Catherine Withrow was walking her bike.
A construction truck killed her at a Fort Collins intersection. In the federal crash data she will probably not be counted as a cyclist at all.
On the evening of Monday, July 6, Catherine Lee Withrow was walking her bicycle across East Mulberry Street at Whedbee Street in Fort Collins, Colorado, with construction personnel present. A semi truck working as part of the construction project struck her. She died at the scene. The Larimer County Coroner gave the cause of death as multiple blunt force injuries and the manner as accident. She was 53, and she lived in Fort Collins. The driver remained at the scene, and the Fort Collins Police collision reconstruction team investigated.
The counting problem
Federal crash data classifies a person by what they were physically doing at the moment of the collision. A person on a bicycle is a pedalcyclist. A person pushing a bicycle across a road on foot is a pedestrian. By that rule, a death that happened during a bike trip, to a person with both hands on her handlebars, will very likely appear in the pedestrian column and never in the cyclist one.
That is not a clerical complaint. Every headline of the form “cyclist deaths rose 5 percent last year” is built on that coding rule, and the rule systematically drops the moments when riding is impossible. Dismounting because the crossing has no ramp. Walking the bike through a work zone because the lane ends at a barrier. Pushing across a stroad because there is no legal way to ride across it. Those moments are manufactured by infrastructure, they are part of the risk of choosing to travel by bike, and the official cyclist fatality count cannot see any of them.
We have argued before that crash data is a poor instrument for measuring danger. This is a second, quieter version of the same problem. It is not only that the count arrives too late. It is that the count is narrower than the thing everybody uses it to describe.
Work zones and large vehicles
The other half of this is the vehicle. Construction trucks operating inside a live urban corridor combine the worst properties available: a high cab with substantial near-field blind spots, frequent low-speed manoeuvres, drivers focused on a site rather than on through traffic, and a temporary traffic pattern that puts people on foot and on bikes into paths neither they nor the driver use habitually. Direct-vision cab standards and side guards exist and are mandated in parts of Europe and in a few US municipal fleets. They are not the norm on a private contractor's truck in most of the country.
What our data will never show you
Rydiq measures vehicles overtaking a moving bicycle on a road. That is the whole domain. A person walking a bike across a work-zone intersection is outside what our pipeline sees, and no volume of ride uploads will ever produce a number that describes what happened to Catherine Withrow.
We would rather say that plainly than let a close-pass distribution get mistaken for a picture of cycling risk. It is a picture of one large, badly measured slice of it. The rest of the slice needs different instruments, and in this case it needs a coroner's report, because nothing else was recording.