Australian Liberals Blame Pollsters For Their Lack Of Vision
Bad polling isn’t the problem—policy without vision is.
In the recent Australian federal elections, the Liberal party headed by Peter Dutton suffered a devastating and humiliating defeat, with Mr. Dutton being the first federal opposition leader to lose his seat.
In the aftermath, the Liberals are now trying to process the blow and find a way forward. Unfortunately, it seems that rather than soul searching, they’re mostly engaged in a blame game.
As Sarah Elks reports on The Australian:
Another senior Liberal source said the internal polling by Freshwater Strategy was “way out” and had given Mr Dutton’s team, strategists and MPs flawed information throughout the campaign.
“The polling was a huge problem … it was way out,” the senior Liberal told The Australian.
“If we’d have known (the truth) we would have changed course overnight.”
The problem isn’t flawed polling—it’s that polling had such a central role in the first place!
Blaming pollsters for the campaign’s direction reveals a deeper issue: lack of vision. And maybe that’s why many Australians didn’t trust Dutton and his team with their vote.
Vision is a type of explanation—a coherent plan for how to bring about a desired future, grounded in values and logic
David Deutsch highlights the importance of explanations in politics in chapter 13 of The Beginning of Infinity:
In reality, the voter is choosing between explanations, not checkboxes, and, while very few voters choose to affect the checkboxes themselves, by running for office, all rational voters create their own explanation for which checkbox they personally should choose.
And later in the same chapter:
In particular, what voters are doing in elections is not synthesizing a decision of a superhuman being, ‘Society’. They are choosing which experiments are to be attempted next, and (principally) which are to be abandoned because there is no longer a good explanation for why they are best. The politicians, and their policies, are those experiments.
A political party should propose experiments, in the form of policies, each aimed at testing an explanation for how to improve the country, state, or municipality.
With that foundation, polls become almost irrelevant. Poll results cannot alter the explanations behind the proposed policies. They can only indicate how many people in the sample agree with the explanation. Good explanations don’t depend on public opinion—they either work or they don’t.
Blaming polls for the result of a bad campaign shows a deep lack of underlying explanations, or at least of integrity to stand by them.
And yes, I’m oversimplifying politics and drawing a strong conclusion from a snippet of reported conversation. But the goal isn’t to scold the Liberal Party; it’s to stress the difference between using a vision to interpret data and using data to invent a vision.
When tech companies over-rely on A/B tests to drive feature development, they risk ending up with a product optimized for short-term gain but lacking a strategy for long-term survival. Likewise, a political party that develops policy starting from polling data will eventually have to reckon with a majority of voters seeing through their lack of vision and voting for someone else.
The lesson from the Liberals’ collapse is that thinking cannot be delegated to data crunchers, be they polling agencies or large language models. Feedback from data is only useful if you have an explanation to interpret it, so you can decide what is signal and what is noise.