Poll chasing doesn't work

There is a not insignificant number of party consultants for centrist and “progressive” political parties who believe electoral savviness consists of broadly polling the electorate on a handful of non-economic issues (many of which are decided not out of organic interest but by cable news and Facebook memes), and then adopting one or several positions that appeal, based on polls, to a majority of voters (often more than 50 percent)—even if those positions are not ostensibly progressive.

These are the same types who write op-eds in the Atlantic and the New Statesman, arguing that liberal political parties are losing elections because they are “too woke.” The thinking here is almost stultifyingly simple: to win, nominally economically progressive political parties need only adopt one or two popular right-wing positions, even if those positions do immense harm to a small subset of vulnerable people. That’s how you “broaden your base.”

We’re seeing the results in the UK, where Prime Minister Keir Starmer has recently declared that “trans women aren’t women,” in response to a recent UK supreme court ruling more or less agreeing.

The problem is that it doesn’t seem to work (Starmer’s favourability is -17 points). This approach to politics makes the classic mistake of assuming polls reflect an ideologically coherent and consistent electorate.

Most importantly, most poll results on these cultural wedge issues are rarely the result of earnest self-reflection on the part of respondents, but rather from what these people hear from others around them, whether it’s their friends, the man on the radio, or the woman on TV. That is, people who have worked to convince people to adopt these positions, exactly the role politicians are supposed to play.

Politicians ideally are supposed to lead, they are supposed to make clear, impassioned and morally compelling arguments for their views, even if some don’t poll as popularly as others. No one wanted the New Deal before FDR proposed it. Many Americans would have been fine with a compromise on slavery before Lincoln refused it in favour of a bloody, years-long civil war. Imagine had either president had based their policies on poll results.

In fact, basing a platform entirely on polling violates Goodhart’s Law, which states that as soon as a measure becomes a target, it ceases to be an effective measure. Polls are a snapshot of broad opinions at every given time; not all positions are as fixed as others. It is a politicians job to move these needles in favour of what will serve the public good, not to chase the needles to win elections.