Fatigue cracking is a primary asphalt pavement distress and various models have been developed to predict the fatigue life of asphalt mixtures using laboratory tests. One such model is the simplified viscoelastic continuum damage (S-VECD) model, which has been implemented in the pavement performance prediction program, FlexPAVETM. The S-VECD model test protocols (AASHTO TP 133 and AASHTO T 400) and data processing tool (FlexMATTM) are widely used around the world. Over the past three decades, this model has been continuously improved and refined. However, questions remain on the model’s ability to predict the material response when stress or strain are used as the model input. Also, different fitting procedures for the model calibration were found to affect the model’s prediction accuracy. In this study, analysis was conducted using data for four typical North Carolina mixes based on single replicate tests and multiple replicate tests to compare the model’s prediction accuracy based on stress versus strain as the model input and by considering fitting errors in the different calculations. The results show that using strain as the model input, which automatically incorporates portions of permanent strain, yields more accurate predictions compared to using stress as the input, regardless of the fitting algorithm. Additionally, in the analysis of individual test data, which is not affected by replicate specimen variability, the model’s predictions match the measured data well, as long as the fitting errors of the damage characteristic curve are controlled. When the data from replicate tests are analyzed together, although specimen variability compromises the S-VECD model’s prediction accuracy, failure can still be reasonably determined when strain is used as the model input.