Federal redistributions in Australia rely on two sets of numbers – the actual enrolment numbers at the time that the process commences, and an estimate of how many voters will be enrolled at a future point, about three and a half years after the conclusion of the process. While electorates need to be drawn within 10% of the average enrolment at the start of the process, the projected numbers are far more important, since seats need to be drawn within 3.5% of the average on those figures.
These projected numbers ensure that electorates are drawn so that faster-growing seats have smaller populations, and slower-growing seats have larger populations, and thus population change reduces malapportionment, rather than increasing it.
Unfortunately, if those projected numbers are no good, the whole thing is undermined.
The projected figures for the Victorian federal redistribution were released in October, and I posted about them here.
There have been a number of people raising concerns about those figures in the comments, and earlier this week the AEC acknowledged the issue, saying that “The AEC has been informed by the ABS that there was an error in the initial enrolment projections supplied for use in the redistribution of Victoria.” They have now released this corrected data, and it does change the distribution of population in a way that will favour outer suburban growth areas, which will be drawn with smaller electorates than if the original projections had been used.
I haven’t personally investigated the problem with the projected numbers, but as an example Zac Gross posted this graph, showing that almost all Victorian SA1s were assumed to have growth of almost exactly 10%, whereas in other redistributions the growth rates vary (as you’d expect).
First up, I’ve modified the following table that I posted in October which breaks down electorates in Melbourne into different parts of the city north and south of the Yarra River, and breaks rural Victoria into east and west.
The original projections had the 26 seats of Melbourne collectively about three-quarters of a seat under quota, but that deficit was spread out across the city. The 16 seats south of the Yarra were about half a seat under quota, while the 10 seats north of the Yarra were about a quarter of a seat under quota. It was particularly surprising that the six seats of western Melbourne, an outer suburban area where you’d expect fast growth, was projected to barely gain any population relative to other parts of the state. This looked very different to the trends in the NSW redistribution, where Western Sydney is set to gain a seat while the eastern half of the city loses two.
But this picture looks different with the new figures. The ten seats north of the Yarra are pretty much spot on quota, although the central city seats are under quota and will probably need to expand to take in surplus growth in the western suburbs. The 16 seats south of the Yarra are now 84% of a seat under quota, rather than 50%. Indeed the eight seats I defined as “eastern Melbourne” are almost half a quota under themselves.
This makes a huge difference to the implications for the redistribution. It’s now clear that the seat to be abolished will be in the eastern suburbs of Melbourne.
This doesn’t mean that other areas won’t be affected – overall the seats north of the Yarra (urban and rural) fall short of their quota by 17% of a seat. It’s possible this deficit could be spread out amongst the 19 seats on the north side and thus doesn’t require a seat to cross the Yarra, but there will definitely be a need to distribute population differently within that area.
Another way to look at the figures is via this map. It has two layers. The first shows the revised projected quotas for each seat, while the second shows how much each seat’s projected quota was changed by the revisions.
The first map now makes a lot more sense. Most seats in Victoria are under quota (as you would expect when a seat has been abolished), but the outer suburban fringe on the north-west and the south-east both tend to be over-quota. This was not the case on the original figures.
When you toggle to the second tab, it’s very clear that the new projections have favoured outer suburban areas.
La Trobe, Lalor and Calwell were all projected to be under quota but are now projected to be well over quota. The change in La Trobe was 13.2% of a seat’s population!
It’s quite unfortunate that this mistake was made after the first two rounds of submissions, which would have been made based on those projections being correct. The mapmakers will be able to use the correct figures, but will be relying on public submissions based on entirely different numbers. But I’m not sure the alternative of allowing further rounds of public submissions would have been viable. At least the problem has been identified and fixed. It would have been far worse to continue with incorrect numbers, that would have likely led to fast-growing outer suburban areas being under-represented.
Western Australia was also affected by this issue, and I’ve got a blog post coming up covering WA this afternoon.
Trent, if you upload the .csv file you already have, then I can put it into the right format.
I think this about the only redistribution we all have different views on
There are two ways I found to do Holt
1. Just remove Pearcedale SA2
2.
– Remove Pearcedale SA2,
– Remove the section of Narre Warren South – West SA2 that is north of Ormond Rd
– Remove the section north of Pound Rd
– Add the Clyde North – South SA2 section that is between Tuckers Rd and Hardy’s Rd.
The problem is the Clyde North SA2’s are so fast growing that it fucks up whatever seat you put them in.
@Trent The way I drew Holt was to first transfer Pearcedale – Tooraddin SA2 from Holt to Flinders, then transfer the remainder of Narre Warren South – West SA2 from Bruce to Holt. Current enrolment of the new Holt is 108992, and the projected enrolment will be 130754.
Yeah the Clyde North SA2s are really the issue.
I also removed Pearcedale-Tooradin and sent it to Flinders as well, but my objective with that area was to try to create Holt as a Cranbourne Line based growth area seat (Cranbourne, Lyndhurst, Clyde North type of area) and La Trobe as a Pakenham Line based growth area seat (Narre Warren, Officer, Pakenham).
On the original incorrect numbers it actually worked pretty well, but on the new numbers, the growth in Clyde North is just so huge that to accommodate it being under projected maximum, it ends up being below the 2023 requirement.
Looking at the SA2s on a map it makes perfect sense because so much of that area is still vacant and will no doubt be filled in with development by 2028.
I think I’ll have to redraw La Trobe and Holt so that rather than being based along the train lines, Holt just remains based a north-south seat running from Narre Warren to Cranbourne while Clyde North remains with Pakenham & Officer in La Trobe.
I do fear though that with both Clyde North and Officer being major growth areas, the same issue is just going to happen in La Trobe as well. I feel like as long as both Clyde North SA2s are united, there will be a massive discrepancy between 2023 & 2028 numbers, but I also don’t feel like splitting the 2 Clyde North SA2s into different seats would make much sense either. It wouldn’t be a good boundary, but the numbers might necessitate it.
It’s interesting looking at Clyde North SA1s on a map, some of the numbers in the projections look a little weird.
SA1s like 2155620 and 2155636 have the most empty space so you’d think would have the most growth, but their projections are both only 1.57%.
Then you look at 2155607 which has a projected growth rate of 126% (from 185 to 418) and it’s a tiny SA1 that is already fully developed with houses, no other space to grow. Unless those houses are full of 14-17 year olds who will turn 18 before 2028, it doesn’t make much sense.
Others, like 2155621 which has projected growth of 86% makes more sense because if you can a lot of the SA1 are vacant lots that have already been subdivided.
The reason I looked into this, was because I thought perhaps the best solution would be to include all the northern & western SA1s in Clyde North that are already developed, and leave out the southeastern SA1s which are more vacant, assuming that’s where the growth would be. But it turned out the more vacant ones almost had no growth.
Furthermore, looking at 2155621 specifically and counting the houses & lots (because it is small and easy to do), there are currently 86 houses and 36 vacant lots in that SA1. All detached suburban houses.
Firstly that’d indicate a growth rate of under 50%, but even if a lot of those 86 houses weren’t occupied at the time of the 2023 numbers, it’s hard to see how a maximum of 122 houses in that SA1 (there is no room for any more) could translate to 418 adult voters. Unless, like I said, those houses had a high number of teenagers that would turn 18 prior to 2028.
Sorry, 2155607 specifically.
(Sorry for the 4th message in a row, just had to correct that)
@drake I’ve added Clyde north – north sa2 from la robe and removed both Hampton parks to Bruce this puts it just under on actual enrollment but they can easily fix that up by topping it up with parts of narre warren from Bruce.
@joseph my solution put its at just under actual enrollment at 104xxxx and actual enrollment of 128xxx. That cab solved by putting a tiny bit of narre Warren in from Bruce
@rent see above. I imagine holt will take all of Clyde north at a future redistribution and Bruce will take in parts of holt more and shed the rest of Dandenong making it solely based in Casey Dandenong can then be split between issacs and hitham
I personally would like Pearcedale-Tooradin removed if possible. It is in the Green Wedge and quite demographically different from the heavily ethnic and growth areas around Cranbourne. It was previously in Flinders and is more demographically like Hastings etc.
@nimalan I doubt it will since Mt Eliza is more then likely to be partially moved into it
I will see what maps everyone comes with and make objections accordingly. i would prefer if Mount Eliza just stays in Dunkely i dont personally like suburbs divided in half in two electorates. I note you would prefer Dunkley moving north up to the Paterson River. I am ok if that is absolutely required but again it would cross an LGA boundary and split Paterson Lakes in two.
@nimalan geographical features like rivers make for better boundaries then suburbs and lgas. M9ving Flinders into holt does the same thing splitting Casey into a 4th division. There’s also the issue odpf the shape of the division. Given Flinders is under quota there won’t be any choice
Also Mt Elizabeth is part of the Mornington lga so that’s slit too.
Dandenong – South SA2 should move to Bruce as it has a large Albanian and Afghan community and is the very different community compared to the ones in Frankston. I think that would be possible if Hotham was abolished allowing the Issacs to move North.
@ John
You do raise excellent points which i respect. Unfortunately, we will not be able to draw perfect electorates. Interestingly, Casey is one of the most populated LGAs in the country almost 400k residents and will reach 0.5 million soon so it has very different communities of interest. There some areas of Casey LGA which are green wedge Fishing villages, very wealthy areas like Lysterfield South and some of the poorest such as Doveton.
I have Dandenong – South SA2 in Bruce in both my pre-and-post revised number scenarios, that’s a given for me and near the top of my priority list.
I almost have La Trobe, Holt & Bruce working now. Bruce & Holt are within range and La Trobe is still 2617 over quota but that’s not as bad as what I had before.
I reverted the three-way boundary there a lot closer to what it is now. I still transferred some of Bruce into La Trobe (the two Berwick SA2s) but have kept most of the Narre Warren ones in Bruce.
As for Holt, I did as others have suggested and only removed Pearcedale-Tooradin (as I had always done anyway) and added the remainder of Narre Warren South – West from Bruce.
So now I just need to find somewhere to send this 2617 excess from La Trobe..
If Dandenong – South SA2 remained in Isaac’s and Mark Dreyfus recontest, there could be a big swing against Labor because that area is Muslim Majority (50.1%) but Dreyfus is Jewish and obviously pro-Israel.
@ Marh,
I think they could be a big swing but it will not affect 2CP as the voters will choose Libs they will likely go Socialists or Green
trent ive transferred clyde north – north from la trobe to hole making la trobe at quota then transfer hapmton park sa2s to bruce then transfer noble park to hotham
How far did you guys push moving Corio into Lalor?
Lalor being way over quota needed to lose a lot of electors, and Corio being considerably under means it can take a lot of them. All the more empty/rural areas surrounding Wyndham Vale and Werribee are logical and make sense, and actually help consolidate Lalor into a more purely built-up suburban seat too, but there aren’t really enough of them without actually encroaching on areas like Wyndham Vale, Werribee South or Manor Lakes themselves.
Semi-suburban areas that are a lot more attached to Werribee than Geelong though seem like a real stretch to put in a Geelong-based seat.
I’ve already transferred bits of Lalor to Gellibrand, but both Gellibrand & Gorton are near capacity now so can’t really take anymore, plus I’m happy with where those boundaries are. I only have 1166 left to get rid of from Lalor but can’t really send any more to Corio without encroaching on the suburbs I mentioned above.
I just put the Dandenong part of Dandenong South SA2 into Bruce and put the rest in Dunkley. The other parts aren’t actually that diverse, and I think fit reasonably well in Dunkley. The actual suburb of Dandenong South is very different to Dandenong. It’s such a small amount of voters that it doesn’t really matter where you put them, I just think Dunkley makes nicer boundaries.
Things I think the AEC will definitely do
– Pearcedale SA2 into Flinders
– Dunkley moving up to Patterson River
– Woodend into Bendigo
– Kilmore into Nicholls
– All of Yarraville into Fraser
– Hughesdale into Higgins
– All of Brunswick East into Wills
Most of the regional areas have such logical fixes that I’d be surprised if they do anything unexpected there.
There’s a lot of times a logical boundary can’t be made because of the fast growth of certain areas. Clyde Nth can’t fit into Holt, Mickleham can’t go into McEwen. You are left trying to balance growth areas with low growth rural areas. That’s why I have Eltham in McEwen.
As someone very familiar with Geelong the only part of Lalor that could go into Corio is Little River. After that it starts to become a mess. I really can’t see the AEC trying that. Especially when there is a very obvious solution of putting Bannockburn into Corio which helps with Corangamite too (and still leaves it with enough to fix Wannon).
Putting Lara into a Werribee seat would maybe work but I don’t really think it works that well the other way around.
@trent corio will push in corangamite as they generally use the greater geelong lga as a boundary. lalor will shed territory to gellibrand, gellibrand can then shed territory to fraser. gorton stays the same.
@drake disagree on peacedale im leaving it in holt
agree on patterson river into dunkley
ive moved bendigo north instead into gunbower sa2
agree on kilmore
agree on yarravile
left hughesdale in hotham
agree on brunswick east
holt can fit the north – north sa2 of cylde north after it sheds hampton park, mickleham stays in calwell calwell sheds the rest of gladstone aprk to maribyrnong along with a tiny bit of broadmeadows to bring calwell under quota, mcewen sheds mernda north to sculling and takes in kangaroo ground, research and diamond creeek from jagajaga and warrandyte north north of the river from menzies
Agree Drake and John, the solution of having Corio push east into the ‘inner’ parts of Golden Plains shire surrounding (and inclusive of) Bannockburn is a better way of resolving the shortfall compared to moving westwards towards Werribee and Melbourne’s outer suburbs.
Although this creates a ‘messy’ district in geographical terms, I believe this configuration is also used for the state seat of Geelong (which also forms an elongated shape, stretching northwards to include Bannockburn when it was redrawn prior to the 2022 election).
For Wannon – rather that it having to absorb more of the Surf Coast around Torquay (as the state seat of Polwarth currently does), I believe the better option is for it to absorb the rural parts of Golden Plains Shire from Ballarat. These areas were previously in Wannon prior to the 2022 election, and it would allow Ballarat to move towards Melbourne by gaining Ballan and surrounding areas in Moorabool Shire.
The structure of the SA1s between Corio’s boundary and Werribee are what make it really difficult.
Little River is in a tiny SA1 (21305146808) but the area immediately outside that very small boundary also has quite a lot of houses, which would be enough to get both Lalor & Corio within tolerance in my scenario.
BUT, they are in a giant SA1 (21305146843) that stretches all the way to the suburban areas of Werribee and Wyndham Vale!
I feel like the perfect boundary would actually be using You Yangs / Little River Road.
I managed to do what I feel is a slight improvement to both Gellibrand and Fraser, and that allowed me to move the northern part of the Lalor/Gellibrand boundary slightly west from the creek to Derrimut Road. It’s a good strong boundary and the vast majority of Tarneit is still in Lalor, the part that moves to Gellibrand is actually more connected to Truganina (all of which I already had united in Gellibrand now) than it is to Tarneit.
Combined with moving the tiny Little River SA1 into Corio, I now have Corio only -372 short and Lalor only 291 over.
There’s where being able to move the remainder of the Little River area, south of Little River Road, into Corio would be a perfect solution and almost certainly have just enough electors. But, it’s not its own SA1.
@john, I don’t like splitting the Clyde North SA2s, and just putting the North one into Holt creates a really odd, awkward shape, it’s not a good boundary.
@Drake: I agree on your whole list of what you think the commission will do on those areas and I have done exactly the same with all of them myself. They all make perfect sense. With the Yarraville one, I think they’ll also move the remainder of Brooklyn into Fraser, that makes the freeway the new continuous boundary between the two (at least until the major interchange at Laverton).
Also on the previous numbers, Wills couldn’t take all of Brunswick East without Melbourne falling below quota and needing to gain something else (eg. Docklands or Kensington) but now it can on the revised numbers so that’s such a simple, logical change.
@yoh an yes I’ve had wannon take in the parts of golden Plains from Ballarat which in turn takes baccas marsh surrounds from hawke. Hawke would be within quota after that so doesn’t need to move any further
@trent unfortunately it cannot absorb the lower part forst due to that part being in even higher growth and would put la trbpe short. It would be a temporary fix that can be solved next time.
Melbourne in my proposal takes in the rest of Yarra council from cooper
I ended up with a small shortfall in Higgins compared to my original proposal.
I really liked my Higgins boundaries too – particularly using North Road as the southern boundary and Warrigal Road as the eastern boundary – so I didn’t want to change that.
The easiest solution was for Higgins to retain a small part of Prahran that I originally had moving to Macnamara, that being just its north-eastern corner bounded by Williams, Commercial, Orrong & High so that’s what I have done.
Having lived in Prahran, that particular pocket is the wealthiest and a bit different to the rest, it’s actually bounded by Toorak to the north and Armadale to the east (the other two sides just bordering more of Prahran). That section is almost entirely houses rather than apartments/flats too, so I think it fits well remaining in Higgins with Toorak & Armadale.
The alternative would have been to put the part of St Kilda East that’s north of Inkerman & east of Hotham into Higgins to unite it with the Glen Eira part south of Inkerman, but it was actually too many people and would have put Macnamara under quota, plus I prefer to keep Port Phillip united in Macnamara, whereas Stonnington was already split between Macnamara & Higgins anyway.
Bannockburn into Corio might look weird on a map but it fits way better with Corio. I also added Moolap which aligns the border completely with the state seat of Geelong. Also helps to give Corio a few fast growing areas.
If the AEC is going to put Winchelsea into Wannon then I don’t see why it shouldn’t go further and put up to Moriac into Wannon. Those areas are not that similar to the now very coastal seat of Corangamite. Corangamite is very quickly becoming: growth suburbs outside Geelong + coastal communities. The rural parts of the Surf Coast and Golden Plains shire no longer fit that well.
Though we are starting to get to the point where Wannon can no longer creep much further into Corangamite. Going any further into Bellbrae/Torquay is a very big problem that luckily we don’t have to deal with yet.
I’ve put Bannockburn into Corio (along with the SA1 covering the southeastern tip of Ballarat’s share of Golden Plains – South), and I’ve put Corangamite’s share of Golden Plains – South into Wannon.
Then with Winchelsea it got complicated, because my Corangamite has a similar issue to Holt (projected population near the top end and 2023 population near the bottom end) if I put all of it into Wannon, Corangamite’s 2023 total was a little below the minimum.
So basically, the SA1s north of Mt Duneed Road and those in the west covering Gherang and Modewarre have been moved to Wannon, and the remainder has stayed in Corangamite. It actually forms a pretty neat boundary and dramatically shrinks Corangamite to entirely exist to the south & east of Geelong now.
I do now have Corio, Corangamite and Wannon all meeting the required numbers and with boundaries I like.
My only problem is just that Lalor is still 511 electors over. If only the SA1s made more sense around Little River that would probably be easy to fix by just moving the section south of Little River Road.
@trent i put tarneit north along with the rest of truganina and laverton from lalor to gellibrand
I already put a little bit of Tarneit – North in Gellibrand but didn’t want to put the rest. I feel like I found the most suitable boundary. I have the numbers working now anyway and Lalor is within tolerance.
Every seat is now within range of quota except Holt, La Trobe & Bruce (but all 3 are much closer now) and for the most part I think I’ve improved on the boundaries compared to my original proposal. I just need to finish playing around with where & how Holt, La Trobe & Bruce all meet to get a set of boundaries I like that are within range.
@ Trent
is it possible to put parts of LaTrobe into either Casey or Monash?
@trent the ones that are out of whcka are gellibrand, calwell, deakin being slightly over with holt,flinders being slightly under on actual number but ahve recommended they make small boundary adjustments to fix these as i dont have the SA1 data to do it,
@nimalan monash is at quota as is gippsland and indi so it woud be hard to without upsetting 3 already at quota divisions. tbh i wanted la trobe to take the rest of cardinia from casey to unite that lga but thats not happening this time arround
i think casey can be made into 2 divisions (holt and bruce) with maybe a bit of overahang in a 3rd and dandenong can all fit into one division (hotham) at a future redistribution
Casey and Monash have already taken all the non-suburban (and even some kinda suburban like the Officer South part of Koo Wee Rup SA2) parts out of La Trobe in my scenario!
My current problem with La Trobe is actually the opposite. It’s within quota for 2028 but about 600 short for 2023 (probably due to having Clyde North!)
The obvious solution is to give something like Officer South back to La Trobe, but since both Holt and Bruce are a little bit over now after some changes I made, better to just transfer from them. The numbers aren’t so much the issue, more just trying to find the boundaries I think work well.
@trent I’m the opposite my holt is short on 2023 numbers due to having the north part of Clyde north in it lol
I’m worried the aec are gonna release the redistribution when I head overseas
Since it seems like we have a bit of time to kill, how did you guys interpret the AEC’s tweets in terms of the timeframe?
I think what Ben said is correct (The AEC media team likely don’t know beyond it being later in Q2, or at least can’t say they know anymore) but it’s fun to speculate.
So when are we predicting? Does ‘a little bit down the road’ mean a month away? or June?
@henry according to Anthony green “not this month” I’d say may.
“The advice to me is not this month.” Based on the indicative timetable and previous redistributions giving 1 month for onpbjections then 2 weeks for comments before its augmented in q2/3 I’d say we can expect it to be released on the 10th of mayat the earliest
Given that the AEC’s indicative timetable has the augmented committee meeting before the end of Q2, surely the proposal would be in the first fortnight of May
Itsonly estimated timeline but I agree
Here’s an revised map of the Melbourne-based seats from my proposal using the new numbers (all seats within tolerance). I haven’t included the regional ones in a map yet but they are all within tolerance too.
https://ibb.co/SPw8mQm
For the most part I’m very happy with it and in fact in most cases I’m happier with these boundaries than the original ones, it let me undo some things I did out of necessity last time that weren’t needed this time.
The boundaries where Deakin, Menzies & Casey all meet are still an area I’m not happy with but I haven’t touched that; and the only area I’m less happy with is the Bruce, La Trobe & Holt area but the numbers necessitated undoing the nice proposal I had before. The shapes of Bruce & La Trobe are a bit awkward – La Trobe especially looks suspiciously phallic – but from a community of interest perspective the boundaries work best and that was more important to me than shape.
Having a guess what what seats might notionally change hands:
– Deakin would certainly be notional Labor;
– Higgins could potentially be around 50/50 so possibly be notional Liberal (hard to tell);
– McEwen (not on the map) could very likely flip to notional Liberal losing all of its suburban areas;
– On the other hand, La Trobe could get close to 50/50 losing all of its regional areas;
– Macnamara would either be notional Greens, or if it remains Labor would change to an ALP v GRN margin. Note that my new proposal keeps keep the Liberals’ best Prahran booth (Prahran North East, the only one where they won the primary vote) in Higgins so that could make it more likely than my original proposal for the Liberals to drop to third in the 3CP (Drake had calculated they only remained in second by about 0.1%)
That’s about all that will or might change hands notionally I think. Obviously based on 2022 results, not what might happen at the next election.
@ Trent
Good map. I am pleased that Hotham is still able to abolished. With McEwen and La Trobe it is very hard to predict if they have switched parties. Being growth areas there will be entire housing estates that did not exist in 2022. McEwen will still have areas that are increasingly considered suburban under your proposal, it will have Kalkallo, Beveridge, Donnybrook and Wallan which are all fast growing and may become an extension of Cragieburn which is solid for Labor.
@Trent I am curious as to how did you redraw McEwen. After I transferred the remainder of Mernda – South SA2 to Scullin, Woodend SA2 to Bendigo and the remainder of Kilmore – Broadford SA2 to Nicholls, McEwen was 16.56% below current quota and 8.82% under projected quota. In order to fill the shortfall McEwen has to gain voters from the already under quota seat of Jagajaga, which will leave Jagajaga no choice but to go south of the Yarra for extra voters. I wonder how did you manage to avoid this. I guess you must have put some parts of the Yarra Valley into McEwen, right?
Trent
Monash doesn’t need to change and you seem to be adding a lot of extra electors. Casey also seems to be gaining a lot as well. Won’t they both end up way over quota?
@Redistributed: In Trent’s proposal, Casey gains a large number of voters from La Trobe after shedding a large number of voters to McEwen.
I agree that Monash doesn’t need to change as it is already within tolerance. In fact, the Redistribution Committee is highly unlikely to draw Monash and La Trobe in any way similar to what Trent has proposed. Since regional seats in the state’s corner will be drawn first, Gippsland and Monash will likely remain unchanged as they are already within tolerance, therefore the Redistribution Committee is more likely to redraw La Trobe in the way that Angas has proposed, which is shedding all of Emerald-Cockatoo SA2 except Beaconsfield Upper and Pakenham Upper to Casey while gaining the area within Princes Highway – Princes Freeway – Clyde Road from Bruce.
As I have said, after shedding voters to Scullin, Bendigo and Nicholls, McEwen will be well below quota and need to gain voters elsewhere. In order to fill the shortfall, McEwen has to take voters from either Jagajaga (as Angas and Drake have proposed) or Casey (as Trent has proposed). If Monash remains unchanged and La Trobe is drawn in the way as described above, Casey will not be able to gain enough voters from La Trobe to be able to afford to shed voters to McEwen. Therefore, I think McEwen taking some voters from Jagajaga and Jagajaga taking some voters from Menzies south of the Yarra is more likely.