Knock-limited engine operation is one of the most important constraints on fuel efficiency and performance that must be considered during the design, control algorithm development and calibration of spark-ignition engines. This research evaluates the accuracy of model-based knock prediction routines and their applicability for control-oriented applications over various engine operating conditions using commercial fuels. Two common methods of knock prediction, a generalized chemical kinetics model and an empirical induction-time correlation, are evaluated and compared against experimental data. The experimental investigation is conducted using a naturally aspirated 3.6L V6 engine, retrofitted with cooled Exhaust Gas Recirculation (EGR). Data are acquired from spark timing sweeps under knocking conditions at different engine speeds and loads in an engine dynamometer cell. The knock prediction models utilize inputs derived from experimental in-cylinder pressure data, without the implementation of multidimensional combustion modeling. Selected chemical kinetics model parameters are calibrated based on experimental data. Multiple combustion phasing parameters are also examined in an effort to provide a deterministic threshold of comparison for knock onset timing, and to distinguish between light and heavy knock events. It is concluded that no single combustion phasing parameter provides a universal threshold value for knock determination. However, knock onset location is identified as the most effective parameter, with different threshold values depending on engine speed and fuel quality. Finally, evaluation of model performance regarding knock borderline prediction through comparison with experimental data in different engine speeds, loads, EGR levels and gasoline fuel quality, reveals that the chemical kinetics model has higher sensitivity to end-gas temperature and better knock prediction than the induction-time correlation method. Over the validation range, the chemical kinetics model presents an average knock borderline prediction error of one crank angle degree of spark timing, with a maximum error of three crank angle degrees.