Speed-time data obtained in two large-scale studies by following randomly selected vehicles in a number of cities with an instrumented car are analyzed here by dividing the data into trip segments between successive stops and computing values of traffic variables for the individual trip segments. Results from the two studies are found to be in good agreement. The analysis focuses on variables previously shown to be related to fuel consumption, particularly on the relationship between energy used to accelerate the vehicle, energy dissipated in braking, and mean traffic speed. Braking and acceleration are found to play a major role in determining tractive energy requirements in low speed urban driving, since about half the energy supplied to the wheels is used to accelerate the vehicle, and about two-thirds of the resulting kinetic energy is dissipated in braking. The amount of energy dissipated in braking is largely determined by the energy used to accelerate the vehicle, suggesting that drivers do not currently reduce braking in order to save fuel. Traffic characteristics in different metropolitan areas are compared based on the computed traffic variables. Acceleration distributions at different speeds are computed.