Computational algorithmic thinking (CAT) is the ability to design, implement, and assess the implementation of algorithms to solve a range of problems. It involves identifying and understanding a problem, articulating an algorithm or set of algorithms in the form of a solution to the problem, implementing that solution in such a way that the solution solves the problem, and evaluating the solution based on some set of criteria. CAT is an important scaffolded on-ramp as students develop more advanced computational thinking capabilities and apply computational thinking to solve problems that are more constrained and require greater expertise. Supporting Computational Algorithmic Thinking (SCAT) is both a longitudinal between-subjects research project and a free enrichment program supporting and guiding African-American middle school girls over three years as they iteratively design a set of complex games for social change. This article explores Scholars’ reflections about the difficulties they faced while using CAT capabilities as they engaged in collaborative game design for social change over those three years. We particularly focus on how these difficulties changed over the course of three years as well as new difficulties that emerged from year to year as Scholars become more expert game designers and computational algorithmic thinkers.