In this paper we utilize data on the head-to-head loss rate for students accepted at Williams College, but who opt to enroll elsewhere. For example, we employ data that measure the fraction of students admitted to Williams and to Amherst (or Harvard or Yale, etc.) but who opt to attend Amherst (or Harvard or Yale, etc.) instead of Williams. We then model this head-to-head loss rate using data from a variety of sources. A better understanding of the head-to-head loss rate can assist an institution in the competition for high quality students. Importantly, it can also shed light on the degree to which some part of the loss rate might be due to "intangible" differences between the schools being compared. These intangibles (positive or negative) might grant a school greater success (or failure) in the market for students than an objective accounting of its characteristics might suggest. Such an advantage (or disadvantage) is closely aligned with the business concept of "goodwill." We present preliminary evidence on how a quantitative measure of educational goodwill can be computed.