by Fred Oswald, LCI #947
Blunders in Planning, Engineering & Facilities
How would you feel about driving a car on a road designed by engineers who do not know how to drive and do not understand the rules of the road? Does this sound insane? This is the situation that faces people who drive bicycles. Such misunderstanding causes serious mistakes, including
- Contradicting safe practices
- Failing to encourage (or even discouraging) cyclists to follow standard rules of the road
- Misunderstand bicycle operation in traffic
- Building bicycle sidepaths (multi-user paths beside roads) with dangerous intersections at cross-roads and driveways.
- Building bike lanes in the door zone of parked cars
- Placing bike lanes on steep hills
- Routing bike lanes to the right of right-turning traffic
- Encouraging left turns from the right edge of road
- Encouraging cyclists to pass slow traffic on the right
There is an inherent fallacy in the practice of building separate bikeways (sidepaths and bicycle lanes) alongside roadways. The fallacy is the assumption that such facilities make cycling safe for those who do not know how to do it properly. The fallacy is dangerously wrong for at least four reasons:
- All of the skills required for riding safely on the roadway are also needed for riding on bikeways.
- Because the facilities complicate the roadway and create intrinsic hazards, more knowledge is needed for riding there safely than is needed on the roadway.
- The facilities often include explicit hazards due to defects in design, construction and maintenance.
- The presence of separate bikeways makes people feel “comfortable” while riding dangerously. This makes them less willing to learn safer methods.
Intrinsic Problems of Bikeways
Separate bicycle facilities treat cyclists as though they are rolling pedestrians, rather than drivers. This treatment makes them feel they are separate from traffic and encourages them not to follow the rules of the road. Violating the rules of the road often leads to crashes. As it is, too many cyclists on the road fail to follow basic traffic law, such as stopping for red traffic lights.
The problem is much worse for those who ride on sidewalks and sidepaths. In Fig. 1 we see two cyclists who were riding on a multiple-use path in the median strip of a park roadway. The path has a pedestrian “Walk” signal, controlled by a push button. Like most path riders, they did not wait for the signal but rather crossed illegally, dodging turning traffic from the road.The blue bike lane shown in Fig. 2 represents the best of a bad practice. The lane seems to be far enough from the curb, thus it avoids the door zone of parked cars. By running left of the right turn lane in the distance, it puts straight-through (but not turning) cyclists in the correct location. Finally, it has adequate room.
However, even a “best” bike lane encourages mistakes. The bike lane stripe encourages cyclists to stay to the right and motorists to stay left, even when the rules of the road require otherwise. If a fast cyclist (perhaps descending a hill) catches up to a slow car, there is a tempting clear channel for passing in the motorist’s blind spot. This can lead to a collision if the motorist turns into a driveway or parking spot while the bicycle is passing.
Separate facilities attract beginners. (This is one of the reasons that “bicycle advocates” push for the facilities.) Beginners often turn left directly from a bike lane without first merging to the left turn lane and without yielding to overtaking traffic (the “shooting gallery” approach). We have also heard of beginners making right turns from this type of bike lane by swerving across right-turning traffic.
Even knowledgeable cyclists, who know enough to merge to the proper place on the road to make a turn or to avoid hazards, experience trouble from separate bicycle facilities because they encourage motorist resentment. Some motorists become vigilantes, harassing any cyclist not in “his place”. The existence of a designated “bike route” on one road makes it very difficult to convince city officials to make improvements or repairs on a parallel route. Separate facilities make educating cyclists much more difficult.
Explicit Bikeway HazardsMany separate bicycle facilities have explicit dangers, in addition to the intrinsic hazards discussed above. The photo in Fig. 3 shows a marked bike lane that routes cyclists across a “free-running right” at a highway entrance. Cyclists are encouraged to swerve across the path of traffic on the ramp. Motorists using the ramp are expected to yield to traffic crossing the ramp. This is intensely dangerous because both cyclists and motorists are directed to violate standard traffic rules.
The proper cyclist’s route, staying in the through lane and avoiding the ramp, is shown by yellow dots in the figure. This avoids swerving across the path of motor vehicles. Because this is in Oregon, it is illegal to use the safe route shown above. One must choose between what is safe and what is legal. You can see more blue bike lane problems in “Blue lane”: really a lane, or a diagonal crosswalk?
The City of Portland, Oregon has been experimenting with painting bike lanes blue to warn travelers of “high-conflict areas”. Although the colored paint may help alert some motorists to the non-standard intersection, it does not make bad engineering safe. It would be much better to simply follow sound practice and avoid creating the conflicts. The Portland explicit bicycle blunders include routing cyclists on the wrong side of turning traffic and steep downhill bike lanes.
Another frequent source of hazards is bikeways that are crammed into inadequate space. Fig. 4 at right shows a very narrow shoulder (varies from 1½ to 3 feet wide) that was made into a dangerous bicycle lane in a Bicycle Friendly Community, Carrboro, North Carolina. The width of this shoulder is much less than the 4 feet minimum width for a bike lane as specified in the AASHTO Guide . Ironically, the traffic on the road was hardly hazardous to start with. It is a one-way, 16ft wide, rural section, residential street five blocks long with a 4-way stop sign at each block. The posted speed limit is 20 mph.
Our goal is to investigate the effectiveness of colored pavement markings in reducing conflicts. To do this, we have selected ten conflict areas with a high level of cyclist and motorist interaction and about which area motorists and cyclists had complained. We then chose seven for initial testing, and painted them blue. … The conflict area in all cases had already been defined with dashed lines, as well as, in most cases, signs indicating the need for motorists to yield to cyclists. The painted area and its accompanying sign are intended to remind motorists that they are crossing a bicycle lane and need to look first and yield to any through cyclists. They are also intended to caution cyclists to be careful in the conflict area. 
There are other safety problems with this bike lane, including the adjacent ditch with no recovery area. The crown of the road and broken pavement at the edge increases the likelihood of riding into the ditch. Concrete culverts at driveway crossings create fixed object hazards. This street is adjacent to an elementary school; presumably the dangerous bike lane was installed for children.
The Bicycle Friendly Community program promotes facilities like this without warning about or even acknowledging hazards introduced by poorly planned facilities. The only hazard considered is overtaking traffic (passing from behind). As we show in part 7, overtaking traffic is a very remote risk. Fortunately, Wayne Pein of the North Carolina Coalition for Bicycle Driving was able to convince the city to remove the bike lane signs.
Bikelanes often encourage mistakes by both cyclists and motorists — mistakes that cause crashes. Fig. 5 shows a bike lane in Berea, OH that passes to the right of a lane primarily for right turning traffic. Cyclists are directed to a route that makes them cross paths with turning vehicles. In the foreground is a sidestreet and a driveway to a drive-through pharmacy that will produce more conflicts with bicycle traffic.
The bike lane stripes discourage motorists from complying with Ohio traffic law §4511.36(A), which says: Approach for a right turn and a right turn shall be made as close as practicable to the right-hand curb or edge of the roadway. The stripes are a traffic control device that encourages violation of this law.
Notice in the photo both the car making a turn and the truck behind were turning improperly because drivers followed the lane markings. The green squares added to the photo show the correct path for turning traffic. The yellow dots indicate the proper route for a straight-through cyclist — avoiding conflicts by staying near the center of the dual destination lane.
This intersection essentially violates recommendations in the Manual on Uniform Traffic Control Devices , which says (page 9C-4): A through bicycle lane shall not be positioned to the right of a right turn only lane. The MUTCD also says: An optional through-right turn lane next to a right turn only lane should not be used where there is a through bicycle lane. If a capacity analysis indicates the need for an optional through-right turn lane, the bicycle lane should be discontinued at the intersection approach.
An intersection with bike lanes striped right to the corner is referred to as a “coffin corner” because it is so dangerous. You can read more about this and several other blunders that occur in just 1/2 mile of roadway (including door zone bike lanes) in Berea’s Bicycle Blunders.
The Door Zone
Bicycle facility design is serious business. Done wrong it kills. Some communities squeeze bike lanes into roads that do not have enough room. They create serious safety compromises.
Believe it or not, most streets are better off without bike lanes. Bike lanes have a lot of physical requirements to be safe. If you bend these requirements too far, you wind up with a bicycling environment more hazardous than the same street without the bike lane. These requirements include factors such as sight distance, driveways, intersections, pavement quality, edge drop-offs and slope. But the most common factors that lead communities to bend the requirements are intersections, overall available width and parking.
Simply stated, if you shoehorn a bike lane into a narrow street with parking, the bike lane will be in the door zone, where car doors can — and do — fling open at unpredictable moments. These accidents are dismayingly common. Quite a few fatalities have been documented.
For more than 30 years, the safety literature has warned cyclists to avoid riding in the door zone. But now we use our tax dollars to pay for traffic control devices that instruct us to ride in the door zone! This is a sad situation.
The city of Cambridge Mass. installed door zone bike lanes on several busy streets with narrow lanes, including Massachusetts Avenue, despite warnings that they were unsafe. The bike lane shown in Fig. 6 proved fatal for Tufts University graduate student Dana Laird in 2002. If you look carefully at the photo, you can see that the open car door blocks almost the entire bike lane. There is not room to avoid the door without encroaching into the motor lane. (Remember, a bike is about 2 feet wide. There is less than two feet between the open door and the left stripe.) You can see a larger copy of this photo from the photographer’s website.
Unfortunately, many other cities also install bike lanes in the door zone (see Fig. 7) and some these also cause causalities to cyclists. You can read accounts of 16 recent door zone casualties (11 fatal) in an article called The Door Zone Project.
One of the most inexcusable examples of carelessness in bike lane design is illustrated by this statement: The City of Chicago installs bike lanes on streets as narrow as 44 feet wide with parking on both sides. Chicago published a Bike Lane Design Guide  that includes plans showing how they achieve this claim. In their zeal to install bike lanes despite inadequate space, they distorted the size of the cars and trucks depicted in the drawings.
At first look, this drawing (below) doesn’t look too bad, but that’s only because the cars and trucks depicted are far, far smaller than in real life. We have the noted cycling author John Forester to thank for pointing this out. In 2002, Forester added dimensions for the vehicles depicted in  by scaling from the drawing. Forester determined that these vehicles are about 20 percent undersized — some cars and trucks are depicted as though they are less than 5 feet wide – narrower than a VW beetle.
Note the tiny utility truck in Fig. 8, which is limited to “OFF ROAD USE ONLY”. This truck is less than 5 feet wide — almost exactly the width of the diminutive truck depicted in Fig. 9 at lower-left. The real width of a full-size pickup is over 6 1/2 feet as we have shown with a red rectangle in Fig. 9.
This is shocking dishonesty and blatant engineering malpractice. Most motor vehicles are around six feet wide, and many commercial vehicles are 8.5 feet wide. Those are what you have to design for. Pretending that you are doing otherwise is dishonest, and it courts dooring accidents.
In Fig. 9 below the red rectangles depict the true sizes of typical cars and trucks. Parked vehicles are shown at a one foot distance from the curb and with open doors. Car doors are shown in both 4-door and 2-door configurations. The diagram shows that the door of even a small 4-door Ford Focus will block essentially all of the supposedly “safe” space of the bike lane. The “stickout” of two-door cars is much greater. (Also note many motorists park further than one foot from the curb and some states allow up to 1.5 feet.)
This striped bike lane serves as a buffer space to keep cars from driving there. However, cyclists, who are much more vulnerable, are encouraged to ride in the danger zone.
For another look at Chicago’s deceptive practices, see John Allen’s article “Honey, they shrunk the cars!”Installing striped bike lanes next to parked vehicles is, at best, a bad practice. If it is done anyway, the space required for parking space and bike lane combined is eighteen feet (see diagram below). The Chicago and Cambridge installations allow only about twelve feet, the minimum that meets AASHTO Guide standards. Other plans have been even worse. A proposed bike lane in Austin Texas would have allowed only 10 feet including the parking area.
Some cities combine a parking lane with a bike lane. This not only leads to riding in the door zone, it also encourages the dangerous practice of cyclists weaving in and out between parked cars. This “peek-a-boo” riding style can easily lead to collisions.
Communities that feel the need to encourage bicycling with pavement markings have a safer alternative. They can use the shared-use marking, which is newly approved by the National Committee on Uniform Traffic Control Devices, and which will appear in the next edition of the Manual on Uniform Traffic Control Devices. Just make sure the markings are placed a minimum of 13 feet from the curb face on streets with parking. For more information about better alternatives, see part 9 of this series: Appropriate Facilities & Best Practices and Best Practices of Cycling Advocacy.
For more information about bicycle operation next to parked vehicles, including a unique idea of using “parking crosses” to mark a door zone buffer, see the article Bicycling and On-Street Parallel Parking. The parking crosses (also shown in Fig. 10) improve safety by discouraging door zone operation. They also encourage motorists to park within the parking lane. However, even if adequate space is provided, a bike lane stripe creates the intrinsic hazards described above.
Even More Dangerous Bicycle Facilities
Generally the most dangerous bicycle facility is a sidepath (a path beside the roadway). Such a path is essentially an asphalt sidewalk and it presents the user with the dangers of riding on sidewalks: a potential crash scene at every road crossing and at every driveway.
Note the sign in Fig. 11 at right that instructs the cyclist to walk his bike (across a driveway leading to a picnic area). There is no stop sign for traffic on the road in the background, nor are motorists asked to push their vehicles through any intersection. This author estimates that less than one cyclist in 1000 stops at the sign. Its purpose is apparently to deter lawsuits (blame crash victims).
Riding on sidewalks and paths also introduces conflicts with pedestrians. Pedestrians are much more maneuverable than cyclists. They can stop or turn in a stride or even step sideways or backwards. Bicycles are vehicles; and they must maneuver accordingly. People walking on such a path do not expect high-speed traffic; they expect to be able to relax while taking a leisurely stroll.
Many bicycle facility designers introduce hazards because they simply do not understand bicycle operation. Often these designers are landscape architects, not trained in roadway design. They produce paths that twist and wind around and under trees, with sight lines obscured by foliage and other obstacles. These paths may look pretty, but they are unsafe except at walking speed.
In Fig. 12, if the cyclist brushes his front wheel against the curbstone, the wheel will be diverted to his left. This would dump him into the roadway to his right, possibly under the wheels of a passing car. In addition, at the curve in the background is a guard rail that has long bolts protruding towards the path. One of these could cause a nasty gash if brushed by a leg.
Bicycle advocates are often enamored of European practices, wishing to re-create Europe in America. They neglect the significant differences between the old world and the new. Recently even Europeans are questioning some of their practices.
The city of Copenhagen published the study Road safety and the perceived risk of bicycle facilities in Copenhagen. The study found that after installation of “cycle tracks” (sidepaths) “an increase of 9-10% in accidents and injuries has taken place.” The largest accident increase occurred to women. Cycle lanes (bike lanes) “has resulted in an increase in accidents of 5% and 15% more injuries.”
Ironically, a survey of Copenhagen cyclists found that they perceived greater safety on the segregated facilities even though these facilities are actually less safe.
Unfriendly RoadsIn a later section of this article, we show how a well-designed road that is adequate for the traffic carried is an excellent bicycle facility. In other words, every lane is a bike lane and every road is a bicycle facility. Unfortunately, some roads are poorly designed, badly maintained, or overcrowded with traffic (Fig. 13). Inadequate roads are less comfortable and less safe than they should be for all users, including those driving bicycles.
The most frequent road defects involve poor repair (potholes, cracks, etc.) Because bicycles are balanced vehicles, often with hard, narrow tires, surface defects affect cyclists more than other users. Other problems include confusing situations that induce mistakes, non-responsive vehicle detectors at demand actuated traffic signals and wide intersections with inadequate time for traffic to clear when the light changes.
In some developing areas, road improvements lag development. This causes extreme congestion as the roads are used beyond their capacity. Heavy, high-speed traffic on what were once rural roads squeezes out cyclists and causes conflicts, particularly at intersections. You can read a discussion of this problem in the Triangle area of N. Carolina in Urban Traffic on Rural Roads by Steve Goodridge.
Hiding Behind Standards
Standards can not replace knowledge and even competence by those who design bicycle facilities. This is especially critical when the standards have dangerous flaws such as in the Guide for the Development of Bicycle Facilities . Engineers are expected to do professional work as experts in their field . However, we often see planners and engineers hide negligent bicycle facilities work behind weak standards.
Imagine if bridges were designed like bikelanes. Let’s say a two-lane bridge is to have a load limit of 40 tons. The engineer, working according to official guidelines, would make the bridge strong enough to support one 40 ton truck, but not two trucks on the bridge at the same time or even one overloaded truck. After the bridge collapsed, everyone would blame the truck drivers for the accident. No professional engineering organizations would question this practice or correct the standards.
The outside edge of the bike lane that killed Dana Laird (see above) was 12’5″ from the curb. It actually exceeded the minimum width of 12 feet for the combined width of bike lane and parking space (although not the “preferred” 14′) specified in the Guide. Critical parts of the AASHTO bike lane standards are dangerously inadequate . It may take large legal judgments to make these standards safer.
Where there is not enough room to install bike lanes without adding explicit hazards (in addition to the intrinsic danger of separate facilities) then there is no excuse for installing bike lanes.
Professional Engineers are required to conform to a code of ethics: The National Society of Professional Engineers NSPE Code of Ethics for Engineers includes the following:
Engineers, in the fulfillment of their professional duties, shall:
[…] 2. Perform services only in areas of their competence.
Rigorous professional certification for bicycle planners and engineers is sorely lacking. Joining the Association of Pedestrian and Bicycle Professionals or taking a bicycle facilities design course from the Federal Highway Administration does not ensure competence about bicycling. We have suggestions for professional training in section 9 of this article.
If an engineer designed a bridge according to the methods often seen in bicycle facilities he would likely have his engineering license suspended or revoked for malpractice. At this time, we know of no engineer disciplined for designing dangerous bicycle facilities. The current state of public knowledge for bicycle matters is so poor that malpractice is normal and expected.
We should acknowledge that the AASHTO Guide, as badly flawed as it is about installing bike lanes next to parking and at intersection approaches, does at least specify minimums. Cyclists can and do cite the AASHTO Guide to stop some of the worst proposals for dangerous bike lanes and shared-use paths.
The traditional schoolyard bike rack that holds a bike by its wheel was designed for children’s bikes with cheap steel wheels. Most adult bicycles have lightweight aluminum wheels that can easily be damaged by these “wheel benders”. Knowledgeable cyclists refuse to use them. Fig. 14 shows a concrete block with wheel slots. This is much worse than a schoolyard rack. The cyclist has instead hitched to a bench that supports his vehicle better.
Recommended Reading About Bicycle Facilities
- Bike Lanes and Safety Research
- Bike Lane Contrarian slideshow
- Articles on Bicycle Facilities (John Forester’s critiques)
- A Realistic Look at Bicycle Facilities, Laws and Programs
- Lessons of the Vassar Street Sidepaths
- Wide Outside Lanes are Superior to Bike Lanes, Why segregation is bad
- Bike Lane Stripes: Do they improve conditions for cycling?
- Critique of FHWA Bike Lane Versus Wide Curb Lane Study–Faulty design and analysis plague this often-quoted study
- Implications of Universal Access Principles for Bicycle-Specific Roadway Markings
- Dilemmas of Bicycle Planning
- Bikeway Activists in the Wrong Lane
- Cycle path safety: A summary of research (British article)
- Cyclecraft, book by John Franklin
 “Blue Bike Lanes for Greater Safety”, article on the Portland, Oregon web site.
 Wayne Pein photo.
 Guide for the Development of Bicycle Facilities, American Association of State Highway and Transportation Officials, 1999
 Robert Winters photo from Cambridge Civic Journal.
 Bike Lane Design Guide, published by the Pedestrian and Bicycle Information Center of the University of North Carolina in cooperation with the Chicagoland Bicycle Federation and the City of Chicago.
 The undersized vehicles were first pointed out in a review by John Forester: Bicycle Facility Selection: A Comparison of Approaches by Michael King, Architect. Bikelane Design Guide by Chicago DOT
 Dan Burden photograph, www.pedbikeimages.org.
 Each state defines requirements for a Professional Engineers license required for certain public safety functions, such as designing buildings, roads and bridges. Portions of the Code of Ethics have been incorporated into state law. For example, the Ohio Administrative Code §4733-35-03(B) states: The Engineer or Surveyor shall undertake to perform assignments only when he and/or his consulting support are qualified by training and experience in the specific technical fields involved.
 For another critique of the “AASHTO Guide”, see AASHTO and Door Zone Bike Lanes.