The quantification of biomechanics is important for making informed decisions when implementing appropriate training interventions for improving performance and reducing injury risk. This study evaluated the viability of movement data provided by wearable inertial measurement units (IMUs) to automate the scoring of the Landing Error Scoring System (LESS). The LESS is an assessment tool used for identifying high-risk movement patterns in a double leg jump landing; however, the LESS is scored by experts using 2D video recordings, which limits large-scale screening. Movement data provided by three IMUs were used to train several out-of-the-box machine learning models, aiming to predict the result of the 17 LESS scoring items individually. Raw movement data was processed and segmented into the key phases of the movement, where additional features were derived from these segments. A supervised learning approach was taken, using a dataset containing 218 LESS scores derived from 40 participants as the desired output for each model. Comparisons were made between various subsets of features, and the features with the greatest importance on accuracy in the best performing models were extracted. Results showed limited improvements to a ZeroR approach, and features with the greatest importance on the best performing models had minimal relevance to the movement involved. Performance of LESS scoring automation using IMU data may be improved with further developments on this approach.