Unmanaged hybrid battery/supercapacitor energy storage systems possess higher performance with lower cost and complexity compared to not only individual cells, but also electronically managed hybrid systems. Achieving full performance requires the understanding of the power distribution and predicting their best combinations. In this work, a semi-empirical modeling methodology is presented that can predict the current distribution and the voltage response of battery/supercapacitor hybrid systems under arbitrary charge/discharge profiles. Results are presented for the assessment of hybrid systems under real life scenarios. The key strength of the presented method is that it is free of any parametrization, fits or subjective inputs. The modeling methodology is validated with experimental measurements for two different Li-ion battery chemistries, namely Lithium Iron Phosphate and Lithium Vanadium Pentoxide, connected in parallel to wide range of supercapacitors. Finally, we outline several design rules for hybrid storage systems for different use cases.