We study the predictive capabilities of magnetic-feature properties (MF) generated by the Solar Monitor Active Region Tracker (SMART: Higgins et al. in Adv. Space Res. 47, 2105, 2011) for solar-flare forecasting from two datasets: the full dataset of SMART detections from 1996 to 2010 which has been previously studied by Ahmed et al. (Solar Phys. 283, 157, 2013) and a subset of that dataset that only includes detections that are NOAA active regions (ARs). The main contributions of this work are: we use marginal relevance as a filter feature selection method to identify the most useful SMART MF properties for separating flaring from non-flaring detections and logistic regression to derive classification rules to predict future observations. For comparison, we employ a Random Forest, Support Vector Machine, and a set of Deep Neural Network models, as well as lasso for feature selection. Using the linear model with three features we obtain significantly better results (True Skill Score: TSS = 0.84) than those reported by Ahmed et al. (Solar Phys. 283, 157, 2013) for the full dataset of SMART detections. The same model produced competitive results (TSS = 0.67) for the dataset of SMART detections that are NOAA ARs, which can be compared to a broader section of flare-forecasting literature. We show that more complex models are not required for this data.