Mapping the Molan BTL vote


Jim Molan grabbed a lot of attention in the election for his quixotic attempt to be re-elected from the fourth spot on the NSW Liberal/National ticket, despite New South Wales having a relatively low rate of below-the-line voting.

He never came close to winning, but has polled over 100,000 votes, equalling about 2.6% of the vote, with a small number of votes yet to be allocated to individual candidates. It’s an impressive result (although the raw numbers are helped by him running in Australia’s largest state).

His result also appears to have led to a spike in below-the-line voting generally in New South Wales while the rate of voting below-the-line appears to have declined in most other jurisdictions.

I’ve mapped out where he got his votes below the fold.

It became clear early on that Molan had no chance of winning, but he is on track for one of the highest results for a candidate who isn’t ranked first in their party group, particularly for a candidate running in a big state.

Almost all votes in New South Wales have been apportioned out as either above- or below-the-line votes. Only 4.5% of the total vote is yet to be apportioned. Molan is currently sitting on 119,827 votes, or 2.6% of the statewide formal vote. This translates to just 0.18 of a quota.

It’s wrong to talk about Molan having the highest ever Senate vote for a candidate: New South Wales is much larger than Tasmania, so a candidate can record a larger raw vote despite having a much smaller percentage. Still, the relatively low rates of below-the-line voting in New South Wales in the past does make it more impressive.

You can see the Molan effect in the proportion of below-the-line votes recorded in New South Wales, compared to other states and compared to 2016.

We don’t know the final below the line rate in any state, since there’s still a lot of votes yet to be apportioned (the final column below shows how many are left to be unapportioned) but we have enough data to get a rough sense. The BTL rate is currently down in six out of eight jurisdictions. It is up very sharply in the ACT (which I’ll return to another day) and is slightly up in New South Wales.

New South Wales had a lower BTL rate than every other state except for Victoria in 2016. This time New South Wales is on track to overtaken Queensland and Western Australia.

State2016 BTL rate2019 BTL rateVotes unapportioned

So where did Molan get his votes?

I’ve analysed the booth results in New South Wales, separating out booths based on whether the Liberal/National group’s votes have been apportioned at each booth. The below map shows what proportion of the total Liberal/National vote was cast for Molan at booths where the votes have been apportioned (booths in grey are yet to be apportioned).

You can toggle the map to see Molan’s vote as a proportion of the total electorate vote.

Molan did best in the Liberal heartlands of Sydney, specifically the north shore and the Sutherland Shire.

He polled over 10% of the total LNP vote in Berowra, Mackellar, Bradfield and Warringah on the north shore, as well as Cook and Hughes in the Shire.

But there were two other surprising seats which were ranked first and third when you look at his vote as a proportion of the LNP total: Eden-Monaro and Hume. 12.9% of Liberal/National votes in Eden-Monaro were cast for Molan, and the figure in Hume was 11.2%. Only Berowra (12.6%) came close to Eden-Monaro. Eden-Monaro’s total Coalition vote was smaller than most of the other seats on this list, so his total vote in that seat doesn’t rank quite so high.

Looking quickly at the map, it appears Molan pulled in more than 10% of all Coalition votes in every booth in the Queanbeyan area, and did remarkably well in Merimbula, with over a quarter of the Coalition vote.

Molan’s worst areas were in traditional Nationals seats, the Hunter and Western Sydney. He only pulled in 3.2% of the Coalition vote in Page, and also pulled in less than 5% of the Coalition vote in Werriwa, Lyne, Fowler, Shortland, Hunter, Newcastle, Richmond, Farrer, New England, Chifley, Parkes, Paterson and Page.

You can download the seat-level data here.

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


  1. Very Interesting Indeed

    I saw at first hand a lot of the Molan-LIB aggression in Warringah. Pushing and Shoving was (literally) the order of the day – the police had to be called at least once at Mosman pre-poll. The Libs also complained that Molan was “stealing their volunteers” – this was a bit rich, since most of Team Tony appears to have been shipped in from outside Warringah (and reportedly got paid $100 a day for it).

    It is interesting that, in the past, Abbott and Molan were allies of a sort – especially on the issues of LIB pre-selection processes (funny about that!). The whole thing arose from the failure of Walter Villatora to be pre-selected for the Manly By-election. We called the people involved the Four Horsemen of the Apocalypse. They clearly had thoughts of undermining the selected candidate James Griffin. Villatora, Molan and Abbott later became part of a group called the Democratic Reform Movement and are said to have paid the attendance fees for their supporters to stack a Liberal Party reform meeting. That meeting passed the “Warringah Motion”, designed to hand more control to local branches and take it away from the Party’s FEC. Ultimately it failed at a later meeting because (guess what?) the FEC had it replaced with the “Bennelong Motion” that postponed any such action until after the 2019 election. I rather think that this might be the reason why the LIBS placed Molan in an unwinnable position.

    The bad blood is still being pumped vigorously around the local party’s circulatory system

  2. Looks like Molan’s voters are those who in US would have been Tea Partiers rather than Trumpers-rightwing of the establishment right-affluent & politically engaged.

  3. Tea Party is correct. There was the following story in the media in relation to the DRM.

    Tony Abbott-aligned Liberal reform group using Tea Party political app.
    By Sean Nicholls
    July 18, 2017 — 9.16pm

    The Tony Abbott-aligned group urging NSW Liberals to attend a party reform meeting this weekend is campaigning with a political app used by the organisation behind the 2010 National Tea Party Convention in the United States.

    The app being used by the group, the Democratic Reform Movement, is created by Right Mobile Pty Ltd.

  4. Molan was a Major at the Infantry Centre in the 1980’s when I was a Captain doing an officers course there. We found him aloof and a bit odd. Typical RMC Duntroon type.

  5. The NSW record for a non-group leader primary vote percentage (since groups were given control over their order) is probably Jack Lang in 1951, who got 3.5% from the ungrouped column (running in the ungrouped column was probably a mistake, a group may have got him more votes and preferences, thus helping him close the 2.8% gap between him and the ALP incumbent who defeated him).

    The Perin Davey BTL BTL campaign does not seem to have been very effective.

    Given there was no need for a Richard Colbeck BTL campaign this time around (since he headed the ticket), the decline in the Tasmanian BTL rate seems relatively small.

  6. If you look at his vote as a proportion of the total Coalition vote, he got 100% at the Special Hospital Team 3 in Hunter – 1 vote for Molan, no other votes for the Coalition. The next best was Glenorie West in Berowra. 496 formal votes, 342 votes for the Coalition, including 212 ATL votes, 1 other BTL vote, and 129 votes for Molan (37.7%).

    Glenorie West is also the booth with the highest Molan vote as a proportion of the total vote – 26% of the formal vote was for Molan.

  7. Thanks.

    The special hunter booth is a bit of an outlier.

    The result at Glenorie West – 129 BTL out of 496 is over 25%. Significant.

  8. There was a letter in the Sydney Morning Herald today 20 Jun 19 from Jim Molan Senator for NSW. Has someone forgotten to tell him?

  9. He’s still a Senator for the rest of the month, Senate terms are fixed to begin on the 1st of July and end on the 30th of June.

  10. Thanks Dryhad. I think that with modern aeroplane travel we need to update terms to be active once all senate seat, like the house of Reps seats, are declared. This delay for the Senate may have been relevant 120 years ago when Senators had to travel by rail or ship first to Melbourne and when parliament in Canberra was completed.

  11. Fixed Senate terms, the reason for the delay, are about restricting the ability to call Senate elections, not travel delays getting to the Senate.

Comments are closed.