In the event, the Labor government was returned with a reduced but clear majority, the size of which is not yet known, while the Coalition has suffered a crushing defeat.
How could the pre-election coverage have been at once so breathless and misleading?
The short answer is because of a combination of groupthink and wishful thinking. Unpacking this requires the disclosure of a few trade secrets.
Two days out from polling day, the Herald Sun published an analysis of some focus-group research by RedBridge Group, carried out over the past two years.
It stated the likeliest scenario on November 26 would see Labor with 43 seats and therefore forced to form a minority government, given it requires 45 seats for a majority. The best-case scenario for Labor was 48 seats and a return to government in its own right.
Earlier in the campaign there had been loose talk in the Herald Sun, based on no particular data, that there could be a hung parliament.
Then in the last week, a Resolve Strategic poll for The Age showed the primary vote for Labor and the Coalition tied at 36%.
It seemed the race was tightening and perhaps a hung parliament or a minority government were real possibilities.
For the media, this is exciting stuff. It suggests drama, suspense, uncertainty – all powerful news values.
So at rival newsdesks, one can imagine an element of consternation. A chief of staff (COS) can be imagined ringing a state political reporter:
COS: “See the Herald Sun has a survey suggesting a minority government?”
Reporter: “Yeah, but some of it’s two years old.”
COS: “Yeah but a minority government. That’s big. I think we have to have something.”
Reporter: “All right. Something.”
COS: “I mean, we’ll look like dills if we don’t have something and it happens.”
Hours later at news conference, where decisions are made on what stories go where, everyone around the table has seen the Herald Sun. At The Age they’ve also seen the ABC pick it up and at the ABC they’ve seen The Age pick it up. Each reinforces the other’s assessment of the story’s credibility.
The chief of staff assures conference that state rounds are on to it. Minority government becomes the story. Its origin in qualitative data, some of which is two years old, stoked up by the Herald Sun as part of its relentless campaign against the Andrews government, is forgotten or overlooked.
Evidence to support the minority-government hypothesis is assembled, especially the Resolve Strategic quantitative data showing the primary votes neck-and-neck.
News conference’s resident Cassandra raises a voice. “What about the two-party-preferred?”
Editor: “What about it?”
Cass: “Every poll we’ve seen so far has Labor ahead by up to ten percentage points. And they’re up to date, not weeks, months or years old.”
Editor: “So you’re saying we should just ignore the RedBridge stuff?”
Cass: “No, but you can’t ignore the two-party-preferred either.”
Editor: “All right. Put in a parachute about the two-party-preferred but lead on the minority government. I mean there could even be a hung parliament. We’ll look like dills if we downplay this.”
Yep. And that’s how you look when wishful thinking and groupthink cloud hard-minded analysis of all the available data. Taken together, the data showed the likeliest (but journalistically least interesting) outcome was the return of the government with a reduced majority.
Not only did the two-party-preferred vote not tighten appreciably, but the primary vote turned out not to be neck-and-neck. This is not hindsight. The discrepancy between the two should have raised a red flag: how could the primary vote be neck-and-neck when the two-party-preferred gap was so large?
In fairness, it was reasonable to suppose this could just be a function of how the minor party and independent preferences would flow, which was unknowable at the time. But this seemed not to enter the discussion about the prospect of a minority government.
And a hard-headed look at the RedBridge focus-group data would have revealed to a dispassionate analyst that once the more far-fetched cases had been eliminated, Labor was likely to end up with somewhere between 47 and 50 seats.
The ABC’s election analyst, Antony Green, is giving Labor 52 seats at this stage, with 68% of the vote counted.
Even more curiously, the hung parliament and minority government possibilities were initially generated by the Herald Sun, which acted throughout as a propaganda arm of the Liberal Party. Why on earth would respectable and usually reliable elements of the media such as The Age and the ABC buy into this nonsense?
The answer is that it is an abiding weakness in newsroom decision-making to prefer the most dramatic possibility, however remote, over the most mundane but strongest probability.
It is a further weakness to wish not to be scooped on the most dramatic possibility, even at the expense of misleading your audience, looking foolish in the aftermath and buying into scenarios created by your most politically partisan and least reliable media rival.
The result was a feverish outburst of speculation in the final week of the campaign that fed into questioning of Andrews about whether he would entertain doing deals with crossbenchers if Labor could not muster the 45 seats necessary to form government in its own right.
He batted it away with his customary dismissiveness, and who could blame him?
This article is republished from The Conversation is the world's leading publisher of research-based news and analysis. A unique collaboration between academics and journalists. It was written by: Denis Muller, The University of Melbourne.
Denis Muller does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.