Support 100 years of independent journalism.

  1. Politics
  2. Brexit
21 June 2016

Are the Brexiteers losing?

The latest polls appear to show movement to Remain, but the picture is complex.

By Peter Kellner

The headline on my blog three weeks ago applies once more: “Remain is still ahead. Probably”. The two latest polls both confirm the weekend surveys. Brexit has lost ground in the past week. ORB’s poll for the Telegraph suggests an eight-point shift in the lead (from “Leave” ahead by one point to “Remain” ahead by seven), while the latest YouGov/Times poll finds a five-point shift (Leave’s lead down from seven points to two).


So it does seem that the surge to Leave around ten days ago has been reversed. But is the current position a comfortable “Remain” lead or a narrow “Leave” lead? Here are three reasons why we can’t be absolutely sure.


Any poll, however well conducted, is subject to sampling fluctuations. YouGov has conducted three polls in the past seven days. They have produced a two-point Leave lead, a one-point Remain lead and, now, another two-point leave lead. In a real, millions-of-votes contest, these are very different results. But in a world of polling samples they are statistically much the same. All three polls are consistent with a 50-50 division. We can’t tell whether the move to Remain reported in the weekend polls has continued or stalled.

Sign up for The New Statesman’s newsletters Tick the boxes of the newsletters you would like to receive. A weekly newsletter helping you fit together the pieces of the global economic slowdown. Quick and essential guide to domestic and global politics from the New Statesman's politics team. The New Statesman’s global affairs newsletter, every Monday and Friday. The best of the New Statesman, delivered to your inbox every weekday morning. The New Statesman’s weekly environment email on the politics, business and culture of the climate and nature crises - in your inbox every Thursday. Our weekly culture newsletter – from books and art to pop culture and memes – sent every Friday. A weekly round-up of some of the best articles featured in the most recent issue of the New Statesman, sent each Saturday. A newsletter showcasing the finest writing from the ideas section and the NS archive, covering political ideas, philosophy, criticism and intellectual history - sent every Wednesday. Sign up to receive information regarding NS events, subscription offers & product updates.


Content from our partners
“I learn something new on every trip"
How data can help revive our high streets in the age of online shopping
Why digital inclusion is a vital piece of levelling up

ORB’s figures have thrown up something curious. They poll 800 people in all, but they apply a strong turnout filter. They base their headline figures on the 500 or so who say they will definitely vote. It’s a number that increases the risk of sampling error. This is the basis on which ORB finds a shift from Leave to Remain.

If we take the views of the 700-plus people in the full sample who takes sides, then a different story emerges. Last week Remain enjoyed a five point lead; this has now fallen to two points. The real change has been the way ORB’s turnout filter works. These are the proportions of people on both sides who say they are certain to vote. Remain: last week 60 per cent, this week 69 per cent. Leave: last week 68 per cent, this week 64 per cent. Suddenly, “remain” voters now tell ORB they are more determined to vote than “leave” voters.

It’s possible that this is a true and important finding, for until now virtually every poll has found that “remain” voters are more reluctant to make the journey to their polling station. If that has now changed, then the “leave” campaign is in real trouble.

However, it maybe that this is a sampling fluke. If so, ORB’s new headline figures may be overstating Remain’s lead. No such transformation in the turnout pattern is visible in any other poll, including today’s YouGov survey.


That said, the difference between online and telephone polls does seem to have reappeared; and it’s a matter of judgement which system is more accurate. My initial blog explored this at a time when the difference was a chasm, with online polls saying the race was neck-and-neck, while telephone polls generally reported large Remain leads. Then the differences seemed to disappear, with both methods throwing up results ranging from narrow Remain leads to large Leave leads.

Now, it looks as if the difference is back. Fourteen polls have been conducted in the past ten days: seven each by phone and online. The phone surveys divide: four Remain leads, three Leave leads. The online polls divide: one Remain lead, one Level-pegging, five Leave leads. The single online Remain lead has been one per cent; three phone surveys have produced Remain leads ranging from 3-7 per cent.


For the reasons I gave when I looked at the online-versus-telephone controversy, my hunch is that Remain entered the final three days of this campaign just ahead. But it can’t be more than a hunch. On Thursday morning I shall assess the final polls and predict the result. Probably.

This blog originally appeared on Peter Kellner’s blog.