Minimum lap time problems are usually solved employing quasi-steady-state models on a predetermined (fixed) trajectory or employing dynamic models on a free (i.e. not predetermined) trajectory. This work describes a third approach, where the minimum-lap-time problem is solved using quasi-steady-state models and free trajectory. The method builds upon g-g maps that can either be derived numerically or experimentally. Such g-g-speed surfaces can either represent the performance of a car or a motorcycle. Both a double-track car model and a motorcycle model are employed as examples of applications. The effect of the free-trajectory vs. fixed-trajectory assumption is also discussed.