Despite many recent studies into the behavior of the complex fractionated atrial electrograms (CFAEs) that characterize atrial fibrillation, there has been relatively little success in analyzing them to reliably identify and ablate the rotor sources of atrial fibrillation in real time. The objective of this study is to propose a new method of analyzing and displaying CFAE signals to help cardiologists identify likely rotors during ablation procedures. This method uses cross-correlation to compare the electrograms collected at individual electrodes, and calculates the time lead/lag for each electrode compared to a local reference. The results of the analysis are displayed on a 3D color map using coordinate data collected alongside the electrogram data. Atrial fibrillation CFAE data was collected from patients with one of two different types of electrode catheter, and a data set from a patient with atrial flutter was evaluated using the same methods. The cross-correlation method was confirmed to be working as designed when the atrial flutter lead/lag color map matched the one created by the cardiologists who collected the data. Several AF wave fronts were identified in the data as a lead to lag shift. By varying the length of electrogram data used through the analysis, it was determined that CFAEs have irregular timing and cannot easily be compared in large time sections versus smaller time sections. The unstable CFAE timing also led to the conclusion that nonlocal references cannot accurately correlate with the data, which may be due to the unpredictable flow of activation waves in the heart. The irregular timing of many of the electrograms points to regularity of CFAEs in an area as a potential indicator of rotor centers. Collecting the data using unipolar electrodes rather than bipolar electrodes was found to give much better results, with many instances of wave front activity found, as well as better agreement between large and small sections in some cases. With its ability to identify AF wave fronts, this method has potential as a tool to be used to locate and identify likely AF sources in real time.