The Cursed Ratios of British Voters

I used to be a commodity trader in the 1990s. This wasn’t my day job, which was to forecast supply & demand, but I had the statistical skills to develop automated trading strategies for the traders I worked with. These were either based on moving averages of futures prices (known as technical trading) or buy/sell signals based on the underlying supply & demand picture (known as fundamental trading). I don’t trade commodities now but I’ve applied the stats skills I learned then to many time series since including those for elections and voting intentions.

Another trading strategy I could have used but didn’t was Chartism. I vividly remember the training course I did on this and coming away gobsmacked. How on earth were people being allowed to buy and sell millions using nothing more than astrology, spiritualism, numerology and other superstitions?! I immediately swore of from Chartism and have managed to stay away from it until this year. I hope you understand why it’s taken me many months to build up the courage to publish two articles where I use chartist superstitions instead of statistics to interpret long term trends in the vote shares of the major parties in the UK.

For this article, I explain why my charts show the Cursed Ratios of British voters are 52 : 26 : 13 : 6.5. For my next article, I will have you fearing for my sanity…

Data used in this article

All charts and tables in this article are based on monthly average vote shares for Great Britain as measured by voting intention polls since 1960. I make no adjustment for house effects by pollster. In some charts I show variation within a month by looking at how the 9-poll rolling average has changed in that month.

The raw data comes from Mark Pack who has systematically recorded every opinion poll published since 1945. Gallup were the only pollster in the 1940s & 1950s but this started to change in the 1960s which is why my analysis starts in January 1960. Since the start of 1962, there has been at least one voting intention poll every month. Between 1960 & 2011, vote share data is only available for the Conservatives, Labour and Liberal Democrats*. All other parties are combined into an Other category. From 2012, vote shares for the Greens and Reform** started to be published separately.

Previous incarnations of the Liberal Democrats* & Reform**

For information, the Liberal Democrats and Reform today used to be known under other names. The following points about the data I will use in this article for these parties should be noted –

The Liberal Democrats were formed in March 1988 as a result of a merger between the Liberal party and Social Democratic Party (SDP). The SDP was formed in March 1981 after the “Gang of Four” Labour MPs broke away from the Labour party. In June 1981, the SDP agreed an electoral alliance with the Liberals and from then on, the two parties together were known as The Alliance. When the Alliance ended with the creation of the Lib Dems, some SDP MPs refused to join and continued as a separate party for a couple of more years. The Lib Dem data I’ve used includes the SDP & Liberals up to February 1988 but afterwards, it only includes the Liberal Democrats and not the SDP rump who moved into the Other category.
Reform is the latest incarnation of a party led by Nigel Farage. He was leader of UKIP (United Kingdom Independence Party) in 2012 when pollsters started to publish vote share data for them. He left UKIP in 2018 and founded the Brexit Party in 2019 which was renamed Reform in 2020. UKIP still exists today but most pollsters stopped publishing vote shares for them after the 2019 general election and the last poll to show UKIP data was in early 2024. When I show data for Reform back to 2012, it is actually the sum of vote shares for UKIP & Brexit Party/Reform. In practice, after the EU elections in June 2019, UKIP only averaged 1% in any poll where their vote share was published.

British Politics in 1 Chart 1960 to 2025

If I’m to be a chartist in this article, I need a chart so here it is. It shows the monthly average of vote shares from all voting intention polls since 1960.

This a historic moment in British politics. For the first time ever, over half of those responding, 50.3% in November 2025, have told pollsters they do not intend to vote for any of the three traditional parties of Conservatives, Labour & Liberal Democrats. These are the only three parties who have ever governed Britain hence why I will refer to them collectively as the Establishment parties. The other historic moment reached in November 2025 for the first time ever is that both Conservatives & Labour are averaging less than 20%.

What is striking about this chart is the rapidity in the growth of Others from less than 20% two years ago, to taking the lead this time last year and then to over 50% today. Many theories have and will be put forward to explain this. What I want to put forward is this is the result of the Cursed Ratios which show themselves when the chart is viewed through chartist eyes.

How the Lib Dems cast the spell of the Cursed Ratios

The phrase “cursed ratio” entered the electoral dictionary when the UK voted to leave the European Union on 23rd June 2016 by the margin of 52% to 48%. Since then, whenever 52-48 rears its head in a vote or poll, you will hear the words cursed ratio. I made the point in my article “How similar are Leave & Trump voters?” that the USA also voted 52-48 in the 2016 US Presidential election which saw a surprise victory for Trump. I observed Clinton got 48% of the vote, Trump 46% and Others 6% but an analysis of voter sentiment showed the 6% Others were much closer to Trump than Clinton in their underlying attitudes.

What no-one seems to have realised until now is ironically, the Liberal Democrats (the most pro-EU party in the UK) set the scene for the cursed ratios. Even today they can’t escape them. To see what I mean, let’s look at the chart again for the Lib Dems only.

I’ve changed the vertical scale gridlines from every 5% to every 6.5% which is half of the Lib Dems current vote share of 13% in the polls. How did I get here?

It all started earlier this year when I noticed Lib Dem surges in the past appeared to peak around 26%. Specifically –

Apr-May 1962 averaging 25.5%
Aug-Oct 1973 averaging 26.0%
Aug-Sep 1994 averaging 26.1%
Sept 2003 averaging 25.9%
Apr-May 2010 averaging 25.5%

That is 5 separate occasions when the Lib Dems surged as a single party to the exact same peak. That can’t be chance alone, surely there must be something in the stars that ordains the Lib Dems cannot rise above 26%? (See what Chartism is doing to me!)

Why have I ignored the 1980s when they broke the 26% ceiling? The answer is this wasn’t the Lib Dems alone, it was the Liberal/SDP Alliance which would become the Lib Dems in 1988. But get this! The Alliance lasted from June 1981 to February 1988 but the merger was proposed by David Steel in June 1987 following the general election. For the 6 years from June 1981 to June 1987, what was the average vote share of the Alliance in the polls? Yes you guessed it, 26%!! Surely you must be a believer in Chartism now!?

I haven’t finished though. 13% is where the Lib Dems are today which is half of 26%. More than that, 13% is another line ordained by the stars or charts. Specifically –

Between the 1962 & 1973 surges, the peak for Lib Dems was 13.6% in Aug 1969.
In 1980, the year before the SDP and thence the Alliance was formed, the Liberals averaged 13.1%
After the SDP folded in 1990, a clear floor was established for the Lib Dems at 13% between 1991 & 2009
In 2025, the Lib Dems have averaged 13.4%.

By now you should be on a roll and have noticed an even lower floor at 6.5%. This is clear for the following periods –

1964 to 1972
1978 which coincides with the end of the Lib-Lab Pact.
May 1989 to Apr 1990 which coincides with the temporary surge of the Green party.
2016 to 2020 which was the aftermath of their time in the coalition government of 2010-2015.

So this is why I say the Lib Dems are responsible for the cursed ratios of 6.5 : 13 : 26. But the origin of the term today is the 52% Leave vote of 2016 and surely the Lib Dems can’t be responsible for that? Apart from 52 being 26 doubled, the chart above says the Lib Dems are blameless or are they?

The 52 : 26 : 13 : 6.5 Ratios

Let’s now look at the chart again but this time for the Conservative and Labour parties only. The gridlines are now set at the ratios of 6.5%, 13%, 26% & 52%.

This is the supernatural power of Chartism (you can’t know the pain I am in writing that!). When the 26% ceiling for the Lib Dems is doubled to 52% we find the floor and ceiling of what were the two main parties up to last year. The only time the 52 : 26 channel did not hold was in Tony Blair’s first five years as leader of the Labour party when 52% became the floor for Labour. Even in the 2020s, both parties nudged the 52% ceiling in April 2020 for the Conservatives and October 2022 for Labour.

In 2024, the Conservatives fell through the 26% floor but picked themselves up to get to 24.4% in the general election and back over 26% when Kemi Badenoch took over as leader. At the time, it looked like a temporary relapse for them not unlike what they experienced in the mid-90s. But 2025 brought devastation and for the last 12 months they have been well below 26%. Labour dipped below 26% in February and have since headed straight down to join the Tories in their attempt to meet up with the Liberal Democrats at the next level of 13%.

Will Other parties bow down to the Cursed Ratios?

At first sight, I’d say Others broadly obey the cursed ratios but not as cleanly as the three establishment parties do.

That shouldn’t be a surprise. Others is a composite of many parties. From a statistical perspective, when you are looking at a data set which must add up to 100%, one of the categories has no “degree of freedom” since its value will automatically be known once you know the values of the other categories.

Saying that, I will make a prediction now that Others will not decisively break through the 52% ceiling between now and the next general election. I think the power of 52% is too strong to ignore.

Ideally, we would like to see the vote shares for the separate parties making up Others on this chart. As I noted earlier, we only have separate data for Reform** and the Greens since 2012 which is a relatively short time frame. Regardless let’s go ahead and look at the monthly trends by the 5 major parties of today sine 2010.

The first thing I notice is the Greens are at 13% today! Are they in thrall to the cursed ratio? They have seen a relatively rapid rise since they elected a new leader but I suspect this is a bounce. For now, I think 13% will be their ceiling especially with the launch of Your Party which competes for the same voters as the Greens. If I am right, that means we have 4 parties jostling to meet up at 13%, which is just unprecedented in British politics at the national levels. However, it is not without precedent in specific regions such as Scotland and Northern Ireland.

That leaves Reform who are averaging 30%. I’m going to predict that 26% will be their floor between now and the next election. If so, that leaves ~15% of voters for the major parties to compete for which I arrive at as follows.

26% floor for Reform
13% floor for the Conservatives
13% floor for Labour
13% ceiling/floor for the Liberal Democrats
13% ceiling/floor for the Greens
6.5% ceiling/floor for all other parties combined
which adds up to 84.5% leaving 15.5% free.

Your guess is as good as mine when it comes to working out where they will end up at the next election! I am intending to publish another article about house effects in the polls today which I think will be the basis of my election forecasting model for the next election.

Harry Potter & The Cursed Ratios of British Voters

If only ~15% of voters are still in play, this has an immediate implication for the Conservative and Labour parties. Their ceiling between now and the next election has to be 26% when looking at the trends through Chartist eyes?

The time when Labour & the Conservatives would fight it out as the two top dogs in the 52-26 channel looks to be over. They are now playing in the 26-13 channel as ordained by the cursed ratios uncovered by Chartism.

I repeat, this is a painful article for a statistician to write! So far though, I’ve only used the most basic tool of Chartism which is to look for floors and ceilings. In my next article, I go further and run the risk of looking like a numerologist instead of a statistician…

— Would you like to comment on this article? —-

Please do leave your comments on this X/Twitter and/or this LinkedIn threads.

—- Want to read more about polls? —-

More articles on polling can be found by clicking on the Polling tab at the top of your screen.

— Subscribe to my newsletter to receive more articles like this one! —-

If you would like to receive notifications from me of news, articles and offers relating to Polling & Elections, please click here to go to my Newsletter Subscription page and tick the Surveys and/or Elections category and other categories that may be of interest to you. You will be able to unsubscribe at anytime.