MRP: Starmer wins ‘best prime minister’ in 389 seats to Sunak’s 127

Patrick EnglishDirector of Political Analytics
October 24, 2022, 4:23 PM GMT+0

The new Tory leader isn’t seen as a better prime minister than his Labour rival in any of the ‘Red Wall’ constituencies

Rishi Sunak faces an uphill battle to restore public confidence in the Conservative party and in himself, new YouGov modelling shows.

Based on responses collected from 12,000 people over the weekend, the public tend to believe that Keir Starmer would make a better prime minister than Rishi Sunak.

MRP modelling is most often associated with estimating constituency vote shares, but it can be used to model all manner of public opinions and preferences – including their preference for prime minister.

We asked, “Which of the following do you think would make the best prime minister?”, with respondents offered choices between Keir Starmer, Rishi Sunak, and ‘not sure’.

Across the full modelled sample, 38% said Starmer would make the best leader, 29% thought Sunak would do a better job, while 32% were not sure. The latter figure highlights how despite a commanding headline lead for Keir Starmer, there are still large amounts of uncertainty and apathy about these two candidates to lead Britain at this moment in time.

This near 10-point lead for Starmer breaks down particularly worryingly for the new prime minister and his party. Across 389 constituencies, Starmer leads Sunak. The soon-to-be prime minister wins in just 127 constituencies, with ‘not sure’ prevailing in almost as many (116).

Rishi Sunak does not triumph in a single one of the 50 constituencies that form part of YouGov’s ‘Red Wall’ definition. Starmer wins in 31 of the 50, including Blyth Valley (+10), Burnley (+12), and Wrexham (+9). In the other 19, ‘not sure’ is the most common response.

Similarly, Sunak starts his premiership in a bad position relative to Starmer in the ‘Blue Wall’, winning in just 10. While Sunak just about leads in Harrow East (+4) and Ruislip, Northwood, and Pinner (+3), Starmer is the victor in 43 of the 53 seats, including strong leads in ‘true blue’ constituencies such as South Cambridgeshire (+13), Winchester (+9), and Wycombe (+8).

Support for Starmer over Sunak is particularly strong in both Scotland and Wales, with the new Tory leader failing to win a single seat north of the border and in just one in England’s westerly neighbour.

There is further worry for Sunak in those seats where ‘not sure’ leads, as well. Of these 116 constituencies, 49 prefer Starmer by a margin of over 5 points – including 23 that go for the Labour leader by a margin of 10+ points. Only seven back Sunak over Starmer by 5 or more points.

YouGov interviewed over 12,000 voters across the country between 21st and 23rd of October 2022. Constituency-level forecasts were estimated using the same method which correctly predicted the 2017 and 2019 General Elections. MRP constituency forecast models first estimate the relationship between a wide variety of characteristics about prospective voters and their opinions – in this case, ‘who would make the best prime minister – in a ‘multilevel model’. It then uses data at the constituency level to predict the outcomes of seats based on the concentration of various different types of people who live there, according to what the multilevel model says about their probability of believing either Starmer or Sunak would make a better prime minister (this is the ‘post-stratification’ component). In this instance, 1000 draws from the posterior distribution of the multilevel model were used to predict the constituency probabilities, which ran for 8000 iterations across four parallel chains. The precise multilevel model equation for GB constituencies has been benchmarked to correctly estimate the 2017 and 2019 General Elections to within a handful of seats of each party’s actual performance in those elections.

Explore more data & articles