Objective To analyze the origin, structure, relationships, and recent admixture in Mexican Native groups based on 15 STRs commonly used in human identification. Methods We analyzed 39 Mexican Native population samples using STR databases based on the AmpFlSTR® Identifiler kit (n = 3,135), including Mexican‐Mestizos (admixed), European and African populations, as reference. Results Based upon effective population size (Ne) differences, Native groups were clustered into three regions: i) Center‐Southeast groups, characterized by larger Ne, migration rate (Nm), genetic diversity (He), and relative homogeneity principally in the Yucatan Peninsula; ii) Isolated southern groups from Chiapas and Oaxaca, characterized by lower Ne, Nm, and He (i.e. higher isolation and genetic differentiation); iii) North‐Northwest groups, which are similar to the previous group but are characterized by generating the widest gene flow barrier in the Pre‐Hispanic Mexican territory, and currently by elevated admixture in some northern Native groups. Despite the relative congruence between genetic relationships with cultural, linguistic, geographic criteria, these factors do not explain the present‐day population structure of Native groups, excepting in those linguistically related to the Mayan that show higher homogeneity. The Isolation by distance model was demonstrated at long distances (>1,500 km), whereas geographic isolation stands as a determining factor to avoid both non‐indigenous admixture and bottleneck processes. Conclusions Different dynamics of gene flow and drift were observed among Mexican Native groups, highlighting the geographic barriers (mountains, canyons and jungle regions) as the main factor differentiating Pre‐Hispanic populations, and eventually helping to avoid Post‐European contact admixture and population bottleneck. Am J Phys Anthropol 160:298–316, 2016. © 2016 Wiley Periodicals, Inc.