The paper addresses the bus crowding prediction problem based on real-time vehicle location and passenger count data and evaluates the performance of a data-driven lasso regression prediction method. The problem is studied for a high-frequency bus line in Stockholm, Sweden. Prediction accuracy is evaluated with respect to absolute passenger loads and predefined discrete crowding levels. When available, predictions with real-time vehicle location and, in particular, passenger count data significantly outperform predictions based only on historical data, with accuracy improvements varying in magnitude across target stations and prediction horizons.