Circuity: How Road-Distance Estimates Work
What circuity means
Circuity is the ratio between the distance a road actually travels and the straight-line distance between its endpoints. A perfectly straight road between two cities would have a circuity of 1.0, and no real intercity route achieves it: roads curve with terrain, serve intermediate places and meet at interchanges arranged for a network. Measured across many corridors, typical intercity circuity in the United States falls roughly between 1.1 and 1.4. Mountain crossings, peninsulas, bays and sparse desert networks push toward the top of that range; a direct interstate across flat farmland sits near the bottom. The concept matters because it converts a number we can always calculate â the straight line â into a defensible estimate of a number we cannot always route.
Why this site uses 1.25
Where a pair has no measured OpenStreetMap routing figure, Miles Between Cities estimates driving distance as the straight-line distance multiplied by 1.25, and estimates drive time by dividing that road distance by an assumed average of 50 miles per hour. The factor sits in the middle of the typical range deliberately: it will modestly overstate a very direct corridor and modestly understate a very indirect one, and across a whole index of pairs those errors roughly balance rather than systematically flattering or alarming the reader. Both the factor and the assumed speed are published here and on the methodology page, because an estimate whose recipe is secret is just an opinion wearing a number's clothing.
Where estimates are weakest
Estimates deserve extra caution in exactly the places circuity varies most. Short urban hops can have circuity well above 1.25 when a river or bay forces a bridge detour, and their drive times depend more on city traffic than on any average speed. Very long routes that funnel through a single mountain pass or a single hub city can also bend further than the factor expects. Island and roadless destinations defeat the method entirely, which is why this site publishes no driving figure at all for pairs with no road connection, instead of manufacturing one. As measured OSRM routes are added for more pairs, estimates are replaced â a measured figure always displaces an estimated one on the pair page and in the hub tables.
How to use an estimated figure well
Treat an estimate as the centre of a range, not as a point. For fuel and time budgeting, allow roughly fifteen percent either side of an estimated driving distance until a measured route or a live directions check tightens it. For comparing two possible destinations from your city, estimates are usually good enough, because the same method was applied to both sides of the comparison. And for the final plan â the departure time, the overnight stop, the arrival promise â confirm the specific journey with a current routing check on the day. Estimates exist to make early planning honest and quick; they are labelled on every page so that nobody mistakes them for measurements.
Keep Exploring
Put the ideas on this page to work in the distance index, browse the New York hub or the Los Angeles hub for full tables, or read how we calculate every figure we publish. Each pair page applies the same method, labels measured routes and estimates differently, and names its sources, so the numbers you plan with are the numbers you can check.
Sources: Coordinates: U.S. Census Bureau Gazetteer files; driving data: OpenStreetMap/OSRM â retrieved October 2026.