One of the major technical obstacles to the practical use of biodiesel fuel is its cold flow properties. Although attempts have been made to determine the correlation between the cold filter plugging point (CFPP) and the fatty acid methyl ester (FAME) profiles, the proposed models are valid only for certain combinations of feedstock oils. In this study, the contributing coefficients of individual saturated FAMEs used in predicting the CFPP were quantified statistically for the first time. Quantification was based on 303 most widely used biodiesel blends (125 from this work and 178 collected from previous studies) of 15 edible, non-edible, or low-molecular-weight oils and animal fats. Results based on a stepwise multiple regression method (Model 1) indicate that the amounts of myristic (C14:0), palmitic (C16:0), stearic (C18:0), and arachidic (C20:0) acid methyl esters significantly influence the CFPP. Considering unconverted monoglycerides as another independent variable for the stepwise analysis, the results (Model 2) indicate that the statistically significant variables are the same as those in Model 1. In order to improve the predictive power of and to reduce the number of parameters in the Models 1 and 2, several modified correlations (Models 3-5) were also established by stepwise analysis, especially for blends containing babacu/coconut methyl esters or a large amount of rapeseed methyl esters. Through these correlations, the optimum FAME profile and blends of common biodiesel feedstocks that result in a satisfactory CFPP can be determined from their C16:0, C18:0, and C20:0 content. (C) 2017 Elsevier Ltd. All rights reserved.