The purpose of this article is to focus on the objective functions for traffic signal timing split optimization and present the comparative results of a related simple signal timing experiment. The feature does not compare specific traffic signal timing optimization software methods being applied today, but rather the underlying methods such software might employ. This was done to concentrate on the theory of optimization objective functions and not perform a comparison of software. The basic premise of this article is that when measuring the quality of service of signal control, the profession debated, made a conscious decision and agreed that delay of individual movements is to be used as the primary measure of effectiveness (MOE), as stated in all versions of the Highway Capacity Manual since 1985. Therefore, this same measure should be used as the basis of split optimization, and a movement delay method should be selected for optimization, not a volume-to-capacity (V/C) method.