Do the Lib Dems win more seats by winning more votes?
Welcome to the 192nd edition of The Week in Polls (TWIP), which takes a look at the relationship between seats and votes for the Liberal Democrats.
That is followed by a summary of the latest national voting intentions polls, seat projections from MRPs, and the most recent party leader ratings.
Subscribers who pay for the full edition can also read a set of ten insights from the latest polling and analysis, including:
The story behind the decline in Labour’s support in Wales;
New polling on assisted dying; and
Who people say they would never vote for.
If you are not yet a paid subscriber, you can read all ten by starting a free trial now:
For more polling news as it comes in during the week, you can also follow The Week in Polls on Bluesky.
And with that, on with the show.
Was ‘seats, not votes’ just a freak Lib Dem result in 2024?
Regular readers of this newsletter will be familiar with my defence of the value of national voting intention polls, even a long way out from a general election, such as:
Polls are rather like a company’s share price, an analogy used by George Osborne and others. (I am not sure who first used it. If you know, please do let me know.) That is, they are both a short-term public measure of how things are going.
Longer-term fundamentals may get obscured by too much focus on the short-term public measure and its volatility.
Yet also there is a substantive impact if that short-term measure plummets - neither companies nor parties nor governments can close their minds, cordon off their morale and isolate their nerves completely from the numbers heading down. Bad numbers have a real world impact.
Yet those who hear me talk about the polls with my Lib Dem hat1 on will often hear me say that the party’s standing in national voting intention polls is a very poor guide to the party’s prospects at a general election.
The mantra for the party’s highly successful 2024 general election campaign - one that resulted in the most seats for a century - was ‘seats, not votes’, a reference in part to a wish to avoid previous results such as that in 1983 where the Alliance soared in vote share - and flopped in seat numbers. The Alliance scored double the vote but only a third of the seats compared with the Lib Dems in 2024.
At most, you might recall me say that Lib Dem national vote share figures are a trailing, rather than leading indicator. That is, they do not foretell election results but they follow them, with the party’s vote share rising after it has done well in elections (and hence the party’s vote share at this stage in the Parliament now being significantly up on the last Parliament).
But how well does the data support holding both these views at the same time? Are the Lib Dems really somehow different from Labour and the Conservatives in having such a divorce between the number of Lib Dem MPs and their vote share?
Here is how the data in all the general elections from 1945 looks, comparing vote share in Great Britain with the percentage of MPs2 from Great Britain elected as Liberal Democrat, or before that SDP/Liberal Alliance or Liberal Party (which I collectively call Lib Dem below for convenience).
Ignore the right-hand half of the graph for the moment.
Over in the left-hand half, you can see there is a clear relationship between votes and seats, but with two important caveats.
First, these are the elections where few Lib Dem MPs were elected - at most 3.5% of the House of Commons, which is 23 MPs at the current size. When doing poorly, there is a relationship between votes and seats.
Second, that vote/seat relationship does not extrapolate up to success. You can draw a trend line through those points on the left-hand side and you get a trade-off in round numbers of an increase in vote share of 10% (a massive increase on the scale of general election vote share changes for any party), increasing the number of MPs by a bit under 1.5%, or a gain of a mere 9-10 MPs in current currency.
As the Alliance discovered in 1983, a huge vote share increase can result in paltry results when it comes to counting seats.
Now look at the right-hand side of the graph, circled this time:
While the left-hand side are elections returning between 6 and 23 Lib Dem MPs, these on the right saw between 46 and 72 MPs secured, including the big surges at the elections of 1997 and 2024.
The successful ones are not an extension of a trend from the less successful ones. They are off doing their own thing, with even among themselves not much of a relationship between seats and votes.
If next week’s opinion polls were to show the Liberal Democrats suddenly up 10 points, of course I’ll find reasons to post on social media ‘ooh look, how interesting’.
What the contrast between the two halves of this graph shows, however, is that the difference between success and failure in seat numbers has not been about national vote share. It’s about getting something else right (targeting, under first past the post).
That point is reinforced if instead we look at the change in vote share and the change in seat tallies at elections:
Again, there’s a bit of a relationship there, especially on the downside (hello 2015, down in the bottom left).
But again, it’s a limited one and one that does not point to vote share gains being a particular route to success. The best seat increases have not come with impressive vote share increases (the two points on the right), and the best vote share increases have not come with great seat gains (the two points at the top).
Dramatic success for the Lib Dems has come not from raising vote share but from breaking the relationship between seats and vote share. Which is rather ironic in the context of internal Lib Dem debates, as some people argue that the ambitious route to unprecedented success is to focus on national messaging and national vote share - yet as the data above shows, that’s playing in the part of the data that produced weak seat results. It’s breaking the relationship that produces success.
There is of course a threshold beyond which, particularly thanks to first past the post’s struggle with multi-party politics, where vote share increases are likely to translate to dramatic seat increases for the Liberal Democrats.
But on the sort of variations we’ve seen across all the varied politics from 1945, success in seat tallies has come from playing a different game from chasing national vote share.
Skip the boring socks.3 Instead, this Christmas give the gift of polling insight, crosstab wisdom and raised eyebrows at media coverage with a gift subscription to The Week in Polls.
The gift is immediately available, handy for anyone who has left present shopping a little late. Not that you have, of course.
You can buy a gift subscription here.
Elsewhere from me…
If you spot a factual error…
Borrowing an idea from Stuart Ritchie and others, if you spot a factual error in an edition of this newsletter, I will give you a 12-month free subscription to the paid-for version of this newsletter (or an extra 12 months on your existing one). Factual rather than grammatical errors, and factual errors rather than disagreements over interpretation. Though messages about both of those are always very welcome.
Voting intentions and leadership ratings
Here are the latest national voting intentions figures, which include a new poll from Verian. The poll is probably the closest we have to a ‘gold standard’ poll these days. That is partly because of Verian’s pedigree over many electoral cycles (often overlooked both due to company name changes and also due to it being pollsters from other firms who dominate media punditry). It is also because its more expensive methodology is the closest we get in modern polling to the purity of random sampling.
Although previous ‘gold standards’ have sometimes had their moments of revelation as being made of iron pyrite, the Verian figures - better for the Conservatives and Lib Dems, less good for Reform and the Greens - are a strong hint that those other pollsters getting similar results are the ones getting it right and that those painting a rather different picture are more likely to be found out when we get real election results. A strong hint, though, rather than conclusive proof (though also one that matches the pattern from council by-elections).
A strong hint too that those other pollsters showing Labour losses to the Lib Dems and Greens being a much bigger factor than their losses to Reform and the Conservatives, as the Verian data shows this too.
The table is also online here. It is updated regularly through the week with new polls.
Next, the latest seat projections from MRP models and similar, also sorted by fieldwork dates. As these are infrequent, note how old the ‘latest’ data is.
The one update to the table this week is the full information being added for the Stack Data Strategy MRP over the summer. I have also kept in the JL Partners POLARIS figures as they intend to produce more figures from this model in the future.
Finally, a summary of the latest leadership ratings, sorted by pollster name.
The table is also online here with additional details and is updated regularly through the week as new data comes out.
For the historic figures for both voting intentions and leader approval, including Parliamentary by-election polls, see PollBase.
Catch up on the previous two editions
My privacy policy and other legal information are available here. Links to purchase books online are usually affiliate links which pay a commission for each sale.
Quotes from social media messages are occasionally edited lightly for punctuation and clarity.
If you are subscribed to other email lists of mine, please note that unsubscribing from this one won’t automatically remove you from the others. If you wish to be removed from all my lists, simply reply to this email to let me know.
Photo of newspaper by Andrys Stienstra from Pixabay.
Views on rejoining Erasmus, and 9 other polling insights you shouldn’t miss
The following 10 findings from the most recent polls and analysis are only for subscribers who pay for the full edition, but you can sign up for a free trial to read them straight away.
Even 2024 Reform voters support “the UK rejoining the Erasmus study scheme, whereby UK students can spend a year studying at European universities as part of their UK degree courses without paying extra fees, and vice versa for European students.” The exact figures are
