An open letter to Kevin Bonham


Dear Kevin

As arguably Australia’s best known and most inaccurate amateur psephologist, I have always admired your excellent work as a real one.

I know we’ve exchanged a few harsh words in the past, largely over a secondary role of mine at The Bug, namely to take the mickey out of #Newspoll from time to time. For that reason, you may not believe me when I declare I am in thrall of your knowledge of the study of political trends.

I can only imagine the academic honours you have obtained to reach that level of professional evaluation, the techniques and the complicated computer algorithms you apply, the endless hours of research you undertake and how you apply all of that information to interpret opinion polls and, more broadly, how parties are performing during campaigns and whether or not their messages are being absorbed by voters.

As you probably know, as arguably Australia’s best known and most inaccurate amateur psephologist I go largely by what my ever-increasing gut feeling tells me. My assessments and predictions aren’t too much more scientific than reading tea leaves or using a stick to examine the entrails of a freshly killed chook or watching Scott Morrison’s inability to get through a media call without lying at least a half-dozen times.

But having said that, as Malcolm Fraser would say, let me say this.

I’m rather proud of my record as an amateur psephologist. I’ve predicted a few election outcomes over the past half-century that many, many experts have missed.

A Queensland-based online news service was bragging the other day about how their federal politics scribe had so accurately predicted the outcome of the 21 May poll.

Indeed, I’m also very proud of my forecasts for that poll, based on the big swinging dick electoral pendulum I invented for this most recent federal election.

You don’t have to read any of the text books I’ve never written to know that in the months leading up to Election ’22 and then the long six-week campaign proper that followed, my pendulum (copyright pending) looked only at the primary vote of the Liberal and National Parties.

I did that for a number of reasons: I didn’t have the time or the skillset to do it any other way, and you’d be aware as a real psephologist that several of the major pollsters after the 2019 federal poll dropped publishing a two-party figure on various bullshit grounds.

It wasn’t rocket surgery on my part that the LNP had fuck-all chance of winning on 21 May if their combined primary vote refused to grow above the 35-36 per cent it was stubbornly stuck around for month after month, poll after poll.

I first formulated that technique of looking specifically at the LNP primary vote several federal elections ago when I heard Barrie Cassidy first argue that LNP needed a primary vote back then of around 42 per cent to win elections.

At the 2016 election, I was rather proud to be sitting in a north London pub providing comfort to increasingly dejected Labor/Labour supporters as that count came in. Watching the primary figures, I kept telling them the result would be very, very close. And it turned out to be exactly that. It also helped that I knew what a FIZZA Malcolm Turnbull had turned out to be.

Just for your interest, Kevin, I made only one prediction, constantly repeated, in the two years leading up to the 21 May poll: that the election would be held on 21 May 2021. Indepth research? Not at all. Just a gut feeling that Morrison’s approval surge during COVID would not last and the knowledge that bullies are also cowards: Morrison would go the full term, buoyed in his own mind, supported by the mainstream media, that he was a master marketer/campaigner who could win elections whenever he wanted to call them.

I’m having trouble locating it, but I only put in writing once, at the start of the campaign, what I thought that election result would be and from memory I put Labor at 78 seats, the LNP at a few more than they ended up with and the Greens and independents a few less than they ended up with.

I take no credit at all as arguably Australia’s best known and most inaccurate amateur psephologist for getting the Labor seats tally so close. I really did think Labor would also get about 35 percent primary vote so I also predicted, without a seats tally, that Labor would win easily. They didn’t.

And once again, my seats prediction had no great indepth research to it. I don’t think I even examined all the marginals in great depth. I thought Kristina Keneally would win Fowler; I thought Andrew Constance would win Gilmore. Labor won more than I thought they would in metropolitan Perth; in Queensland they lost a seat and didn’t pick up the handful I thought they would.

In fact, I now appreciate that my patented big swinging dick pendulum needs some fine tuning.
It’s now clear with the increasing vote for third and minor parties and independents that that original first rule of mine – and Barrie’s – of the LNP needing 42 per cent primary to win no longer applies. They can now win with a lot less. As can Labor.

I’m going to have to tighten the balls on my big swinging dick electoral pendulum and make adjustments to its bellend. It might be a little too swollen to pinpoint at-risk seats accurately.

Finally, let’s get back to Newspoll. Over recent years, you have tut-tutted at The Bug for having fun with that pollster.

I gleaned from our most recent social mediocre exchange that your concern over people who take the mickey out of Newspoll rests largely with such satirical attempts that are clever enough to be mistaken for the real thing.

And therein, Kevin, lies the problem. The best satire is that which can be mistaken for the real thing. Besides, do you really think any harm is done if people momentarily get tricked into thinking briefly that findings bodgeyed up for comic relief are real? It’s possible that some mug punters could rush off and waste their hard-earned with the online bookies. More fool them, I suppose.

I really do think, Kevin, that the best solution is for you to lighten up a bit. Newspoll isn’t sacrosanct. By their own admission, their template for the 2019 poll was flawed. Besides, we’re not having fun with domestic violence numbers or road toll statistics.

Anyway, why should we take a polling outfit too seriously if it got the final ALP primary vote more than 3 per cent wrong on election day?

Do you think they’d mind if I sent them some tea leaves to look at and a stick and some freshly killed chook entrails as they try to work out why they got that result so wrong that it rests right on the outer edge of a margin of error?

Or perhaps you as a real psephologist can try to explain to Bug readers how, between that last poll being taken and when voters walked into polling booths, 3 per cent of Labor voters miraculously changed their minds and decided to vote strategically in a handful of seats contested by teal-coloured candidates.

There surely must be a doctorate there for someone who can explain that.

Don Gordon-Brown