There were 21 seats in 2025 where an independent candidate made the two-candidate-preferred count. For today’s post, I want to explore what we can tell about those voters from their other voting patterns. For voters who ended up with the independent in the two-candidate-preferred count: how did they cast their primary vote, and where did they rank Labor or Coalition?
For some of this analysis, it is important to analyse seats where the independent was primarily opposing Labor or primarily opposing the Coalition separately.
This first chart shows the 14 independents who made the 2CP count against the Coalition.
They have been ranked by the size of their 2CP vote. The six lowest-ranked candidates did not win.
For most candidates there is a strong relationship between the size of their primary vote and their 2CP but there are some exceptions. Andrew Gee in Calare polled a lower primary vote than seven independents who achieved a lower 2CP than him, including three who ended up losing. This is mostly driven by 12.8% of the vote for Kate Hook that flowed to Gee. Hook’s total primary vote was 15.8%, so her preferences favoured Gee by more than 4:1.
The size of the Labor vote was relatively larger in Bradfield and Flinders. The primary vote of the independents in Cowper and Wannon would have been almost enough to win but for the smaller pool of preferences, with both Labor and the Greens polling worse than in more urban seats.
It’s also worth noting that not one of these candidates was elected on their own primary vote. Helen Haines came closest, but her primary vote was in the low 40s.
This next chart shows the seven seats where independents made the 2CP against Labor. This group is much newer – there were only two seats in this category in 2022.
Andrew Wilkie was miles out in front, and almost won a majority of the primary vote. Dai Le was the only other to win, but two others came close.
Wilkie and Fowler gained much smaller Liberal vote shares than Price or Hulett. This mostly reflects the relatively smaller Liberal primary vote overall, but surprisingly Price did better on Liberal preferences than Le.
Carly Moore has a notably lower primary vote than the other independents, but did better thanks to the ‘others’ vote (mostly the two other prominent independents).
This next chart compares seats where the same independent made the 2CP in 2022 and 2025, either against Labor or Coalition. Worth noting these haven’t been adjusted for the redistribution.
In most cases, independents strengthened their primary vote, even if they didn’t improve their 2CP. Alex Dyson had a much better primary vote in Wannon, but he barely improved his 2CP position. His growth mostly came at the expense of Labor and the Greens. Monique Ryan’s primary dropped, but a lot of that could be explained by the addition of parts of Higgins, that had a sitting Labor MP. The drop in Warringah would partly be due to the redistribution removing Steggall’s suburbs and adding in parts of North Sydney.
Suzie Holt shocked most by coming from fourth place to make the 2CP in Groom in 2022. This year she has consolidated her primary vote, but is not much closer to winning, having mostly taken those votes from others who preferenced her.
Now let’s look at how those same voters split on the 2PP, between Labor and Coalition. You can derive this statistic by comparing the 2CP and 2PP both to the 3CP.
This chart contains both IND vs Coalition and IND vs Labor seats, although it can’t include Calare or Indi where Labor didn’t make the 3CP.
One thing worth noting is that the voters who are part of an independent’s 2CP coalition will depend a lot on who they end up in the count against. Andrew Wilkie would easily win a 2CP against either Labor or the Liberal Party, but the make-up of that group would differ quite a bit in its 2PP.
I’ve sorted seats by the proportion of the 2CP that favoured Labor over the Coalition. This has naturally sorted the IND vs ALP seats all to the bottom.
Over 87% of Monique Ryan’s 2CP vote went to Labor over Liberal. Indeed three of the top four were in the Greater Melbourne region. Regional seats like Farrer, Cowper, Groom and Wannon all ranked lower, but a substantial majority still preferenced Labor over the Coalition. It’s likely a winning independent there would need to win over more Coalition voters, and thus this metric would go down. It certainly was much lower when Tony Windsor was winning in New England.
The 2PP flow in Clark was much higher than the other independent vs Labor seats.
Less than half of the 2CP for the independents in Bean and Fremantle flowed to Labor, while Franklin was higher. I was surprised that the rate in Watson was so low.
It’s also worth noting that we now are starting to collect data on how Greens voters preference when they face a Labor vs Independent choice. Understandably the Greens overwhelmingly favoured independents against the Coalition, ranging from 81.9% in Farrer to 92.5% in Wentworth. But in contests against Labor, only Andrew Wilkie gained a high proportion.
Famously, the distribution of Greens preferences in Calwell was a massive setback for independent Carly Moore. It’s not surprising the conservative independent Dai Le didn’t do great, but Muslim independent Ziad Basyouny didn’t do much better. Even in Fremantle, Bean and particularly Franklin, where independents could arguably be seen to be quite aligned with the Greens, they barely gained 50% of Greens preferences despite being ranked high on the Greens how-to-vote card. The preference to Labor is quite strong amongst Greens voters.
Finally, this chart shows the 2PP make-up of the independent 2CP in six urban teal seats in both 2022 and 2025.
In every case, the independent 2CP has become more Labor-leaning from 2022 to 2025, sometimes notably so. This could reflect the redistribution in some seats, but it happens everywhere.
This is more evidence for a thesis that Australian politics is polarising further into a Labor-leaning bloc that includes the Greens and the teal independents, and a Coalition-leaning bloc that includes One Nation. A funny thing about the preferential voting system is it lets voters do this directly, whether the parties want it or not.
I’m surprised Green -> Basyouny flows were so poor. I’d have thought he’d have been higher on their HTVs, and a lot of Greens voters somewhere like here would do it anyway?
They preferenced Basyouny #2.
I doubt they had a good level of coverage in a seat like this but the bigger issue that Greens voters don’t pay attention.
I presume Bradfield is missing from the later charts because there is not yet a proper 2PP count?
Boele’s primary vote did not strengthen as much as the chart indicates in reality, as she had all the Tink votes from the addition of North Sydney areas to sweep up (and the Teal like IND from Bradfield in 2022 didn’t run this time)
Ben where do you find the information about the ‘Primary Origin’ seat by seat. I am right, for example, in assuming that in Goldstein, the 11.0% of those who voted 1 for Labor and gave a preference to Zoe Daniel represents 80.3% of those who voted 1 for Labor, in other words roughly 12696 out of 15812 who voted 1 Labor. In other words, can I say for sure that 3116 Labor votes preferred Tim Wilson?
It is notable that a large minority of Green voters will preference the ALP directly, even against a green-friendly Independent. At the exclusion of the Green in Forrest, the vote split evenly between ALP and Independent. In Curtin nearly 40% of Green preferences went first to Labor rather than to Kate Chaney.
In the March contest for the State seat of Fremantle, Kate Hulett was defeated by a considerable leakage of Green preferences to the ALP, offsetting an equally large leakage of Liberal preferences to Independent – both sets of voters ignoring HTV cards.
In contests where every ballot paper is scanned (NSW LGA, NSW HoA: 2017 to 2024) I have found that only about 6% of Greens voters will follow a GRN ticket.
How’s THAT for discipline?
@Jeremy Buxton , are those Greens voters in Forrest, Fremantle, etc (even in Fisher & Grey), aware that preferencing a Labor candidate (who might do relatively less well in the final two-candidate count against the Liberal) earlier than an environment-aware Independent (who might do relatively better in the final two-candidate count against the Liberal, and thereby be better placed to defeat the Liberal) has real-result consequences, or aware but don’t care, or aware and actively dislike environment-aware Independents more than Liberals?
Hi Stephen,
If you go to the results website, the House downloads page, the correct file is “Two candidate preferred flow of preferences by candidate by division”.
In the case of Goldstein, you can immediately see that 80.67% of Labor voters gave their 2CP to Daniel and thus 19.33% went to Wilson.
You can also see the raw numbers, so if you multiply those by the total formal vote for that seat you’ll get proportions of the overall vote. I then stack those so they add up to the total 2CP for that candidate.