It has been recognized for some time that calculations of potential encounter rates based on the speed and lateral distributions of ships using sea routes provide only a partial estimation of the actual collision risk, as all encounters are usually given equal rating, and no attempt is made to include factors such as the degree of domain penetration or the relative speed of the ships. One possible solution to the problem consists in attempting to weight the calculated potential encounter rates according to the most relevant factors, and thus replace the calculated rates by weighted values to provide a relative estimate of collision risk. Goodwin, Lamb and Kemp reported a first attempt at such a procedure in their calculation of weighted encounter rates based on the observed traffic distributions and flow rates in the Sunk Lightship area in the neighbourhood of Harwich. A two-part index was used in the work; the first part employed a weighting factor based on the ratio of the time required for the completion of a manoeuvre involving a specified angle of course change, to the time available as measured from the ‘decision-distance’ at which the manoeuvre was deemed to commence. The second part used a factor based on the time required to pass and draw clear after completion of the manoeuvre.