The problem of measuring distances in migration is a nontrivial, but important problem, e.g., for delimiting the concepts of population migration and residential mobility. With limited access to detailed spatial data, researchers address this problem in different ways. Only a few countries, e.g., Sweden, have the ability to calculate migration distances between point locations using Euclidean distance. This article examines the measure of correspondence of Euclidean distances (measured taking into account the curvature of the globe) to real distances along transport routes for the case of Russia. For this, 3407 pairs of distances were calculated for randomly selected 23 municipal districts in Russia directly and along transport routes—roads. Also, 411 pairs of distances were calculated for population centers with no road connections by using air service. These distances give an idea of migrations of different distances in Russia. As a result, for hypothetical relocations within the selected municipal districts, real distances on roads exceed Euclidean distances by 51%; for relocations between the centers of districts and urban okrug s within a region, by 40%; for interregional relocations, 33%. The air service used “straightens” distances, but even taking this into account, real distances (mainly long-distance, interregional) exceed Euclidean distances by 28.5%. The calculations give an idea of the deviations of real transport distances from the relatively simple Euclidean distances, which can be used in analyzing migration distances in Russia.