In this work we calculate a net pay map and run an uncertainty analysis on the Frontier Formation of the Teapot Dome dataset. First, we perform a preconditioning of the seismic focusing on reducing errors from noise attenuation, spectral differences, residual move out and relative phase differences, and use those results to generate the partial stacks, perform the seismic well tie, wavelet estimation and run an elastic deterministic inversion. The volumes resulting from the deterministic inversion are used as input of a Deep Feed-forward Neural Network to predict volumes of porosity, clay, and water saturation. We then use those results alongside net pay at well locations, to run a multiple attribute map analysis and collocated cokriging to estimate the net pay map of the zone of interest. Finally, these results are used for the uncertainty analysis process where we perform stochastic methods to simulate several probable model maps to make quantitative predictions. These advanced analytics methods for seismic characterization allow us to go beyond the inversion results, providing us with valuable information about the reservoir productivity.