Hi Andrew. No, my model does not explicitly account for duplicates other than the fact that I take advantage of the correlation wherein the number of applicants to Boston is fewer than the total number of BQ's that were run during the qualification period.