While polling can give us an idea of what the public thinks, there is nothing like real votes to test the mood of the country. With elections across large areas of England, plus Scotland and Wales tomorrow, we’ll finally get some real votes to compare recent polling against. While the results in Scotland and Wales will be roughly in line with party vote shares, as they use voting systems that reflect how people vote, local council elections in England are a different story.
England’s First Past the Post (FPTP) voting system is likely to deliver numerous highly disproportional results, with parties possibly taking control of councils where they didn’t win the most votes and councillors elected on tiny vote shares. First Past the Post simply cannot cope with this new reality of multi-party politics.
As can be seen below, we have hit another milestone in terms of electoral fragmentation, with no fewer than five parties separated by just 14.2 percentage points in our April polling averages data – a new record.
Scroll down to see the polling for April 2026
First Past The Post is designed for a two-party environment. When people express their political preferences in a wider variety of ways the system starts to splutter and breakdown, producing increasingly chaotic outcomes that do not properly reflect how people have voted. We will be watching closely as the English local elections results come in and will report back on what happens.
Two sets of numbers to keep an eye on in the aftermath of the English local elections are the Projected National Share (PNS) produced by Professor Sir John Curtice, for the BBC and the National Equivalent Vote (NEV) produced by Professors Rallings and Thrasher, for Sky News.
Both the PNS and the NEV have been published every year for around the last 45 years. What these data attempt to do, via slightly different methodologies, is to estimate the vote share each party might have got if local elections had taken place across the whole country, rather than just in certain parts of it. This makes it possible to compare local elections vote shares from one year to the next, even though the types of places where English local elections take place varies significantly from one year to the next.
In May 2025, the PNS recorded five parties on over 10% of the vote for the first time and placed those five parties within 19 percentage points of each other. This year we expect to see both the PNS and NEV reflect an even more fragmented political landscape, in line with how the opinion polls have moved over the last year.
April UK General Election Polling Averages
The average (mean) vote shares from the most recent April poll by each of the ten polling companies who published a UK general election poll during April, is as follows:
Reform UK: 26.4% (-0.7*)
Labour: 19.1% (+0.3)
Conservatives: 18.6% (+0.3)
Greens: 15.6% (-0.2)
Liberal Democrats: 12.2% (+0.1)
Others: 8.1% (+0.2)
* Compared with March’s average – Each month a different combination of pollsters will publish polls, so this change is not strictly comparing like with like, but gives a general sense of change
What do these mean?
- The party with the highest vote share has the support of only just over a quarter of voters.
- There is only a 14.2-point gap between the 1st and 5th placed parties, the smallest we have seen since we started collecting this data.
- The combined Conservative and Labour vote share is just 37.7%, a significant drop on the combined 57.4% they achieved at the 2024 general election, itself a record low.
- Reform UK’s vote share (26.4%) represents its lowest monthly average vote share in the last 12 months, since they recorded 24.9% in April 2025. They are down 5.4 points from their monthly average vote share peak, which was 31.8% in September 2025.
April’s polling average was compiled using data from the following pollsters – BMG; Find Out Now; Freshwater Strategy; Good Growth Foundation; Ipsos; J.L. Partners; More In Common; Opinium; Techne; YouGov.
With growing fragmentation, it becomes harder and harder to work out how polling will translate into seats in Westminster, at UK general elections. Something that pollsters have tried to get round with MRP polls. These MRP polls take data from a bigger than normal sample of people. For example, the sample for YouGov’s most recent UK general election MRP was 13,000.YouGov’s MRP model looks for relationships between people’s characteristics and how they intend to vote. It then combines these relationships with information about the characteristics of people living in different constituencies, in order to produce estimates of voting intention in each constituency.
Both YouGov and More In Common have published MRP polls for this week’s London local elections. However, both companies have stuck to estimating vote shares for each party, in each London borough, rather than attempting to estimate the number of council seats each party will get in each borough. This highlights just how difficult it is becoming to understand how party vote shares will translate into representation in elected chambers, under FPTP.
Do you agree we need a voting system where every vote counts?
Add your name