How is the Lib Dem campaign going?
A good starting point to answer that is the big YouGov/MRP election poll, released last week showing the Lib Dem vote share doubling compared with 2017 but the party’s seat tally only going up by one.
Here are five key points about that finding…
1. It’s only one poll and this particular MRP model has only been used in one previous general election. That said, it’s a big poll, the overall voting intentions are in line with other polls and the one time this MRP model was tried before, it got things spectacularly right.
2. The more detailed figures are also generally plausible – such as the Conservatives winning over Leave voters but shedding some Remain voters. How detailed though can MRP get with specific constituencies? Here’s my rule of thumb (which Ben Lauderdale, one of the brains behind this MRP model tweeted his agreement with):
My super-rough rule of thumb is to look at the average sample size in each seat (about 130 in this case); see how many multiples you need to get up to 1,000 (8 in this case); and then concluded that if a factor is present in that many seats then MRP should be able to cope.
— Mark Pack 🔶🎒 (@markpack) November 28, 2019
There’s definitely a difference in what Liberal Democrats think is happening in some areas (backed up by constituency polling results) and what the YouGov/MRP data shows. But given these margins of error for very specific local circumstances in such MRP analysis, it’s possible both that and the constituency polls will turn out to be right.
3. Overall, the picture shows the problem the Liberal Democrats face in starting first-past-the-post elections with such a small core vote compared with other parties. It makes the party far more prone to being squeezed, as I warned about before the election.
The big challenge for the party is to start future elections with a larger core vote.
Even so, the party is on double the level of national support than last time, though this turns into only one extra seat thanks to first-past-the-post.
Interestingly, one point John Curtice made when I did a podcast alongside him last week was that he thought the Liberal Democrat manifesto, or at least the way the party is promoting its contents, does not do enough to emphasis the other distinctively liberal aspects of the party’s position in addition to Brexit.
4. However, in several dozen seats the current vote shares – Conservatives ahead, Liberal Democrats not that far behind, large Labour vote still in play – are of the sort that, in a Parliamentary by-election this far out from polling day, would make you think the Lib Dems have an excellent chance of winning. Particularly because when the tactical vote squeeze works really well, it often also happens late in such a campaign. But will this general election play out like a series of Parliamentary by-elections? (2015 of course, very notably didn’t.)
For Liberal Democrat supporters, of course, the way to help make that answer to be ‘yes’ is to double-down on helping in target seats.
Overall, the MRP model puts the Liberal Democrats in first or second place in 134 seats compared with 50 in the 2017 general election.
5. The MRP results are not a prediction of the general election result for two reasons. First, they are based on what the public was saying at the time of the polling rather than a projection forward to what the public will do. But also, second, parties, candidates and voters can react to the results and change what they do.
In particular, it’s worth noting that Labour is planning to emphasise it is offering a route for Britain to leave the EU while Conservative Remainers may now be less worried about the prospect of Prime Minister Corbyn. That, in turn, could change people’s support for parties and so change the election result.