Carbon anodes are made by baking a compacted mixture of calcined petroleum coke, recycled anodes and butts, and coal-tar pitch. Pitch is utilized as a binder for the dry aggregate. Good interaction between coke and pitch is essential for the generation of a satisfactory bond between them, and the contact angle is a measure of this interaction. A program was developed using the linear multivariable analysis to predict the contact angle for a given coke/pitch pair at different times. Normalized coke and pitch properties and their corresponding contact angles were used to train the program. The precision of the model depends on the amount of data available for training. The value and the sign of the weighting factors indicate the effect of the input parameters on the contact angle. The model gives information on the effect of raw material properties such as composition and impurities on the contact angle.