There was a time when I believed that striped bicycle-specific lanes were a good idea, but this was before I became familiar with the problems with bicycle-specific roadway markings in the real world. As I began to study the various bicycle-specific facility designs that have been proposed and implemented, I discovered that what I really wanted was an improvement over narrow travel lanes with heavy, fast traffic, but without creating new operational/social problems or making existing problems worse. I have spent considerable time studying the causes of car-bike collisions, including analysis of car-bike crash data compiled by NCDOT and review of the actual police reports from Cary for a six-year period. I have also studied alternative vehicles such as Neighborhood Electric vehicles and various types of motor scooters. All of this has led me to believe that wider outside through lanes are a simple, elegant, effective solution for providing safe, comfortable, and efficient accommodation of a diversity of vehicle types with varied speed capabilities and widths on urban streets. Wider outside through lanes are especially good for bicyclists. I believe that bicycle-specific segregation markings, however, have operational and social disadvantages that likely outweigh any possible operational or social advantages when employed in most urban areas.

The idea of bicycle-specific lanes/markings is alluring to many people, especially to inexperienced cyclists. But the professional standard applied to installation of traffic control devices is that the operational advantages must be estimated to exceed the operational disadvantages, regardless of popular demand. I have therefore spent a great deal of time studying the existing research on bicycle lanes, car-bike collision data, and pro-bike-lane advocacy articles in attempts to ascertain what operational advantages may be provided by the stripe, and to determine where striping may be warranted. My efforts to make conclusions about the advantages of the stripe have been thwarted because of the following observation:

All of the available publications and advocacy materials promoting operational and safety benefits for bicycle-specific lane markings fail to properly differentiate the effects of striping from the effects of total shared pavement width, motor traffic characteristics, or both.

It is only natural for surveyed cyclists to say they feel more comfortable and safe on roads with bike lanes when the only roads chosen for bike lane striping have lower traffic volumes and/or extra-wide pavement. But this doesn't mean the segregation marking itself is a net benefit. When I have reviewed actual car-bike collision data, what jumps out right away is that the vast majority (>90%) of urban car-bike collisions involve right-angle junction hazards that cannot possibly be reduced by bicycle-specific roadway markings. Rather, these collisions may be exacerbated by cyclists positioning themselves by vehicle type rather than as their destination or speed requires. Meanwhile, the rare overtaking collisions offer no correlation with striping or lack thereof. Rather, the overtaking collision events I have looked at in Triangle correlate with travel lanes narrower than 12 feet wide with a posted speed limit of 45 mph, or gross negligence from the motorist or cyclist that could not be affected by a bike lane stripe. In my own experience, close (especially uncomfortably close) overtaking correlates with cycling on the edge of narrower lanes; wider lanes (14' to 16' or even wider) effectively minimize this. Another way to eliminate close overtaking is to ride in the center of narrow lanes, but I understand that some cyclists prefer not to do this due to some motorists' expression of displeasure at such, and so I support initiatives to increase the pavement space on important roads.

I believe that 14' wide outside through lanes provide reasonably good separation between normal-width cars and normal-width cyclists on 4-lane streets where overtaking drivers can move safely up to the left lane line to pass. I believe 16' or wider lanes are preferable on busy two-lane roads in order to provide extra shy distance between opposite-direction lanes of traffic. 16' or wider outside lanes may also be desirable on multi-lane roads where there is significant tractor-trailer, bus, or bicycle-trailer traffic. Note that the extra lane space should be incorporated into the through-lanes at intersections, not the right-turn-only lane, since the speeds and volumes are usually highest in the through-lane.

I believe that bike lane striping may have operational advantages in some cases, but that we cannot conclude what these are and how important they are. One potential advantage is that the least problematic bike lane stripe installations make it easier to tell upon initial inspection that a roadway provides adequate overtaking space without motorists changing lanes. This increases both motorist and cyclist confidence, which may or may not be a good thing. On the same pavement width without the stripe, both cyclists and motorists sharing the lane will use greater caution initially, and so there is no net increased danger, and possibly a decreased danger. Meanwhile, I believe that bicycle-specific lane striping creates the following operational and social problems:

1. Bike lane markings very frequently conflict with safe and legal cyclist and motorist positioning at intersections and other junctions, where most car-bike collisions occur.

2. Bike lane markings conflict with safe and legal cyclist positioning when cyclists are traveling at significant speed.

3. Bike lane markings usually increase the rate of accumulation of debris at the right side of the road by directing farther away the normal sweeping action of cars.

4. Bike lane markings tell motorists that bicyclists should ride somewhere other than in the normal travel lanes, inviting increased harassment of competent cyclists who operate in normal travel lanes in order to position themselves appropriately based on context or to avoid debris or the door zone of parked cars.

5. Bike lane markings reinforce inexperienced cyclists' unwarranted fear of overtaking collisions and foster psychological dependency on separation by stripes that offer no known safety benefit.

6. Bike lane markings confuse the public about a basic principle of the state vehicle code, which is that every street and every travel lane is a bicycle facility where cyclists operate according to the rules for drivers of vehicles.

Bicycle facilities are not something that a road engineer can add to a roadway, in any legal sense. By law, the bicycle facility is already there. Every travel lane of every street already provides a uniform, reasonably easy to use bicycle facility. What the road engineer can do is provide improved overtaking facilities on those street sections where motor traffic speed and/or volume warrants it. In my opinion, promotion of bicycle-specific roadway markings and segregation by vehicle type on urban streets confuses the issue of protecting lawful cyclists' legal right to safe access to every destination on every street. I believe it distracts public attention from things that can provide real improvements in safety, comfort, and efficiency, such as:

* Better design and maintenance of smooth, clean roadway surfaces
* Detection of bicycles by demand-actuated traffic signals
* Provision of adequate overtaking clearance space on the most popular roads for transportation, in particular those with the highest average travel speeds
* Education and enforcement about best bicycle driving practices and the rights and responsibilities of all road users


Thank you for your time and attention,
Steven Goodridge, Ph.D.
Member, Tau Beta Pi engineering honor society
Webmaster, https://humantransport.org