This dataset provides comprehensive prediction market data for the 2024 US Presidential Election, sourced from Polymarket, a leading blockchain-based prediction market platform. It offers a unique glimpse into real-time probabilistic forecasts for election outcomes across all 50 US states.

About Polymarket

Polymarket is a decentralized information markets platform that allows users to trade on the outcomes of events, effectively crowdsourcing predictions. For political events like the US Presidential Election, these markets can offer valuable insights into public sentiment and expectations.

Dataset Content

This collection includes CSV files for all 50 US states, covering various time granularities:

• Minutely data: From 2024-10-04 to 2024-11-04
• Hourly data: From 2024-07-05 to 2024-11-04
• Daily data: From 2024-03-08 to 2024-11-04
• Weekly data: From 2024-03-14 to 2024-11-04
• Monthly data: From 2024-03-18 to 2024-11-04

The data was collected up to October 5, 2024, providing a rich historical context leading up to the 2024 election.

Data Structure

Each CSV file contains the following columns:

• Date (UTC): Timestamp of the prediction
• Republican: Probability of a Republican win
• Democrat: Probability of a Democratic win
• Other: Probability of any other outcome

Potential Uses

This dataset is invaluable for political analysts, data scientists, and researchers interested in:

• Tracking the evolution of election predictions over time
• Comparing prediction market data with traditional polling methods
• Analyzing state-by-state variations in election forecasts
• Studying the impact of political events on prediction markets

Limitations and Considerations

Prediction markets reflect the collective beliefs of traders, not necessarily the actual probabilities of outcomes. The data starts at different times for different states and frequencies, which should be considered in analyses. Polymarket data is subject to market dynamics and may be influenced by factors such as liquidity and trader behavior.

Data Collection Method

The data was systematically collected from Polymarket using web scraping techniques, ensuring consistent and reliable extraction across all states and time periods. This dataset offers a unique opportunity to dive deep into the dynamics of political prediction markets, providing insights into how perceptions of the 2024 US Presidential Election evolved over time across all 50 states. Whether you're interested in political forecasting, time series analysis, or studying the behavior of prediction markets, this comprehensive collection of Polymarket data serves as a valuable resource for your research and analysis.

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.

According to an October 2024 survey, young Americans were much more likely to vote for Kamala Harris in the November 2024 presidential elections. Of those between the ages of 18 and 29, 60 percent said they were planning on voting for Harris, compared to 33 percent who said they planned on voting for Trump. In contrast, Trump was much more popular among those between 45 and 64 years old.

Surveys from swing states conducted the day before the 2024 election indicated an extremely close contest between Trump and Harris. Trump held a slight lead over of Harris in the majority of swing states.

According to post-election polling, Minnesota had the highest voter turnout among residents between 18 and 29 years old, with 62 percent voting in the 2024 presidential election. In comparison, Oklahoma and Arkansas saw the lowest youth voter turnout, with 33 percent voting in the presidential election.

On July 21, Biden announced he was ending his bid for reelection, later endorsing Kamala Harris, who is the official Democratic nominee as of the Democratic National Convention in August. Although approval of Harris was once generally low, favorability of the vice president has spiked since announcing her presidential bid. Although the race is certainly closer since Harris began her campaign, polling has fluctuated, with support for Trump increasing just days before the election. National polling indicated that the two presidential hopefuls were 0.1 percentage points apart on November 4, 2024, making it nearly impossible to predict the results. While presidential polls are generally reliable in measuring national trends, they are not infallible, particularly in close races or predictions of Electoral College outcomes.

We present Iowa Electronic Markets (IEM) forecasts for the popular vote shares in the 2024 U.S. Presidential election. We discuss the differences between IEM forecasts and polls, the impact 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.

This article considers both presidential approval and party brand differentials, as measured by the generic ballot, to forecast the 2024 U.S. presidential and congressional elections. While 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 while, 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 both win back the White House this November. By contrast, our model forecasts control of both chambers of the U.S. Congress to be essentially a tied contest.

2024 National: Trump vs. Harris | RealClearPolling

Utilizing 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.

The majority of polls in swing states Arizona, Georgia, Michigan, Nevada, North Carolina, Pennsylvania, and Wisconsin showed Kamala Harris leading former President Donald Trump in a number of the seven states surveyed. According to a November survey, Harris had a slight lead in Nevada, securing support from 48 percent of registered voters in the state, compared to support for Trump standing at 47.7 percent.

Abstract: This repository contains the full dataset and model implementation for the analysis of voting patterns in Romania's 2024 presidential elections, focusing on the relationship between territorial economic structures and electoral preferences. The models estimate vote dominance at LAU level using sectoral, demographic, and regional predictors, including spatial autoregression. Particular attention is given to the overrepresentation of Bucharest in national-level FDI statistics, which is corrected through a GDP-based imputation model. For reproducibility, the repository includes: Cleaned and structured input data (LAU, NUTS3), all modelling scripts in R, Tableau maps for visual analysis and public presentation.File DescriptionsLAU.csvThis dataset contains the local-level electoral and socio-economic data for all Romanian LAU2 units used in the spatial and statistical analyses. The file is used as the base for all models and includes identifiers for merging with the shapefile or spatial weights. It includes:- Electoral results by presidential candidate (2024, simulated),- Dominant vote type per locality,- Sectoral employment categories,- Demographic variables (ethnicity, education, age),- Regional and metropolitan classifications,- Weights for modelling.NUTS3.csvThis dataset provides county-level economic indicators (GDP and FDI) over the period 2011–2022. The file supports the construction of regional indicators such as FDI-to-GDP ratios and export structure. Notably, the file includes both original and corrected values of FDI for Bucharest, following the imputation procedure described in the model script.model.RThis R script contains the full modelling pipeline. The script includes both a model variant with Bucharest excluded and an alternative version using corrected FDI values, confirming the robustness of coefficients across specifications. It includes:- Pre-processing of LAU and NUTS3 data,- Imputation of Bucharest FDI using a linear model on GDP,- Survey-weighted logistic regression models for vote dominance per candidate,- Multinomial and hierarchical logistic models,- Seemingly Unrelated Regressions (SUR),- Spatial error models (SEM),- Principal Component Analysis on SEM residuals,- Latent dominance prediction using softmax transformation,- Export of predicted latent vote maps.Maps.twbxThis Tableau workbook contains all final cartographic representations.The workbook uses a consistent colour palette based on candidate-typified economic structures (industry, services, agriculture, shrinking).- Choropleth maps of dominant vote by candidate,- Gradients reflecting latent probabilities from spatial models,- Maps of residuals and ideological factor scores (PCA-derived),- Sectoral economic geographies per county and per locality,- Overlay of dominant vote and sectoral transformation types.

There is a vast literature on determinants of electoral turnout that allows us to forecast which groups of the population will turn out to vote, and which will not. In this study, we report on a rather unique forecasting experiment on the individual level. In June 2024, elections were held in Belgium with compulsory voting. In October 2024, another election was held, but this time without compulsory voting. Simultaneously, a panel survey was conducted, spanning from April to November 2024. The information in the first two waves of the panel were used to forecast the likelihood of individual respondents turning out again in October, which we pre-registered. The forecasting models were indeed successful in predicting who would turn out to vote, but they tend to give relatively elevated turnout likelihood scores to non-voters. The prediction models tend to underestimate the effect of political interest in explaining actual electoral turnout.

2024 Ohio Senate - Brown vs. Moreno | RealClearPolling

The gross domestic product (GDP) of the United States amounted to 27.7 trillion U.S. dollars in 2023, making it the largest economy in the G20 and the largest worldwide. China was the second largest economy in that year, with a GDP valued at 17.8 trillion U.S. dollars. It is worth noticing that while the U.S. GDP was forecast to increase by around five trillion U.S. dollars until 2027, China's GDP is forecast to grow by around 4.2 trillion U.S. dollars in the same time.

2024 Montana Senate - Sheehy vs. Tester | RealClearPolling

2024 Nevada: Trump vs. Harris | RealClearPolling

2012 General Election: Romney vs. Obama | RealClearPolling

Global Public Opinion And Election Polling market size is expected to reach $10.29 billion by 2029 at 3.7%, segmented as by mode, online surveys, paper surveys, telephonic surveys, one-to-one interviews

Broadcast TV has been the largest recipient of political advertising spending in the United States. In the 2024 presidential elections, around *** billion U.S. dollars in ad buys were forecast to be devoted to the medium. On the other hand, digital video spending was forecast to amount to *** billion dollars, down from *** billion in 2022. Connected TV (CTV) was expected to add up to *** billion dollars. The 2024 U.S. presidential election was projected to drive an all-time high in political ad spending in the country. Digital is not to be discounted in the electoral race Despite broadcast TV’s might in the political ad arena, the trend towards digital media in political advertising keeps growing. For example, governmental and political advertisers registered by far the highest ad spending on Meta between November 2018 and April 2022, for a total of more than half a billion dollars. When considering political ad expenditures on Google, the top spender between 2018 and 2023 was the Biden for President group, investing more than **** million dollars in the measured period. Political ads are flush with cash, but struggle to win voter trust The American political advertising scene shows no sign of slowing down, with ad spending for the 2024 election cycle projected to surpass its predecessors. Despite the surge in investments, voter attention to ads remained aloof during the 2022 midterms, with ** percent of eligible voters saying they tend to tune out or ignore political ads. In addition to that, around ** percent of voters considered political advertising unethical, while nearly ** percent believed consumer data should never reach the hands of political advertisers.