The resting electrocardiogram (ECG) and stress ECG are established tests in the array of cardiovascular diagnostic modalities. In addition to their diagnostic value for structural heart disease and rhythm disorders, ECGs at rest or during stress also contain prognostically relevant information. Several ECG abnormalities, e.g., left ventricular hypertrophy (LVH), Q waves, ST segment changes, left bundle branch block, atrial fibrillation or QT interval prolongation, were shown to be associated with cardiovascular events. Differences in study design, the cohorts of investigation and morphological definitions of ECG abnormalities may in part be responsible for the abnormalities not being implemented in risk stratification algorithms. The non-ST-segment-related variables in stress testing, e.g., functional capacity, chronotropic (in)competence, heart rate (HR) recovery, and the HR/ST index and slope, could be identified as prognostically relevant markers in population-based studies. For many of these resting and stress ECG-based abnormalities, associations with the extent of subclinical atherosclerosis in persons without established coronary heart disease were observed, indicating a preclinical relationship between epicardial atherosclerosis and myocardial pathology. The resting and the stress ECG provide a number of prognostically relevant indices that can easily be obtained in routine clinical practice, but have thus far found little acceptance for risk stratification of asymptomatic individuals.