Submitted by John Allen, CRW Safety Coordinator
“Real improvement in safety and mobility requires data-driven planning.”
— Mighk Wilson, planner at MetroPlan Orlando (Florida), utility cyclist and CyclingSavvy co-founder
This applies to planning, but also, your own understanding of how to be safe, and your confidence in choosing bicycling for transportation and recreation. So, please read on!
Paul Schimek’s Boston study
Bicycling crashes in Boston have been the subject of an excellent study, conducted by Dr. Paul Schimek and based on crash reports and emergency-room data. Schimek didn’t have bicycle use data, but he could determine the relative frequencies of different kinds of crashes. They may not be quite what you think:
-
dooring was the most common crash type reported to police;
-
the motorist left turn (“left cross”) came in second;
-
following this were the motorist drive-out and motorist right turn (“right hook”).
All of these are more likely when cyclists are riding at the far right side of the street – in the door zone and passing on the right. Most bike lanes in the Boston area encourage this behavior.
Next came bicyclist errors:
Last and least were overtaking crashes. Surprisingly, there were more overtaking crashes with bicyclists rear-ending motor vehicles than with motorists rear-ending bicyclists.
Schimek looked into hospital data and determined that police reports miss 76% of emergency-room visits and 60% of hospital admissions. These are overwhelmingly crashes that do not involve a motor vehicle.
What lessons does Schimek’s study offer? It’s important to be predictable, obeying the rules of the road – but also to be predictable by using defensive driving techniques – and that can mean staying out of the bike lane. Good bike-handling technique and situational awareness of potholes and other road hazards help too.
A summary and link to the detailed research paper are online on Schimek’s site.
Mighk Wilson’s research
Mighk Wilson’s cutting-edge research is able to establish crash rates. He also has taken a wider look at nationwide bicycling trends.
Wilson collects data with Miovision cameras. They are mounted on poles like the controversial Flock license-plate readers, but instead count road users and record aggregate behavior in traffic. Wilson uses the cameras to count cyclists and track their direction of travel, speed, and position (e.g. travel lane, bike lane, paved shoulder, sidewalk, sidepath, or shared use path).
Before the camera technology became available, determining bicycling crash rates was very cumbersome. Checking data from the cameras against records made this easy.
Some of Wilson’s findings can be summarized as follows:
-
Cyclist speed is a top factor in crash rates. Wilson found twice the crash rate on streets as on sidewalks, surprisingly, but the underlying factor was speed, not facility type: sidewalk cyclists were much slower, on average. Wilson observes that “[i]t’s not that the bike lane or sidewalk makes the cyclist safer, it’s that the cyclist makes the facility safer because they are slower. When a motorist violates their right of way, they don’t need the stopping distance.”
-
In a related finding, downhill cycling on sidewalks proved to be very hazardous due to the speed.
-
Higher posted speed limits and nighttime cycling proved to be highly important risk factors in car-bike collisions (though most nighttime crash reports didn’t indicate whether a cyclist was using lights).
-
As already documented in other studies, bicycling opposite the direction of traffic results in about 5 times the rate of car-bike crashes as bicycling with traffic. Wilson was able to establish that this holds whether the cyclist is riding on the street, sidewalk or sidepath.
The speed issue bears on the utility of bicycling, because a main advantage of cycling is in decreasing travel time compared with walking, and—for many urban trips—compared with public transportation and motoring when you include walks to and from transit stops or parking.
Wilson conducted his research in the Orlando urban area, but he also reviewed data about rates of bicycle use in the ten cities around the USA with top “bicycle network scores”: measures of construction of special infrastructure. He came to a rather striking conclusion. Despite the construction of many more special bicycle facilities, bicycle commuting has increased only in two: New York and San Francisco. In other cities with top “bicycle network scores” it has declined. These include Portland, Oregon, Washington, DC and Seattle, cities which have made large infrastructure investments. Bicycle crash numbers also have increased very substantially in all of these cities except New York and San Francisco.
What makes those two cities different? Wilson notes that street connectivity, convenience and safety of bicycle parking, expense of auto parking, college age population, low income and weather prove to be more important than the extent of special bicycle facilities. Wilson makes a telling comparison between Orlando and Amsterdam, with a much smaller urban area and secure parking at train stations.
Wilson has recently given a seminar online about his research, now released as an online video. You may find his written introduction and the embedded video here. If you open the video in its Vimeo platform, you’ll find a description including a hot-linked table of contents to its sections.
Summary
Wilson’s and Schimek’s studies are the two most significant ones on urban cycling in the USA of which I know. They are indispensable reading if you want to have a solid background on the topic.
