The Australian Electoral Commission yesterday published the official statistics that will be used for federal redistributions in South Australia, Tasmania and the ACT. This gives us a bit more insight into what is likely to happen in those redistributions.
Electorates must be drawn so each seat is within 10% of average enrolment at the start of the process (in this case, August 2025), but also so that seats are within 3.5% of the average for a projected enrolment three-and-a-half years after the end of the process (in this case, April 2030). That second rule is much stricter, and thus ends up being more important in the drawing of new seats.
In a previous post, I looked at the population patterns based on June enrolment data. But we didn’t have projected data until yesterday afternoon. So for this post I will look at that second set of datapoints.
Not one seat deviates from the first quota by more than 10%. Nine seats deviate from the projected quota by more than 3.5%: Barker, Bass, Bean, Canberra, Clark, Lyons, Makin, Mayo and Spence.
Firstly, South Australia.
The biggest deviations are in the northern suburbs of Adelaide. Spence is projected to be 12% over quota, while the neighbouring seat of Makin is almost 8% under quota. Mayo is also about 8% over, while Barker is almost 4% under quota. Pretty much every other seat is slightly under quota.
It seems most likely that Spence will shrink, and the surplus voters will mostly go into Makin. Mayo will also need to shrink, likely giving some of those surplus voters to Barker.
In Tasmania, the Hobart-area seat of Clark is projected to be 10% under quota. So Clark will need to grow. Franklin is 3.2% over quota, so can absorb some of that growth, but it will be necessary for Clark to expand north into parts of northern Hobart that are currently contained in Lyons. That central seat is almost 10% over the projected quota. Bass is also 4.4% under quota, so will also need to take more voters from Lyons.
Right now Lyons includes outer suburban areas in both the Hobart and Launceston areas, and they will be the first to be cut, making Lyons more of a rural seat. It’s hard to see any scenario that doesn’t make Lyons more favourable for the Liberal Party.
There just aren’t that many different ways to shift populations in these small jurisdictions with a small number of electorates, and that is particularly true of the ACT. The growth in the ACT has been fastest in the south, with the southern electorate of Bean 7.6% over quota. While the northern seat of Fenner is slightly under quota, it’s the central seat of Canberra which is most under. So I expect we’ll see Canberra expand south, probably taking in parts of Woden or Weston Creek. Fenner can be mostly left alone, but it could gain some more of the Belconnen area to produce more equal numbers.


Preity much exactly sums up what I’m looking at Ben.
I started working on South Australia and have basically done that, although I’ve added Mitcham council ban info Boothby too.
I think in the ACT, Canberra cleanly takes the eastern part of Woden, Hume and Symonston, although I might look at what a shift in the Molonglo Valley looks like before submitting my “idea”.
Finally Tasmania, I think it depends on whether it’s viable to try to fix the split in Franklin, and move Clark southward. I’m not sure, as we know, and you mentioned in your previous article, how averse to majority changes the Committees seem to be lately.
Looking forward to seeing ideas from everyone.
The non contiguous seat has to be fixed
Thanks Ben. Any reason to suggest Canberra would gain from Bean in Woden or Weston Creek rather than Molonglo Valley? Molonglo Valley is arguably more connected to Canberra – Cotter Rd and Coppins Crossing Rd are the major arteries and both go to Canberra rather than south to other parts of Bean.
I suspect Makin will become stronger for Labor and absorb parts of Spence. Spence will weaken for Labor but not to the extent it will be under threat.
I think Kingston should absorb Aldinga from Mayo maybe even Sellicks beach. Kingston can move Southwards.
Canberra should be drawn to the Molong river
And Hindmarsh drive solved. Leave fenner as is. Solved act in 5 secs
Agree Nimalan, Spence although being considered as a replacement for the old Wakefield district is more like the previous seat of Bonython that was a Labor stronghold centred around Virginia, Parafield Gardens and Munno Parra.
Spence does extend further north to Gawler, but it now has the same sort of demographics as its neighbouring suburbs and is no longer the rural township it once was.
An expansion of Parliament may well see the two previous seats (old Wakefield and Port Adelaide) re-created as all existing seats contract.
@ Yoh An
Correct Spence is a bit weaker than the old Bonythorn due to Gawler. As you said correctly Gawler is urbanising and increasing part of Adelaide Metro. We can see how state seat of Light has changed from Safe Liberal to Safe Labor. I do accept once parliament is expanded there will be two rock solid Labor seats in Northern Adelaide (recreated Port Adelaide and a smaller version of Spence)
Tasmania – put the rest of wilderness east into Lyons (it only has 2 electors so dead ing on where they are that basically a 0 transfer. Lyons takes everything west of pelverata road from franklin.along with the west coast lga from Brandon. Franklin gains all Sorell Tasman and Clarence from Lyons clark gains everything west of the north west Bay river from franklin . This makes the non contiguous parts of franklin reduced to just the remainder of kingston. Bass takes everything north of scarmander river from Lyons.
John I agree that non-contiguous Franklin is a blight on the electoral landscape. But I don’t like the chances that they’ll actually do anything. It’s been that way since 1992 and no one, except for us, really seems to care that much.
Frank, and John, I’m definitely thinking it’s worth exploring Canberra expand toward Molonglo Valley district. I’m going to look at it a bit closer in the coming days, but my rough and ready solution is Phillip, Isaacs, O’Malley, Hume, Symonston and Jerrabomberra into Canberra.
It would be good to be able to join all of Belconnen together, likewise for all of Woden, but I can’t see any way that’s possible.
Nimalan, I looked at Kingston taking some of Aldinga from Mayo, but it’s too much for the whole area down to Sellicks Beach. I mean, without making significant changes elsewhere.
Looks like these will be three very minor redistributions with not much change, unless the Tasmanian committee buys the contiguity argument about Franklin. I’d hazard a guess that they won’t.
Sa mayo gains territory up to the okapringo River from Kingston. Kingston’s gains the parts of the city of okapringo north of the river from mayo. And parts of boothby to sturt road and southern expressway. Boothby gains the remainder of mitcham from mayo. Sturt loses city of unley to adelaide barker gains the northern parts of the Adelaide Hills, boothby Stirling sa1, sturt Summertown and mayo retains woodisde/ Lobethal. Barker also gaing barossa council from spence. Hindmarsh u chnaged. grey gains light council from spence. Makin gains everything south of little para river from spence
@darren while i doubt they ll remove it entirely i will make the case for as much as possible. I ve drawn it so it’s now onto the parts of kington that are non contiguous. Solve the rest at the next one. I solved canberra in 5seconds simply move it to the river and Hindmarsh drive
@ Darren ok maybe up to Sellicks Beach does not work. I do feel Aldinga fits better with Kingston than the Adelade Hills. Mayo has some growth areas so needs to shrink
I reckon Kingston works better with parts of boothby. The okapringo River is the perfect boundary. Boothby works better with adelaide hills. The torren river also seems like a good future boundary. Couldn’t justify doing anything major to adeliade or Hindmarsh this time around
@Nimalan and John, Re-checking, it is possible to move all of Aldinga SA2 into Kingston, then by moving either Hallet Cove or Sheidow Park-Trott Park (or finding a way to split them) into Boothby, and in turn, partially straightening that angled Boothby-Adelaide boundary with Goodwood-Millswood going into Adelaide. I thought I had it all figured out neatly in one day.
I’m going to have to re-evaluate it to see where the better Community of Interests lie.
@John, I’m also going to advocate to make Franklin contiguous, and probably take the opportunity for a name change given the significant changes, and the fact John Franklin was merely a British military officer who had an extensive career outside Australia.
Darren, those changes appear to be a partial reversal of the 2017 redistribution for SA’s federal seats which abolished Port Adelaide. As part of that redistribution, Aldinga was moved from Kingston into Mayo, with Kingston gaining Hallett Cove from Boothby. The Aldinga change did attract some criticism and objections from the local community, who argued they have better connection with the wider Seaford area in Kingston.
Thanks Darren
i think Boothby can pick up areas such as Hallet Cove or Sheidow Park i would like Mayo to be focused on Adelaide Hills and more of a peri-urban rather than Metro seat.
Ive made mayo more metro I’ve pushed it up to the river on the south side and retracted it to the river on the north side. Kingston moves to sturt road. Boothby sturt and barker then move into the Adelaide Hills. I was wanting to remove the hills entirely but it didnt work out. It doesn’t work out with adelaide and Hindmarsh being close to quota but I’d like to see a future redistribution use the Torrens river as a boundary. That angled boundary is the city of unley in though about that but divided to leave it in to keep boothby at quota due to other changes.
@darren. I’ve made it more contiguous so a bit of compromise between the status quo and massive change. Franklin gains Clarence sorresna d tasman from Lyons Lyons gain Huon Valley and Clark moves up to the river on the west side. That way only parts of kingston are left on the opposite bank.
You I’ve solved that problem by putting seaford into mayo
Interesting ideas shared so far.
I’ll probably not submit something, being a non-resident of these states, but it’s still an interesting exercise.
If Lyons gains Huon Valley, would that not make that seat non-contiguous, in reality if not on the map?
There would be no way for those residents to drive to the rest of Lyons without going through Clark and Franklin.
Not true real talk. There are some back roads that lead there. Besides they already have to do that
Im guessing this redistribution will notionally flip bean to an ind seat on paper. That should worry David Smith and encourage Jessica price
What backroads?
There are no state highways that lead from Huonville to New Norfolk (for instance) without having to use the A6 or Highway One through Kingston and Hobart. There is a national park to the west of Huonville.
“Besides they already have to do that”
Nobody from Huonville needs to drive north of Hobart to see their federal member.
I honestly don’t really understand the zeal to eliminate the discontiguity of Franklin when all the supposed “solutions” to it produce boundaries that are at least as bad.
Where are the roads from the Huon Valley to the rest of Tasmania that don’t go through Hobart, other than dirt trails?
And in relation to the more “radical” approach – if Clark spans both sides of the Derwent but doesn’t have a connection at the Tasman Bridge – and the Hobart CBD is in the same electorate as the Huon Valley – is that really any better than the existing Franklin?
Whys that isn’t their member in franklin on the other side of the river?
If you zoom in yes there are.
Seats are meant to be non contiguous. The only exception to this other then franklin is fenner
Really, nothing needs to happen in SA except Makin taking Salisbury south of the river from Pearce and Barker taking Adelaide Hills from Mayo. If you want, you can have Kingston and Boothby gain from Mayo too to make Mayo’s western boundary align with the Adelaide metro area around Craigburn Farm and Hawthorndene.
The ACT is solved by moving Phillips north of Hindmarsh Ave from Bean to Canberra. If you want to improve it, you can also move Whitlam (I.e. Molonglo north of the river) from Bean to Canberra.
In Tassie, Bass so solved by taking Blackstone Heights and Prospect Vale from Lyons. Around Hobart, I would have Clark move south into Franklin around Kingston, and Franklin gain from Lyons around Old Beach-Otago as far north as Gagebrook. I don’t see how Clark can extend north without crossing the Derwent, which in my mind is less appropriate as it makes two non-contiguous divisions, if it weren’t for the Bridgewater bridge. But I’m more open to suggestions on that.
For what it’s worth, my changes to Tasmania would be pretty extensive, and therefore probably highly unpalatable to the one in five Tasmanians who are in a new division.
Braddon is unchanged.
Bass gains St Helens-Scamander from Lyons.
Lyons is otherwise unchanged.
Clark and Franklin would be effectively abolished.
The new version of the seat of Clark begins at the City of Hobart northern boundary and extends southwards to include the entire ‘west shore’ of the former seat of Franklin.
The new seat of [Insert More Appropriate Name Than Franklin Here] is simply the City of Clarence and the City of Glenorchy, or the northern part of the old Clark and the east shore of the old Franklin. There is a road link via the Bowen Bridge between the two sides of the electorate which is otherwise split by the River Derwent.
Given the inherent conservatism of federal redistributions, I don’t expect this to get up.
Anyhow, Tasmanians would know better than me what works and what doesn’t.
John, I seriously suggest you consult an atlas. Or ask a local.
Or just tell me
In case some haven’t seen it yet, just wanting to plug my hopefully-useful redistribution tool one last time:
https://auredistribution.neocities.org/
Even better, Tally Room contributor JWood has done some amazing work at building an ‘enhanced’ version with some excellent UI improvements and the ability to save and load your proposals. Certainly much more comprehensive that what my limited Javascript skills would allow for, and we’ll see what we can do to get the two codebases unified and more features/datasets to the community more quickly.
As always, if you have any feature suggestions or run into any issues, just let us know!
JWood’s Redistribution Tool can be accessed here:
https://auredistribution.com/
Massive props to Angas for developing his tool, which so many of us have played with.
I got very excited, so made some additions over the past month, building off his great efforts.
We’re having a think how we can bring them together, and keep it getting better and better.
I’m very excited about what it will allow us to achieve as a community.
In South Australia, the first thing I would investigate is if Spence’s share of Light LGA and Barossa LGA is enough to get Grey and Barker, respectively, up to tolerance. If so, there’s a strong case for leaving the Barker/Mayo boundary as is. This has benefit of avoiding splitting regional LGAs and allowing the most suburban parts of Mayo to shift into Kingston and Boothby. The downside is it would mean more changes in metropolitan Adelaide than would otherwise be the case.
In Tasmania, the shortfall in Bass provides the opportunity to add in parts of suburban Launceston. The removal of these areas (Prospect Vale et al) was contentious last time.
Ive gone the other way moving bass in north eastern Lyons and then having Lyons taking the western shore of braddon.
What an incredible gift to everyone, JWood and Angas. Thank you both sincerely.
John, I have told you. You’re welcome to create your own conclusions, but you can’t create your own geography. Anyhow. Life’s too short to argue with strangers on the internet. 🙂 Enjoy your flight.
Yes but the non contiguous seat cannot continue forever. Sooner or later the aec has to fix it. Unfortunately the numbers don’t currently exist to fix it without radical change. Hence I’ve gone for something in the middle.
Great tool indeed Angas. On these numbers, Adelaide Hills LGA will have to be split between Mayo and Barker. The split of Light LGA between Grey and Barker will also need adjustment. Makin will assume most of the Salisbury area; the boundary with Spence will more closely follow Little Para River. Boothby will pick up a suburb in the foothills from Mayo (probably Hawthorndene). The other electorates can remain unchanged.
@JWood @Real Talk
It’s great to have another (and dare I say better) set of hands pushing things forward! Plus borrowing on some ideas from other incarnations shared previously (by I think by James and Kevin?), as well as the valuable support and feedback from everyone here at the Tally Room. And thanks to @Ben Raue for providing a forum for these discussions which I know can sometimes get a little off track!
I know some of us do enjoy poking around in Excel/QGIS every now and then, but my hope is that tools like this can save people a lot of time. I’m sure some people are particularly interested in seeing how the recently mooted parliamentary expansion might end up looking.
@David
Thank you!
Have been looking over South Australia as well and it seems to be the most straightforward of redistributions this round:
1. Transferring everything south of Little Para River solves Makin and Spence in one go
2. Barker can simply take in the remainder of Barossa Council in Spence (although this brings it right up to the edge of Gawler), and additionally the Adelaide Hills SA2 if willing to split the Adelaide Hills Council
3. Regardless, Mayo still needs to lose electors and its small share of Mitcham Council seems to be where it most encroaches into the Adelaide Metropolitan Area so that should go into Boothby
4. If required, Boothby’s share of Unley Council can go into Adelaide
All up between 1.59% and 2.62% of electors moved.
I can see why people might want to change Kingston’s boundaries around Aldinga/Hallett Cove/Seaford, but I don’t see much value at this point in time. Kingston’s boundaries are currently placed at reasonable gaps in the urban area, plus Boothby’s northern boundary is about as solid as it can get which leaves limited space to move.
Ive split Adelaide Hills 4 ways.
Real Talk
I agree that the Franklin split has to go as the community of interest is nothing beyond being in the same federal / state electorate. My starting point was Hobart City plus the Kingborough part of Clark and the western part of Franklin. Unfortunately that doesn’t give enough electors so there will be some need to go partly into the City of Glenorchy. Any movement of Clark southwards to get quota make the Franklin split worse. If the AEC are determined to keep the split Franklin then Clark may need to cross the Derwent and take in Bridgewater and Gagebrook and whatever else is needed to make the quota.
@Angas and @JWood: Thank you so much for creating the redistribution tools. I would like to point out that there are some non-negligible differences between the current and projected enrolment figures listed in your tools to the degree of a few hundred electors per electorate. For example, the current and projected enrolment figures for the electorates of Barker, Grey, Mayo and Boothby in SA listed in the tools of both of you differ by a few hundred electors per electorate. I think the main cause is that @JWood handled some SA1s split between multiple divisions inappropriately, causing some SA1s from one electorate to be completely surrounded by SA1s from another electorate. For example, the SA1 40201102503 is split between Spence, Barker and Grey, with the vast majority of its population in Grey, while the Spence part of SA1 40201102503 has only one current and projected elector. Yet @JWood allocated the entire SA1 to Spence, causing the SA1s of 40201102501 and 40201102502 in Grey to be completely surrounded by 40201102503 in Spence. @JWood should have handled it like @Angas did by allocating the entire SA1 40201102503 to Grey. The SA1 40501111125 is split between Barker and Grey, with the vast majority of its population in Barker, while the Grey part of SA1 40501111125 has only 9 current electors and 10 projected electors. Yet @JWood allocated the entire SA1 to Grey, causing six SA1s in the town of Freeling in the electorate of Barker to be completedly surrounded by SA1 40501111125 in the electorate of Grey. The SA1 40501111125 should have been allocated entirely to Barker like @Angas did.
Hopefully this can be fixed soon @JWood.
@Angas @JWood
Some minor feedback regarding the Qld federal page.
– Swap Capricornia and Dawson in the seat menu (they’re in the wrong geographical category)
– Some of the SA2s in the Townsville area appear to be non-contiguous.
Aside from that, thanks once again.
A quick play with the Qld 2026 redistribution page has confirmed for me that it will not be a straight forward process
here is no state redistribution for Tasmania. State boundaries are drawn to match the federal ones.
@Joseph Great spotting! Really appreciate the detective work on that. It’s these sorts of irregularities that can be really frustrating. I’ll get these fixed up now, and consolidate those split SA1s.
@Angas and I are discussing straggles of dealing with these split SA1s in our tools.
For now, assigning to the largest proportion seems safest.
I’ll conduct some thorough QA on each of the new states, to ensure numbers are adding up as expected.
@redistributed
I’ve done exactly as you described (City of Hobart, Kingborough and “Western Franklin”) and got a division -2.86% under current quota and -2.65% under projected quota, without using Glenorchy. The remainder of Clark that isn’t in the new division went with the eastern side of Franklin. Effectively I merged two divisions and split them again.
With respect, I struggle to see how splitting Clark across the Derwent is preferable to the status quo of non-contiguous Franklin.
@Darth Vader That will be true for the House of Assembly, but I believe there will be a redistribution of the 15 seats of the Legislative Council in next couple years.
With split sa1s I try where practice to combine them and then provide an estimate range. Occasionally I split them when they cross rivers or lgas.