Thursday, November 7, 2019

Braddon And Bass 2019: Another Rec Fishers Preferences Beatup

One of the eternal tropes of Australian media electoral reporting is the breathless expose of how the preferencing behaviour of some obscure party or candidate could swing or did swing an important contest or in cases an entire election.  And, on a day when there was quite enough going on for election buffs to look at in the Chisholm and Kooyong signs challenge (see my updated coverage of that on the link) it has unfortunately broken out again with the sensational headline "CFMMEU-funded independents helped Liberals steal two key seats".  (CFMMEU = Construction Forestry Maritime Mining and Energy Union).  The article goes on:

"The CFMMEU helped the Coalition win two key seats back from Labor at the May election by funding the campaigns of two independents who sent 1757 votes between them to the Liberal Party."  It then refers to independent recreational fishers Todd Lambert (Bass) and Brett Smith (Braddon).

The piece has been widely criticised on the Twitter psephosphere, but not everyone uses Twitter, and especially as the seats in question are Tasmanian I think it's worth posting a detailed explanation here of why this piece is incorrect.  I note that no psephologist was interviewed for the article.

Firstly, the basic facts on Lambert's and Smith's preferences on a two-party basis.  Lambert polled 2607 votes (3.79%) in Bass and his preferences ultimately split 1370-1237 (52.55%) to Labor, in a seat Bridget Archer (Liberal) won by 563.  Smith polled 1203 votes (1.72%) in Braddon and his preferences split 683-520 (56.77%) to Labor, in a seat Gavin Pearce (Liberal) won by 4239.

So there's the first problem for this narrative right away - while 1757 votes flowed from these recreational fisher candidates to the Liberal Party, 2053 flowed from them to Labor.  So even if we assume that nobody who voted for those candidates would have otherwise voted at all, the conclusion to be drawn from that would be that these candidates helped Labor slightly, not hurt them.

But of course the vast majority of voters for these candidates would have voted anyway had these candidates not stood, they just would have not voted 1 for those candidates.  To argue that the CFMMEU helped the Coalition by having these candidates run, one needs to argue that had these candidates not run, voters who voted 1 for them and ultimately preferenced the Liberals would instead have voted for or preferenced Labor.  This is despite these candidates actually preferencing Labor, so ...

The idea is simply nonsensical.

What actually happened here is just a typical story of low-polling candidates, especially independents, in federal electorates.  Low-polling independent candidates tend to have weak preference flows between the majors.  They tend to attract a lot of "random" voting:

* voters who know the candidate personally and vote for them irrespective of the voter's normal politics
* voters who dislike all the other candidates, sometimes for idiosyncratic reasons, and vote for the candidate by elimination
* voters who like independents on principle
* donkey votes if first on the ticket

(and so on).

As well as these "random" factors, low-polling candidates and campaigners for them have little impact on their own preference flows.  People who vote for low-scoring candidates are generally independently minded in their politics and more likely to make up their own minds.  Low-polling candidates tend to have small support bases and limited abilities to distribute how-to-vote recommendations anyway.

The other thing to note here is that while Lambert was one of several candidates whose preferences together determined the very close Bass outcome (had the flow from him been 64% to Labor, Labor would have retained), in Braddon the Liberals were easily home without needing a single Smith preference.  So the idea in the headline that Braddon was "stolen" would be pejorative nonsense even if the CFMMEU had caused some of Smith's preferences to flow to the Liberals.

2016 Beatup Redux

The article reminds me of a similar beatup following the 2016 election, where the Liberals (Eric Abetz in particular) tried to blame the Recreational Fishers (then a party) for the Liberals' defeats in Tasmanian seats.  On the surface, this seemed to have some plausibility - the Rec Fishers had polled strongly in these seats, polling between 4.9% and 6.3% with between 60% and 67% of their preferences flowing to Labor.  But it turned out on closer scrutiny (see my detailed article on this and Abetz's other excuses for the 2016 result) that these voters were mostly not voting for the Rec Fishers out of positive enthusiasm, but rather because they considered them the least worst option on the Reps ballots.  This was proven by the Rec Fishers sinking without trace in the Senate, where they polled just 0.7% statewide.  Moreover, only 44 voters in all of Tasmania followed their Senate how-to-vote card! So the idea that the Rec Fishers were an effective preference harvester for Labor in 2016 was nonsense.  Rather, voters who voted 1 for the Rec Fishers in 2016 and preferenced Labor would generally preferenced (or voted for) Labor anyway.  They didn't cause Labor wins in 2016, and off smaller votes they are hardly causing Liberal wins now.

The individuality of voters for the 2019 minor candidates shows up nicely when you look at where their preferences actually went first when distributed, though this is slightly messed up by votes passed to them from other candidates also being included.  Smith was excluded with 1234 votes (31 from another candidate) and these split as follows:

Brakey (IND) 42.9%
Labor 18.3%
One Nation 12.2%
Liberal 8.8%
Green 6.5%
UAP 5.9%
National 5.4%

Lambert was excluded with 3212 (605 of these from two other candidates) and these split as follows:

UAP 37.2%
Liberal 24.3%
Labor 23.9%
Green 14.6%

So in the first case we have a lot of voters voting for one independent and preferencing another.  In the second, it's interesting that the UAP did well on Lambert's votes having done poorly on Smith's, and a probable reason for this is that by Lambert's exclusion the UAP were the last remaining party that was none of Liberal, Labor and Green.  These big three had held every Tasmanian lower house seat for 21 years until a recent defection of sorts, and there is some level of sentiment out there that is sick of all three of them.

There is no real question of the CFMMEU's financing/encouragement of these candidates driving preferences to the Liberal Party.  The first question really should be whether a union supporting small-party or independent candidates who preference Labor on their cards is an attempt to harvest preferences for Labor.  If it is, it is clear it doesn't work as the preferences of obscure candidates are largely undirectable - but it might also be that a union with "maritime" in its name is genuinely more interested in promoting fishing issues by supporting candidates who will make noises about them.  The second question is whether such a union would support its objectives more effectively by simply donating straight to Labor.  All I can say is that if the objective is to harvest votes for Labor then running obscure candidates doesn't work - but I can't say if that's actually the aim or not.


  1. Union with "fishing" in its name, Kevin?

  2. Nothing will ever beat the “Communist Party preferences saved Menzies from losing government” trope. Like most political myths, a pile of self-serving rubbish that people swallow cos they like the sound of it. (or don’t care enough, or - most often - because ‘professional’ journalists keep repeating it, so others reasonably enough then assume it must be true.)

  3. Have you got any idea what has happened to emrs polling in Tasmania? What has happened to state opinion polls in Australia in general? Is this the new normal i.e no more polls until the pre-election period begins?

    1. EMRS has continued to poll every three months, but unfortunately in recent years it has sometimes not released results until many months after they were taken. Thus, the March and May EMRS polls were not released until the July EMRS poll was released in August. An EMRS poll has been reported as in the field but it is unclear when it will be released.

      State polling has in general atrophied over the past few years. The Courier-Mail's Queensland YouGov (formerly Galaxy) series has continued but other state polling has been very rare, and I suspect will remain so with so long until the next elections.