Federal, June 2026

01

The Coalition loses five points and the poll can finally say who is third

Four parties are unchanged within sampling error and the Coalition is not. Its fall is what separates One Nation and Labor from the field, which neither of them managed by rising.

The poll excludes 7 per cent of respondents who were undecided. The change intervals use the two effective samples, 754 in May and 871 here, and each party's own share. The election baselines are reconstructed from the swings DemosAU states and sum to 100.0. No two-party preferred figure is published; the report records the calculation method as not applicable.

The May wave could not order any of its top three parties: two points between One Nation and Labor, three between Labor and the Coalition and five between One Nation and the Coalition, each smaller than its own interval. June resolves half of that question. One Nation's twelve points over the Coalition and Labor's nine both clear their intervals of about ±4.5 and ±4.4, so the Coalition is established in third place. One Nation's three points over Labor, against about ±5.0, are no more conclusive than May's two.

That ordering arrives because one party fell rather than because another rose. One Nation's two points and Labor's one are both well inside their intervals, and the Greens did not move at all. A reader looking for what changed between the two waves has exactly one place to look, and a reader wanting to know who leads the federal vote still has no answer from this instrument.

Against the May 2025 election the Coalition is now 13.8 points down, which is the largest fall recorded in the three monthly federal waves and five points further than a month earlier. One Nation is 23.6 points up, Labor 7.6 down and the Others 3.0 down, all of them many times the poll's own interval because an election result is a count rather than a sample. The Greens' 0.8 remains the one movement since 2025 that this instrument cannot distinguish from none.

02

Five points of vote take two-thirds of the Coalition's projected seats

The report publishes a projection of the 150-seat House as a range for each party, beside the previous month's, and states no method for either. The Coalition's range falls from 16–28 seats to 4–11.

A majority of the 150 seats is 76. The top of Labor's range crosses it, where in May the top of its range was 74.

A five-point fall in primary vote removes roughly two-thirds of the Coalition's projected seats at either end of its range. That is the behaviour of a party sitting just below the level at which single-member districts start returning seats at all: between 23 and 18 per cent it stops being the leading candidate in most of the places it was still winning, and there is no second tier of marginal seats underneath to cushion the fall. One Nation and Labor each gain about seven seats at both ends of their ranges, which is where the Coalition's share goes.

The Nationals component is worth separating out. The published split puts the Coalition's range at Liberal 4–10 and Nationals 0–1, against Liberal 16–25 and Nationals 0–3 in May. A projection in which the Nationals' top of range is one seat is a projection in which the regional vote has gone to One Nation, and section 05 shows that happening directly: One Nation takes 42 per cent in regional and rural Australia.

The cautions from the previous wave still apply and apply harder here. The report states no method for the projection, and a range is not a point: nothing published says where inside 4 to 11 seats the projection is most likely to fall, and the difference between those two numbers is the difference between a party with a parliamentary presence and one without.

03

Changing the Coalition's leader in the question changes nothing

The report put a second voting-intention question to the same respondents, offering the Coalition with Tony Abbott as leader instead of Angus Taylor. The Coalition's share under that wording is 18 per cent, the same figure it records as asked.

Nothing in that table moves outside its interval, and the party the question was about does not move at all. Set beside the previous wave's hypothetical, where removing Pauline Hanson from the One Nation line cost that party seven points and returned four of them to the Coalition, the comparison is instructive about where this contest's personal vote sits. One leader's name is worth seven points to the party that carries it. The other's is worth nothing to the party that would carry it.

Each bar is the net rating, the positive share less the negative share, in points. The note gives the three underlying shares and the change in the net since May, which the report publishes itself.

All three leaders are net negative, where in May two of the three were net positive. None of the three changes clears its interval on its own: Taylor's seven points sit against about ±7.1, Hanson's six against about ±8.5 and Albanese's three against about ±8.2. What can be said is the level rather than the movement, and the level is that no federal leader in this question commands more positive opinion than negative.

On preferred prime minister Albanese takes 35 per cent, Hanson 28, Taylor 19 and 18 per cent name none of them. Albanese's seven points over Hanson clear an interval of about ±5.3 and Hanson's nine over Taylor clear about ±4.5, so this question orders all three where the voting-intention question orders only the third place. Taylor's four-point fall since May sits right on its interval of about ±4.0 and should be read as the edge of what the sample can see rather than as an established decline.

04

Forty-three per cent of the Coalition's 2025 voters now say One Nation

The report breaks this wave's vote down by how each respondent says they voted in May 2025. The Coalition's retention falls from 60 per cent to 51 and the share going to One Nation rises from 37 to 43.

The report draws this table as stacked bars and prints a figure only where a segment is wide enough to hold one. An en dash marks a cell the report leaves unlabelled; each of those rows balances to 100 across the cells shown plus the unlabelled remainder, and this briefing does not write a figure into a cell the report does not print.

The Coalition's row is the mechanism behind sections 01 and 02. Of the people who say they voted Coalition thirteen months ago, 51 per cent would vote Coalition again and 43 per cent would vote One Nation, where a month earlier those figures were 60 and 37. No other party in the table loses anything like that share to a single rival, and the direction is the same one the May wave showed, only further along.

Two other rows are worth noting. One Nation's own retention is 94 per cent, four points below May's 98, which is the only sign in this wave of any movement away from it. And 51 per cent of respondents who do not recall voting in 2025 or did not vote say they would vote One Nation now, against 39 per cent in May: the party is drawing more heavily than any other on people outside the 2025 electorate, and those are the respondents whose stated intention is hardest to rely on.

05

Labor leads by eleven points among men and trails by eleven among women

DemosAU publishes no cohort sample sizes for this wave. If a two-way split holds about half the effective sample of 871 each, a gap between the halves carries roughly ±6.1 points, and a gap between thirds about ±7.5.

Every row sums to 100 in the report. The undecided share of 7 per cent sits outside these totals. The report also publishes a breakdown by personal income, which runs from 31 per cent One Nation and 24 Labor below $45,000 to 30 and 23 above $125,000, with the Coalition at 15 and 31 across the same two bands. Cohort figures rest on subsamples and carry wider margins than the ±3.3% national topline.

Labor takes 32 per cent of men and 21 per cent of women, an eleven-point gap against an interval of about ±5.9, and it is the only party in the table with a gender division this wave can resolve. One Nation's 32 among women against 29 among men and the Coalition's 19 against 18 are both well inside theirs. The arrangement is the opposite of the one these cohorts are usually described as holding, and it is the single most citable figure in the wave.

The other divisions that clear are the familiar ones in this series, and June's are the widest of the three waves. One Nation takes 42 per cent in regional and rural Australia against 19 in the inner metropolitan areas, a twenty-three-point gap. The Greens take 29 per cent of under-35s against 5 of the over-55s. Labor takes 35 per cent of respondents who speak a language other than English at home against 25 of those who speak English, and One Nation runs the other way at 21 and 33.

The Coalition's position in this table is what 18 per cent looks like from underneath. It clears a quarter of the vote in two places only, 25 per cent among home owners and 31 among those earning more than $125,000, and it sits at 12 among renters and 12 among the under-35s. What is left of its support is concentrated in older, wealthier and outright-owning Australia, which is the condition the seat projection in section 02 describes in a different language.