The predictability and probabilistic forecast skill of tropical cyclogenesis have been examined using 3 years of ECMWF ensemble forecasts. Quantitative criteria for tropical cyclone (TC) formation were determined from model analyses based on threshold values of lower-tropospheric circulation, local thickness anomaly, and minimum sea level pressure.  A binary verification was then performed for all ensemble forecasts with initial-time tropical disturbances.  The equitable threat score was small and positive, with skill mostly lost after 5 days. The reliability diagrams for 1–5-day forecasts were monotonic increasing, though an overly large number of short-range ensemble forecasts predicted a low probability of a TC when a TC was verified.  Qualitative investigations revealed case-to-case variability in the probabilistic predictions, and the predictability of associated variables.  For about half the cases examined, the predicted circulation was directly related to the predicted favorability of the environment.  For the remainder of the cases, predictability was more directly associated with the strength and location of the pre-depression disturbance.  Overall, the results show promise for the prediction of genesis out to a week, though some limitations remain.