We are heading into the most exciting election in decades. With everything potentially hanging on a few thousand votes in key states, the election has never been more uncertain.
That’s why it’s more important than ever to be able to distinguish the best bollards from the best bollards.
Meet the Daily Mail/JL Partners forecasting model. Based on our experience modeling elections around the world and in the US, we have developed a model to take on the race for the White House.
Here’s how it works and why we believe it’s the best in the industry.
It starts with a fundamentals model. This predicts how many votes each candidate will win based on historical data. This model consists of two parts: the long-term model and the short-term model.
The race between former President Donald Trump and Vice President Kamala Harris couldn’t be more exciting. You can follow the best election model available
The long-term portion uses presidential election data from the past 76 years to estimate vote shares. We take into account party in power, economic data, and presidential approval ratings.
All of these factors then give us an initial estimate of the final vote share. For example, if Biden’s approval ratings are low, this will reduce Harris’s previous vote share.
But this history alone is not enough. We delve deeper into the elections since 1996.
The US political environment has become more polarized over the past 30 years and our short-term model reflects that. We add variables such as consumer confidence and congressional approval to form our short-term fundamentals.
For example, if confidence in the economy has declined over the past year, this will affect the share of votes received by the incumbent representative.
We combine these two estimates to produce a best-effort estimate, based on history alone, of the number of votes each candidate could receive.
This gives us a historical picture, but does not take into account the contours of 2024. To do this, we also collect all the polls conducted across the country, at the state and national level.
Compared to other models, we use only polls that use full-ballot testing. That’s because third-party candidates, from RFK Jr. to Jill Stein, matter a lot.
In 2000, Green Party candidate Ralph Nader received more votes than the margin between George Bush and Al Gore. Jill Stein’s votes likely cost Hillary Clinton the 2016 election, by more than Trump’s margins in Michigan, Wisconsin and Pennsylvania.
Your browser does not support iframes.
Then we do something else. Not every third-party candidate has access to the ballot in every state. We use our independent research to determine how votes flow from these third-party candidates to the two main contenders—and answer the question of how their supporters would vote if they were not on the ballot. We do this for every state.
Our model can therefore take into account the impact that external candidates have on the race.
That unique algorithm ultimately helps Trump in states where RFK Jr. isn’t on the ballot, but Jill Stein and Cornel West — who have more support from Harris — are.
Furthermore, in our model we give less weight to polls of only adults or registered voters than to polls of likely voters. Furthermore, the contribution of a poll to the model is weighted based on the historical accuracy of a pollster.
Your browser does not support iframes.
Your browser does not support iframes.
Furthermore, we give less weight to online-only polls that do not have a rigorous methodology to verify the identity of their interviewees and match them to the voter base. Instead, we give more weight to polls that use mixed methodology approaches or have a robust methodology – such as that of the Daily Mail.
The final step is the most technical part. We use a highly sophisticated autoregressive Bayesian algorithm that continuously updates our predictions with real-time data from states and national polls. We make crucial adjustments to reflect who is on the ballot in each state, providing an accurate, up-to-date picture of the nation’s voting intentions.
Let’s be clear: Any pollster who tells you with certainty what will happen in this election is selling you bullshit. But we believe our model — which synthesizes both historical trends and current polls, and takes into account third-party effects — is the closest to reality.
Skip the guesswork or dubious partisanship this election. The Daily Mail/JL Partners model is your best bet to see where the real battle lines are drawn.
The authors are: Callum Hunter, Data Scientist at JL Partners in London; Landon Wall, Partner in Modeling and Analytics in Arizona; James Johnson, Co-Founder in Washington