Environmental pollution is a negative externality of urbanization and is of great concern due to the fact that it poses serious problems to human health. Pollutants, such as heavy metals, have been found in urban road dust; however, it is unclear whether the urban form has a role in its accumulation, mainly in cases where there is no dominant unique source. We collected 482 samples of road dust, we determined the concentrations of five heavy metals (Cr, Cu, Pb, Zn, and Ni) using inductively coupled plasma optical emission spectrometry (ICP-OES), and then we derived the pollution load index (PLI). After estimating the mostly anthropogenic origin of these pollutants based on global levels of reference, there were two main aims of this study. Firstly, to analyze the spatial correlation of heavy metals, and secondly, to identify the main factors that influenced the heavy metal concentrations in the road dust of Mexico City. We did this by using a spatial autocorrelation indicator (Global Moran’s I) and applying ordinary least squares (OLS) and spatial regression models. The results indicated low levels of positive spatial autocorrelation for all heavy metals. Most variables failed to detect any relationship with heavy metals. The median strip area in the roads had a weak (significance level of 90%) but consistent positive relationship with Cr, Cu, Ni, Pb, and the PLI. The distance to the airport had a weak (significance level of 90%) and inverse relationship with Pb. Manufacturing units were associated with an increase in Cu (significance level of 95%), while the entropy index was associated with an increase in Ni (significance level of 95%).