(This is on a separate page as once you've read it once, you probably don't want to be scrolling past it all the time.
This site is intended to be interesting during election night (July 4th-5th 2024) and potentially shortly afterwards, for those wishing to get an at-a-glance view of the state of play as the night goes on, and also to compare reality with the various predictions over the course of the campaign. The "2019 notional results" are the 2019 election results, redistributed to the new constituency boundaries for 2024 as calculated by Michael Thrasher and Colin Rallings, as listed on Wikipedia. A very few aspects of the site are down to the judgement of the author (Jon Skeet) - I'm not a political analyst by any stretch of the imagination:
The results are supplied by the Democracy Club API. These are provided on the basis of a CC BY 4.0 licence. Many thanks to the Democracy Club for providing such an excellent service - you may wish to make a donation to the cause.
Predicted result times have been compiled from the Press Association declaration times by Matt Singh. See his post for more details.
Everything else is either the 2024 results or the work of the pollsters and analysts listed below.(Descriptions for pollsters and analysts are adapted from their web sites or Wikipedia articles.)
There are five views available:
The default "mode" for the site is for the results and predictions to be shown just in text. An optional "colour mode" is available, which changes the background of the relevant table cells, according to the following colour key:
| Safe Cons | Likely Cons | Lean Cons | Toss-up Cons |
| Safe Lab | Likely Lab | Lean Lab | Toss-up Lab |
| Safe LD | Likely LD | Lean LD | Toss-up LD |
| Safe SNP | Likely SNP | Lean SNP | Toss-up SNP |
| Safe PC | Likely PC | Lean PC | Toss-up PC |
| Safe Green | Likely Green | Lean Green | Toss-up Green |
| Safe Ref | Likely Ref | Lean Ref | Toss-up Ref |
| Safe DUP | Likely DUP | Lean DUP | Toss-up DUP |
| Safe UUP | Likely UUP | Lean UUP | Toss-up UUP |
| Safe SDLP | Likely SDLP | Lean SDLP | Toss-up SDLP |
| Safe SF | Likely SF | Lean SF | Toss-up SF |
| Safe Alliance | Likely Alliance | Lean Alliance | Toss-up Alliance |
| Safe Ind | Likely Ind | Lean Ind | Toss-up Ind |
| Safe Other | Likely Other | Lean Other | Toss-up Other |
Some users may find this easier to read at a glance; others may find it too jarring and with too little contrast. Links for both text-only mode and colour mode are at the top of each page.
Sam Freedman is a senior fellow at the Institute for Government and a senior adviser to the Ark Schools network. He runs the Comment is Freed Substack with his father, Laurence Freedman. During the election campaign of 2024, there are daily election briefings, including seat predictions with analsys by Sam. These seat predictions (linked in the tables in this site) are paywalled content, but I (Jon Skeet) thoroughly recommend taking out a subscription; all the content is truly excellent. I'd also recommend the podcast which he hosts with Ayesha Hazarika, The Power Test. Follow Sam on Twitter/X at @Samfr.
Survation is a polling and market research agency based in London. Survation have been conducting research surveys since 2010. Surveys are conducted via telephone, online panel and face to face as well as omnibus research for a broad range of clients including television, newspapers, charities, lobby groups, trade unions, law firms and political parties.
Abbreviations:
More in Common was founded in the aftermath of the tragic murder of Jo Cox MP in 2016. More in Common takes its name from Jo’s maiden speech in Parliament where she said: “We are far more united and have far more in common than that which divides us.” Their hope is that through their work, their honour Jo’s memory and legacy.
Abbreviations:
YouGov is an international online research data and analytics technology group. For the 2024 election it has published polls and MRPs.
Abbreviations:
The Financial Times publishes a model which predicts seat numbers based on overall vote shares. The sets of predictions below are based on polling on the specified date. Many thanks to Peter Inglesby for his work in scraping the data. See https://inglesp.github.io/apogee/ for more details.
Abbreviations:
Ipsos Group S.A. is a multinational market research and consulting firm with headquarters in Paris, France.
Focaldata is a London-based research company. From the Mission part of their "about us" page:
Our mission is closing the gap between what organisations believe about the public (consumers, voters, citizens) and reality. We call this the "understanding gap". This is our focus because an imperfect understanding of people is the root cause of fundamental social, political, environmental and economic problems — from low growth to inequality.We believe better understanding of how and why people act and think can help foster a more dynamic, more generous world.
Britain Elects was co-founded in 2013 by Ben Walker and Lily Jayne-Summers. Initially set up as a Twitter account to collate and analyse council by-elections, it has since produced trackers on public opinion and contributes extensively towards the understanding of UK attitudes to politicians, politics and issues more generally. With the help of data jounalists, its poll aggregation is used to generate election forecasts that are demonstrating an increasing form for accuracy in the field.
Ben Walker writes on the state of UK and international public opinion, and has previously worked as a political consultant for campaigns and candidates advising on message promotion and voter targeting.
Britain Predicts is the New Statesman's election model. It was built by Ben Walker of Britain Elects and others, and uses Britain Elects poll tracking data to forecast how the country would vote if an election was held today.
Abbreviations:
(From their web site): "WeThink aims to be one of the most trusted names in public polling, a place where people and our clients can turn to for accurate, reliable data that brings intelligent and clear insight in an often crazy world." WeThink has published their MRP in collaboration with The Economist.
From their web site:
We formed JLP after three years running the research programme for 10 Downing Street, in the heart of politics. Through a range of quantitative and qualitative methods, we know which questions to ask to get the richness of understanding that you need.
J. L. Partner has been working with The Rest Is Politics to provide polling information throughout the election campaign. Their SRP (Stacked Regression and Poststratification) poll provides per-seat predictions in a manner similar to (but distinct from) MRPs. See their post for far more details around methodology.
Each prediction has a "strength" of safe, likely, lean or toss-up. This is calculated differently for different providers:
The boundaries for the majority/probability-based descriptions may be adjusted over time.