The election prophecies of the Canberra psephologist Malcolm Mackerras are really just harmless entertainment. Last week he predicted that the Greens candidate Clive Hamilton would defeat the Liberal Kelly O’Dwyer in the byelection for the Melbourne seat of Higgins on Saturday, and prophesied that the Liberal Paul Fletcher would be forced to preferences in Bradfield on the north shore.

Of greater concern are the byelection predictions of some social science academics who are employed to teach politics to fee-paying students at taxpayer-subsidised universities. Both O’Dwyer and Fletcher increased the total Liberal vote, after the distribution of preferences, over that which was obtained in the 2007 election.

Labor did not run candidates and the Greens were not able to match the combined anti-Liberal vote of two years ago. Yet some academics predicted not only a dismal showing for the new Liberal leader, Tony Abbott, in his first electoral test but also the demise of his party.

In a bizarre article in The Australian on Friday, Robert Manne, a politics professor at La Trobe University, canvassed not only a victory for his friend Hamilton but also “the destruction of the Liberal Party” this week. Manne acknowledged some of his views were “fantasy” but it was difficult to work out what part of his article was fantasy and what was academic analysis. Most teachers would fail a paper like this if it were presented as a university essay.

Manne also made his position clear on the Liberals, referring to the party’s “troglodyte-denialist wing” and Abbott as the “troglodyte-in-chief”. Such language seems acceptable in the La Trobe University politics department.

Judith Brett, Manne’s professional colleague, did not throw the switch to fantasy or engage in labelling. Even so, her analysis was very similar to Manne’s. Writing in the Herald on Saturday, she said that “the Liberals risk becoming a down-market protest party of angry old men in the outer suburbs”. She also said the Liberals were “the natural party of the big end of town and of the big producer groups”.

In fact, big business and the big producer groups are willing to co-operate with whichever party is in government. The core of the Coalition’s support turns on medium to small business, farmers and middle-income earners.

According to O’Dwyer, the Liberals gained votes in such suburbs as Carnegie and Murrumbeena, which are not the high socio-economic parts of Higgins, where she received strong support from young married women. So much for Brett’s analysis. Or perhaps fantasy is a better word.

Manne and Brett are not alone. Brian Costar, a professor of political science at Swinburne University, said he expected Higgins would go to preferences. And Paul Strangio, a member of the Monash University politics department, wrote in The Age that “Abbott’s leadership will need emotional intelligence – a quality in short supply in the Liberal Party in recent times”.

Manne, Brett, Costar and Strangio are all left-of-centre or leftist academics who comment on the Liberal Party as part of their professional career. A reading of their analyses this week reveals the pitfalls of projection. Manne, Brett, Costar and Strangio dislike Abbott’s social conservatism and his rejection of the Rudd Government’s emissions trading scheme. They made the familiar error of projecting their views on to the voters in Higgins and Bradfield.

There is also an unpleasant double standard here involving Tony Abbott’s Catholicism. On Friday Manne wrote that “very many Australians will not vote for a Catholic party leader whose religious convictions fashion their politics”. Manne was the chairman of The Monthly when it ran Rudd’s essay on the German theologian Dietrich Bonhoeffer in 2006, and enthusiastically endorsed Rudd’s religious convictions at the time. The views of Rudd and Abbott on social issues are not far apart. Yet it seems, according to Manne, Rudd’s religious convictions are acceptable while Abbott’s are not.

Come to think of it, the fantasy surrounding last Saturday’s byelections has not been confined to academics. This year, the Radio National program Breakfast has been giving publicity to Fiona Patten’s new Australian Sex Party. As recently as last Friday it was suggested on Breakfast that the party could win a seat in the Senate. Not on Saturday’s vote it couldn’t. Patten scored 3.3 per cent of the primary vote, finishing behind the Democratic Labor Party candidate John Mulholland. This is a breakaway from the original DLP, which was formally wound up three decades ago.

Few would expect that Abbott could lead the Coalition to victory in next year’s election. His task will become more difficult following the decision of Malcolm Turnbull to adopt the stance taken by such former Liberal leaders as John Gorton, Malcolm Fraser and John Hewson and become a public critic of the party he once led.

Turnbull’s announcement that he would cross the floor and support Labor’s emissions trading scheme is a blow to the Coalition. But it does not overturn the fact that, based on last week’s Liberal Party secret ballot, 75 per cent of Coalition parliamentarians support Abbott’s approach on climate change.

The Liberal vote at the weekend indicates that Abbott is capable of at least stabilising the Coalition vote at the level of the 2007 election and perhaps increasing it somewhat. Moreover, Abbott’s approach may attract support among the lower socio-economic groups who elected Robert Menzies in 1949, Fraser in 1975 and John Howard in 1996. This is a fact that the left-of-centre academy has invariably been slow to appreciate.