Topline summary of General Election polling assessments

Sir John began with a presentation on what the polls are currently reporting, focusing on Labour’s consistent lead over the Conservatives, and the increase in figures seen by Reform UK since 2019. The sentiment in the room, as well as in the polling explained by Sir John, is that this election is one driven by how badly the Conservatives are losing, rather than how much the Labour Party is winning.

Sir John detailed the ways in which polling can be a useful tool, MRP polls in particular, when looking at geographical shifts in voting, especially relevant for the Conservatives in this scenario. However, he was clear that polling reliability should not be overstated as there are some factors that can impact the outcome, such as the data exclusion of those who do not know how they will vote, or the assumptions made about their voting intentions based on historic polling.

Here is a summary of the key points from Professor Curtice’s talk:

Key points

• Polling is useful in an election campaign for a variety of reasons, notably to analyse change, separate the “knows” from the “don’t knows”, and to measure sentiment across the country. However, their reliability should not be overstated – polls should be “taken and not inhaled”.
• The “don’t knows” provide a certain level of uncertainty in the campaign polling, which is likely to last through to 4 July.
• Two key reasons driving the demise in polling figures for the Conservatives are Boris Johnson’s untenable relationship with the truth following the Covid inquiry and Liz Truss’s short-lived economic policy.
• While initially polls showed ex-Conservative voters tended to sway to Labour, recent trends indicate that these figures have gone down, as more instead are now shifting to Reform.
• Age is the biggest differentiator in terms of voting demographic in our society.
• Geography will play an important part in shifting votes away from the Conservatives in this general election:
  • You Gov’s MRP poll showed a predicted 140 seats for the Conservatives and 422 for Labour, whereas the Survation MRP poll showed a tougher election day for the Conservatives, putting them at 80 seats, and Labour at 443.
  • The conclusion one can draw from the Survation poll as opposed to YouGov’s poll is that the Conservative vote loss could be much higher in seats that are deemed “safe”, rather than marginal.
• MRP polls can of course still shift, but they are reliable insofar as we have MRP data from other elections to direct them.
• This aligns with polling in the local elections, where the Conservatives fell in many historically safe seats.
• One key issue for the Conservatives is that Reform has increased their polling position by approximately 10% and the Conservatives have done little in terms of producing a strategy to combat this threat from the right.
• Are we really facing an unprecedented election in terms of result and campaign?
  • The campaign trail itself is not unprecedented.
  • The level of Tory support we come to see on polling day could be unprecedented.
  • The share of votes won by the two largest parties may set a new record low.
  • We may end up with five UK-wide parties in Parliament, each with at least five seats.

‍