Thomas Chadefaux

I am a Professor in Political Science at Trinity College Dublin. My research focuses on the causes of interstate conflict and on their prediction.

I rely on large amounts of fine-grained spatial and temporal data (e.g., newspapers, satellite images, financial markets) to reveal early warning signals for war. My work combines methodological innovation with novel data sources to understand patterns of conflict emergence and escalation.

Forecast Lab

I direct the Forecast Lab, a research initiative focused on geopolitical risk forecasting and conflict prediction. Our lab combines transparent methods, rich data, and clear communication to support early warning and decision-making.

We provide monthly conflict risk forecasts for over 180 countries and 259,000 sub-national areas worldwide, with a six-month forecasting horizon. Using machine learning and pattern recognition, we predict conflict fatalities at both national and sub-national levels.

The lab operates as part of the PaCE project, funded by the European Research Council (grant 101002240). All our forecasts, datasets, and methodology documentation are publicly available through interactive dashboards at forecastlab.org.

Teaching

Trinity College Dublin

ETH Zurich

University of Rochester