Mysticism and how people think the world 'must make sense': a brutal take-down of the failures of modern polling

A modern-day druid at Stonehenge. Photo: Daniel Berehulak/Getty Images

You’ll probably recall that one of Malcolm Turnbull’s key justifications for challenging Tony Abbott as prime minister was that the Coalition had lost “30 Newspolls in a row”.

Politicians roll out all sorts of cliches about opinion polls: that they’re only a snapshot of opinion at a particular point in time; the only poll that matters is the one on election day; rather than paying attention to them they’re getting on with the job of making policy.

Based on the polls, nobody believed Donald Trump would become US president.

Based on the polls, nobody believed Britain would vote to leave the EU.

Even though the polls were tight in Australia, only a few believed Malcolm Turnbull would be returned with anything other than a comfortable majority in the House of Representatives, and yet there he is, in power by a single seat that was only finally resolved days after Australians voted.

In Western Australia, polls had Pauline Hanson’s One Nation running at 9-10% support. Its primary vote in the lower house came in at just under 5%. (In the Legislative Council, it registered 8.15% — a much stronger result, but still below what the polls had indicated.)

In this environment there are plenty of dark opinions about the profession of opinion polling, particularly the type that involves calling or approaching people online and asking them which way they plan to vote. And yet as each poll is issued, sage commentaries pour forth on the implications and reasons for each point move.

And they are becoming incredibly frequent. As Coalition strategist Mark Textor, in a column for Business Insider last year, noted:

Data obtained from Emeritus Professor Murray Goot of Macquarie University, show that in the UK, from the dissolution of parliament to election day there was a remarkable 3.5 polls per day published and force fed to voters. The same analysis shows that in Australia in 2013, despite having a significantly smaller voting population, there was an equally remarkable 3.2 polls per day from the proroguing of Parliament to election day.

Whilst this number includes some state polling and a flurry of marginal seat polls published towards the end of the campaign, the frequency is still remarkable….

Hundreds of polls will have been published since the last election and by the end of this campaign. And will the public be any wiser because of these? No, because like the process that produced foie gras, it’s the poor geese that get covered in shit. As my business partner points out: “If you think a campaign should be about ideas and communication with voters to give them a sense of empowerment and understanding of issues, then I think we really had to question the role that they [the polls] started to play in [campaigns].”

When you have this incredible frequency and focus on published polls it is the polls and their (usually small) vote movements that become the most frequent story rather then the issues.

Campaign professionals know there is limited value in the headline results of the high-frequency published polls. But while it may be a blunt tool, far removed from the deeper analyses of voter concerns that the political parties receive, at the same time it is the only metric publicly available to track the voter “marking” of the major parties and their leaders.

But after what happened in the UK and the US last year, there’s at least a more robust cynicism developing around the polling industry. In this month’s edition of US periodical The Baffler, Sam Kriss has written a brutal critique of some of the problems, and wonders if much deeper questions need to be asked about our obsession with trying to predict the future. Starting with the catastrophic misreading of electoral outcomes in abroad last year, Kriss writes (my emphasis added in bold):

If one pollster had failed to accurately predict one result, you could conduct a fairly simple investigation: Had they chosen their samples incorrectly? Had they asked their questions misleadingly? When every poll gets it wrong, with increasing and alarming frequency, the problem is no longer methodological but metaphysical. There are, of course, some perfectly reasonable explanations. For starters, there are anything-but-surveyable patterns of voter suppression and voter lethargy, together with steady influxes of new, never-before-surveyed voters in the electorate. There’s also the huge methodological difficulty that most polling is traditionally carried out via phone, and large sectors of the electorate are no longer happy to answer the phone when unknown numbers appear on the screen.

But these second-order obstacles aren’t enough to explain the current collapse of poll-driven political certainty. They’re just excuses, even if they’re not untrue. Something about the whole general scheme of polling—the idea that you can predict what millions of undecided voters will do by selecting a small group and then just simply asking them—is out of whack. We need to think seriously about what the strange game of election-watching actually is, in terms of our relation to the future, our power to choose our own outcomes, the large-scale structure of the universe, and the mysteries of fate. And these questions are urgent. Because predictions of the future don’t simply exist in the future, but change the way we act in the present. Because in our future something monstrous is rampaging: it paces hungrily toward us, and we need to know if we’ll be able to spot it in time.

Kriss recounts the practices of divination and voting in the ancient world through the use of casting pebbles and various other rituals. “There’s a single principle running through all this,” he says, “one that the ancients were mostly certain of, but that’s proven harder and harder to accept in the twenty-first century. For any kind of divination to work, from haruspicy to opinion polling, the future has to already exist within the here and now. In other words, the world must make some kind of sense.”

The problem is that the rules of cause and effect in modern politics are changing. The post hoc, ergo propter hoc fallacy is more dangerous than ever (emphasis added in bold):

induction will tell you that effect always follows cause, because it’s always done so in the past and we have no reason to think it’ll stop doing so in the future. If what we’re talking about is the sun rising in the morning or billiard balls on a table, you can probably go on thinking like that. But things aren’t always the same. Bertrand Russell uses the example of a chicken who expects the farmer to come and feed it every day, until the day he comes to wring its neck (and possibly, depending on when this farmer lives, spill out its viscera to build a map of the universe). And in politics this expectation is fatal, because in politics nothing is ever the same.

The day before the Brexit referendum, it seemed utterly impossible that the United Kingdom would vote to leave the European Union. The vast majority of polls had consistently predicted a win for Remain, but it wasn’t just that: a lifetime of unconscious induction had convinced us that things would be, more or less, the same forever. Britain might leave the EU, in the same way that your workplace might be obliterated by an asteroid impact, or that you might awake to find yourself transformed into a gigantic insect.

And anyway, people are smarter about polling now, and simply asking someone a question about who they plan to vote for is fraught with risk:

Besides, asking someone how they plan to vote is inherently flawed. Something is lost. Between the intention and the act, between the opinion and the result, between the poll and the prediction, falls the Shadow.

As any psychoanalyst could tell you, the statement “I plan to vote for Donald Trump” has any number of meanings — “I hate my mother,” “I’m afraid of death,” … but none of them can tell you who’s going to win.

In a warning that underlines some of the weakness of establishment politics, including in Australia, he notes:

Politicians, especially, live their lives in fear of the sibyls and soothsayers. They adjust their policies, their messaging, even their personal appearances on the basis of what the polls tell them people want. Court astrologer is still a deeply respectable vocation. It’s hard to imagine a Democratic campaign so torpidly complacent without its whispering crowd of prognosticators. If neither Clinton nor the public had been so convinced that she would win, maybe she would have. It’s possible that in the United Kingdom, Labour’s slow autophagy might be less fatal if the party’s old guard couldn’t point to the mystical charts in every newspaper, echoing their blind insistence that any kind of socialism is electoral poison. We’re not likely to abolish polling… [but] we should be aware that there are always other futures, streaming and effervescent, that the priests of empty, heterogeneous time can’t see.

This week Malcolm Turnbull has seen, finally, something of a bounce in the polls, pulling back to 52-48 in this week’s Newspoll.

He has announced his “nation-building” $2 billion Snowy Hydro project and today moved to amend the Racial Discrimination Act, an issue that’s been a consistent gripe of the conservative wing of his party. At the very least, these make it likely that at least a associated uptick in the polls makes a future with Malcolm Turnbull as leader of the Liberal Party much more likely.

Beyond that, what’s certain is we are in an era of a new-found caution around the signals from published polling. You can read the full, in-depth (and in places very amusing) piece here at The Baffler.

Business Insider Emails & Alerts

Site highlights each day to your inbox.

Follow Business Insider Australia on Facebook, Twitter, LinkedIn, and Instagram.