Walk Score

A while back, I was introduced to walkscore.com. At the time, it seemed little more than a Google Maps application that linked types of local amenities with a "walkability" score. Pluggin in "Ann Arbor" returned a score of 91/100. Not too shabby, if I say so myself. However, the website offers a lot more now than just this "get your walkability score" feature.

Since a year ago, the website has increased its analysis (by a lot) and has rated 40 cities by "walkability" of their neighborhoods. There are several no-brainers, based on the criteria of walkability: San Francisco, New York, Boston, and Chicago rank 1-4, respectively, while Indianapolis, Charlotte, Nashville, and Jacksonville rank 37-40, respectively. However, a few are a little odd (having lived in or visited these cities):

• #9 Los Angeles (score: 67)
• #11 Denver (score: 66)
• #23 Detroit (score: 52)
• #25 Tucson (score: 51)
• #28 Phoenix (score: 50)

Looking at all of these cities, one can say that the reason why the walkability scores are "off kilter" are due to the scale at which WalkScore decided to investigate each city. Having live in Denver, and visited Los Angeles, I can tell you that these are not "walkable" cities - in the context of their metropolitan areas. True, their down town regions are quite walkable, but getting to it usually involves a lot of sitting in a car in slow-moving traffic. Looking at the maps for both of these high-ranking walkable cities, you can see that their "boundaries" of consideration are much smaller than the total metropolitan area. If you were to expand the analysis into these areas, it will be quite likely that both of these cities will see their overall score (and likely their position) decline markedly. For example, if Los Angeles included the cities of Compton, Lynwood, and Carson, it would be linked to the score for Long Beach. However, even this addition would not likely bring the reality of the metropolitan area's walkability. Without including areas like Santa Monica, Beverly Hills, Burbank, Glendale, Pasadena, Alhambra, Monterey Park, Huntington Park, Inglewood, Hawthorne, Manhattan Beach, Redondo Beach, Torrance, Palos Verdes, Bellflower, Lakewood, Cypress, Los Alamitos, Garden Grove, Fountain Falley, Huntington Beach, Santa Ana, Anaheim, Fullerton, Yorba Linda, Irvine, Lake Forest, Mission Viejo, Laguna Niguel, as well as (possibly) all the cities up through San Bernadino and Moreno Valley.

Similarly, Denver's walkability is much higher than I "know" it should be. True, the walkability score refers only to Denver's city limits. However, the metropolitan area includes the cities of Englewood, Aurora, Littleton, Glendale, Columbine, Wheat Ridge, and Arvada, Westminster, Thornton, and Broomfield. Without including these commuter cities to Denver's total, it is not really possible to get a sense of the true walkability of the contiguous metropolitan area. Why, for example, include the corridor out to the Denver International Airport (when very few people live out there) as a "neighborhood" and add it to Denver's walkability score? If you were to include that, then why not include the city of Glendale, which is completely surrounded by Denver?

Similarly, I was shocked to see Detroit and Phoenix located so high in the pecking order. However, looking at their maps, you quickly see that both of these cities' analyses do not include their outlying regions and contiguous cities. For example, the highly unwalkable Scottsdale, Glendale, and Peoria aren't included with Phoenix (but for some reason the unwalkable town of Paradise Valley is). In fact, if the mostly unwalkable (outside ASU campus) Tempe were added to the analysis, then the unwalkable Mesa (rank: 30, score: 48) would automatically be folded in, as well, making us wonder why Chandler, Gilbert, and Apache Junction weren't included in the analysis... In my estimation, Phoenix metro area should be WAY lower than a walkability score of 50.

Detroit is a similar case. As of July 18, 2008, it was sitting pretty at 23rd position, and the analysis didn't take into account anything outside the actual city limits of Detroit, including the Detroic metro-area cities (in Oakland County alone) of Allen Park, Belleville, Dearborn Heights, Dearborn, Ecorse, Flat Rock, Garden City, Gibraltar, Grosse Pointe Farms, Grosse Point Park, Grosse Pointe Woods, Grosse Point, Hamtramck, Harper Woods, Highland Park, Inkster, Lincoln Park, Livonia, Melvindale, Northville, Plymouth, River Rouge, Riverview, Rockwood, Romulus, Southgate, Taylor, Trenton, Wayne, Westland, Woodhaven, and Wyandotte as well as the developed-but-unincorporated areas of Brownstown Township, Canton Charter Township, Grosse Ile Township, Grosse Pointe Township, Huron Charter Township, Northville Charter Township, Plymouth Township, Redford Charter Township, Sumpter Township, Van Buren Charter Township. If you were to add the highly unwalkwalbe contiguous Detroit-area metro cities of Wayne County and Oakland County, then D-town's walkability score (similar to Phoenix) would decline by a LOT. In fact, if you take a look at Jacksonville's score again, it takes into account a LOT of the outlying areas of the city (which D-town's analysis fails to do, let alone take into account a majority of the contiguous urban areas outside the city limits).

Conversely, Tucson is not really noted for its lack of walkability. So what's it doing below the Motor City? Well, looking at the map for Tucson, you see that Davis Monthan Air Force Base is included in the analysis (not really part of the city), as well as Rita Ranch and Houghton. If these neighborhoods (and military bases) are added to the score for Tucson, why aren't larger metro areas included for LA and Denver?