One Nation's Rise Upends Australian Election Predictions

One Nation’s spectacular rise from a distant 6% of the vote in the last election to first or second in some recent polls has upended Australian politics. It has also made it a lot harder to predict what exactly will happen at the next election.

Traditionally, pollsters and election experts would look at how preferences flowed in previous elections when estimating two-party preferred numbers, or translating polling into seat projections. This was fairly predictable when almost every seat would come down to a contest between Labor and the Coalition.

But, given the rise of One Nation, we are now in “unknown territory”, says George Hasanakos, head of research at DemosAu.

In last year’s federal election, One Nation was one of the final two candidates in only two seats and made it to the final three in just 26 seats. But the numbers in DemosAu’s big MRP poll in March suggest that, on current numbers, a One Nation candidate would make it into the final three in well over 100 seats.

Pauline Hanson has said she has the ability to do the job of prime minister, and has considered running for the House of Representatives in 2028.

Despite this, One Nation may still have a serious impediment to taking government: inner metropolitan electorates. Hasanakos notes that One Nation is still doing poorly in these seats. This means the party will likely have to win three-quarters of the 107 seats in outer metropolitan and rural and regional areas to attain government.

One Nation has had a steady rise to the top of the polls since getting just over 6% of the vote in last year’s election. In May and June this year, the party was getting primary vote shares in the high 20s and low 30s in polling, slipping slightly behind Labor in a couple of recent polls from Redbridge and Newspoll.

Nevertheless, this is a huge shift. For those trying to figure out how it translates to seats in the next election, most pertinent is where the newfound One Nation support is coming from. This is because of Australia’s preferential voting system, where we cast our vote by ranking candidates in the order of our preference. If no candidate has a majority, then the one with the least votes is eliminated and preferences are used to redistribute their votes.

Only a handful of seats in the 2025 election were decided on first preferences, meaning all the others were affected in some way by how preferences flowed throughout the count. When most seats came down to a contest between Labor and the Coalition, pollsters could more easily estimate flows by looking at what happened in previous elections.

But on current numbers, the majority of seats will come down to some combination of Labor or the Coalition v One Nation, and that has rarely happened before. And a significant share of One Nation’s support now comes from former Coalition and Labor voters, meaning preferences may not flow like they used to.

“We’re in a situation in the polls now where One Nation is the lead conservative party,” Hasanakos says. “But they’ve taken a large portion, perhaps half, of all the conservative voters that used to vote for the Coalition.

“And so the proportion of Coalition supporters who are not fans of Pauline Hanson, all of a sudden they’ve become a larger share of the remaining Coalition voters. So this could very well impact preference flows.”

Polling shows support for the Coalition and Labor sinking across many demographics, including homeowners and gen X voters. Accent Research and Redbridge asked voters to rank issues by how salient they are, showing stark differences for subjects such as housing affordability and immigration:

How-to-vote cards in focus

Pollsters do ask people to rank parties, which feeds into their models. But there are several wrinkles here, including that the candidate who is running in each seat may not be finalised until closer to the election, especially for independents. Many parties also have volunteers handing out how-to-vote cards at election booths.

“Those how-to-vote cards will have some influence,” says Murray Goot, an emeritus professor and politics expert at Macquarie University.

He says a “significant” number of voters will simply follow the cards, which means polls that ask people for their preferences may be “misleading” to some extent.

Data from Victorian and South Australian elections suggests about 40% of major party voters follow their party’s how-to-vote cards. But is the Coalition still a major party? And if the decline solidifies, will it have enough volunteers to staff every booth?

All of this adds up to a lot of uncertainty about how polling support will translate into actual seats. Goot acknowledges that One Nation are doing well in polls. “But the serious matter is seats,” he says. “You’ve got to be able to translate votes into seats, and there’s no evidence that One Nation is winning as large a proportion of the seats as it is of the votes.”

To illustrate this we’ve put together an interactive seat map based on DemosAu’s March MRP poll. The pollster’s model found 84 seats where Labor and One Nation were the final two candidates, with the Coalition third. But how would those Coalition voters direct their preferences in such a contest?

The default is the preference flows that DemosAu used, but given how many tight races there are, even slight changes have a big impact:

It also striking how relatively few of these Labor v One Nation contests are in metropolitan areas – where there are almost 90 seats. Without a clear path to government for One Nation, the “interesting question”, Goot says, is “whether a combination of One Nation seats plus Coalition seats might amount to something close to a majority”.

“And for that, the Coalition has to win some metropolitan seats.”