FED25 Election – New Seat Margins and Electoral Pendulum

At the 2025 Federal election, more than half of seats in the House of Representatives will be contested on new electoral boundaries.

In this post I summarise the state of play going into the election and provide estimated margins for all seats affected by the redistribution. You can find a listing of margins for all seats at this link.

The above document lists traditional two-party preferred margins for all seats, but also alternative two-candidate preferred margins in seats that were not contests between Labor and the Coalition in 2022.

There were a record 27 seats in 2022 that did not finish as contests between Labor and Coalition candidates. There is every chance this number could increase in 2025, with analysis of results further complicated by the new electoral boundaries.

I’ve tried to create a new electoral pendulum that summarises the party contest for 2025. You can find the pendulum at this link.

The pendulum groups seats by holding party, simplified to Labor, Coalition and ‘Other’ groupings. Within each party grouping, seats are listed in margin order. Second placed party is also shown for seats that were not Labor versus Coalition contests in 2022.

The problem with this layout is that it mixes two-party and two-candidate margins. The ordering allows you to identify marginal seats, but makes it impossible to apply uniform two-party preferred swing to try and predict seat numbers. It is best to skip over the ‘Other’ contest seats in applying two-party swing.

On a third page of the pendulum document I’ve added mini-pendulums that set out margins for Labor versus Others and Coalition versus Others contests. For several electorates where Labor and the Greens compete on first preferences to make the final contest versus the Coalition, I have also provided an estimated primary vote margin.

There’s more about the redistribution and information on how the new margins have been calculated inside this post.

Why have there been Redistributions?

A redistributions in a state or territory is triggered seven years after a set of boundaries has been put in place, or whenever the number of House seats a state is entitled to changes according to a formula set out in the Constitution.

Changes in seat entitlement have triggered major redistributions in three states. New South Wales and Victoria have each lost a seat, while Western Australia has gained one. A minor redistribution for the Northern Territory was triggered by the seven year rule.

The changes reduce the size of the House of Representatives by one, the 2025 election to be a contest for 150 seats. Majority government will need Labor or the Coalition to win 76 seats.

Estimating New Margins

After every election, the Australian Electoral Commission (AEC) publishes detailed results by vote centre and for various categories of declaration vote. The AEC also publishes roll mark-off data by SA1. An SA1 is the smallest unit used by the Australian Bureau of Statistics to measure population and collect census data. All addresses exist in an SA1, so the number of electors from each SA1 that voted at a particular centre can be accumulated.

Using the AEC’s data, it is possible to allocate a proportion of votes by party from each polling places to an SA1. After a redistribution, you can then add up the votes by party for all the SA1s in a new electoral division to come up with an estimated vote by party and margin for the new division. This is how I have calculated two-party preferred margins for all seats.

Calculations of this type can only provide an estimate of new margins. Accumulation by party cannot take account of a candidate’s personal vote. Personal vote will be diluted by the transfer of new areas into a seat, or the transfer of votes for the candidate out of the seat. Redistribution calculations also cannot take account of local factors that intensified the party contest in particular seats.

Estimating margins is easier when all areas in a new seat finished with the same final pairing of party candidates. If all areas of a new seat were Labor versus Coalition contests, or Labor versus the Greens, you can simply combine the final two-candidate votes across the new seat to produce an estimated margin.

But what about when a new seat takes in parts of seats where the final pairings of candidates differ? This is especially the case with Independents. How do you asses the support for an Independent when a re-drawn seat includes new areas the Independent did not contest in 2022? For example, the re-drawn Melbourne seat of Kooyong includes 26,000 electors from the abolished seat of Higgins who did not have Dr Monique Ryan in their ballot paper in 2022.

At the 2022 Federal election, 27 of the 151 divisions did not finish as traditional two-party preferred races between Labor and the Coalition. Many of these seats present problems in estimating new margins.

The method I have adopted is to try and adjust party shares based on a comparison of House and Senate vote shares. The method is imperfect, but so are all alternatives. The aim is to come up with a margin that can also be applied to polling place history as a basis for analysing results on election night in 2025.

The Political Impact of the Redistributions

Two seats have been abolished and one created by the redistributions. In Victoria, the Labor-held inner-Melbourne seat of Higgins has been abolished, but Labor’s loss has been balanced in Western Australia by the creation of a notionally marginal Labor seat called Bullwinkel in Perth’s east.

In New South Wales, Independent Kylea Tink’s seat of North Sydney has been abolished and divided between three neighbouring electorates leaving Tink no obvious seat to contest.

The redistribution has also changed the notional party status of two seats. Having absorbed large parts of abolished North Sydney, the marginal Labor seat of Bennelong is now a notionally a very marginal Liberal seat. In Melbourne, the abolition of Higgins has affected the margins in seats across eastern Melbourne, with the marginal Liberal seat of Menzies now having a notional Labor majority.

At the 2022 election, the final result was Labor 77 seats, Coalition 58 and crossbench 16. Post-redistribution those numbers become Labor 77, Coalition 58 and crossbench 15. If you include Labor’s gain of Aston (VIC) at an April 2023 by-election, the major party numbers are now Labor 78, Coalition 57.

As a rough guide, a swing of under 1% would deprive Labor of its majority. But the swing would have to be around 3.6% for the Coalition to finish with more seats than Labor, and 5.3% to achieve majority government in its own right.

But these numbers treat the contest as a two-party race and ignore that one-sixth of contests were not two-party races. Were Labor to win back any Green seats, its electoral buffer would increase. If the Liberal Party were to recover any seats lost to Independents in 2022, down comes the required swing for the Coalition to win more seats than Labor or achieve majority government.

Simply reducing the election to a two-party contest misses the increasing complexity created by elections where the first preference vote splits three-ways between Labor, the Coalition and third parties plus independents.

Notes on Individual Margins

The tables attached to this post treat Aston as Labor held with its by-election margin of 3.6%. Come the next election the result will be analysed compared to the Liberal two-party result in 2022, but the seat will be classed as Labor held.

In three other seats where there have been by-elections, Fadden, Cook and Dunkley, I have used the 2022 election margins, adjusted to new boundaries in the case of Cook and Dunkley.

Two Coalition members have left their party since 2022, former National Andrew Gee in Calare (NSW), and former Liberal Russell Broadbent in Monash (VIC). On current information, both will re-contest their seats as Independents. I have left both seats categorised with their 2022 winning party and margin. If either member is re-elected in 2025, their seats will be flagged as Independent gains.

The seats listed below have changed enough to significantly alter the their political importance to the result of the 2025 election.

• Bennelong (NSW) – the abolition of North Sydney drags Bennelong eastward, picking up traditional Liberal territory and turning this Labor 1.0% seat into a marginal Liberal seat with a tiny 0.04% margin.
• Bradfield (NSW) – gains large parts of Independent held North Sydney cutting Bradfield’s Liberal margin versus an Independent from 4.2% to 2.5%.
• Eden-Monaro (NSW) – boundary changes with Hume and Riverina reduce the Labor margin from 8.2% to 6.1%.
• Greenway (NSW) – shifts further north into Sydney’s Hills district, reducing the Labor margin from 11.5% to 7.9%.
• Hughes (NSW) – loses Liberal voting areas in the Sutherland Shire, replaced by Labor voting territory south of Liverpool. The Liberal margin halves from 7.0% to 3.5%.
• Parramatta (NSW) – loses districts south of Parramatta to Blaxland, Labor’s margin reduced from 4.6% to 3.7%.
• Paterson (NSW) – loses Kurri Kurri to Hunter reducing the Labor margin from 3.3% to 2.6%.
• Wentworth (NSW) – this traditional Liberal enclave in Sydney’s eastern suburbs expands to take in surrounding Labor voting suburbs. That translates into a boost of Independent member Allegra Spender, her margin versus the Liberal Party increasing from 4.2% to 6.8%.
• Chisholm (VIC) – the abolition of neighbouring Higgins halves the Labor margin from 6.4% to 3.3%.
• Goldstein (VIC) – gains some Labor voting territory from neighbouring Hotham and Isaacs, increasing the margin of Independent Zoe Daniel over the Liberal Party from 2.9% to 3.3%.
• Kooyong (VIC) – 16,000 voters from the abolished seat of Higgins have been absorbed into Kooyong. This makes it difficult to calculate a new margin of Independent Monique Ryan. My best estimate is that Ryan’s margin versus the Liberal Party slip from 2.9% to 2.2%.
• Menzies (VIC) – the re-arrangement of boundaries in eastern Melbourne wipes out the 0.7% Liberal margin for Menzies, leaving the seat with a notional Labor margin of 0.4%.
• Wills (VIC) – gains parts of Melbourne, the safe seat of Greens Leader Adam Bandt, reducing Labor’s margin in Wills versus the Greens from 8.6% to 4.6%.
• Canning (WA) – Liberal two-party preferred margin cut from 3.6% to 1.2%.
• Hasluck (WA) – loses the Darling Range suburbs east of Perth to the new seat of Bullwinkel while gaining the rest of Ellenbrook and nearby Labor voting suburbs. The Labor margin increases significantly from 6.0% to 10.1%. Labor margin boosted from
• Lingiari (NT) – loses eastern parts of Palmerston increasing the Labor margin from 0.9% to an estimated 1.7%.
• Solomon (NT) – gains parts of Palmerston cutting the Labor margin from 9.4% to 8.4%.