This project takes a look at the variation of the parameters related to soil erodibility (fractions of clay, silt, fine and coarse sand; organic matter, permeability, and structure) coming from soil pits from the Community of Madrid's soil map (Spain), according to Soil Taxonomy at subgroup level. It draws the conclusion that map erodibility shouldn't be estimated from a soil map because the K factor obtained does not present significant differences among the different types of soil. One or more key factors related with soil erodibility must be taken into account if erodibility maps are to be drawn. This research has shown that silt and structure could be considered key factors for erodibility maps of the area, but not significant differences have been found in important factors such as clay or organic matter due to the wide range of data variance. In order to elaborate erosion risk maps the use of the K factor from the physiographical map is a good alternative. When data are grouped according to these criteria significant differences in K factor are shown. Erodibility was greater in soils developed over gypsic material, with a value of 0.63+/-0.28, than in high plateaus (locally know as alcarrias), with a value of 0.40+/-0.18. In order to adequately represent soil erodibility, a kriging geostatistic technique is used, which reduces the variation of the factors considered when they are found to correlate, as is the case with the parameters considered to calculate K factor.