The model of EU legislative politics developed by Garrett and Tsebelis has served as the main reference point for scholars studying the EU legislative process. It stipulates that the preferences of actors in the EU legislative process (the member states, the European Parliament and the Commission) are distributed one - dimensionally along a more - less integration continuum. Furthermore, the Garrett - Tsebelis model predicts that under the rules governing the co - decision introduced at Maastricht, the European Parliament will be systematically worse off than under the provisions of the co - operation procedure and will, as a result, invariably accept any legislation proposed by the pivotal Council member state. This article argues that the Garrett - Tsebelis model falls short of explaining why the European Parliament sometimes rejects legislation. It also demonstrates that the European Parliament may actually be more influential under the co - decision procedure than Garrett and Tsebelis suggest. By analysing different directives that are subject to the co - decision procedure - thus keeping the institutional rules of co - decision constant - it can be demonstrated that actors' time horizons (their levels of impatience) and the type and salience of issue - dimensions (more or less integration, left-right, changes in institutional rules) exercise independent effects on legislators' influence and on legislative outcomes. This article thus explains cases otherwise unexplained by the 'standard version'.