The constitutional amendment mechanism of the Mexican Constitution of 1857 (reproduced in the Constitution of 1917) and Article V of the US Constitution are very similar in design. Both require a two-thirds majority of each of the houses of a bicameral Congress and ratification by the states (half of the state legislatures in Mexico and three-fourths in the United States). Both articles were the result of an experiment aiming at striking the right balance between rigidity and flexibility. Yet, while characterized by similar levels of formal rigidity, these experiments have had the exact opposite effect. While the US Constitution has been described as one of the world’s most rigid and has only been amended twenty-seven times, the Mexican Constitution of 1917 has gone through over 700 amendments. Why are the amendment rates so divergent? This article argues that Mexico’s amendment practice offers an opportunity to deepen our knowledge about how non-institutional factors condition the way amendment provisions work and, thus, to dispel the idea that amendment difficulty is institutionally determined. In particular, there are at least three lessons that may be drawn from the Mexican case: (i) constitutional scholarship needs to shift its attention to political parties and party systems; (ii) unwritten rules influencing the behavior of party members need further study; and (iii) we must carefully look at the agency of constitutional decision makers, specifically regarding the choices they make among different means to advance their interests and agendas.