Measuring the corrosion rate of a corroding metal is of interest in many situations, including monitoring industrial processes and undertaking fundamental research. The corrosion rate of a metal can be measured electrochemically by determining its polarization resistance, which is inversely proportional to the corrosion rate. What is described in this article is a technique for mathematically extracting from electrochemical noise (EN) data the polarization resistance as well as a measure of the frequency of anodic and cathodic transients. The theoretical framework for self-linear polarization resistance is based on a time-domain analysis of an electrical circuit model of an EN experiment. The analysis indicates that the polarization resistance for one electrode can be interpreted only if the second electrode alone is generating current transients during a given time record. One advantage of this approach, compared with other techniques for obtaining a polarization resistance from EN data, is that short time records, i.e. less than one minute, can be assessed. The self-consistency of the polarization resistance can be assessed with a correlation coefficient. Another advantage is that the nature of localized corrosion events can be attributed to either anodic or cathodic current transients from one of the electrodes.