Bicycle Guidelines and Crash Rates on Cycle Tracks in the United States

Anne Lusk and her collaborators have published a paper in the American Journal of Public Health claiming that “the risk of bicycle–vehicle crashes is lower on US cycle tracks than published crashes [sic] rates on roadways.” You can read the full article, available open access, on the AJPH website (direct to PDF version).  AJPH has also published my letter about the article, and Lusk et al.’s response.

Lessons from Tragedies

Peter Furth is spreading the misinformation (as has Pete Stidman before him) that I prevented the installation of bike lanes in Boston. This is absolutely not the case. The Mayor was opposed to bicycling and bicyclists at the time, and of course the default in any road design project is no special bike lanes. I believe that bike lanes should only be installed when they don’t encourage dangerous behavior, but I never had the opportunity to make this argument in relation to an actual bike lane proposal during my 23 months working for the City of Boston.

Professor Furth’s comments came in the context of the discussion of the tragic death earlier this month of a bicyclist near the intersection of Huntington Avenue and Forsyth Street in Boston. Many bicyclists are claiming that Huntington Avenue is especially dangerous as shown by this and earlier bicyclist fatalities, and the city has started to make some small changes in reaction.

Despite Furth’s assertion that I was behind the lack of bike lanes on Huntington, I was not working for Boston in 1995 when Huntington Ave was redesigned nor in 1999 when it was rebuilt (I worked for BTD August 2001 to July 2003). During the design process I supported removing on-street parking on Huntington Ave in order to have enough room for motorists to pass bicyclists without changing lanes (at least 14 feet of usable space). The city agreed, but the state insisted on a 2 ft left shoulder, narrowing the right lane.  An insider connection to the Highway Department got the state to modify the striping slightly — but not to the design I would have preferred. Also, along the long stretches where parking was retained the lanes remain very narrow lane (8′ parking and 10-11′ travel). Cyclists must ride in the middle of such lanes not to get doored (shared lane markings would still be welcome here, in the middle of the lane).

People think Huntington Avenue is very dangerous because they will get run down by high-speed traffic, and the four fatalities over the past 12 years seem to be proof of that. Therefore a bike lane — or better yet, a separate bike sidewalk called a cycle track — is needed, they say.  Well, it doesn’t seem to me that these fatalities, tragic as they are, tell that story:

* Dr. Ruth Michler, November 2000: Bicyclist was run over by a construction vehicle of a type that is not allowed to be on the roads because the driver cannot adequately see in front of him. Lesson: Provide more supervision for construction and harsher penalties for criminally negligent behavior such as the driver’s. In this case the state refused to prosecute the driver, I think, because he was not operating a motor vehicle as it is officially defined. I think they were wrong about the law, and if they were right about it, it’s a crazy Catch-22.

* Gordon Riker, April 2007. “The bicyclist . . . was riding between two lanes of traffic on Huntington Avenue when he was clipped by a taxicab” and then fell under the rear wheels of a dump truck.  Lesson: don’t overtake between lanes of moving traffic. Tell cyclists to stay behind other vehicles unless traffic is stopped and can’t move (and then, pass only slowly and carefully).

* Eric Hunt, April 2010. Bicyclist fell due to the trolley tracks (near South Huntington, where they are in the passing travel lane, not in the protected median) and then was hit by an approaching bus (or fell into the side of the bus). Lesson: Post warning signs about the dangers of tracks (removing them in JP has dramatically improved bike safety & convenience); teach cyclists to stay far away from them.

* Kelsey Rennebohm, June 2012: Bicyclist was apparently riding on the sidewalk when she fell into the roadway and underneath a passing bus. Lesson: Don’t ride on the sidewalk. Tell cyclists about the dangers of sidewalk riding, and make streets more bicycle-friendly.

All of these four fatalities had completely different causes, only three of them were close to the same location, and not one of them was caused by a cyclist operating normally in the roadway getting run down by a motor vehicle driver approaching from behind (the construction vehicle was not legally a motor vehicle and was not allowed to be on the road). So let’s get our stories straight, and take the right lessons from them.

Out of The Frying Pan

In a recent column, Derrick Z. Jackson of The Boston Globe makes this observation about city cycling: “But for the most part, the streets still belong to daredevils willing to risk their lives on painted lanes in between whizzing cars on the left and parked cars on the right, where a deadly door could be opened into their lane without warning.” The irony is that these “daredevils” may believe they are cautious and prudent because they are in bicycle lanes, which have been designed to “Increase bicycle safety” (from Mass Ave Bike Lane Presentation).

The safe cyclist always rides far enough away from parked cars to stay out of the door zone. On Mass Ave in Boston, before bike lanes were added in December 2011, this meant using all of the 10-foot lane. Since the parking lane was removed on one side, there is obviously no door zone problem on that side, but much of the bike lane on the other side is within door range. Derrick Jackson’s preferred solution, a cycle track (a misnomer for a sidewalk for cyclists) would move from the frying pan to the fire by creating conflicts between straight-through bicyclists and right-turning motorists, by making the (narrowed) road effectively off limits for cyclists, and by increasing conflicts with pedestrians, who view all space where cars are prohibited as walking (or standing) space.  Among other problems.

And yet, according to the article, “A physically separated bike lane takes about 8 feet,” 3 more than the 5-foot bike lanes. Getting this extra 6 feet (3 feet on each side) would require removing the remaining parking lane. Yet if you remove all parking, you wouldn’t have any “deadly door” to worry about.

In the Mass Ave case, Boston could have made sufficient space by making the bike lane on the side without parking 4 feet instead of 5, and then making the bike lane on the parking side 6 feet instead of 5. With these dimensions (8 foot parking lane next to 6 foot bike lane), you can comfortably ride just inside the bike lane and still be out of the door zone. Now, it’s not intuitive to beginners that “cyclists should stay in [sic] to the left side of the bike lane in order . . .  to stay outside the radius of the doors of parked cars” (according to the City of Boston’s official advice). The common sense assumption is that you travel in the middle of marked lanes.  But the Mass Ave design is still much preferable to the most common situation in Boston, a 5 foot bike lane next to a 7 foot parking lane, which requires the bicyclists to be (at least partly) outside the bike lane to be safe.

When there is enough room for 14 feet to be divided between bike lane and parking, I would rather see a bike marking at least 11 feet from the curb — as is the standard for shared lane markings — so that even beginners are encouraged to ride outside the door zone.

Jackson pines for “bike tracks that have become iconic throughout Europe.” But if the current experiment on Western Avenue is a guide (and it is only one block without driveways or intersections), we are more likely to get something altogether less pleasant, like this:

Still a Need to Train Police Officers

An amazing story from New York City:  bicyclist Christina Thede passes a double-parked car, is nearly doored by its driver, complains to him, and goes on her way  — but is chased and assaulted by the door-opening motorist. Only it turns out the driver is an off-duty police officer, who arrests her — actually brings her to the police station — for bicycling recklessly (and disorderly conduct for good measure). How was she reckless? By deciding to pass a car double-parked in her lane! You can read more and see a picture here (someone should really get her a nicer bike). Here is the story in the cyclist’s own words:

“The first part of the incident is the part that no one witnessed (that I know of) and everyone has been filling in why that cop was chasing me.  What happened was I was riding up Amsterdam on the right hand side.  The black car that we now know was the police car was double parked in the right lane.  I started to pass the car on the driver’s side, when the driver opened the door into my path.  As a community of cyclists, I know you all must know this is a big danger and concern for those of us who bike in traffic. Anyway, I braked suddenly to avoid a collision.  A delivery bike behind me ran into me from behind because I had to stop so suddenly.  The driver of the vehicle and I had a brief verbal exchange, but I continued on in the right lane.  It was then that he got back in the car and pursued me.  I realized he was following and got scared, worried that this might be a severe case of road rage and who knows what he might do.  I crossed over to the left side of the street in order to shake him off.  This is where the witnesses’ stories pick up.  He skidded his car perpendicular to traffic, blocking my path.  Ok, now I’m REALLY scared.  I got off my bike and started to walk it up to the side walk between the parked cars. That’s when he got out of the vehicle and grabbed me from behind.   I started screaming for help and struggling thinking I was being assaulted or that he was trying to take my bike.  Suddenly, more cops showed up, and at that point I was turned over to the uniformed officers.  I was frisked, cuffed, and put in a police van, driven to the Central Park precinct (where the original plain clothes officer was from), and held for about an hour. I am not accused of running a red light.  The charges against me are disorderly conduct and reckless operation of a bicycle.”

AAA Bike Safety

The Sept/Oct issue of American Bicyclist has a feature article, “Bringing Bicycling into the Mix: The New AAA” (not yet on the LAB website) consisting of an interview with Rhonda L. Markos, a Traffic Safety Specialist with AAA. She acknowledges that “AAA’s involvement in bicycle safety has targeted predominantly school-aged children.” I kept waiting for the part where she says that AAA will partner with the LAB to improve motorist awareness of bicyclists’ rights (in addition to their customary wagging of  fingers at naughty child cyclists). Continue reading →

Centre/South Bike Markings Plan

Bicyclist narrowly avoids opening door -- Centre St, Jamaica Plain, Mass.
Bicyclist narrowly avoids opening door -- Centre St, Jamaica Plain, Mass.

This bicyclist had to swerve as the oblivious motorist flung open her door on Centre Street in the heart of the business district where parking turnover is high and traffic is slow.

This Thursday the City of Boston will present its proposal for bike markings on South Street and Centre St from Forest Hills to Jackson Square in Jamaica Plain. The proposal will be:Continue reading →

Door Zone Bike Lanes Again

John Ruch finally published his story about bike lanes in Jamaica Plain, only half of which made it to the print edition. As usual, Ruch mangled the quotes. The worst one came up front:

That little stripe is not going to protect you,” Schimek said, explaining he is worried the bike lane message will be, “‘You novice cyclists, come out here. You see that bike lane, and you’re safe because that’s what a bike lane is for.

I believe I was trying to say that that the bike lane tells new bicyclists that they will be safe riding in it, even though in the proposed design they will be smack in the danger (door) zone. And this summary also got mangled:

Bike lanes suggest that bicyclists have to use them, when in fact they do not, Schimek said. That confusion can lead to more car-bike conflicts, and some research shows it leads drivers to come closer to bikes in the lane, he said. Bike lanes also lock riders onto the right-hand side of the street, when they should be on the left to make left-hand turns safely, he noted.

What I was saying is that most people act as if there is a legal requirement to use bike lanes, even if there is not. It is the bike lane stripes that can cause more conflicts, by encouraging motorists to keep left when turning right, encouraging bicyclists to pass on the right, and even suggesting that bicyclists should turn left from the bike lane. And then there’s this one:

An illusion is exactly what 5-foot bike lanes on Centre Street would be, Schimek said. He blasted the idea that bike lanes are good because they make potential riders feel safer as a “backhanded…indirect and disingenuous argument.”

The argument that I “blasted” is the one repeated in the article: even if door-zone bike lanes are more dangerous, having more bicyclists on the road will make all bicyclists safer.

Another mangled point:

Schimek noted that the city’s own design illustration for Centre Street bike lanes showed people riding down the middle of the lanes. In fact, riders are supposed to ride along the left-hand line in a bike lane to avoid dooring. Without education, new riders attracted to the bike lanes are set up for disaster, he said. And even if there is education, the lanes may be so narrow that riders are in the dooring zone anyway, he said.

Ruch had a copy of the illustration showing bicyclists in the door zone with bike lanes and outside the door zone without (see previous post). He did not communicate that point. Riders are not “supposed to ride along the left-hand line in a bike lane,” they are supposed to ride in the middle of the bike lane, following normal lane use expectations — and several studies show this is what they actually do. I don’t use the meaningless term (in this context) “education,” I prefer to talk about learning from experience (i.e., being doored, or narrowly avoiding a door), and training (from friends, books, or formal classes). The final point is that riding outside the door zone will probably put the bicyclist at least partly outside the bike lane. And thus we end up with the paradox that the only safe way to use these bike lanes is not to use them.

Ruch also makes this blooper:

The dispute is over methods—especially whether it makes sense to slightly reduce the width of car lanes on Centre and add 5-foot bike lanes.

Where are those “car lanes” on Centre Street? All I see — and all the law sees — is a road shared by cars, trucks, buses, motorcycles, and, yes, bicycles.

News about the famous Hampshire Street Bike Lane study coming soon.

Save the Bike Lanes on Concord Ave!

Concord Avenue "Bicycle Lanes"
Concord Avenue 'Raised Bicycle Lanes' - Not Part of "Roadway"

Concord Avenue north of Fresh Pond in Cambridge has perfectly nice, five-foot bike lanes. They’re not even in the door zone, since there is no on-street parking. But Cambridge is planning to take them away, narrow the roadway, and force bicyclists to ride on the sidewalk. The bike lanes were striped 10 years ago as part of a successful “road diet” project. John Allen has posted a history and more information about the proposed project. The current project, whose limits are Fresh Pond Parkway and Blanchard Road, includes water main installation and started this week.  The planned “raised bike lanes” are at sidewalk level and in most places will be separated from the travel lanes by a 6-inch high curb. A bicyclist who strays too close to the edge of the “bike lane” risks falling into the travel lane. In other words, the new “bicycle lanes” are really “sidewalks”. A sidewalk by any other name is still a dangerous place to ride a bike at anything faster than walking speed. Look at the above excerpt from the construction drawings. They show clearly that the “roadway” includes the “travel lane” but not the “bicycle lane.”  This turns out to be an important detail.

MassDOT’s Engineering Directive on Bicycle Accommodation says that “Bicycle accommodation shall be in accordance with Chapter 5 of the [Project Development & Design] Guide and the 1999 AASHTO Guide for the Development of Bicycle Facilities.” What do those documents say? Chapter 5 of the Project Development & Design Guide says: “Bicycle lanes are portions of the traveled way designed for bicycle use. .  .  . Bicycle lanes should be designated by a 6-inch solid white line on the right edge of the motor vehicle travel lane [sic].” (section

The 1999 AASHTO Guide for the Development of Bicycle Facilities is more precise. It has the following definitions: “BICYCLE LANE or BIKE LANE—A portion of a roadway which has been designated by striping, signing and pavement markings for the preferential or exclusive use of bicyclists. ROADWAY—The portion of the highway, including shoulders, intended for vehicular use.” It further says that “A bike lane should be delineated from the motor vehicle travel lanes with a 150-mm (6-inch) solid white line.”

In other words, the official design manuals say that bike lanes are part of the roadway, not the sidewalk, and are separated from other lanes by lines, not curbs. The AASHTO bike guide also includes this helpful guidance: “Sidewalks generally are not acceptable for bicycling. However, in a few limited situations, such as on long and narrow bridges and where bicyclists are incidental or infrequent users, the sidewalk can serve as an alternate facility, provided any significant difference in height from the roadway is protected by a suitable barrier between the sidewalk and roadway.” The Concord Ave project clearly does not meet the requirements for exceptional designation of a sidewalk as a bike facility, and even if it did, the design guidelines require a barrier to prevent bicyclists from falling into the roadway — an important detail that Cambridge has not followed in this project.

Well, what if Cambridge just edits the drawings so that they say “bicycle path” instead of “bicycle lane”? One problem is that there is no design standard for a bicycle facility on the sidewalk that prohibits pedestrians; the only standards are for “shared use paths” — shared by pedestrians and bicyclists. Moreover, the Project Development & Design Guide says this: “Shared use paths are facilities on exclusive right-of-way with minimal cross flow by motor vehicles. Shared use paths should be thought of as a complementary system of off-road transportation routes for bicyclists and others that serves as a necessary extension to the roadway network. The presence of a shared use path near a roadway does not eliminate the need to accommodate bicyclists within a roadway” (Section If Cambridge were to call the “raised bicycle lane” a “path” it would not be able to narrow the travel lanes to only 12 feet, as proposed, but would have to leave at least another 3 foot shoulder to provide the minimum 15 feet of room for motorists to comfortably pass bicyclists according to MassDOT’s Engineering Directive previously mentioned.

Ok, but does Cambridge really have to follow these design guidelines? To pay for this project Cambridge applied for and apparently received a grant from the Public Works Economic Development (PWED) Program, formerly administered by the Executive Office of Transportation, now part of MassDOT. So it would seem that MassDOT’s own Engineering Directive should apply.

If not, there is a higher authority. All traffic control devices in the U.S. must conform with the Manual on Uniform Traffic Control Devices published by the Federal Highway Administration. The manual has the following definitions:
Bicycle Lane—a portion of a roadway that has been designated for preferential or exclusive use by bicyclists by pavement markings and, if used, signs.
Roadway—that portion of a highway improved, designed, or ordinarily used for vehicular travel and parking lanes, but exclusive of the sidewalk, berm, or shoulder even though such sidewalk, berm, or shoulder is used by persons riding bicycles or other human-powered vehicles.”
In other words, a Bicycle Lane that is not part of a roadway does not comply with the manual.
Noncompliance with engineering directives and design manuals could mean that Cambridge would be held liable by a court in the event of a lawsuit by an injured bicyclist.

The Livable Streets Alliance endorsed the project, saying, “The Concord Avenue reconstruction plan pushes for best practice for street design. It includes well-designed and inviting facilities for pedestrians, bicyclists, and motorists alike. Of particular note, there is a raised bicycle lane (cycle track) included as part of the project, a type of facility that has been gaining interest among the public, but which we currently have very few of.” WalkBoston endorsed it as well.