A method for evaluating safety performance of an autonomous vehicle including comparing first sensor data characterizing a driver-operation of a vehicle to a first threshold to obtain a first driving quality value, comparing second sensor data characterizing an autonomous-vehicle-operation to a second threshold to obtain a second driving quality value, and determining, based at least in part on a comparing of the first and second driving quality values to each other, a safety performance of the autonomous-operated vehicle relative to the driver-operated vehicle, and a corresponding system.