Both stopped and overall delay are useful measures of intersection performance. While stopped delay is easier to measure, overall delay reflects better the efficiency of traffic signal operation. Ratio of overall to stopped delay was assumed to be constant and equal to 1.3 in the 1985 “Highway Capacity Manual.” Recent improvements of the HCM formula were also based on the same assumption. It is demonstrated that for the uniform delay component, the delay ratio is in fact a function of red‐period duration and deceleration‐acceleration delay. The value of the latter parameter was measured and found to have a mean of 8.6 s. Given this value, the delay ratio is close to 1.3 only for red periods longer than 60 s for random arrivals. For the overflow delay component, the delay ratio depends also on cycle time and degree of saturation. In case of platooned arrivals, the relationship between overall and stopped delay is more complex. This case is analyzed using the step arrival flow model. New equations are propo...