This page is easier to use with a wider screen. Rotate your device or make the browser window wider for the best map and controls experience.
The Bike Friendliness Index estimates how easy it is to get around by bike in hundreds of communities around the world. It uses OpenStreetMap data on roads, bike infrastructure, and destinations cyclists may want to reach to estimate how well each community’s bike network allows cyclists to comfortably go where they want to go.
Other bikeability ratings tend to focus on cities famous for biking, or only compare cities within a specific country. Most also treat city limits like hard walls and don't measure how well infrastructure connects to destinations and the rest of the local bike network. That can miss what it actually feels like to live in a suburb or a smaller or slightly less bike-friendly city. This index is built to work across many communities using publicly available data, while estimating how far someone can realistically bike without relying too much on stressful or risky riding.
Because bikeability depends on why you ride and what you are comparing, each location gets several types of scores. Infrastructure Focus emphasizes recreational biking, Destination Focus emphasizes transportation and errands, and Balanced Focus weighs both equally. Scores are also shown as “City Wide” and “Best Area.” City Wide averages the quality of the cycling network at many points across the city. Looking at the whole city is the traditional way to score cities, but City Wide differs because it includes easy-to-access infrastructure and destinations outside of city limits as well. "Best Area" takes things one step further and only averages the higher scoring areas, which helps to avoid penalizing cities just because their official borders include large car-dependent suburban/rural areas that most people would not choose when looking for a bike-friendly neighborhood. Best Area also is a more useful metric when deciding where to live.
This was a lot of work, so I hope you enjoy! This site is still a work in progress as I tweak things and add new locations. If you have any suggestions or questions, feel free to contact me at comments@bikefriendlinessindex.org. Check out the Methodology to see more in depth how the program works, and check out "Want To See More?" to see the full power of the analysis program for localized insights. Thank you to Momentum Mag for the coverage.
The Inspiration
This project started with a simple problem: I wanted to find a genuinely bikeable place to live. I was disappointed with existing bikeability ratings because they typically focus on a select few very bike-friendly cities, or focus on a single country. Additionally, if they evaluate infrastructure at all, they tend to look strictly within municipal borders. That can miss what biking somewhere is actually like. For example, a quiet neighborhood with minimal internal infrastructure might sit directly adjacent to a world-class, multi-city trail system. To address these blind spots, this tool tries to measure what it is actually like to get around a community you live in by bike.
Data Sourcing
The index is built from crowdsourced OpenStreetMap (OSM) data collected through the Overpass API. Beyond just roads and paths, all relevant data is collected, from speed limits to intersection signage to path surface materials. By compiling this raw data, the tool builds a complete graph of the local cycling network.
Stress Scoring
With the network graph, the algorithm assigns a stressfulness penalty to every segment, with over a dozen possible penalties factored in. High-speed arterials receive severe penalties, while painted lanes reduce stress, and protected cycleways are considered extremely low stress. It also estimates the danger of an intersection based on speed limits and how it is controlled, factoring that into the cumulative stress.
Pathfinding
The core simulation uses Dijkstra's algorithm to model a cyclist's journey outward from a starting node. The tool attempts to simulate riding in all possible directions, measuring the stress along the way. If the cumulative stress of navigating fast traffic, dangerous intersections, or poor surfaces exceeds this limit, the path is cut off. This is used to calculate several metrics, such as how far a cyclist can travel in a single direction and what percent of their route is bike infrastructure. While all paths start inside the city, they may leave the city while simulating the cycling, incorporating data that alternatives do not.
Destination Value and Amenity Scoring
A safe bike network is only as good as the places it connects. The tool cross-references the safely reachable network with a database of local Points of Interest (POIs). Amenities are weighted by how important they are likely to be to the average cyclist: grocery stores, schools, transit hubs, and bike shops carry heavy weight, while restaurants and parks offer moderate value, and a car mechanic is completely excluded. By prioritizing locations that serve daily livability over tourist destinations, the tool outputs a "Destination Score" that quantifies the practical utility of the local network.
Scoring
Because bikeability means different things to different people, the program generates nine distinct scores for every city. It uses three weightings: Infrastructure prioritized (for recreational riding), Destination prioritized (for commuting and errands), and a Balanced approach. It applies these across three scopes: a traditional City Wide average, a Best Area score (highlighting the top 20% most bikeable neighborhoods), and a High Point score (the 95th percentile, as the highest point or two can sometimes be an unrealistic living location). There is an “Address Mode” where you can score any single address, and the main "City Mode," that runs dozens to hundreds of Address Mode calculations spread out across the city and averages the scores together.
Contextual Accuracy in Practice
The strengths of the methodology are visible when comparing locations on alternative indexes. For instance, Mackinac Island (a small tourist town that bans cars) scores a perfect 100 on other indexes, while a small inner suburb like Grandview Heights, Ohio scores poorly due to minimal bike-specific infrastructure within the suburb. However, Mackinac Island has very few amenities, limiting practical daily living. On the other hand, Grandview Heights borders a world-class, grade-separated path offering stress-free cycling to a major downtown, university, and surrounding suburbs, making the suburb more practical for the typical cyclist.
To make this massive dataset accessible, this site currently visualizes data for many of the analyzed communities, with a focus on cities with at least 200,000 people. You can explore the findings using three distinct viewing modes:
The program does have some inherent limitations. Its accuracy relies heavily on the completeness of OSM data. When metadata like speed limits or surface types are missing, the algorithm must use estimates. Some cities are currently lacking full compatibility for analysis, as they do not have a boundary on OSM (such as Albuquerque and quite a few cities in developing countries) or because their boundaries are very large (a common issue particularly with Chinese cities). Additionally, because the tool just measures physical infrastructure and destinations, it cannot include cultural factors such as driver awareness, local cycling volume, or municipal policy. Variables like weather and elevation are currently excluded, although they may be possible to eventually integrate.
Despite these limitations, the index is one of the best ways to compare bikeability across many places around the globe.
While the public database provides an overview of the data with a focus on cities over 200,000 in population, the Bike Friendliness Index can also process other areas or provide additional data for a location upon request. You can request the addition of any compatible city, suburb, town, neighborhood, county, or other small political region to the database. To verify compatibility, simply search your desired location using Nominatim. If the search returns a polygon boundary roughly the size of a town or city, the area can be processed. If it just returns a point, a very small boundary (like the size of a house) or a very large boundary (like the size of a state), it is not currently supported. I'm currently accepting one free request per person. If you are requesting multiple locations, I ask for a donation of $5 per additional location. Email comments@bikefriendlinessindex.org with the Nominatim link(s) for the boundary/boundaries you want added, or for questions/help with this process.
Requesting a Custom Analysis: You can request a Full City Analysis for any city currently on the map or a newly requested area. This deep dive provides all the data for the location. This includes the value for the ~2 dozen stats, both overall and for every individual calculated point (typically at least 1 point per square mile). These points can be viewed on a map to get an idea of which neighborhoods are best to live in. It also includes the comprehensive data overlays. These overlays display:
You can also request an Address Analysis for any specific address or set of coordinates worldwide, regardless of boundary compatibility (provided it is adjacent to a bike-permissive road or path). Address Mode includes the four main data overlays from City Mode, plus three specialized visualizations unique to Address Analysis: a cumulative stress map detailing how difficult it is to reach surrounding areas, a tree visualization highlighting ideal corridors to take, and a boundary visualization of the total safe area within your reach.
Custom analyses require a moderate amount of work for me to prepare, so I ask for a donation for a Full City Analysis or Address Analysis (minimum $10 per Full City or Address Analysis) to help cover the time needed to process areas, prepare custom results, and keep the project running. If you are on a computer, you can check out a handful of sample cities at bikefriendlinessindex.org/viewer. Email comments@bikefriendlinessindex.org with questions/requests.
I also greatly appreciate donations from anyone who just wants to support the project and help keep it running!