Although the Late Pleistocene–Holocene fluctuations in the Black Sea have been studied over many years, no overall picture of this difficult natural phenomenon has as yet been developed and described. Moreover, in the literature, many researchers have provided mutually exclusive opinions about many of the processes occurring within this time interval. The variation in conceptualizations about the filling of the Black Sea basin is explained by the following reasons. (1) Fluctuations of sea level in the basin were never properly studied as physical processes under conditions of increased freshwater runoff, change in Bosporus strait depth, increase in World Ocean level, etc. Therefore, some conclusions about sea-level changes were not verified by appropriate calculations and have not been physically substantiated. (2) The study of sea-level fluctuations are conducted according to modern sea depth markers and sediment age, but as the sediment was formed under varying depth conditions, and could have moved relative to modern markers by the agency of various geological processes, it is possible that that every core may give a different level curve. (3) Processes of sedimentation, erosion and redeposition of sediment occurring in the sea have been studied insufficiently. For this reason the same geological material has been interpreted differently by various researchers. Therefore, mathematical modeling of sea-level fluctuations based on various physical laws seems to be of great importance. In this paper, the value and sign for the Black Sea freshwater balance during the Upper Pleistocene–Holocene are discussed. The hypothesis suggests that during the specified period of time, the freshwater balance of the sea was always positive, the sea was filled with water up to the sill mark in the Bosporus strait, and a river flowed into the Sea of Marmara along the strait bottom. In the present paper, geologic evidence of the river flow in the strait during the glacial age and later is presented. Other points of view about the sea's freshwater balance change are considered. A mathematical model of the Black Sea basin filling with freshwater in the Upper Pleistocene–Holocene is suggested. It considers that change in water volume inflowing to the Black Sea to be a result of ablation, neotectonic processes in the strait, bottom erosion, and sediment accumulation. Black Sea level change calculations have shown the following. As a result of repeated increases in river runoff during glacial melting, the level of the Black Sea started to rise from −80 m, probably to −20/−30 m, making, in the process, numerous secondary fluctuations. The reason for this phenomenon was the circumstance that the water volume brought by the rivers could not flow to the Sea of Marmara through the narrow canyon of the Bosporus strait at the time. Therefore, water collected in the sea, raising its level. At approximately 12 ka BP, the World Ocean level rose to the river surface and began to increase its depth. As a result of the increase in strait depth, the accumulated water in the Black Sea flowed out, thereby lowering its level to a mark close to the Ocean level. According to calculations, this occurred ca. 11 ka BP. Thereafter, the Black Sea level rose together with the Ocean level. Water within the Sea of Marmara has flowed to the Black Sea since approximately 9 ka BP in the form of a bottom counterflow. With a small time delay, theoretical change in the Black Sea level practically corresponds to data obtained from geological investigations. The theory offered here of fluctuations in Black Sea level is self-sufficient; it explains all processes from the melting of the glaciers, by physical laws and does not demand the application of any additional hypotheses. In the conclusion of this paper, the debatable questions about the Black Sea's possible lowering and the subsequent flooding of the basin are considered.