Sunday, March 4, 2018

2018 Tasmanian Postcount: Denison

Denison 2 Labor 2 Liberal 1 Green
Re-elected: Bacon (ALP), Archer (Lib), O'Connor (Green)
New MP: Hickey (Lib)
Within party contest: Haddad (ALP) vs Ogilvie (ALP).  CALLED: Haddad will win.

There was much interest in the final week of the campaign in Denison.  The possibility sprang up that the seat might rip up the script by delivering three seats to Labor.  This has, in the end, fallen about a further 6% swing short of happening, so it really wasn't all that close.  The Liberal vote, bolstered by the preselection of Sue Hickey to replace Matthew Groom, has held up well clear of two quotas.

Scott Bacon has topped the poll with a small surplus.  Elise Archer is currently just short of a quota but may be brought up to quota during the remaining counting.  If not, she will be put over quota as soon as a minor Liberal is excluded.  The remaining Liberal votes will gradually pool with and elect Sue Hickey, with a small surplus.  Green votes will gradually pool with Cassy O'Connor, who will eventually be elected (currently with no surplus to speak of).

Archer, I do mention, has polled outstandingly well.  I thought that Hickey with her huge profile as Hobart Lord Mayor would get as many votes as Archer or thereabouts, and that Archer could be at risk if the Liberals struggled.  Not the case!

The other star of the Denison show is Ella Haddad, who currently leads Ogilvie and Tim Cox in the race for the second Labor seat, which will be resolved by cutting from the bottom up with the odd small surplus along the way.  At present, Haddad leads Ogilvie by 932 votes.

Cox needs to pass Ogilvie (who he trails by 425 votes) to stay in business but has really only the votes of Sherlock (1837), the Bacon surplus (c. 1100), a few minor candidates, the Liberal surplus (c. 2479, but c. 60% of that will exhaust) and leakage from the Greens to do it on.  His high profile should assist him, but some Liberal voters may favour Ogilvie.  If Cox is out, then his votes become the major factor in the Haddad-Ogilvie contest.  If he catches Ogilvie, Ogilvie's preferences decide matters between him and Haddad.  It should be noted that the Liberal preferences that will become available to Labor will be the votes of Kristy Johnson, as those of Dean Young and Simon Behrakis will be used up electing Hickey.

My initial feeling is that Haddad's lead is just a bit too much for her opponents, but it is possible we will see flows between candidates strong enough for one of them to catch her.  Further enlightenment may appear when the TEC throws the Bacon surplus early in the cutup.

Update Wednesday:  After today's extra votes and rechecking, Haddad leads Ogilvie by 862 votes while Ogilvie leads Cox by 459.  It is possible Ogilvie does better on postals because she has more appeal to older voters (this could also explain how she closed the gap late on election night).  That's still a solid lead for Haddad but it is not over yet.  The other development possibly in Ogilvie's favour is that the Liberal surplus is now notionally 3310 votes, meaning that we could see something like 1300 non-exhausting Kristy Johnson votes being split between the Labor ticket (note: these will also include some Young and Behrakis votes that go to Johnson ahead of Hickey).  An issue for any chance Cox might have is that he needs to stay ahead of Johnson to access this lot (which might help him catch Ogilvie), but it's rather touch and go if this will happen. 

I have not had any scrutineering figures on this one yet.

Update Friday: Something weird has happened with the published data in Denison.  On Wednesday the count showed 94.89% counted, which seemed exceptionally high for that stage given that Denison hasn't broken 95% turnout at the end of the count since 1996.  Now the figures show a more realistic 91.34% counted, with every candidate down on primaries and only informals up.  My assumption is that the Wednesday tally had some kind of error.  Haddad's lead is back to 942 (about where it was on the night) and Ogilvie leads Cox by 483.  Also the Liberals are on 2.26 quotas, not the 2.3 of the Wednesday update.  I have crossed out the Wednesday update under the circumstances, but some comments from it still apply: there will be a Liberal surplus of about 2850 votes and this will be mostly Kristy Johnson primaries and primaries she receives from minor candidates (mostly Young and Behrakis), available after they put Hickey over quota.  However, 60% or so of this surplus will exhaust.  It's still unclear if Cox would have access to this surplus or if he would have been excluded by then.

Tuesday morning: Just another 72 votes added for final primaries and Haddad leads Ogilvie by 948 while Ogilvie leads Cox by 480.  Just the Bacon surplus to throw and then exclusions from the bottom up, with a small surplus for Archer, eventually O'Connor and near the end of the count Hickey along the way.   The most important early-ish exclusion is Sherlock.

Tuesday lunch: The Bacon surplus is thrown and the minor exclusions have started.  Ogilvie made a significant gain on the Bacon surplus, closing the gap to Haddad to 797 votes and stretching her lead over Cox to 529.

1:20 pm: A minor exclusion, but an eyebrow-raiser: Ogilvie closes 18 votes off leakage from Greens candidate Aaron Benham.

4:10: Surprisingly little progress in Denison with no update for a few hours.

4:30: Well that worked, Denison updated - after a few more minor exclusions (two Greens and a Liberal) Ogilvie has made no further gains and is still 780 votes behind.  Elise Archer has just crossed the line with a very small surplus.

Overnight: Here's a booth data graph that might offer some insight into the fate of Tim Cox's preferences:

The graph shows that, in general, if Tim Cox polled well in a booth, then it was also likely that Ella Haddad polled well relative to Madeleine Ogilvie.  That suggests that Cox and Haddad were getting their votes in basically the same places.  I had already noticed Haddad was strong compared to Ogilvie in left-wing inner-city booths and around Moonah/Lutana while Ogilvie was strongest in the richest and poorest booths.  It seems Cox is strong on average in the same types of booths as Haddad.  That makes it more likely Cox's voters will preference Haddad or at least not break strongly to Ogilvie, all else being equal.


10:50 While we're waiting (and it's a long wait in the case of Denison where things are going more slowly than Franklin) I've done the same analysis as above for Sherlock.  There was basically no pattern in her case.

11:08 Ogilvie has not surprisingly made a modest gain on the preferences of the lone Shooters, Fishers and Farmers candidate and trails by 713 votes.  She is 571 ahead of Tim Cox and we now have the preferences of Sherlock (2331 votes).  This isn't quite the last throw of the dice for Cox since it looks like he will stay in the count long enough to access the departing Liberal preferences (to do this he needs to stay ahead of Kristy Johnson).  However he would need a big gain on Ogilvie here for any chance.

1:17 Haddad has made a huge gain on the Sherlock preferences and leads by 1125.  Ogilvie leads Cox by 457.  Now Helen Burnet will be excluded putting Cassy O'Conner over the line with a small surplus which will probably stretch Haddad's lead further.

2:53 More gains for Haddad - plus 64 over Ogilvie on leakage from Burnet.  Now we have an O'Connor surplus worth 374 votes, 30-40% of which will probably exhaust.

5:30 Game over for Cox who to my surprise falls 43 votes short of getting a swing at the Liberal votes.  Also game over for the day - come back tomorrow for the throw of 5245 Tim Cox votes with Haddad holding a 1245 vote lead.  Should that fail to settle matters clearly, about 1400 non-exhausting votes will probably come out of the Liberal ticket.


11:55 Ogilvie has come back slightly on the Cox preferences and is 912 behind.  We now have the throw of Kristy Johnson then the Sue Hickey surplus.  I think there might be more like 1600 non-exhausting Liberal votes to go and if that's so Ogilvie is going to need a 78% split off those to win.

1:55: Ogilvie has made further gains on leakage from Kristy Johnson and is now 828 behind.  However there are now only 2779 Liberal votes remaining, of which most will exhaust, leaving maybe 1250 if that live between the Labor candidates.  I expect Ogilvie to make further gains but Ogilvie would now need something like 83% of the non-exhausting votes to win. Called.

4:31: All over.  Haddad wins by 731. This is the sole case this election of a candidate winning by ousting an MP from their own party.


  1. Hi Kevin, hope you are well. Looking at the latest TEC figures from Denison as at the 28th count, is it within the realm of faint possibility the Labor could yet emerge with 3 Seats? As a 'New-(ish) Tasmanian' I'm not well versed in the fascinating intricacies of Hare Clark, so I don't know whether it's Labor, or Liberal more likely to emerge victorious as the former has 2.63 quotas & the latter has 2.32 quotas? In the interests of openness & transparency I should disclose that I am a very proud Member of the Tasmanian Labor Party, but I am not a staffer, I am however paying especially forensically close attention to the happenings in Lyons. It is my sincerest hope that the current order of things is maintained. Perhaps I am biased, JL is my local Branch Secretary, whereas JB isn't local at all, & thus I've never had the pleasure of meeting her, just JL & DC!

    1. It's impossible for Labor to win three in Denison as Hickey needs only a third of Kristy Johnson's votes to reach quota thus winning the second Liberal seat with O'Connor having already taken one too.
      Hickey will get at least 80% of Johnson's votes.

    2. Thank you for both publishing my comment, & for your reply. I suspected that you'd say what you've said, but hoped you'd say the opposite none the less! What else would you expect of a partisan animal such as myself? I haven't been Tweeting much lately, I've had to move for reasons of safety, long story, throat cut & death dodged by four millimeters last June, & unrelated punch in nose/smashed window in January. So much raw violence, too much raw much violence, let alone for a toughened up once upon a time growing up in 2327* mummy's boy like myself, I very nearly, almost, thought I was a broken man. Are these just very late rites of masculine passage? If so, enough already I say! Alternatively & more disturbingly, my experience of life in the underclass of a small Tasmanian town could be quite common, I sincerely hope not!? TasPol has been just simply most excellent throughout & I am also extremely pleased to say that Mark Shelton's office was very helpful in getting Minister Petrusma's office involved to facilitate a rapid transfer to another Housing Tasmania property, in another electorate. It wasn't easy for this very proud Labor Member to walk into a Liberal MHA's office asking for help, I don't imagine very many Liberal Members would EVER find themselves in a comparable situation, but needing help from a Labor MHA's office, do you? I don't want to jump the gun here Kevin, but I suspect that unlike most of the rest of the rotting GOP-Lite Liberal carcass across the Nation, the Tasmanian Liberal Party, uniquely, still seems to have a soul. Certainly the few Liberals I've met since being a Tasmanian have all came across as decent people! Sometimes it's rather nice to have one's preconceived notions, i.e. '#AllLibzREvil'proved incorrect, isn't it? I do so wish I'd read your pinned Tweet & subsequent article before I voted, instead of today Kevin. Silly me, though an #Auspol tragic hadn't realized how to maximise the power of a Hare Clark vote. Silly me just voted for all 6 of Labor's candidates in Lyons, I'll leave it to you, & ANY of your readers, (whomever they are, & wherever they may be, friends without, or enemies within, greetings to all of you) of this comment's imagination as to whom it was I voted for 6th, i.e. last. It ain't rocket science Kevin, have a nice evening! ~Ian..:)
      *Kurri Kurri, of all the places in Australia, is home of the Mullet-as in hair-Festival, God help us.


The comment system is unreliable. If you cannot submit comments you can email me a comment (via email link in profile) - email must be entitled: Comment for publication, followed by the name of the article you wish to comment on. Comments are accepted in full or not at all. Comments will be published under the name the email is sent from unless an alias is clearly requested and stated. If you submit a comment which is not accepted within a few days you can also email me and I will check if it has been received.