This paper proposes a simple electroanalytical methodology for classification of wine samples according to their grape variety (Verdejo, Airén, Albariño and Godello) employing flow injection analysis with electrochemical detection together with chemometric analysis. To this aim, a glassy carbon electrode (GCE) modified with carbon nanotubes (CNT) film was developed from optimized aqueous dispersions of this nanomaterial using polyvinylpyrrolidone (PVP) as dispersing agent. This CNT-based electrode was incorporated to a flow injection system, showing excellent electrochemical performance for the amperometric detection of polyphenols commonly present in wine. The adequate sensitivity and excellent signal stability allowed a high sample throughput analysis of white wines and the estimation of electrochemical indexes (EI) based on the applied detection potential, that are related to different polyphenol fractions in wines. The valuable information of these EI supplement that provided by other spectrophotometric methodologies developed for total polyphenol content. Finally, the hydrodynamic voltammograms of the analyzed samples were subjected to principal component analysis and discriminant analysis for classification purposes. It was demonstrated that these amperometric signals possessed some latent information that characterize the wine depending on its grape varietal. Indeed, the discriminant analysis model built using those signals obtained at +0.30, +0.40, +0.50, +0.60, +0.80 and +1.00 V was able to assign successfully the grape varietal of the analyzed samples, with a mean classification rate of 97.0% in random subsampling validation assays.