Sunday, January 25, 2015

2014 Ehrlich Awards For Wrong Predictions

This supposedly annual end-of-year feature is a little overdue, which I blame partly on the snap Queensland election and partly on the field being uninspiring compared to the 2012 and 2013 ensembles.  But while I wait for another poll to shed light on public response to what seems to be a ragged week for the LNP in the Queensland campaign, here we go. This site awards the Ehrlich Award early every year (unless I decide not to) to the "wrongest" public prediction I observe in or relating to the previous calendar year.

There are a few groundrules, for instance predictions need to be meaningful (in terms of being able to assess factually whether they have come to fruition), and predictions that carry a stated chance of falsehood are not included unless that chance proves to be ludicrously low.  The first of these, for instance, excludes Tony Abbott's pledge that in 2014 his government "every day [.. ] will keep building the stronger, more prosperous country that all Australians want and deserve" - not only are the terms of such platitudes undefinable, but those making them will continue maintaining they were true.  Likewise, I have so far been unable to find an empirical unit for measuring whether or not the President of Russia has been "shirtfronted".

Some election-watchers, including me, had a narrow escape from predictive disaster when we declared the Liberals busted in the South Australian by-election for the seat of Fisher following the revelation of a result-reporting error that meant Labor had over 52% of the 2PP vote on the night.  The Liberals did in fact lose as predicted, but only by nine votes.  (It's surprising how often this sort of thing - that a called seat springs to life then ends up with the originally called outcome anyway - happens.)

I did have at least one nomination-worthy prediction that was actually incorrect though: in looking ahead to the Tasmanian Legislative Council seat of Huon without any polling data I had "no hesitation at all" in predicting Liberal Peter Hodgman (uncle of the new Premier Will) would win.  Oops. Not quite the same thing as declaring it a certainty in advance, but nonetheless way overconfident.  This prediction was by analogy with Labor's string of Upper House wins after they had won the 1998 state election.  However it turned out that the Liberals' huge win in the 2014 state election was more a vote against Labor, the Greens and PUP than one of mass enthusiasm for Liberal rule, and therefore history didn't repeat. Hodgman snr led on primaries but by not nearly enough and was gobbled up by Huon Council mayor Robert Armstrong on preferences.

The 2014 Tasmanian lower house election (easily won by the Liberals as expected here) was one of the year's richer veins of predictive fools gold.  I've told the story of a certain social media swimming wager too many times on here to go through it all again, but some mention should be made of Palmer United Party claims, supposedly based on polling, that they would win government in their own right,.  This was later revised down to 3-5 seats. Their end result was zero and not all that close to even one.

PUP's nomination must be marked down for their well-known habit of making any noise imaginable with no quality control at all just to get media attention, and Bob Ellis' contribution was also merely true to form (he did get the Labor seat tally right, but nothing else).

 Perhaps very slightly more interesting (and I'm a little biased on this) was the failure of Tasmanian Times, a supposed alternative to mass media down here, to contribute much other than nonsense to advance expectations of the 2014 state election.  In the absence of its former house pseph for reasons noted when this site commenced, the best TT could come up with was this spectacularly wrong piece which asserted as fact that the Liberals "just won't" win a majority.  The piece (whose pseudonymous author did predict the party breakdown in every electorate correctly in 2010) asserted that Joan Rylah's hope had "disappeared" (she won), that Sarah Courtney "hasn't a chance" (she won) and that Tim Morris "should easily pick up a quota and top the poll" and was by implication the only incumbent who was "safe".  Morris polled below half a quota, was sixth on primaries, lost and was the only incumbent in that seat close to losing - a shock result admittedly, but one assessed here as an outside but realistic chance.

Similar examples could be found in the SA state election too, where plenty would have completely written off the struggling Labor government that pulled off an unexpected minority win.  But for this year's winner I turn to federal politics.  Last year's prize went to a former Labor PM and a former Labor Treasurer, and this year's goes to the other side of politics.

On 8 March 2014, Malaysian Airlines flight MH-370 disappeared in still mysterious circumstances, having apparently changed course for unknown reasons without any indication of danger or further communication from the crew.  Complex reconstructions of a range of satellite data suggested that the plane flew way off its intended course for several hours and then crashed somewhere in the southern Indian Ocean. (Alternative "theories" will not be debated here.)

Australia joined the search for wreckage - which to this stage has yet to produce anything - and it wasn't long before there were promising noises.  Following the detection of some underwater "pings" by Australian and Chinese search forces, the Prime Minister Tony Abbott held a press conference a month after the plane disappeared.  He declared that "We are very confident that we know the position of the black box flight recorder to within some kilometres" and "We have very much narrowed the search area and we are very confident the signals are from the black box".

It is true that Abbott did not specifically predict that wreckage would be recovered and indeed warned against inferring one thing from the other.  But his comments can be taken as implying a range of other predictions, including that there would be agreement that the search area was correct and that the signals would be accepted as being from the black box.

Neither of these things turned out to be the case.  No further evidence was found in the target area, searching has since continued over a much wider area and the pings in question have since been widely dismissed by scientists, among other things because the frequencies involved don't match.

Further interest in the PM's "very confident" noises arose from the apparent lack of any official advice to the PM that actually went quite that far - it appeared he went further than the briefings presented to him out of desire to score a political win at a time when he needed it.  The announcements came at a time when Abbott's government had been trailing narrowly in the polls for five months and was just about to deliver what has turned out to be an extremely unpopular Budget.

Yes, it's a mediocre field this year and it's a bit of a stretch to have to go to inference from what could be taken to be more a claim of fact about the present than one about the future.  But Tony Abbott's overconfidence in jumping in and implying that the net was closing on the wreckage of the missed airliner when he was relying on areas of science he knew nothing about, is just interesting enough to make him the winner of this year's gong for getting it wrong.

He may not have too many other wins this year.

No comments:

Post a Comment