On the relationship between intelligence and creativity, a classic theory is that intelligence is necessary but not sufficient for creativity. Graphically, this theory is represented by a triangular shape of bivariate scatter between the two. As conventional linear methods are known to be inappropriate, a long-standing problem has been how to substantiate this theory. One innovation purported to solve this problem is the use of Necessary Condition Analysis, a method that confirms the relationship on the basis of an empty upper left corner in the scatterplot. The present article elaborates a novel take on this methodological problem. What it takes to account for necessity and sufficiency is tackled, and it is argued that Necessary Condition Analysis is not an appropriate method. As an alternative, a probability model of creativity as a function of IQ was posited, in particular for double-bounded creativity variables. Using the model proposed, intelligence versus creativity data from Jauk, Benedek, Dunst, and Neubauer (2013b) were reanalyzed. A formal hypothesis based on the theorized relationship was supported for one of the two creativity variables analyzed. (PsycINFO Database Record (c) 2018 APA, all rights reserved)