Quantitative evaluation of cardiac function is very important in the clinical application of a ventricular assist device. This article reports a new evaluating method of E max, which is the most reliable parameter to evaluate cardiac function. Fluctuation in the E max time series data was evaluated by the nonlinear mathematical analyzing method including chaos and fractal theory. Experimental goats were anesthetized with halothane inhalation, and left ventricular volume and pressure were measured with other hemodynamic parameters to evaluate E max during various drug administrations. E max was evaluated by two methods. One was the conventional pressure volume loop evaluation and the other was the parameter optimization method without left ventricular volume data. As a result, E max evaluated by the parameter optimization method correlated well with the E max with conventional PV curve. Furthermore, interesting results were obtained. There were rhythmical fluctuations in the E max time series data. By the methodology of Takens, E max time series data was embedded into the phase space and a strange attractor was observed. These results may be important when considering E max evaluation during left ventricular assistance.