Batman 2018 – the map tells the story


I love a good electoral map, that can tell a story about how a seat’s population votes in different ways and how the balance of those communities decides who wins an election. Batman has produced a fascinating map at recent elections, with the Greens dominating south of Bell Street while Labor rules the roost to the north.

The question at this election was whether the Greens could breach Bell Street and take more of the vote north of the road. The answer is: yes, they can. But they did so while losing support in their heartland, and the swing to Labor in the south was stronger than the Greens swing in the north.

The Greens gained swings in a majority of booths north of Bell Street, but most of the swings were small. The biggest swings took place at the northern fringe of the electorate in suburbs like Bundoora and Reservoir.

But they were dwarfed by Labor swings in the south – swings of 7% in the Thornbury area, and swings of 8% and 11% in Northcote.

This map shows the swings to Labor or the Greens across Batman:

I feel like this swing map is one of the most illuminating electoral maps I have ever made, and can be interpreted in many ways to tell the story of this by-election: Labor’s shift to the left from Feeney to Kearney, the Greens’ concerted push into the north and efforts to court more conservative voters with their last-minute message on dividend imputation, and a more equalised vote, with less difference between the strongest Labor and Greens areas.

Labor defeated the Greens by just 1% in 2016, so this differential swing was enough to make a big difference at this by-election.

This by-election hasn’t erased the gap between the north and the south. The Greens still won most booths south of Bell Street, and Labor won all but one booth north of this line.

To finish this post, here is the map showing the two-candidate-preferred vote for the winning party at each booth. You can compare this to the 2016 result here.

Liked it? Take a second to support the Tally Room on Patreon!
Become a patron at Patreon!


  1. Invaluable work. But I don’t think it shows that The Greens can win votes north of Bell Street when the Liberals didn’t run.

  2. I’m confused by this. I presume the green colour is for the Greens, the red for Labor, but what is the black for?

  3. And overall the sad but at this point in time inevitable result was an organised political gang ( a.k.a. political party ) won. I live in hope that one day the voters will realise that their representative is duty bound to represent them and not a party and their mongrel donors!!!

  4. Peter, I can’t see any black numbers? The darker red means it’s better for Labor (over 60% on the 2CP map, over 5% swing on the swing map), the darker green means it’s better for Greens (same figures).

  5. Hello Ben

    Good stuff, as usual

    My question is, however, about the discrepancy between AEC swings and Antony’s swings, which are
    Party AEC ABC
    ALP +7.4% -5.4%
    GRN +4.1% -9.5%
    In total, Antony’s swings add up to -34.6%, when it should be zero?

    I can’t see how Antony gets those swings – something to do with ignoring the LIBs altogether?

    Geoff Lambert

  6. Hi Ben,

    Do any/all of the three booths that recorded big swings to the Greens line up with known Greek or Macedonian community pockets in the seat? Much was made of that factor in the lead up.

  7. Your map shows that “every” booth bar one swung to Labor south of bell road… perhaps the black wiggle isn’t so appealing after all 🙁

  8. Yeah there’s a bug there. The first preference ABC swings are just wrong. The funny thing is they correct themselves when you click ‘list all candidates’.

  9. Yeah there’s a bug there. The first preference ABC swings are just wrong. The funny thing is they correct themselves when you click ‘list all candidates’.

  10. Ben, one other interesting factor I haven’t seen reported on yet – despite the lack of a Liberal Party candidate, the dismal showing of the Australian Conservatives candidate. Even in such a progressive seat as this, the lack of primary votes for them in the complete absence of a Coalition candidate bodes well for their irrelevance at the next Federal election surely?

  11. Sorry for the duplicate comment. Feel free to delete one.

    This is an excellent result for Labor. One would have reasonably thought the absence of the Liberals and their preference recommendations alone would have been enough to change the result. (You could argue that HTVs are overrated, but the tractability of major party voters is well established. Perhaps most Liberal voters just naturally prefer Labor to the Greens anyway.)

    Additionally they’ve also combatted the usual hostile by-election atmosphere for major parties against third parties in their own seats. Think Cunningham, or more recently Northcote. (A booth comparison with Northcote would be interesting.)

    Maybe the simplest explanation is the best. Feeney was a dud and Kearney is a star.

  12. A generous concession speech by the Greens candidate election night saying the ALP’s wining candidate cares about people etc. With all the pre election fighting within the Greens, which is an immature and silly think to do after the candidate is selected and the campaigning has started, I wonder of she will to a Cheryl Kernot and defect to the ALP?

  13. Arky – The Australian Conservatives name is a problem for them and voters. They should have stuck with Family First which is a much more friendly name.

    The Family First Senator who moved to the Liberal when the original Family First Senator was disqualified was a smart move. She is very funny when door stopped last week about Cash’s comments on ALP women in Shortens office – “Well…Well… next question, see ya”

  14. David Walsh – Yes Kearney is not part of the ALP’s deranged right wing like Feeney is too.

  15. An incredible win for Labor. Ben, your map highlights that strong support for Labor south of Bell was important to their holding Batman. (I also love a map!) David Walsh, the booth comparison with the Northcote By-election four months prior is interesting, and eliminates the significance of the Liberals not running as a factor. Labor had good results in Thornbury East (up 5% compared to Northcote By-election) and Thornbury South (up 5% from Northcote By-election), alongside strong showings at two early voting places located 450 metres apart in Thornbury (up 9% on Northcote By-election early voting). On top of the Greens’ self-sabotaging and the ‘Feeney factor’, this localised support was crucial in getting Labor over the line in a high-stakes election.

  16. As someone with “red-green” colour blindness, I can only guess what the numbers are saying. Can you do it with plusses and minuses please Ben?

  17. Ben, re Peter’s comment about the black numbers. Quite apart from possible colour-vision problems, different screens will have different contrast and “gamma” settings and some of them will make dark colours indistiguishable from black. Another argument for plusses and minuses.

  18. Hi Jack, you raise some good points about colour blindness. I’m not sure what the best solution is that doesn’t add a lot of clutter and make the map unreadable for everyone. Ironically the original swing map did have minor symbols but these were removed because I had a number of people getting confused by them (did the green numbers with minus symbols mean a negative swing for the Greens?).

    You can see the original map with the negative symbols in this tweet:

  19. Eeeek! Even more confusing! I wouldn’t advocate symbols as_well_as colours – just that when you do red vs green you do it as swings towards and away from one party or the other, and don’t use colours at all. Red vs blue is ok except for some really tiny proportion of people – red vs green confuses about 8-10% of males and 1% or so of females (at least those of Euro descent). Fortunately I’ve picked up the gist of what happened in Batman from the commentary.

Comments are closed.