Beyond the sliding doors election

"[T]his election…is a sliding doors moment for our nation," declared Peter Dutton, and maybe he was right, and in ways he didn't quite intend.

Sliding doors refers to the 1998 British film of the same name. The premise is that a seemingly minor event or decision can dramatically alter the course of a person's life.

The election results suggest the sliding doors analogy was accurate, at least as far as the Liberal Party was concerned. Polls late last year and early this showed them possibly narrowly winning government. With Labor dominant, and the Liberal Party routed, the doors slid and the alternate reality arrived.

And accurate too in the way voters treated the election - casually as though nothing much was at stake.

If Dutton had a chance of winning this election he needed to build a sense of crisis, but he didn't. It needed to be framed more like the Kokoda Track, from his perspective, than Sliding Doors.

In contrast Anthony Albanese needed to get people focussing on the future and convince them that he was a safe pair of hands and Dutton a risk, and that his stewardship was sound in a world less certain than it was 6 months ago.

On our polling, taken in two tranches, one finishing on April 3, right at the beginning of the campaign, and the second an "exit" poll taken over the weekend of the election and finishing at midnight on Monday May 5, it appears that Peter Dutton lost the election more than Anthony Albanese won it.

The deterioration in the Coalition vote and Dutton's standing is more dramatic than the increase in the Labor Party's and Albanese's.

I'm not saying Albanese didn't do well – he did – just that the Liberal Party did atrociously. True, the National Party held all their seats, but we don't have enough National Party respondents in our samples to make a comment on them.

Dutton was never that popular with our respondents, but his approval went from a net -19% to net -42% (a change of -23 percentage points) over the course of the election. At the same time Albanese went from -20% to -6% (a change of +14 percentage points).

Asked whether they believed the Opposition deserved to be the government, respondents agreed net -21% at the beginning of the campaign, but almost exactly mirrored Dutton's fall by ending up 21 points lower on -42%. On the same metric the Labor government went from -11% to +1%, a 12 point increase.

You might be beginning to see why voters were casual about this election at the beginning. Neither side appealed at the beginning, although they made a decisive decision by the end.

On the question of preferred Prime Minister, while Albanese went up from 50% to 57% Dutton fell from 45% to 38%, 7 percentage points either way, so a kinder judgment than that on performance, but again, very decisive.

And our two-party preferred vote was 57% Labor to 43% Coalition, better for Dutton's side than the preferred PM figures. (This result is skewed to the left, showing that despite our best attempts to adjust for political bias, the sample still had some).

A number of things appear to have turned possible change of government into a landslide towards the government. First has to be Peter Dutton.

He went into this election with the attribute of being strong. Some people also linked this to him being divisive and racist, and likened him to Trump, but these were voters who had never voted Liberal or Nationalist in their lives – they weren't issues that would swing an election.

When we took our second poll, the word "strong" was still associated with Dutton, but this time as a qualifier. "He doesn't appear to be as strong and naturally commanding as Albanese," and "He wasn't strong enough and didn't have good policies," are two representative comments from Coalition voters.

Dutton had walked back policies like work from home (which didn't register as an issue in our sample, so it was probably consigned to particular seats), not branded the government with its record, talked about where he was going to live when he was elected, or Welcome to Country ceremonies, rather than cost-of-living issues.

The issue that could have made a difference to either side, if they'd leveraged it properly, was housing – a cost-of-living issue that affects almost everyone under the age of 50 or 60 because they are paying-off a mortgage at suddenly massively higher interest rates; can't afford a deposit to get into a house where they couldn't afford the repayments anyway; have to pay high percentages of their income on rent; or have children, grandchildren, and friends who fall into any or all of those categories.

This issue was the most important to most of our respondents, but it was a side issue for both the major parties. While both had policies to address it none of them penetrated.

The issue Dutton was most clearly identified with was nuclear. At the end of the election it was most likely to be raised when making a personal judgment on him, and it was obvious, even from Coalition supporters, that he had not made out the case for its necessity.

Albanese was most closely associated with health, but despite the clear $8 billion on Medicare, and the promise to build more urgent care clinics and provide telehealth centres, they weren't specified by voters – they just referred to health in general.

The same thing happened with education, where forgiving HECS debt, or free TAFE weren't mentioned, but the issue of education was.

Those voting for Albanese were also very likely to cite interest rates and inflation. There was a sense that Australia had turned the corner and it was the government's doing, even though times were still tough. Labor didn't appear to attract the blame for the problems, so there was no need to punish them. Or if they did attract blame, so did the Morrison/Frydenberg stewardship.

But as noted earlier, emotions were not really that strongly engaged on policies. Decisions were as likely to be made on the basis of who they didn't want as prime minister. In this situation the hate economy was more or less equal.

However, Coalition voters were twice as likely to cite Peter Dutton as a positive reason for their vote than Labor voters were to cite Anthony Albanese as the positive one for theirs. That is not good for Albanese, no matter that his voters think he is "decent", because their support is not strong.

The most startling figure from our polling was that while having voted Labor in with a landslide, fully 40% wanted a Coalition win, versus 42% who wanted Labor, while 19% favoured a hung parliament. You get what you vote for, not what you want, but the difference between the fact and the desire is striking.

Issues that stayed with their tribes were "climate change", strongly correlated with Greens and Cosmopolitans, but not so much Labor; the "economy" and "immigration", strongly correlated with Coalition and Nationalists, along with "security" and "defence".

In the end, this result might be neither sliding doors, nor Kokoda, but a tribute to electoral volatility. The minor party votes are at more or less record highs, and this masks the underlying strength of each major party's votes. Take renewables out of the mix, and the Teals seats could look quite different, for example.

The victors need to be humble in victory, and the vanquished are beaten, not terminated.