More in Common's MRP model released today predicts Labour will win 406 seats in the General Election on 4 July - a majority of 162, while the Conservatives are expected to hold just 155 seats.
Luke Tryl, Executive Director of More in Common UK, said:
"The fact that this projection showing the Conservatives barely holding 150 seats is one of the most favourable to the Conservatives shows how deep a hole the party finds itself in - with barely two weeks to go for them to change the dial. Far from the narrowing in the polls many expected to see by now the Conservatives position instead appears to be getting worse and only a small move away from them could see them reduced to 107 seats.
Labour on the other hand look set to inherit a historic majority while still remaining largely undefined in the eyes of the electorate. While creating such a broad electoral Coalition, that will span from Blue Wall Worthing to Blyth in the Red Wall is a good problem to have in the short term, it points to potential difficulties in creating a governing agenda that unites such disparate tribes - especially when electoral cynicism is so high."
Key findings
- Labour to have a majority of 162 gaining over 200 seats across England, Scotland and Wales - More in Common’s MRP projects Labour will win between 395 and 416 seats.
- Conservatives facing their lowest seat total in modern history rivalling their historic 1906 defeats as they are expected to hold just 155 seats. Conservatives facing wipeout in Wales and losses in every nation of Great Britain.
- Liberal Democrats to recover to pre-Coalition levels and overtake the SNP as the third largest party in the Commons. The model predicts the Liberal Democrat’s will increase their seat total four-fold from 2019, recovering to the level of seats they held in the 2000s before entering the coalition.
- The SNP’s long reign of dominance in Scotland is set to end with their projected seat total reduced to 18 seats, and Labour set to become the biggest party in Scotland with 33 seats.
- 112 seats to have a majority of less than 5%
- Five cabinet ministers projected to lose their seats
The model is based on voting intention data collected between 22nd May 2024 and 17th June 2024 from 10,850 adults in Great Britain.
Labour set to gain more than 200 seats
Based on this model, Labour would receive 44% of the total vote share and 62% of the 650 Parliamentary constituencies in the UK, while the Conservatives would win 24% of seats with 28% of the vote. The Liberal Democrats are set to reach a total of 49 seats, while the SNP are set to be reduced to 18 seats in Scotland.
Big Names at Risk
The MRP projects the following Cabinet Ministers are set to lose their seats:
- Chancellor of the Exchequer Jeremy Hunt would lose Godalming and Ash to the Liberal Democrats
- Grant Shapps, Defence Secretary would lose Welwyn Hatfield to Labour
- Alex Chalk, Justice Secretary would lose Cheltenham to the Liberal Democrats
- Mark Harper, Transport Secretary would lose Forest of Dean to Labour
- David Davies, Welsh Secretary would lose Monmouthshire to Labour.
Over a hundred seats currently too close to call
In 112 seats, the second candidate’s vote share is 5 points or less behind the winner. Accounting for uncertainty within the model, these races can be thought of as currently too close to call. Of these marginal seats, in 96 of them the Conservatives are in first or second place. In the extreme case that the Conservatives win all these seats, they would be on 203 seats, whereas if they lose each of these they would be down to 107 seats.
Labour set to be biggest party in Scotland with significant SNP losses
The SNP’s long reign of dominance in Scotland is set to end with their projected seat total reduced to 18 seats, and Labour set to become the biggest party in Scotland with 33 seats. 20 seats in Scotland are within a 5% margin.
Our MRP model in England and Wales uses census data from 2021. Scotland's latest census data has not yet been released on a constituency-level, so for some demographic information in Scotland we rely on 2011 data.
Range of outcomes possible, but a Labour majority in each
In 500 simulations of possible General Election outcomes using the results of the MRP, Labour wins between 395 and 416 seats.
The best scenario for the Conservatives sees them winning 165 seats, and the worst scenario gives them 145 seats. While these are possible values and should not be thought of as predictions, they highlight the limited set of possible outcomes facing the Conservative Party.
FAQ
What is an MRP?
‘Multilevel Regression with Post-stratification’ (MRP) uses data from a voting intention poll to model how people will vote based on their demographics, voting behaviour and information about their constituency. These results are then applied to the demographic and electoral makeup of each constituency to make a constituency-level prediction. The model is 'multilevel' because it uses both individual and constituency-level data.
How is this different from your normal voting intention poll?
The voting intention regularly published by More in Common is a national estimate based on a representative sample of at least 2,000 people. It indicates roughly how many people in Great Britain intend to vote for one party or another. This is simple to calculate and allows us to track changes through time.
But if you want to project a national seat count, this isn’t as useful. No political party performs equally well in every seat, because their supporters are not evenly spread across the country. For example, a 70-year-old man who didn’t go to university and lives in a small village has a higher likelihood of voting Conservative than a 25-year-old woman renting a flat in a major city. The benefit of MRP is the ability to use information about the different people who live in every constituency across the country to predict how many people will vote for each party.
How does the model account for those who don't know how they will vote?
When we ask people their voting intention, some people say they don’t know. We push them to say who they would vote for if they were forced to choose, and we use this response as their expected vote. Some people, when asked to imagine that they were forced to choose, still don’t know who they would vote for. Using our MRP model, we’re able to make a better guess at how these “double don’t knows” might end up voting. When training the model to predict people’s voting intention based on their demographics, voting behaviour and information about their constituency, we excluded the responses of people who didn’t know who they would vote for (after the squeeze) from the training data. When we apply the model to all the voters in the constituency, it effectively means we estimate the votes of people who don’t know, according to how people like them (in terms of demographics and past voting behaviour) but who do know, intend to vote. So if someone lives in a rural area, is over 75 and voted Conservative in 2019, the model uses the fact that most over 75s in rural areas who voted Conservative in 2019 and do know who they’ll vote for say they will vote Conservative, to guess that if they do vote it will likely be for the Conservatives.
How does the model account for who will vote?
When we survey the public about who they will vote for, many of the responses come from people who will not vote in the General Election. Since past election turnout is one of the strongest and most consistent predictors of future turnout, we factor in whether people voted in 2019 into our predictions of who will turnout in this election rather than relying solely on self-reported voting likelihood.
Is this a snapshot or a projection?
Some voting intention models attempt to take a snapshot of who the public would vote for if an election was called today. So they don’t factor in things that could affect election results on a future date, such as which parties stand in different seats, who might vote tactically, and who might make up their mind during the campaign. Others take factors like these into account - a projection of the General Election rather than a snapshot of today. This introduces uncertainty and requires making assumptions, but may give a clearer picture of how the public is likely to vote, based on where we are today. That is why in our model we have included historical information - such as who actually voted in the last General Election - as well as current information - such as who people say is their second choice, or whether they’d vote differently if their candidate was unlikely to win. While we can only model people’s responses as they have given them in the last few weeks, we attempt to account for how tactical voting and other dynamics might change behaviour on election day. We will release one more model output before the end of the campaign to provide a final up-to-date prediction.
Is this your final model?
This is our second model and we will have another final model based on fresh data gathered between now and the 4th July. Given that our voting intention methodology has changed, we have aligned this model with our new voting intention methodology since the release of our first MRP model on June 3rd.
How do you deal with smaller parties?
MRPs can be less effective at predicting the seat distribution for smaller parties. This is made even more uncertain by the fact that smaller parties did not stand candidates in every constituency in 2019. In particular, as the Brexit Party withdrew candidates from Conservative seats at the last election, we have less information about voters’ previous preferences and behaviour.
To address this, we have taken into account additional information about historic vote efficiency, local presence and the relationship between opinion polling performance and realised election results. In our model this has the effect of boosting the Liberal Democrat seat total and reducing the implied level of national support for Reform UK.
When creating any standard voting intention or MRP projection, it’s necessary to make subjective choices about which variables to include and what to adjust. Not making adjustments or including relevant information is one such choice. We have instead tried to include fair assumptions to better represent the true picture.
Why is your fieldwork range so large?
Training a model to predict someone’s voting intention requires lots of information about lots of people. More in Common gathers data for our MRP model over a longer period of time. For this model, we’re using data from 22nd May 2024 to 17th June 2024. We do this for two reasons: firstly, we want to ensure that we have balanced data that is not overly influenced by fielding dates. Secondly, we want to capture a broader group of people, including those who might not ordinarily respond to a survey the day it is released.
Why does your model not show Reform winning any seats?
Reform UK’s voter base is spread out more evenly across constituencies, making it harder for them to win as many seats as parties such as the Liberal Democrats, which are on a similar national vote share, but have votes more concentrated in target seats. For example, many of Reform’s best performances in our model are in so-called “Red Wall” seats that will be safely won by Labour, making their vote share particularly inefficient.
In addition, the longer fieldwork dates on this poll might not account for all of the recent rise in Reform’s national vote share. Our final MRP release will likely capture more of this, and will incorporate over-sampling of seats with unique local dynamics, such as Clacton where Nigel Farage is running.
What has changed since the publication of our first MRP?
Since we published our first MRP, we’ve made some methodological adjustments to better capture the changing electoral context including the finalisation of ballot papers and the confirmation of candidates earlier this month. We’ve also made some changes to align with those made to our headline voting intention where we now ask the public to choose who they will vote for based on the candidates who will be on the ballot paper in their constituency come polling day, and in this MRP model we use the same voter turnout methodology as we do in our updated headline voting intention.
Other updates to the model itself include the work we’ve done to examine additional factors that are predictive of voting intention to help improve the accuracy of our estimates.
Our first model included data collected before and after the announcement of the General Election date. This new model includes only data from after the short campaign began, in which we ask respondents a series of tactical voting questions to better understand who might vote tactically and where. As such, we now include data about tactical voting in our model.
MRPs have performed well in recent elections, but current polling suggests record levels of proportionality in swing patterns for some seats, taking these models beyond historic election cycles where MRPs have accurately predicted results. To account for this and to avoid extreme values that go beyond proportional swing, we have developed a proportionality check to adjust our posterior predictions and bring them closer to historical swing patterns.
Why does the model show X party winning in Y constituency?
MRP models are very good at indicating how the parties might perform across different constituencies based on their demographic makeup. But they don’t account for local factors that impact a small number of constituencies, such as a popular incumbent, well known or controversial council policy Therefore it would be a mistake to draw too much from the projected vote share in an individual constituency.