Banking crises tend to follow periods of financial excess. Can financial variables serve as early warning indicators to inform monetary policy and financial regulation? In this paper we define an empirically-based approach to measure financial cycles and study their leading indicator properties using Early Warning System models and regressions. We find that financial variables’ predictive power is on par or better than that of real sector variables. Equity prices in particular offer the best signal, followed by property prices. By contrast, credit, though widely seen as a strong crisis predictor, does not offer the clearest signals. Aggregating financial cycle information in an overheating index tends to improve prediction in real time and underscores the need to look at a wide range of indicators.