This study examines the relationship between walkability estimates including Walkscore and a 10-year sample of hotel transactions in New York City. Using a Hedonic pricing model, ordinary least squares (OLS) regression applied citywide initially produced significant positive relationships between walkability estimates and transaction value. However, the associations became more obscure once submarket fixed effects were introduced to control for unobserved differences between neighborhoods. More granular analysis of walking accessible destinations revealed that accessibility to certain destination categories like entertainment can have a negative impact on hotel value. The results suggest that builtenvironment pedestrian friendliness more consistently benefits hotel value compared to accessibility-based walking potential. This study also finds that while high value hotels are often found in areas with high walkability, hotel value premiums in these areas may not be attributable to walkability and can arise from other unobserved neighborhood characteristics. The study concludes by questioning the ability of current walkability estimates to accurately measure walking behavior of travelers.