Lies, Damn Lies, and Wishful Thinking

Death of Pauline Hanson’s One Nation greatly exaggerated.

Writing about the Queensland election results I am struck by the similarities with the press coverage of the 45th Presidential elections in the U.S.A. which I wrote about previously. Here, as there, there have been multiple, almost 180 degree shifts in predictions about the election result.

Here, as there, nearly everybody got it wrong. In this case, they’re still getting it wrong.

Before the election we were treated to article after article predicting that Pauline Hanson’s One Nation (PHON) was resurgent, and would possibly hold the balance of power with as many as ten seats.

The actual result is that they contested 61 seats, and will win ONE.

Now we’re seeing a flurry of navel-gazing articles asking “how did we get it wrong”. We’re also seeing articles explaining how Labor preferences “saved” LNP candidates in some seats, and likewise how LNP preferences “saved” Labor candidates in others.

Then there are lots of articles rejoicing in the “death” of PHON, because of their failure to win seats.

What nobody seems to be doing is connecting the dots between these articles and pointing out these bleeding obvious truths:

  1. PHON’s statewide support is far higher now than it has been for many years. Far from being dead, it’s actually on the rise again.
  2. PHON failed to gain the seats it hoped for not because its vote was down, but because the compulsory preferential voting system favours the major parties when protest votes for minor parties are forced to preference the majors.
  3. You can be sure that this will only mean further moves to the right by the LNP, and probably by the ALP, with possible further Green wins.

The following table (derived from data the Electoral Commission Queensland website) demonstrates these points.

QLD PHON Election results 2


(The 2017 results were still undeclared at the time of writing, however 76% of votes had been counted and there was no indication of any significant change in the results for PHON. For simplicity, LP and NP results have been totalled in years before the amalgamation.)

The first, and most obvious result from the table is that percentage of first-preference votes is not a reliable predictor of seats won. You can win government with 36% or 37% of first-preference votes as the ALP appears to be doing in 2015 and 2017, or you can lose with 42% of first-preference votes, as the LNP did in 2009.

By the same token, in 2001 PHON received 8.7% of primary votes but won 3 seats, however in 2017 with 13.7% of the vote it will win only one.

None of this is surprising given that aggregated figures hide individual seats, and preference flows aren’t revealed by first preference counts.

However these results only go to underscore points 1 and 2: PHON’s statewide support is going UP, not DOWN, because using seats won is not a good measure of overall support. Primary vote is by far the best measure, in that it’s a statistically rock-solid major sample (in those electorates) of voters’ first choice. In this case we have – not samples – but the preferences of every elector in 61 electorates.

PHON is not dead, it’s on the rise. Its primary support may be less than the nearly 23% it received (along with 13 seats) in 1998, but equally it has risen sharply and substantially from its 1.5% near annihilation in 2012.

For point 2, there’s quite a lot of material on the change from optional preferential to compulsory preferential voting available. This article from the Brisbane Times points out:

ABC election analyst Antony Green reviewed the 2015 results and found that if compulsory preferential voting had been in place, Labor would have won eight more seats than it did, bringing its number in the House to 52.

It’s not as simple as that, though. In a later ABC article Antony Green re-analysed the 2015 State results using the preference patterns from the 2015 Federal election:

He found that if people preferenced in 2015 in the same way they did in this year’s federal election, Labor would have actually won two less seats than the party did, likely meaning the LNP would still be in power.

In other words, your mileage may vary. What can definitely be said is that compulsory, rather than optional, preferential voting removes voters’ ability to “exhaust” their vote; that is, not specify a preference for candidates they don’t know about or actively dislike. This system always works in favour of the two candidates who will still be standing near the end of the count, and disadvantages third place getters which is where minor parties often sit. Here’s a simple example: on first-preferences in a seat, the results are: PHON 31%, ALP 30%, LP 19%, NP 20%. If everyone just voted “1” then PHON would win the seat. However if preferences are forced then LP preferences will flow to the Nationals, with, say a 2% leak to the ALP. Now we have PHON 31%, ALP 32%, NP 37%.  PHON drops to third place and is eliminated. This is despite it receiving the majority of first-preference votes.

Something else that’s clear from the results is that both major parties’ first-preference support has dropped. In the case of the ALP the movement is probably to the Greens, but for the LNP, as in 2008, their losses are PHON gains.

This brings us to point 3: in 1998 the only way the LNP believed it could regain that lost ground was to move even further to the right, and it did. The same has been seen in right-wing parties all over the world, and there’s no reason to believe it won’t continue.

The ALP is between a rock and a hard place. If it moves to the right in order to compete with the LNP, then the Greens vote will go up and Labor will start to lose seats. Stay still, or move to the left, and regional electorates that are more conservative and support PHON will be lost for sure.

In attempting to please everybody both major parties have displayed a terrible lack of consistency and basic principles that has clearly frustrated the electorate. Unfortunately the result has been an increase in the number of people promoting the politics of blame, hatred and division using simplistic slogans and facile policies, and that’s just the Coalition…

It has also seen the return of PHON.

They may not hold the balance of power for this term of the Queensland State parliament, but that doesn’t mean their malign influence is gone, or falling; far from it. It’s on the rise. Again.


4 thoughts on “Lies, Damn Lies, and Wishful Thinking

    • 😄 Indeed. You had me confused for a moment, as I have been almost completely absent from twitter for a couple of years and done very little blogging in that time also. I’m finally picking up the threads again now, but somehow missed most of the momentous (and interesting, in the Chinese sense, alas) events in your life. Commiserations, and I’m very happy to see you’re still swinging.

      • I’ve been absent too Peter. I am surprised you put two and two together after so long. I was surprised I wasn’t following you on Twitter already, but maybe that was a Twitter error at some point. Life happens, you know. We are both sad, actually. No bad stuff, just could not mesh the cultures at our advanced age. As for your “swinging” comment, I’ll assume you mean “fighting (to be) fit”.

      • Oh, absolutely – in there swinging, an unnecessarily pugilistic metaphor for, um, still fighting, hang on, let’s move metaphors to somewhere more peaceful… Sorry, I hadn’t even considered any other interpretation, but thanks for the presumption of innocence.
        I may have changed twitter handles, although that was some time ago, it’s @infinite8horizo (pre long names). Nobody can hide from Google, though.
        Nice to hear from you, and no doubt we will catch up again!

Comments are closed.