Herein, a method to calculate the total sulfur concentration in petroleum samples from the chemometric modeling of data obtained by positive-ion atmospheric pressure photoionization Fourier transform ion cyclotron resonance mass spectrometry [(+) APPI FT-ICR MS] is described. Analysis by FT-ICR MS provides both a measurement of the total sulfur concentration and detailed molecular-level speciation of sulfur-containing compounds. A total of 30 crude oil samples ranging from 0.2 to 4.6 wt % sulfur were employed to train the sulfur prediction model. The ratios of the percent relative abundance (% RA) between the sulfur classes (S-x) and the hydrocarbon (HC) class were employed as variables for principal component analysis (PCA). The PCA results reveal a highly linear trend along the principal component with the highest explained variance (PC1). Analysis of the loadings plot reveals that the S-1/HC ratio governs the trend in PC1. Values for PC2 are governed by S-1/HC, S-2/HC, and S-3/HC ratios and provide the ability to distinguish between oils with higher total sulfur contents (greater relative abundance of S-2- and S-3-containing compounds for sulfur of >1 wt %). Thus, these results indicate that the sulfur concentration of crude oils can be modeled by a linear combination of variables based on S-x/HC ratio(s). The model successfully predicted sulfur concentrations for 11 test samples within 0.36% standard deviation, as compared to sulfur concentrations obtained from bulk elemental analysis.