Third combined forecast for the 2019 general election

By Stephen Fisher, John Kenny and Rosalind Shorrocks

Once again all three sources of seat forecasts suggest the Conservatives are heading to a comfortable majority, while Labour are on course for a result on par with their previous post-war low of 209 seats in 1983. The Liberal Democrat forecast has been dropping steadily, so that they are now expected to end up with only a few more MPs than the twenty they had when they chose to support the election.

Seats	Betting Markets	Complex models	Simple models	Average
Con	346	354	363	354
Lab	209	211	200	206
LD	30	22	22	25
Brexit		0	0	0
Green	2	1	1	1
SNP	46	46	44	45
PC	4	4	4	4
Con majority	42	58	75	58

The average forecast for Conservative seats has increased only slightly since our first forecast. That is in line with a slight increase in the Conservative lead over Labour, and a slight decline in the forecast Liberal Democrat share of the vote. Relative to forecasts from other sources, some betting markets are slightly more bullish for the Lib Dems on both seats and shares, but they remain more cautious about the size of a Conservative majority without being much different from other forecasts regarding the Tory vote share lead (traditionally the key predictor of seats outcomes). Both main parties have risen in the polls and in the forecasts over the past fortnight, but the Tories have gone up by slightly more than Labour.

% GB Vote	Poll aggregates	Betting markets	Models	Average
Con	40.5	41.5	43.5	41.8
Lab	28.8	30.4	30.1	29.7
Lib Dem	14.9	16.2	13.7	14.3
Brexit	6.6	3.8	4.1	4.8
Green	3.1		3.2	3.1
SNP	3.5		3.5	3.5
PC	0.5		0.5	0.5
Con-Lab lead	11.7	11.2	13.4	12.1

Con-Lab lead 20 Nov

In line with the strengthening forecast for the Conservative seat tally, most of the estimates for the probability of a Conservative majority have risen, so there is now about a two-thirds chance in our combined estimate. Citizens, however, remain unconvinced, with fewer than a third expecting a Tory majority. Intriguingly, thanks to a new poll from Lord Ashcroft, it is clear that the public at large do not appreciate (or are not aware of) the need for Boris Johnson to win a majority (or something very close to it) if he is to continue as prime minister. Commentators largely agree that the Tories have run out of potential coalition partners or backers for a minority administration, having upset the Democratic Unionist Party (DUP). In the Ashcroft poll fully 80% expect Boris Johnson to continue as PM but only 30% expect him to win a majority. So the numbers need to be interpreted with caution.

Probabilities	Betting markets	Models	Polls	Citizen forecast	Volunteered	Average
Conservative Majority	0.68	0.81	0.89	0.29	0.60	0.65
Conservative landslide			0.22			0.22
Boris Johnson next PM	0.79			0.80		0.80
Jeremy Corbyn next PM	0.23			0.17		0.20

Con probabilities Con landslide 20 Nov

For the table and graph above we have changed the way we estimate the probabilities from opinion polls, and revised previous estimates. Instead of using just the two-party swing, we are doing a full uniform change projection for each of the most recent polls from each pollster in the last week. For these we also take estimates for the SNP and Plaid Cymru shares from the most recent polls in Scotland and Wales. Unfortunately, the most recent polls in those countries are already almost a month old. This is extraordinary given that both have disproportionately more marginal constituencies that appear to be vulnerable, Scotland especially, and both have their own devolved politics that mean changes in vote shares since the last election might differ considerably from those in England.

Methodology

The idea of combining forecasts from different sources has a good track record, though it has to be admitted that our attempts for the 2017 general election and the 2016 Brexit referendum did not work out well. See here for a review of our 2017 forecasts. Also worth noting is the experience of the pollyvote.com combined forecast of the US presidential elections.

The basic approach is to combine forecasts by averaging them within each category and then average across categories. Since the different sources do not all present similar figures that can be averaged on a like for like basis there are various judgement calls we have had to make on how to treat the data.

The main changes in method this week are: New models are included from Lebo and Fisher as well as from English and Bailey. Survation is excluded from the citizen forecasts as they do not have one from the previous two weeks, and in fact this week only two citizen forecasts are included: Deltapoll 14^th-16^thNov and Opinium 6^th-8^thNov. We have also changed how we calculate the probabilities of a Conservative majority and a Conservative landslide from the polls, using a full uniform change projection as described above.

We have also made an adjustment to our re-percentaging of the overall probabilities from the betting markets which affects our previous estimates for the Conservative Party – and is reflected in the graphs above – but the overall trends within the betting markets remain the same notwithstanding this.

We haven’t included the headline seats tally from the Best for Britain/Focaldata MRP model mainly because we’re not sure they count as official forecasts from the people that generated them. The same is true of figures from other tactical voting websites so far as we know.

Polls

For vote shares, we use the various available polling averages. There are various different polling averages. Some admittedly are quite sophisticated, allowing for house effects, but they are nonetheless estimates of current public opinion and not future votes. We only use polling averages that have been updated in the last 7 days.

To estimate a pseudo-probability of a Conservative majority from the polls, we use uniform change projections for each of the most recent polls from each pollster in the last week. For these we also take estimates for the SNP and Plaid Cymru shares from the most recent polls in Scotland and Wales. The proportion of these projections with a Conservative majority is taken as a pseudo-probability, and similarly we consider the proportion pointing to a majority of over 100 as a pseudo-probability of a landslide.

The pollsters vary considerably in the lead they attribute to the Conservatives. Since nearly all the pollsters are doing polls at least once a week and since we take just the most recent poll from each pollster, our calculated probabilities of a Conservative majority from the polls should not be, and do not seem to be, much affected by any changes in the composition of our poll lead average.

Statistical models

There are just a few statistical forecasting models this year (but hopefully more to come). We have divided them between simple (poll average plus uniform swing seats projection) and complex (anything more elaborate but not necessarily particularly complex). Within these categories we simply average the available estimates of seats and shares. We have not excluded any models based on quality, but they do have to be statistical models as opposed to personal guesses. We only include models that have been updated in the last 7 days.

Betting markets

For the seats forecasts we are using just the mid points of the spread betting for the markets. Note that the markets might imply fewer or more seats forecast for the main parties than there are in Britain. This is because the markets are separate for each party and do not need to be consistent collectively.

Betting markets for shares of the vote are typically done in bands. We take the mid points of the (up to) three bands with the highest implied probabilities weighted by their implied probabilities from the betting odds, averaging across bookies.

Citizen forecasts

Some polls ask people what they think that the outcome will be on December 12^th. We call these ‘citizen forecasts.’ Different pollsters use different survey questions but they can be combined to generate pseudo-probabilities. We use the proportion of poll respondents who think there will be a Conservative majority, excluding don’t knows and re-percentaging, as the pseudo-probability of a Conservative majority. We similarly use the proportion of poll respondents who think that Boris Johnson will become the next Prime Minister, and similarly generate a pseudo-probability for this outcome excluding the don’t knows. We take the most recent such poll from each pollster in the previous two weeks, although it should be noted that only a few pollsters ask such questions.

In our first 2019 forecast on 6^thNovember, we did not exclude the don’t know responses when calculating these pseudo-probabilities, and so the probabilities of Conservative majority/Johnson next PM are artificially lower in the first forecast compared to subsequent forecasts. If we had excluded the don’t knows on 6^thNov, the probability of a Conservative majority from the citizen forecasts would have been 0.29 (compared to 0.22) and the probability of Boris Johnson as the next Prime Minister would have been 0.66 (compared to 0.47).

Volunteered forecasts

These come from the Good Judgement Project, which encourages people to forecast the outcomes of various events in order to develop and improve their forecasting skills. The people providing their predictions to the Good Judgement Project are not necessarily experts, but nor can they be seen as a representative sample, and so we report their predictions separately as ‘volunteered’ forecasts.

Estimates come from morning on 20^thNovember 2019. All of the polling data used was conducted before last night’s leadership debate.

Sources

Citizen forecasts:

Deltapoll

Opinium

Lord Ashcroft

Forecasting models:
Electoral Calculus

Englishand Bailey