In this study, we perform horse gait classification using mobile phone sensors located in the rider’s pocket. We train a classifier using data obtained simultaneously from a mobile phone and a gait label acquired from a device based on four wearable sensors attached to a horse’s limbs. With this efficient approach to acquire labelled data, we train a gait classification model and demonstrate that data from mobile phones can be used to classify the five gaits of the Icelandic horse with up to 98% accuracy. The result suggests that horseriding activities can be studied at a large scale using mobile phones to gather data on gaits.