Forecast update for the Historical Referendums and Polls based method

by Stephen Fisher and Alan Renwick.

We offer our condolences to the family and friends of Jo Cox MP. We hope that it is not disrespectful of us to post this updated forecast now that the official campaigns have resumed.

Our polling average now has Remain at 49% after setting aside Don’t Knows.

From this we forecast Remain to get 50% of the vote.

The 95% prediction interval is only a little narrower than ±12 points. So Remain are forecast to win between 39% and 62% of the vote.

The probability that Remain will win the referendum is now 52%.

The methods behind our forecast

The method behind this forecast is based on the historical experience of referendum polls and referendum outcomes in the UK and on the EU elsewhere, as discussed here.

Our polling average is constructed by taking the most recent poll from each company within the last two weeks. If a company uses both phone and online modes then both the most recent phone poll and most recent online poll are used. This applies to BMG, ICM and ORB this week. The current average is based on the results of twelve polls from nine companies, of which six were conducted by phone and six online. All polls are adjusted to account for the tendency for phone polls to be more favourable to Remain. This is done by adding 1.75 to the Remain share for online polls and subtracting the same amount for phone polls.

Advertisements

Updated combined EU Referendum forecast

Stephen Fisher and Rosalind Shorrocks.

This week’s forecast shows a sharp drop in the probability of a Remain win to 60.6%, down from 67.7% last week. Similarly the forecast share of the vote for Remain has dropped from 53.8% to 52.7%. For the first time, our poll of polls of polls has Leave ahead, and more of the polls under consideration (see below for details) have had Leave ahead than have had Remain ahead. There has been little or no change in the expert forecasts and non-poll based models because of little or no new data. However there have been substantial changes in the betting and prediction markets, largely in response to the opinion polls. Intriguingly the volunteer forecasters at the Good Judgement Project now 72% chance to Remain, more than 10 points higher than the corresponding figure from the prediction markets: the biggest gap between these sources we’ve observed so far.

Remain % share Leave % share Probability Remain wins
Betting markets 51.5 48.5 60.7
Prediction markets 60.1
Citizen forecasts  52.0  48.0 64.4
Expert forecasts 55.1 44.9  62.0
Volunteer forecasts 54.2 45.8 72.2
Polls 49.6 50.4 45.7
Poll based models 51.0 49.0 59.0
Non-poll based models 55.6 44.4
Combined forecast (mean) 52.7 47.3 60.6

(Individual forecasts collected on 14th June 2016.)

METHODOLOGY Continue reading Updated combined EU Referendum forecast

A 50:50 forecast from the Historical Referendums and Polls based method

by Stephen Fisher and Alan Renwick.

Our forecast has taken a dramatic turn. Last week our polling average had Remain at 51% after setting aside Don’t Knows. It has this week dropped a further two points to 49%. This means Leave is ahead in our polling average for the first time, with 51%.

The forecast share of the vote for Remain has correspondingly dropped from 53% to slightly over 50%.

The 95% prediction interval is still ±12 points. So we are now forecasting that both Leave and Remain will win between 38% and 62% of the vote.

The probability that Remain will win the referendum has fallen from 68% last week to just 51% this week. Continue reading A 50:50 forecast from the Historical Referendums and Polls based method

Combined EU Referendum forecast update

Stephen Fisher and Rosalind Shorrocks.

This week’s forecast shows another drop in the probability of a Remain win, this time to 67.7%. Part of the reason for the drop is the introduction of significant new data sources and methodological changes (detailed below). But it is also true that all of the component methods, including those without methodological changes, have seen drops in the probability of a Remain vote. By contrast the forecast share of the vote for Remain continues to hover around 54% as in previous forecasts.

Remain % share Leave % share Probability Remain wins
Betting markets 53.2 46.8 72
Prediction markets 70.6
Citizen forecasts 64.3
Expert forecasts 55.1 44.9  62.0
Volunteer forecasts 54.2 45.8 74.2
Polls 51.1 48.9 59.1
Poll based models 53.5 46.5 72
Non-poll based models 55.6 44.4
Combined forecast (mean) 53.8 46.2 67.7

(Individual forecasts collected on 8th June 2016.)

METHODOLOGY Continue reading Combined EU Referendum forecast update

Historical Referendums and Polls based Forecast Update

by Stephen Fisher and Alan Renwick.

Remain have continued their gentle slide in the polls. Last week our polling average saw Remain drop from 53% to 52%. Now they are on 51% after setting aside Don’t Knows.

The further one-point drop in our polling average has produced a one-point drop in the forecast share of the vote for Remain, from 54% to 53%.

The 95% prediction interval is still ±12 points. We are now forecasting that Remain will win between 40% and 65% of the vote.

The probability that Remain will win the referendum is now down to 68%.

The method behind this forecast is based on the historical experience of referendum polls and referendum outcomes in the UK and on the EU elsewhere, as discussed here.

Our polling average is constructed by taking the most recent poll from each company within the last two weeks. If a company uses both phone and online modes then both the most recent phone poll and most recent online poll are used. This applies just to ICM this week. The current average is based on the results of eight polls from seven companies, of which three were conducted by phone and five online. All polls are adjusted to account for the tendency for phone polls to be more favourable to Remain. This is done by adding 2.2 to the Remain share for online polls and subtracting the same amount for phone polls.

Updated combined EU Referendum forecast

Stephen Fisher and Rosalind Shorrocks.

This week’s forecast shows the probability of a Remain win reduced to 71.9%, down from 76.1% last week. Once again the forecast share of the vote for Remain, at 53.7%, is similar to all our previous combined forecasts.

Remain % share Leave % share Probability Remain wins
Betting markets 54.1 45.9 73.7
Prediction markets 73.8
Citizen forecasts 68.4
Expert forecasts 56.0 44.0
Volunteer forecasts 54.3 45.7 76.1
Polls 51.4 48.6 63.6
Poll based models 54.5 45.5 76.0
Non-poll based models 51.8 48.2
Combined forecast (mean) 53.7 46.3 71.9

(Individual forecasts collected in the morning of 1st June 2016.) Continue reading Updated combined EU Referendum forecast

Forecast update for the Historical Referendums and Polls based Method

by Stephen Fisher and Alan Renwick.

The small shift towards Remain in the polls that we observed last week has been reversed. Setting aside Don’t Knows, our polling average for Remain has dropped back from 53% to 52%. Despite there being little difference between the headline figures for yesterday’s ICM phone and online polls, our estimate (and corresponding adjustment) for the typical difference between the two modes of interviewing has barely changed.

The one-point drop in our polling average has led to a corresponding one-point drop in the forecast share of the vote for Remain, from 55% to 54%. The 95% prediction interval surrounding this estimate has again narrowed very slightly to ±12 points. So we are now forecasting that Remain will win between 42% and 66% of the vote.

Overall the probability that the Remain vote will be larger than the Leave vote has dropped from 79% last week to 73% now. Continue reading Forecast update for the Historical Referendums and Polls based Method