When faced with unfamiliar reaction space, synthetic chemists typically apply the reported conditions (reagents, catalyst, solvent and additives) of a successful reaction to a desired, closely related reaction using a new substrate type. Unfortunately, this approach often fails owing to subtle differences in reaction requirements. Consequently, an important goal in synthetic chemistry is the ability to transfer chemical observations quantitatively from one reaction to another. Here we present a holistic, data-driven workflow for deriving statistical models of one set of reactions that can be used to predict out-of-sample reactions. As a validating case study, we combined published enantioselectivity datasets that employ 1,1'-bi-2-naphthol (BINOL)-derived chiral phosphoric acids for a range of nucleophilic addition reactions to imines and developed statistical models. These models reveal the general interactions that impart asymmetric induction and allow the quantitative transfer of this information to new reaction components. This technique creates opportunities for translating comprehensive reaction analysis to diverse chemical space, streamlining both catalyst and reaction development.