Beware the phantom swings: Why dramatic bounces in the polls aren’t always what they seem

Beware the phantom swings: Why dramatic bounces in the polls aren’t always what they seem

Recently Nate Silver asked us why our polls don’t bounce around much. In our polling, Clinton had a small lead in September which expanded to five or six points after the first Presidential debate on September 26. Since then a lot has happened – sex tapes, election rigging, WikiLeaks – but our numbers have budged only slightly. Over the past three weeks, our election model and polling for The Economist has shown a consistent lead for Hillary Clinton over Donald Trump of three to five percent. In contrast, some other polls have shown wide swings. For example, the ABC/Washington Post poll had a Clinton lead of two points on September 22, rising to 12 points on October 22-23, and falling back to a single point yesterday.

We believe that most of the bounces seen in surveys this year represent sampling noise that can be reduced or eliminated by adopting by better statistical methodology. We risk a repetition of 2012 where polling swings were largely statistical mirages. The convention and first debate bounces in 2012 were mostly the consequence of transitory variations in response rates. Fewer voters were changing their minds than were changing their inclination to respond to surveys.

Continue reading →


Commenting Policy

We have no tolerance for comments containing violence, racism, vulgarity, profanity, all caps, or discourteous behavior. Thank you for partnering with us to maintain a courteous and useful public environment where we can engage in reasonable discourse.

You may use HTML in your comments. Feel free to review the full list of allowed HTML here.