We analyze the ECB Governing Council’s voting procedures. The literature has by now discussed numerous aspects of the rotation model but does not account for many institutional aspects of the voting procedure of the GC. Using the randomization scheme based on the multilinear extension (MLE) of games, we try to close three of these gaps. First, we integrate specifi c preferences of national central bank presidents, i.e. their desired interest rates. Second, we address the agenda-setting power of the ECB president. Third, we do not simulate an average of the decisions but look at every relevant point in time separately.