Emerging inner diameter high velocity oxygen fuel (ID-HVOF) thermal spraying is increasingly considered for challenging wear applications for space restricted industrial parts. In thermal spray processes, coating properties are dictated by the combination of the feedstock and spray parameter choices. The aim of this case study is to apply LARS, a machine learning approach, to predict coating properties from ID-HVOF process input parameters. The LARS method was chosen here because it can be applied to high dimensional data as often encountered in thermal spray processes. Within the parameter space comprising fuel and oxidant flow rates, standoff distance, feeding conditions and surface speed the prediction addresses coating hardness, porosity and roughness as well as residual stresses. While correlations to inflight particle temperature and velocity and surface temperature will be assessed, the approach presented in this study allows direct prediction of the coating properties when in-situ diagnostics may not be possible.