Legislative Council 2022: Pembroke By-Election Live

Pembroke: CALLED (7:23 pm) Edmunds wins (Labor retain) 

Approx 4.6% 2PP swing to Labor after preferences

----------------------------------------------------------------------------------------------------------

Donations welcome!

If you find my coverage useful please consider donating to support the large amount of time I spend working on this site.  Donations can be made by the Paypal button in the sidebar or email me via the address in my profile for my account details.  Please only donate if you are sure you can afford to do so.

----------------------------------------------------------------------------------------------------------

Live Comments (Scrolls to top)

All updates are unofficial; check the TEC site for official figures

-------------------------------------------------------------------------------

Tuesday: The provisional preference distribution is underway and Edmunds has 43.84% to Brown's 33.29% with Brewer 22.87% to throw; Edmunds will be declared the winner after Brewer's preferences sometime later today.  Brown gained 50 off preferences from di Falco but lost 25 back off Willink so as expected the preferences of those two candidates had little impact.  

EDIT: And it's up, a 63.26-36.74% 2PP to Labor (will change slightly with remaining postals).  A 4.61% swing to Labor for their own vacancy, a very strong result.  It's not really possible to compare this to past Council by-elections because most had incomplete preference distributions and few if any had genuine 2PP swings.  In federal by-elections the average swing for Opposition vacancies is around 1.2% to the opposition.

Spin Corner: In a relatively low level of post-result commentary there's been some extraordinary spin from Michael Ferguson who claimed "the slide in Labor votes is clearly a vote of no confidence in Rebecca White and it’s hard not to notice that Green preferences bring Labor over the line" What's actually hard not to notice is that Labor is 10.7 points ahead before those preferences are thrown; this is hardly a case of Labor needing a big gain from behind (in which case the argument about dependence might have teeth) or even a 50-50 split.  That's four points further ahead than Jim Bacon's Labor were in 1998 when they won the state election outright with hardly any help on preferences from the Greens!  Obviously more than half the regular Greens voters would have voted 1 for Jo Siejka in the 2019 field and the swing against Labor on primaries is more than explained by those people returning to sender with a Green on the ballot.  

There are some reports that we may get a distribution by Tuesday night - not sure if that's correct or confusion with Tuesday week.

Some Sunday History:  Mileage may vary because of the informal nature of party endorsements in the old days but I get this as just the 6th case of a major party retaining their own vacant LegCo seat at a by-election out of 16 opportunities.  I have the following:

Retain: Hobart 1943*, Launceston 1951, Derwent 1995, Derwent 2011, Pembroke 2022 (all ALP), Launceston 1966 (Lib - incumbent reportedly quit party during term)

(* One of three seats being elected from the same division on a rotating basis.)

Loss: Pembroke 2017 Lib, Huon 2022 ALP (incumbent left caucus during term)

Not contested: Pembroke 1999 Lib, Pembroke 2009 ALP 

Unsure whether lost or not contested: Gordon 1947 ALP, Tamar 1960 Lib, Cornwall 1961 Lib, Gordon 1968 ALP, Launceston 1968 Lib, Newdegate 1986 ALP.  (I think Labor did contest the last named but not certain.)  

Info on whether parties contested the unclear seats above is welcome, as are corrections.  

Wrap: Well that was a short on-the-day count and, for Labor, a very sweet one.  The result is a bit of an analogue of the federal election: The Greens did well, the indie did well, Labor has probably done very well on the 2PP despite increased competition on the left for primary votes, and Labor will be delighted after that one.  There is currently a 3.5% primary vote swing to the Liberals and a 5.7% primary vote swing away from Labor, but these will be exposed as meaningless when the preferences of the Greens (who have polled nearly 20% after not running in 2019) are distributed; I doubt the preferences of Willink and di Falco will do much.  I am thinking Labor might get a 2PP swing around 3%, a good result in a by-election for their own vacancy whatever it lands at, and one that again suggests the party brand has not been trashed beyond repair by the party's recent problems.  For the Liberals there is not a lot to like about these numbers.  They do have the excuse of having run a lower profile candidate but it is not as if Luke Edmunds is some kind of super-mayor; he is a first-term local councillor.   Signing off for tonight and will update further results as they arrive.

8:22 An unsurprising 45% for Labor at Warrane and they now sit on 39.5% to the Liberals' 28.8%. And we're at FINAL FOR NIGHT with an impressive 79.85% of turnout counted (it will get up well into the 80s with remaining postals).  The TEC has noted there are fewer than 500 votes to come.

7:57 Even better for Edmunds - wins prepolls by 11.9 points and is now outstripping my earlier projections as a result of his good pre-day performance and pushing towards 40% primary.  

7:49 Warrane (the division's most Labor-y booth) is still to come as well as whatever further prepolls and postals we may get tonight (I am not sure if there will be more postals.)

7:37 Edmunds also winning postals.  In fact he's doing so by about as much as he's winning booth votes.  This is a big win for Labor and they'll be very happy with it.  It's also a good night for the Greens and Hans Willink.

7:23 Edmunds wins Howrah despite the increased primary count caused by relocating Tranmere voters.  CALLED.

7:18 Lindisfarne Village is in and this was rather bad for the Liberals with the Greens almost beating them for second.  My projections now have Labor on 35-37, Liberals around 28, Greens around 21 and the 2PP around 60-40 to Labor.  That might be optimistic but it is looking pretty much callable - just want to run some projections on the size of the on the day vote.

7:14 The Tranmere mystery is solved - the booth was moved to an out of the way church in outer Howrah so I would expect a lot of Tranmere voters to vote at Howrah instead.  That may make the Howrah booth more Liberal than usual.

7:10 Geilston Bay is in and it didn't move the dial on my projections much, save that the Pembroke projection has continued to be a bit pessimistic about Labor's primary (dropping it below 34 with the others on 37).  The pattern so far is that most booths are not wildly below 2019 turnout after Tranmere was way below.  I'll be checking if any booths have moved that might explain that.

7:00 Have had a report that the Greens are running the Liberals close for second in a Lindisfarne booth.  Also the number of votes on the day are decent there and I have also heard of a scrutineering sample of unknown size with an 86% Green to Labor flow.  All pointing to an easy Labor win.

6:52 Two more booths in, Mornington and Montagu Bay and as expected Edwards now takes the primary lead and will be projecting to a strong win after preferences when I enter them in.  But again, the low on the day turnout requires some caution.  

On projections, all have the Liberals heading for about 29, Labor heading for 34-37 and the Greens heading for about 20.  Willink seems to have di Falco covered for fourth place but has dropped into high single figures.  The 2PP projection has Labor winning about 60-40 (that assumes an even split off Willink and di Falco, which might be optimistic).  A very strong start for Luke Edmunds.  Also the Montagu and Mornington booths are much closer to previous turnout so the very low turnout at Tranmere is a bit odd (perhaps just a huge postal vote there?)  I'm going to be cautious about calling this one early, normally would have already done so off this.

6:35 Tranmere is first off the block and (i) the number of on the day votes is very, very low (barely a quarter of 2019) (ii) the Liberals lead off primaries but that's to be expected there and not by enough to win the booth on preferences.  A strong vote for Willink, close to his home area but about four times his 2017 percentage there.  My projections off Tranmere booth put Labor somewhere in the 30s (varying considerably - the federal one is bearish at 31, the others are mid to high 30s), the Liberals around 30 and the Greens in the low 20s (would be an easy win for Labor) but (i) this is just one booth (ii) given the extremely low turnout in this booth, treat with a lot of caution.  There is a substantial primary vote swing from Labor to the Liberals in this booth, but one that would probably wash to nothing or less after preferences.

6:33 Nothing yet.

6 pm: Polls have closed.  There are 10 booths and no mobile booth.  The Mornington booth is significantly smaller than the others.  I'll be tracking the number of votes compared to 2019 to see how many postals/prepolls we might expect as I've so far seen no information on that.  

------------------------------------------

Intro (Saturday 12:45 pm)

Welcome to my coverage of the 2022 Pembroke Legislative Council by-election, caused by the resignation of Labor's Jo Siejka for family reasons.  My guide to the by-election is here.  To outside appearances this has been a relatively low energy contest, overshadowed by the federal election and in the final days the death of Queen Elizabeth II, and having to also compete for attention with the upcoming local council elections.  Nonetheless, the candidates have been campaigning hard (some of them at least!) and one of our five contenders will win themselves just over two and a half years on the red couches in advance of their first defence in 2025.  (The federal, state and Legislative Council elections are all scheduled for May of that year and while one or both of the others might be held early, there's a fair chance that Pembroke gets overshadowed again.)

In a desperate bid to breathe excitement into proceedings I will be projecting tonight's count using not one not two but three previous elections!  The votes for Liberal, Labor and the Greens will be projected off the 2021 state election and the 2022 federal election, and the votes for Liberal and Labor will be projected off the 2019 Pembroke contest (which also featured two fairly prominent centre-right and right independents).  

On the assumption that this will most likely be a Labor vs Liberal contest I will be running 2PP estimates as well, but these will be rubbery.  (If it is not a Labor vs Liberal contest I will hastily put together something else.) There have been very few cases where Labor, Liberal and Green all run for LegCo seats together, and there has never yet been one where the major parties finish first and second with the Greens third.  I would expect that, in the absence of how-to-vote cards, the preference flow from the Greens to Labor should be weaker than the 90.8% seen across Franklin in the federal election, but how much weaker isn't clear.  In the state election, preferences leaving the Greens and continuing to a major party flowed 85.8% to Labor across Franklin as a whole, but 28% of Greens preferences exhausted.  I have assumed Labor will get 82% of Greens preferences.  This will be modified if I receive a good volume of scrutineering data (say, a representative sample of over 100).

Why use three different projections?  Because I don't believe any of the previous contests are necessarily a robust guide to this seat.  Especially the last Legislative Council contest had no Greens and a couple of independents, while the state and federal elections had factors pulling voters to the Liberals and Labor respectively.  It will be interesting to see which one is the most informative, but I'm finding with test data that they usually come out within a few percent of each other anyway.

Labor's Luke Edmunds is generally considered a pretty strong favourite for this one.  A win for him would maintain the current Legislative Council balance in which the Government needs support from either the Opposition or at least one centre or left independent to win votes, but pretty often gets one or the other.  Labor would accept some swing against them for the vacancy (and especially on their primary vote because the Greens are running and ) but wouldn't want it to get very close on a two-party basis.  Unlike Huon where it was unusual that Labor ever won the seat in the first place, a loss for Labor in Pembroke wouldn't be acceptable - unless perhaps there was a massive swing to the state government in on the day voting that was absent from pre-poll and postal voting, the reason for which would be obvious.  (I really don't expect the death of the Queen to be a factor.)

Comments will start soon after 6 pm (slowly at first) and will go through til counting finishes (usually by 10 pm) with a wrapup posted sometime after that.  Refresh frequently after 6:45 see the most recent comments - at the height of counting on average there will may be new comments every ten minutes or so.  When I consider there is no realistic doubt about the fate of the seat the magic word CALLED will appear in the header.  Until then it will contain a brief summary of how the count is going, which may at times be out of date.  

In the event that the result is not clear at the end of tonight the best case scenario is that the top two are clear enough for an indicative TEC preference throw tomorrow or in the next few days; the worst case is the long wait for postals with a distribution on Tuesday week.  

At this stage I don't have figures on the likely level of prepoll and postal voting, which has been high in recent Legislative Council contests.  Any information I get about this will be added.