20.6 percent, with a notable step-change late in the year and only a brief plateau through autumn before resuming its decline.

The clear beneficiary is One Nation, whose primary has surged from roughly 8 percent to 28.5 percent, overtaking both major parties at the very end of the period and tracking above its 6.4 percent election result by a wide margin. The Greens have been broadly stable in a narrow band, lifting modestly to about 12.5 percent, while the "Others" vote has eased from the mid-teens to around 10.8 percent. In short, support has drained out of both Labor and the Coalition and flowed substantially to One Nation, leaving a contest where three forces now sit within roughly seven points of one another rather than the familiar two-party structure. 

Polling Update

Primary votes

While it looks like Labor and the Coalition have each fallen a similar amount in the polls, around 7 and 7.3 percentage points respectively, this comparison uses each party's mid-2025 polling level as the starting point. A fairer benchmark is the 2025 election result, which is what each party actually achieved at the ballot box. Measured this way, the picture changes sharply. Labor won 34.6 percent in 2025 and now sits at 29.5 percent, a loss of about 5 points. The Coalition won 31.8 percent and has fallen to 22.4 percent, a collapse of more than 9 points. The reason for the difference is that Labor was polling above its election result through mid-2025 (a post-Election honeymoon) and only later sank below it. The Coalition, by contrast, was already polling below its 31.8 percent election result from the outset and kept falling from there. So although the two parties have shed a similar amount in the polls since mid-2025, the Coalition's decline relative to its actual electoral support is nearly double Labor's, and One Nation has absorbed almost all of the combined loss, rising from 6.4 percent at the election to 25.3 percent today.

The Slow Death of 2pp

For decades, Australian political watchers have leaned on one figure above all others. Not the primary vote share. Not preferred prime minister. The two-party-preferred number, the 2pp, the share each major bloc ends up with once preferences are distributed. It is two things at once: the headline number pollsters publish, and the figure the electoral commission reports on the night. In most elections it has been the number that decides who governs. That dual role is why pundits reach for it (with exceptions in the top left and bottom right quadrants of the next chart).

A Deep-Dive on One Nation Polling

One Nation's primary vote has surged from ~9% in mid-2025 to 23.5% in my latest poll aggregation. That's nearly 4x their 2025 federal election result of 6.4%.

The trajectory is remarkable. Steady climb through late 2025, accelerating from October onward, then plateauing around 23% from February 2026. The leadership changes (Taylor for the Libs, Canavan for the Nats) coincide with the plateau rather than the surge itself.

Attitudinal Polling

 Preferred Prime Minister

Aggregated Polling

 Two Party Preferred

Primary votes

Aggregated polling

Aggregated Attitudinal Polling

I am now applying the same Bayesian aggregation techniques I use to the opinion polling. I wrote extensively about those techniques here. The results are as follows:

Albanese - preferred leader

Bayesian aggregation

Now that we have enough data (around 40 national polls since the last election), we can start looking at a Bayesian aggregation using a Gaussian random walk to estimate voting intention from day to day.  I will tease with the chart that intrigues me the most: the rise of the One Nation Party since the 2025 election.

Some movement on the betting market