The final boundaries for the ACT have been released. Maps and numbers of voters transferred have not been released, but I have managed to estimate new margins using data published with the draft boundaries.
In short, nothing much changes. The table below summarise the margins for the ACT’s three electorates.
| District | Contest | Old Boundaries | New Boundaries | Shift |
|---|---|---|---|---|
| Bean | 2-Party | ALP 19.3% | ALP 19.4% | 0.1 to ALP |
| Canberra | 2-Party | ALP 26.4% | ALP 26.4% | No change |
| Fenner | 2-Party | ALP 22.1% | ALP 22.1% | No change |
| Bean | ALP v IND | ALP 0.3% | ALP 0.2% | 0.1 to IND |
| Canberra | ALP v GRN | ALP 19.5% | ALP 19.5% | No change |
There’s more details on each electorate inside the post.
Fenner
Fenner is unchanged. The Commissioners abandoned the proposed suburb swaps with neighbouring Canberra.
Canberra
See note on Fenner on the proposed boundary changes being abandoned, the existing northern boundary of Canberra remaining in place.
The map below gives a rough outline of the new Canberra. The old boundaries are shaded red and I’ve tried to show the new boundary with Bean with a hand drawn black line. (Maps won’t be available until October.)
The major changes proposed by the draft boundaries remain. The suburbs of Curtin, Lyons, Hughes and Garran are transferred from Canberra to Bean. The new Molonglo Valley suburbs of Whitlam, Denman Prospect, Wright and Coombes go the other way from Bean to Canberra.
The two changes compared to the draft are that Norfolk Island now remains in Bean, and that the southern boundary of Canberra now extends south to run along the ACT border and include the suburb of Hume.
Bean
See above map and description for Canberra which in reverse explain the changes to Bean.
Bean finished as a close contest in 2025 between Labor and Independent Jessie Price. The difficulty in calculating a new two-candidate preferred margin for Bean is the addition of areas from Canberra that were not contested by Price. Independent Claire Miles contested Canberra in 2025 but only polled half the vote for Price. If I simply allocate Miles’ votes to Price, I estimate the Labor margin in Bean versus Price rises to 1.4%.
However, I think it is better to build a two-candidate margin based on estimating Price’s vote in the areas added to Bean. The method I have used is to compare results in joint polling places and pre-poll centres along the old border between Bean and Canberra. I estimate that the presence of Price lowered the Labor primary vote by 8%, the Green primary vote by 4%, and lifted the Independent vote by 12%.
Applying these adjustments produces an estimated Labor margin of 0.2% versus Price, a slight weakening of Labor’s position.
As is the way with AEC redistributions, maps for the new boundaries are not released until gazetted in three months time. The hand drawn Canberra map above is the best I can achieve.
Change in Votes Table
| Bean | Canberra | ||||
|---|---|---|---|---|---|
| Party | Old | New | Old | New | Fenner |
| First Preference Vote % | |||||
| Labor | 41.0 | 41.7 | 48.3 | 47.7 | 53.8 |
| Liberal | 23.0 | 23.1 | 18.3 | 18.2 | 21.9 |
| Greens | 9.5 | 9.8 | 19.8 | 19.5 | 16.4 |
| Independent | 26.4 | 25.1 | 10.9 | 12.3 | .. |
| Other | .. | 0.3 | 2.7 | 2.4 | 7.9 |
| Two-Party Prefered % | |||||
| Labor | 69.3 | 69.4 | 76.4 | 76.4 | 72.1 |
| Liberal | 30.7 | 30.6 | 23.6 | 23.6 | 27.9 |
| Two-Candidate Preferred Vote % | |||||
| Labor | 50.3 | 50.2 | 69.5 | 69.5 | 72.1 |
| Liberal | .. | .. | .. | .. | 27.9 |
| Greens | .. | .. | 30.5 | 30.5 | .. |
| Independent | 49.7 | 49.8 | .. | .. | .. |
Hi Antony, looking forward to seeing your analysis. Regarding the challenge of transferring votes between Independents in Bean and Canberra that you mentioned in your Draft Boundaries post, I wanted to bring back a thought shared on Twitter/X a few months ago involving Senate Formal Preferences.
One method would be to leverage the fact that Senate Formal Preferences allow us to simulate vote allocation to any subset of senate parties (thus letting us ‘bypass’ Pocock’s massive vote share in 2025).
Bean -> Canberra: one can use Bean’s Senate allocation among only [‘ALP’, ‘LP’, ‘GRN’] to reduce polling booths in Bean to the set [‘ALP’, ‘LP’, ‘GRN’], and subsequently use Canberra’s electorate-wide DoP table to expand out to Ind. Miles/Other candidates.
Canberra -> Bean: proceeding in reverse, one can reduce each Canberra polling booth to [‘ALP’, ‘LP’, ‘GRN’] using its DoP table. Then, we calculate what proportion of votes Ind. Price pulled from each of the 3 majors by comparing Bean’s Senate votes restricted to [‘ALP’, ‘LP’, ‘GRN’] with Bean’s House votes (which include Price). Then apply these electorate-averaged pull rates to the incoming votes from Canberra to make up Price’s incoming primary.
On the draft boundaries, this method leaves Bean as a 0.3% margin (Price slightly up to 26.5% primary), and Canberra margin at 19.4% (Miles’ primary at 10.8%).
Would be curious to hear if you think this is a fruitful way to handle the transfers between Independents.
COMMENT: I’ve had a look at the numbers in the Woden pre-poll and a couple of joint polling places along the boundary and comparing results in those centres for Bean and Canberra. Based on this comparison, I will alter the primary votes for the preference calculation by the following – Labor -8%, Greens -4% and Independent +12%. Doing this produces a Labor versus Independent margin of 0.3% on the draft boundaries. That’s the method I will apply to the final boundaries. Note that while adjust the primaries for the calculation of a two-candidate preferred result, I do not alter the primary votes shown in the tables.
That seems quite appropriate (and results consistent with the method described above). Does a similar approach work for Peter George/Andrew Wilkie in the Tasmanian redistribution, given there are some quite substantial changes, at least in the draft?
COMMENT: You could. The fact that Peter George is now a state MP means it may not be important to do for Franklin. All the calculations do is come up with an estimated margin for the last election. It gives you something to calculate swing against come the next election for modelling purposes but has little to do with who will win next time.
With the rise of PHON and rumors that Mark Parton will run for Bean as a Liberal in 2028, I’m skeptical if Jessie Price will do as well as she did in 2025 come the 2028 election.
PHON is likely to draw more support from Labor than from Jessie Price, reflecting voter dissatisfaction with Labor. That dynamic could benefit Price too as PHON and Greens viewed as too extreme as alternatives to Labor. However, Price may also lose some of the anti-establishment voters who backed her in 2025, as those voters could shift to PHON instead. It will be interesting to see whether One Nation directs preferences to Price or to Labor, and the Liberals appear to be in a difficult position no matter who they run. Parton should remain in ACT parliament, he has little chance to win. Also note Liberals preferenced Price second in 2025.
Mark Parton won’t leave his position in the Legislative Assembly for a run in Bean that will yield the Liberal Party precisely nothing. Price has I think a good chance in Bean in 2028 – after a brief burst of energy the Labor member seems to be drifting back into the invisibility that nearly lost him the seat in 2025.