Data-driven planning for bicycling

Real improvement in safety and mobility requires data-driven planning.

An optimal bicycle plan takes in all 6 the 6 E’s of planning. They are Education, Enforcement, Engineering, Encouragement, Evaluation and Equality. Evaluation is where the data comes in.

I work as a planner at at Metroplan Orlando (Florida, USA). In March of this year I presented a webinar I called Bicycle Planning 101 to our Community Advisory Committee (CAC). The presentation grew out of questions about bicycling that I’d received from CAC members; most notably about how bicycling for transportation differs from bicycling for recreation, and the impacts that e-bikes and scooters are having on our non-motorist crash trend.

Initially this was going to be a two-part presentation made at regularly scheduled meetings, but it was difficult to find the time amidst all the other topics the Committee needed to cover. So instead we set this up as a separate webinar, and doing so gave me the opportunity to expand on the topics I thought the Committee needed to understand. The total webinar length is about an hour and twelve minutes, including introductions and discussion periods, so grab some snacks and a beverage and settle in.

I’ve developed and collected much of the data in this presentation over the past five years. An especially useful data source was the Bicycle Network Score developed by People for Bikes and applied to major and intermediate-sized cities around the U.S. I found it interesting that no-one had looked at the correlations of that Network Score to bicycle commute rates or fatality rates. Perhaps the answers were too disconcerting to be shared by those promoting separated facilities.

Using cameras to track exposure and behavior

Key to our understanding of how infrastructure affects the amount of cycling and the risks of cycling is our recent use of Miovision cameras, which have allowed us to track cyclist use of travel lanes, bike lanes, paved shoulders, sidewalks, sidepaths, and shared use paths in independent rights-of-way.

Other studies of bicyclist crashes in the U.S. have either estimated bike counts indirectly, or just ignored exposure altogether. None have measured the bicyclists’ position in the right-of-way, measured how many cyclists are riding with the flow of traffic versus against, or addressed bicyclist speed.

Findings

Our bicyclist count and speed data have led me to two key findings. First is that land use and demographics are far more important drivers of bicycle transportation than bikeways are. Ultimately, denser land use reduces trip distances, making bicycle trips more attractive; and places that make it harder to drive a car – such as by limiting car parking – end up with much more bicycle travel.

Your strategy as a cyclist must depend on the context, the street characteristics, and how fast you wish to ride.

Second, with a few exceptions, saying “Cycling on “X” is “dangerous” or “safe.”” is misleading. As bicycle driving educators many of us have tended towards “don’t ride on the sidewalk (or sidepath)” when it’s really a matter of speed, direction, and context. Your strategy as a cyclist must depend on the context, the street characteristics, and how fast you wish to ride. Certainly, an 18-mph cyclist riding with the flow of traffic on a sidewalk along a commercial arterial during evening rush hour is going to have a high risk level, but a 9-mph cyclist on a residential collector on Sunday morning will likely have a very low risk. There are too many factors at play to tell a cyclist whether their risk is “too high” or not. More detailed summaries of my crash risk research can be found in three earlier posts, starting here.

All of this serves to support the approach that we have taken with CyclingSavvy over the years, which is rather than tell cyclists “never do this” (well, except for riding to the right of large trucks), we give them an understanding of the risks, a toolkit of principles, and a set of strategies for addressing them.