Now that the election counting for the House of Representatives is done and dusted (with Clive Palmer’s 36 vote victory in Fairfax being the last outstanding datum of import; though see the interesting Senate races still undeclared), there’s been a nice coincidence of fact-based articles which seek to disrupt the narrative that there was some sort of Abbott/Coalition landslide.
In part, this was foreshadowed by a series of polls which overshot the mark (particularly robopolls of individual seats, but also a Newspoll with an appalling sampling technique purporting to show Labor wipeouts in Western Sydney and Queensland which never eventuated). The mistake the media made was to ignore (or not to know) the fact that elections are dynamic not static events. Even had the seat level polls been accurate, you can never just put a few seats in a box called “Lost” (which many papers literally did), and then throw another few in the next time you have some polls to report. Everyone who is actually involved in close contests knows that the lead swings. Undecided voters are normally switching several times, being won and lost all the way up to the polling booth.
It didn’t help that the bored commentariat pronounced the election over the week before and turned their attention to Double Dissolution and Labor Leadership scenarios.
Here’s Phil Doyle in New Matilda:
Whichever way you want to look at the result of the recent Federal election, one argument that doesn’t stack up is that this is some kind of landslide for the Coalition. That’s how the election is being rewritten — but the figures don’t support the story.
Fifteen of the Coalition’s new seats are held on very thin margins. Eleven seats have margins of less than 4000 voters, according to AEC figures as at 24 September:
Barton (NSW) 493 Petrie (QLD) 971 Eden-Monaro (NSW) 1071 Capricornia (QLD) 1381 Dobell (NSW) 1189 Solomon (NT) 1492 Reid (NSW) 1490 Lyons (TAS) 1605 Banks (NSW) 3169 Braddon (TAS) 3398 Hindmarsh (SA) 3541
If Gilmore, Lindsay, Robertson, and Deakin are included, less than 30,000 voters nationally would need to change their minds for the government to change as well.
And Malcolm Mackerras in Crikey:
The Whitlam government was defeated in a landslide in 1975, and the Rudd government suffered a respectable loss in 2013. But what of the defeat of the Keating government in 1996? Having examined all the data I would call it a respectable loss.
So, contrary to The Narrative, it is a fact that there was no landslide (and thus Mr Abbott is eminently beatable in 2016). It is also a fact that the polling, and the results, showed volatility and dissatisfaction rather than some race to the LNP. That’s a problem for a government which seems to be governing as if it has won big – from the right, refighting the culture wars which are of such irrelevance to much of the electorate.
Of course, as Bernard Keane argues today, this government hasn’t really thrown the switch from fact free campaigning to governing yet. It seems to have a continuing problem with fact.