Today, few political news outlets receive universally favorable evaluations from the American public. In retaining broad public approval, local media remain a notable exception that is important to understand. We consider the durability of local news trust to criticism from politicians, focusing on Republican elites because they generally are more willing to attack the media than Democrats. In a survey experiment, we find that exposure to a Republican politician’s attack on a local newspaper dramatically reduces the public’s trust in and intent to use local news. This attack is particularly effective among Republicans, although it also leads Democrats and Independents to negatively shift their views of the local newspaper. Among those exposed to elite criticism of local news, overall trust and partisan divides between Republicans and Democrats resemble those for national media. This shows that the credibility of local news depends on the absence of elite criticism rather than resilience to it.

This Special Issue presents a wide array of election forecasting models for the 2024 US elections. Most of these models generate forecasts for the presidential, congressional, and gubernatorial races. The contributions are characterized by the variety of their approaches: citizen forecasting, electronic markets, large language models, machine learning, poll-based models, and regression analysis. This introduction first summarizes some of the lessons and challenges of election forecasting. We then provide a brief context of the 2024 campaign and a short overview of the articles included in the Special Issue. The forecasts point to a tight presidential race. The two-party popular-vote predictions are almost evenly split, with some favoring Donald Trump and others Kamala Harris. However, among the models that offer an Electoral College forecast, three predict that Harris will win and five predict that Trump will return to the presidency.

Donald Trump’s bid for the 2024 Republican presidential nomination is unique in that no former president since Theodore Roosevelt in 1912 has sought the nomination of their political party, nor has a candidate sought the nomination while facing multiple criminal indictments. With data from previous nomination cycles, we use presidential nominations from 1980 to 2020 to create a forecast for the 2024 Republican primaries. The variables in the equations consist of data from the pre-primary period (e.g., money raised, cash reserves, elite endorsements, and polling results) and a second model with results of the Iowa caucuses and the New Hampshire primary to forecast the remaining primary vote. The models accurately predict Trump’s victory despite the unique nature of his candidacy.

This model generates projections of the national popular vote and Electoral College votes a year in advance of the U.S. Presidential Election, before each party’s nominees are known. It forecasts the Democratic two-party popular vote in each state and the District of Columbia. It uses four independent variables: national head-to-head polling data 13 months prior to the election, the states’ prior election result, a party-adjusted home state advantage dummy variable, and a party adjusted variable simply counting the number of consecutive terms the current incumbent party has occupied the White House. New to this year’s model is a polling average approach that encompasses all possible candidate matchups for whom data is available. This year’s forecast suggests a distinct possibility of an Electoral College misfire benefitting the Republicans.

Voter turnout is a crucial indicator of democratic health, yet forecasting turnout remains an understudied area in political science. This article presents two pioneering models for predicting US presidential election turnout: the national and the state model. The national one, using data from 1868 from 2020, employs lagged turnout as its sole predictor. The state model, covering 1984 to 2020, incorporates demographic and institutional variables to forecast state-level participation. The national predicts 65.3% turnout for 2024, whereas the state model forecasts increased turnout in 41 states compared to 2020. The models’ ability to generate early predictions offers valuable lead time for planning and resource allocation, which has implications for election administrators and political campaigns, as well as for the vibrancy of civic engagement in America.

This article considers both presidential approval and party brand differentials, as measured by the generic ballot, to forecast the 2024 US presidential and congressional elections. Although both variables are leveraged to forecast collective partisan election outcomes, we consider the variables together as distinct determinants of partisan fortunes at both the executive and legislative levels. First, using a novel time series of mass national opinion since 1937, we show that presidential approval and generic brands are distinct conceptual and empirical measures of mass public assessments of collective institutions. Second, in a series of fully specified models validated with out-of-sample predictions, we show that presidential approval is the main predictor of presidential elections, yet, perhaps surprisingly, the vast bulk of the incumbent party’s performance in congressional elections is explained by partisan brands. Lastly, we forecast the 2024 U.S. national elections and find that Republicans are well positioned to win back the White House this November. By contrast, our model forecasts control of both chambers of the US Congress to be essentially a tied contest.

In this article, we build a model to predict the state-level results of the 2024 election. We do so by using both polling from similar points in past election cycles and the results of the previous election. Notably, we update our model over time, and the coefficients of the two variables change as a result: the model puts more weight on polling as the election gets closer. As of September 1, 2024, we find that Kamala Harris is a narrow favorite to win the 2024 election, with a 57% chance of doing so. Currently, the model predicts she will win 289 electoral votes to Trump’s 249. However, there remains significant uncertainty, and the model will continue to be updated as the election nears.

Despite governors’ crucial roles in shaping important policies, including abortion, education, and infrastructure, forecasters have paid little attention to gubernatorial elections. We posit that institutional idiosyncrasies and lack of public opinion data have exacerbated the classic problem facing all election forecasts: there are too many predictors and too few cases, leading to overfitting. To address these problems, we combine new governor and state-level presidential approval data with a machine-learning approach, LASSO, for variable selection. LASSO examines numerous variables but retains only those that substantively improve model performance. Results demonstrate the efficacy of gubernatorial and presidential approval ratings measured two quarters preelection in predicting both incumbent-party vote share and election winners in out-of-sample predictions. For 2022, our approach outperformed the Cook Political Report’s Partisan Voting Index and compared well with 538’s Election Day prediction. For 2024, our LASSO-Popularity model predictions indicate that it will likely be a difficult year for Democrats in gubernatorial contests.

The outcome of the 2016 election made it abundantly clear that victory in US presidential contests depends on the Electoral College much more than on direct universal suffrage. This fact points to the importance of using state-level models to arrive at adequate predictions of winners and losers in US presidential elections. In fact, the use of a model disaggregated to the state level and focusing on three types of measures—namely, changes in the unemployment rate, presidential popularity, and indicators of long-term patterns in the regional strength of the Democratic and Republican parties—has in the past enabled us to produce fairly accurate forecasts of the number of Electoral College votes for the presidential candidates of the two major American parties. In this article, we bring various modifications to this model to improve its overall accuracy. With Joe Biden out of the race, this revised model predicts that Donald Trump will succeed in winning back the presidency with 341 electoral votes against 197 for Kamala Harris.

Using a forecasting model based on economic pessimism and recognizing the difficulties of making such a forecast in such atypical times, the forecasting model predicts a narrow loss for the incumbent presidential party and a loss of 12 seats in the House of Representatives. Even with the unusual nature of politics in the United States over the past decade, this model does a good job of predicting election outcomes. The more pessimistic people are, the worse the incumbent party does in presidential and House elections. Moreover, the power of incumbency shows strongly.

Our political economy model, as it has come to be called, has offered up forecasts of the American presidential election outcome since the early 1980s. The model, based on referendum theory, as measured by the job performance of the president and the economy (1948 to the present), yields a forecast from data available in the summer of the election year. We consider alternative specifications of this parsimonious model, examining the possible effects of other economic measures, Covid-19, and incumbency advantage on forecasting. The current point estimate of the core political economy model predicts the Democratic candidate will receive 48 percent of the two-party popular vote, which translates into a narrow Electoral College loss for the incumbent party. This point forecast, however, comes with a considerable amount of uncertainty. There is an 11-point spread around our point estimate, which effectively means we have a horserace on our hands, with both horses close to the finish line.

Election forecasting in modern democracies faces significant challenges, including increasing survey nonresponse and selection bias. Moreover, there are limitations to the current predictive approaches. Whereas structural models focus solely on macro-level variables (e.g., economic conditions and leader popularity), thereby overlooking the importance of individual-level factors, survey-based aggregation methods often rely on intuitive procedures that lack theoretical foundations. To address these gaps, this article proposes a combined (i.e., both standard and Bayesian) logistic regression approach that leverages voter-level data and incorporates a theory-based specification. By testing these models on recent waves of the American National Election Studies Time Series, this study demonstrates that the proposed approach yields notably accurate predictions of Republican popular support in each election.

We present Iowa Electronic Markets (IEM) forecasts for the popular vote shares in the 2024 US presidential election. We discuss the differences between IEM forecasts and polls, the influence of the first presidential debate, the changes resulting from Biden dropping out of the race, and the degree of uncertainty implied by IEM forecasts. On September 29, the IEM forecast a 9-percentage-point Democratic popular vote margin according to a thinly traded vote-share market and an 85.7% chance the Democrat will receive more votes than the Republican in a thickly traded winner-takes-all market. Using a distribution derived from both markets, the forecasts are for a 6- to 7-percentage-point Democratic margin and 87.0% chance of winning. However, significant uncertainty remains.

The Partisan-Bounded Economic Model forecasts popular vote and Electoral College vote results in the 2024 presidential election based on economic growth, presidential popularity, and shifts in party identification within the electorate. Due to the partisan vote and voter inertia, presidential elections are unlikely to result in lopsided outcomes even in the face of economic turmoil or prosperity. In the Partisan-Bounded Economic Model, outliers in the annualized GDP growth are adjusted to fall within a fixed range of 5 percent to -5 percent. Any GDP growth values exceeding 5 percent or falling below -5 percent are re-coded as 5 percent and -5 percent, respectively. The model predicts that Vice President Kamala Harris will secure 52.3 percent of the party vote, 49.4 percent of the total popular vote, and 59.1 percent of the Electoral College votes.

Statesman and scholar Alexis de Tocqueville (1876) once noted, “History is a gallery of pictures in which there are few originals and many copies.” In other words, history has a habit of repeating itself, and we can deduce cycles and patterns that likely will recur. Such stability and inertia should bode well for prediction. Nevertheless, when it comes to election forecasting, especially in the United States, most prognostications rely on short-term political fundamentals measuring macroeconomic performance or government or leader popularity. In this contribution, we adopt a structural approach but depart from existing literature by focusing on historical party and governance dynamics in the vein of de Tocqueville to establish whether they offer solid guidance regarding the performance of Democrats in US congressional elections. Our ex-post political history models provide solid predictions of which party will control Congress and the Democrats’ seat tally in each chamber between 1946 and 2022. This creates conditions to assume that political history may help us forecast Campaign 2024. Our study applied this political history model to predict the 2024 congressional elections. It forecast Democrats to lose control of the Senate with a net loss of three seats and estimated an exceptionally close race for House control, with the point estimate for the House suggesting that the Democrats would fall short of winning control.

Drawing on the case of American Muslim voter engagement in the 2024 election season, this article argues that election-forecasting models – particularly state-based models – should integrate minority populations into their analysis as crucial variables. This is of particular significance in swing states. By including minority-voter engagement and related variables relevant to them such as pressing policy concerns (e.g., anti-war sentiment and racial attitudes), forecasters can better understand and predict electoral outcomes and address the gaps identified in traditional forecasting approaches. The recommendations presented in this article help election forecasters prepare for unexpected changes, such as the American Muslim shift of support away from President Biden in the 2024 primary election season.

The initial predictions presented in this article confirm that presidential candidate vote-share estimates based on AI polling are broadly exchangeable with those of other polling organizations. We present our first two biweekly vote-share estimates for the 2024 US presidential election and benchmark them against those being generated by other polling organizations. Our post–Democratic National Convention top-line estimates for Trump (47%) and Harris (46%) closely track measurements generated by other polls during the month of August. The subsequent early September (post-debate) PoSSUM vote-share estimates for Trump (47%) and Harris (48%) again closely track with other national polling being conducted in the United States. An ultimate test for the PoSSUM polling method will be the final preelection vote-share results that we publish before Election Day on November 5, 2024.

Our 2020 analysis correctly forecasted Joe Biden’s victory and the outcome of every state except Georgia. That forecast relied on economic data from 125 days prior to Election Day and presidential approval data from 104 days (or more) before the election. Since 2000, our model would have correctly forecasted the winner in 95% of all states. We updated our State Presidential Approval/State Economy Model for 2024. This article summarizes the model and its historical accuracy as well as new data updates. We then generate forecasts for the overall two-party popular vote, each state’s outcome, and the Electoral College winner for the 2024 US presidential election. One hundred days prior to Election Day, our model forecasts a split two-party popular vote (50.3% for Trump, 49.7% for Harris) but a notable Trump advantage in the Electoral College, with slightly less than a three-in-four chance that Trump wins the election. This Republican advantage 100 days prior to Election Day sheds light on Biden’s abrupt decision to drop out of the race and suggests that if Harris wins, she will have overcome extremely challenging fundamentals, and/or that Donald Trump and the Republican Party will have squandered a sizeable Electoral College advantage.

Originally founded in 2004 to improve election forecasting accuracy through evidence-based methods, the PollyVote project applies the principle of combining forecasts to predict the outcome of US presidential elections. The 2024 forecast uses the same methodology as in previous elections by combining forecasts from four methods: polls, expectations, models, and naive forecasts. By averaging within and across these methods, PollyVote predicts a close race, giving Kamala Harris a slight edge over Donald Trump in both the two-party popular vote (50.8 vs. 49.2%) and the Electoral College (276 vs. 262 votes). The forecast gives Harris a 65% chance of winning the popular vote and a 56% chance of winning the Electoral College, making both outcomes toss-ups. Compared to the combined PollyVote, component forecasts that rely on trial-heat polls tend to favor Harris, whereas methods that rely on alternative measures are less optimistic about the Democratic candidate’s chances. The polls may be overestimating Harris’s lead.

Every four years, numerous election-forecasting models attempt to predict the results of the US presidential election. Regardless of the stability of any election system, such as the bipartisan system in the United States, conditions can arise (e.g., candidate resignations) that negatively impact forecasters’ ability to predict electoral outcomes. Citizen forecasting—that is, directly asking respondents who they think will win an election—has a long track record of successfully predicting presidential elections. This study proposes adapting a citizen forecasting measure originally intended for use in multiparty systems to predict the US presidential election in 2024. Using this measure, we created a forecast of the national-level popular vote and vote-share forecasts for seven swing states.