How often do drivers actually leave three feet?
Thirty-plus US states require it. The data on whether anyone follows the rule is grim, and inconsistent.
The first state to put a numeric passing distance into law was Wisconsin in 1973. Pennsylvania widened that requirement to four feet in 2012, and a steady drip of states have adopted some version of the rule in the years since. As of 2025, more than thirty US states and the District of Columbia have a statutory minimum lateral distance for vehicles passing cyclists. The most common figure is three feet.
The next question is the awkward one: does the law work? The honest answer is that we don't fully know, and the studies that exist suggest the gap between law and practice is wide.
What the field studies have found
The best-known instrumented-bike studies have repeatedly measured a meaningful share of passes occurring at less than three feet. Goddard and Dill's 2010 Portland State work used a sensor-equipped bike on Portland streets and found something like one in five overtaking events came within three feet of the rider. Later studies in Texas, Tennessee, and the UK have produced numbers in roughly the same band, though the exact rate depends heavily on the corridor: shoulder width, presence of a bike lane, speed limit, and time of day all push the violation rate up or down.
The qualitative picture is also consistent across studies: rural two-lane roads produce the worst lateral clearance numbers, urban corridors with marked bike facilities the best, and heavy commercial vehicles overtake closer on average than passenger cars do.
Why "average" is a bad metric here
A 95th-percentile pass at six feet is consistent with one pass-in-twenty being terrifyingly close. Riders don't experience the average; they experience the outliers. This is part of why we record every pass on a Rydiq ride, not just the ones that triggered an incident: the distribution of all passes tells a more honest story than just the close-call count.
We'll publish aggregate ride data from the beta cohort as soon as the sample size is large enough to be useful, broken down by road class and time of day. If you want to see your own per-ride distribution today, the analytics drawer in the dashboard plots it for you.