In this paper frequency response, one of the most critical ancillary services, provided by grid-scale battery energy storage systems is studied with manipulating the optimum battery state of charge control area. An algorithm has been developed that primarily provides frequency response service and at the same time optimizes the battery state of charge. The difference between the algorithm developed in this study and the other algorithms is that, apart from the maximum and minimum points of the state of charge level, it has also the optimum highest and optimum lowest state of charge points. Thus, it is aimed that the battery will support the grid at the most extreme points in urgent cases and decides the power output it will provide according to the battery state of charge in the case of non-urgent cases. The algorithm used the one-second resolution 1-day frequency data which has the highest standard deviation in a month obtained from Turkey National Transmission System Operator's database as input data. Assuming a facility in Turkey which has a battery energy storage system that participates in the grid frequency response ancillary service, the output power and the battery state of charge graph are examined.