Blog
23.02.2025

Data Science Meets Democracy: Leading Experts Analyse German Federal Elections

Shortly before Germany's federal election, a fruitful discussion took place at the Hertie School, where leading experts analyzed campaign strategies and forecasted post-election scenarios through the lens of data science

Some unexpected insights at the Hertie School’s pre-election panel, where political scientists broke down what to expect in today’s federal election. Prof. Simon Munzert (Hertie School), Prof. Professorin Andrea Roemmele (Hertie School), and Prof. Thorsten Faas (FU Berlin) shared their latest take on the campagin—right as Germany heads into a potentially historic vote.

Prof. Munzert’s forecasting project, zweitstimme.org, currently sees the CDU/CSU leading with around 29%—but with several parties on the fence of joining parliament, coalition-building remains the big question. Moderated by Anke Plättner, the scientists shared some key insights, including:

  • How unusual the campaign was – very short, with very unpopular candidates, and remarkably little movement in the polls despite major events
  • How young voters differ from older generations in their preferences, and why many of them are likely to be frustrated by the Bundestag election
  • How voters’ behavior at the urns might be less likely to be affected by published forecasts than we might think

With final forecasts now locked in, we’ll soon see how well the numbers match reality. Exciting times for political research and election watchers alike!

  • Simon Munzert, Professor of Data Science and Public Policy | Director, Data Science Lab