The emergence of on-demand ride-hailing service platforms, such as Uber, Lyft and Didi, can provide innovative data sources to understand and model individual mobility behavior. Compared to the previously used data sources such as mobile phone and social media, mobility data extracted from ride-hailing service platforms contain more specific, detailed, and longitudinal information of individual travel mode and coordinates of the visited locations. In this study, using large-scale data extracted from 50,000 ride-hailing service users' mobility records, we apply an input-output hidden Markov model (IOHMM) to predict the origin and destination of the next ride-hailing service trip of an individual. The results indicate that the IOHMM model can achieve 71% accuracy for predicting the origin and 67% for predicting the destination of a trip made for commuting purposes. The IOHMM model can capture the influence of different time periods as the prediction performance of the model varies over different time periods. Since individual mobility behavior shows both regularity and uncertainty, we analyze the performance of IOHMM by investigating the predictability of each mobility sequence. We find that model accuracy is proportional to the predictability of individual movement. Using the concept of predictability, we can determine the limits of the accuracy of mobility prediction models.