We present a new method to describe the dynamics of the beat-to-beat RR time series. The classification of the phase-space plots obtained from RR time series is performed by a calculation of parameters which describe the features of the two-dimensional plot. We demonstrate that every parameter has its specific consequence on the evaluation of the state of the cardiac function. By applying the method to the DIAMOND MI study we demonstrate that these parameters have more prognostic power than previously suggested risk markers. The results suggest that the RR intervals constitute a highly complex time series which necessitates the use of refined mathematical-statistical methods in order to reveal pathologies in the heart rate.