San Francisco has a Walk Score of 89. Boston sits at 83. Portland is 67, Pittsburgh 62, Des Moines 45, El Paso 40, and Raleigh comes in at 31. That's not a small range — it's the difference between a city where most errands work on foot and one built almost entirely around the car.
The number is useful. It's also incomplete. A citywide Walk Score can't tell you whether the sidewalk near your specific address actually connects to anything, whether the crossing near the school has a long signal wait, or whether the route that looks fine on a map is a mess in July heat or January ice. Here's what the index actually does, what real cities show once you compare them, and how to check the parts it can't see.
What Walk Score actually measures
Walk Score's methodology analyzes routes from an address to nearby amenities — groceries, schools, parks, restaurants, shopping — and awards points based on distance, with a decay function that reduces credit for anything farther away until it stops counting past the model's walking-time limit. It layers in pedestrian-friendliness signals too, including population density and street-network features like block length and intersection density, since a grid with short blocks generally offers more usable route choices than a highway-adjacent cul-de-sac.
A city-level score, specifically, is a population-weighted summary calculated across a grid of points, not a simple average of every address. That makes it a reasonable way to compare places at a glance — but it also means a city's headline number can be pulled up by a dense, amenity-rich core even if large parts of the metro area are far less walkable. Our Walk Score glossary entry covers the underlying definition in more detail.
It's worth cross-checking against a second source. The EPA's Smart Location Mapping program publishes a National Walkability Index built from block-group data on density, land-use mix, street-network design, and transit access — a different methodology measuring a related but not identical idea. When the two broadly agree on a place, that's a stronger signal than either one alone.
Seven cities, seven different walking lives
Line up real Walk Scores and the differences aren't subtle:
| City | Walk Score | What it generally reflects |
|---|---|---|
| San Francisco, CA | 89 | Dense pre-war grid, transit-rich, most errands reachable on foot |
| Boston, MA | 83 | Compact historic core, strong transit backbone |
| Portland, OR | 67 | Mixed — walkable inner neighborhoods, car-dependent outer ones |
| Pittsburgh, PA | 62 | Hilly terrain and distinct walkable neighborhoods separated by rivers |
| Des Moines, IA | 45 | Midwestern mid-size city, walkable downtown, car-oriented elsewhere |
| El Paso, TX | 40 | Sun Belt sprawl, built around driving distances |
| Raleigh, NC | 31 | Newer growth pattern, low density, few connected sidewalks outside downtown |
Notice that the range tracks urban form more than it tracks city size or desirability — Pittsburgh and Portland outscore much larger metros because of how their street grids and density are laid out, not because of population alone. That's the value of the index: it's a fast, apples-to-apples screen. Use it to narrow a list of candidate cities before you start looking at specific neighborhoods, not as the final word on any one of them.
Why walkability is only 15% of the WhereAtHome score
On WhereAtHome, a city's Walk Score passes straight through as its walkability component — an 83 becomes an 83-point component score, with no transformation — and it counts for 15% of the overall score when the data is available. That's a real weight, but it's smaller than housing affordability (20%), income-vs-housing (20%), or crime (20%), which means a city can still rank well overall with a modest walk score if it's strong elsewhere, and a highly walkable city can still rank lower overall if housing costs or other factors pull it down.
That's intentional, not a flaw: walkability matters enormously to some households and barely at all to others. If daily errands on foot are a top priority for you, weight it more heavily in your own comparison than the default composite does — the city comparison tool lets you look at the walkability component on its own instead of only the blended total.
That also means two cities with the same overall score can arrive there completely differently. A city with a modest walk score but strong housing affordability and low crime can land in the same range as a highly walkable city with a tougher cost of living — the total tells you they're comparable, not that they're comparable for the same reasons. Always check the component breakdown, not just the headline number.
When walkability data is missing, it's excluded — not faked
Not every city in the WhereAtHome catalog has a confirmed Walk Score yet. Liberal, Kansas, for example, doesn't currently have one on file. When that happens, the scoring model doesn't guess — it drops walkability out of the calculation entirely and renormalizes the remaining components so they still sum to 100%.
With walkability excluded, the default weights (housing 20%, income-vs-housing 20%, commute 15%, crime 20%, walkability 15%, purchasing power 10%) redistribute across the other five: housing, income-vs-housing, and crime each become roughly 23.5% of the score, commute becomes about 17.6%, and purchasing power becomes about 11.8%. The math still adds to 100%, and nothing is quietly treated as an average or "moderate" walkability score standing in for real data. If a city profile is missing a walk score, that's worth knowing before you compare it head-to-head with a fully scored city — check which components actually fed the number on the city comparison tool rather than assuming every city was scored on the same six inputs.
What the score can't see: sidewalks, weather, safety, slope
A high Walk Score doesn't guarantee sidewalk continuity, safe crossings, manageable terrain, or a comfortable walk after dark — and a low one doesn't mean nothing nearby is walkable. Pittsburgh's 62 reflects a genuinely walkable set of neighborhoods, but the city's hills mean a route that's flat on a map can be a serious climb in practice; the index doesn't grade for elevation. El Paso's 40 and Raleigh's 31 largely reflect how those cities were built — wide arterial roads and long block lengths designed around cars — but downtown pockets and older neighborhoods in both cities can still work well on foot even though the citywide average doesn't show it.
Climate matters too, and no index accounts for it. A route that's pleasant in October can be dangerous in a Des Moines ice storm or miserable in an El Paso heat wave. Accessibility is another blind spot entirely: curb ramps, crossing signal timing, surface quality, and rest points matter enormously to someone using a wheelchair, a walker, or a stroller, and none of that is captured in a citywide amenity-distance score.
Field-test your daily errands before you trust the number
Before you let a Walk Score decide anything, check it against the trips you'd actually make:
- List your real destinations — the specific grocery store, pharmacy, school, or transit stop you'd use, not "amenities nearby" in the abstract.
- Walk the actual route, door to door, at the time of day you'd normally go — a Tuesday morning grocery run and a Friday night walk home are different tests.
- Check sidewalk continuity and crossings. Note anywhere a sidewalk ends, a crossing has a long wait, or there's no marked crossing at all.
- Test it in bad weather or repeat it in a different season if the route matters year-round — one sunny afternoon isn't a full picture.
- Rate it against your specific needs — carrying groceries, managing a stroller, or using a mobility aid can turn an easy-looking route into an impractical one.
- Compare candidate addresses, not just candidate cities, on the city comparison tool or by checking an individual city page like /city/pittsburgh-pa — a citywide average can hide a huge amount of block-by-block variation.
A Walk Score is a fast way to narrow a search from dozens of cities to a handful. The three or four routes you'd actually walk every week are what tell you whether a specific address works.
