The King County Right Size Parking Calculator can help support and guide decisions about parking supply and management. It cannot provide definitive answers about specific future policies or developments, but can serve as a resource to inform discussions as users weigh the factors affecting parking use and consider how much parking to provide.
The model used in the calculator is statistically very strong, but it is important to note that it is just a model and there is always error in estimates (the standard error for this model’s estimates is 0.16). Limitations on our data collection also affect the model’s accuracy. Observed parking mostly included supply that was on-site and off-street, unless additional parking provided for residents was noted by property managers. However, the sites selected for the study were screened based on building age and available parking supply, in order to control for potential undersupplied parking that could result in spillover. The result was sites studied whose predominant parking could be measured through parking counts, rather than those where undefined off-site parking would have resulted in an underrepresentation of parking use.
In addition, real-world parking use can and will vary from these estimates for many reasons. Actions can be taken to reduce parking use below the levels predicted by this model. In addition to non-typical development opportunities, transportation demand management strategies, including the provision of residential transit passes, car sharing memberships, and bicycle facilities, have been shown to reduce car ownership and parking use. Some of these strategies were tested as the model was developed, but not enough cases were available to test with statistical rigor, so the calculator does not capture their influence.
To ensure confidence in the model estimates, limits were established for the coverage area. The sample utilized for data collection covered a wide range of built environment characteristics and land uses, but it did not cover the full spectrum found throughout the county. Therefore, the coverage for which model estimates were calculated was limited to range of built environment characteristics found in the data collection sample. In other words, areas of the county that had lower transit service, population, or job concentrations than those found in the sample were removed from the coverage area.