Ipsos MRP has the Tories on 115 seats
It says a lot about the pickle the Tories find themselves in that a MRP showing the Tories on 115 seats feels like good news for them. Beyond Topline makes the astute observation above on why this MRP is better for the Tories than some other MRPs.
I spoke to a political analyst a couple of days ago who said MRPs will always throw out a few constituency results which seem logic defying (Canterbury going Labour in 2017 anyone?) but not to focus on those but the quality of the work put in beforehand.
The Guardian notes
According to the projection, Nigel Farage is on track to overturn a huge Tory majority to win in Clacton while Jeremy Corbyn, standing as an independent, is predicted to lose to Labour in Islington North.
Labour has an implied vote share of 43%, with Rishi Sunak’s Tories on 25%, Reform UK on 12%, the Lib Dems on 10%, the Greens on 6%, the SNP on 3% and Plaid Cymru on 1%.
The model finds 117 seats, the majority of which were won by the Tories in 2019, are now considered “too close to call” as they have a winning margin of less than five percentage points. They include 56 where the Tories are marginally ahead and 48 where Labour is slightly in the lead.
It underlines the extent to which a relatively small number of undecided voters or switchers to smaller parties, or Labour, could alter the outcome for the Tories. With millions of voters still undecided, the party will be hoping to claw back some support by polling day…
…High-profile Tories at risk of losing their seats include Penny Mordaunt, the Commons leader; Gillian Keegan, the education secretary; Grant Shapps, the defence secretary, and Jacob Rees-Mogg.
However, Jeremy Hunt, the chancellor, is predicted to cling on in Godalming and Ash, while Sunak’s Richmond and Northallerton seat is a Conservative hold but on a much-reduced majority, 40% to Labour’s 33%.
What this MRP shows a very different result for the SNP than the Survation one which makes betting awkward. If this MRP is accurate then Lib Dems are a sell on the spreads but that isn’t something I am advocating at the moment.
TSE