Most methods to determine realized total energy savings at national or sectoral level make choices, or neglect problems, which hamper the calculation of sound and useful energy-saving figures. Issues are the choice of the right aggregation level, the appropriate variables to construct a reference energy consumption trend, the energy quantities to be applied and interaction between various effects. Uncertainty margins for results lack in most presentations as well. This paper presents six methods, illustrates the adverse effects of certain choices and problems, and investigates how these methods deal with them. The methods are scored with respect to the issues mentioned above. Finally, a number of improvements are suggested, among which the use of final energy demand expressed in primary energy units, and bottom-up analyses at the level of real saving options. The last option is the more important, as it could provide top-down evaluation results (total savings from decomposition) as well as bottom-up policy monitoring results, both being crucial to new European energy-saving policy.