This study used data from www.walkscore.com to assess walking behavior in four U.S. cities. Walk scores measuring the so-called walkability of neighborhoods are ubiquitous, and although the relationship between walk scores and real estate values has been established, the relationship between walk scores and walking has not. In this research three models were developed to understand the correlation between walk scores (as indicators of walkability; i.e., opportunity to walk) and walking. The models looked at walk scores and walk mode share for different trip types. What changes should be expected with changing walk scores along different parts of the walk score spectrum are illustrated. Results suggest that walk scores may be used as a reasonable heuristic to assist with assessing trip impacts for individual projects. With the universal availability of such data, planners can establish a consistent, cost-effective tool for assessing walking behavior with robust and transferable results.