How to ruin a buffered bike lane

If you’ve followed me for the last decade and a half, you know I’ve never been a big fan of bike lanes. Most of my experience has been with ones that were too narrow, next to lanes that were also narrow. And then there are the door zone bike lanes — an abomination that should not still exist. Compared to lane control, narrow bike lanes decrease passing clearance while increasing intersection conflicts, or at least increasing work load approaching intersections. I never felt that was a fair tradeoff to have my own space.

Turns out, it’s possible to design bike lanes that even I like.

Since moving to California, I have enjoyed riding in wide, buffered bike lanes. They offer a comfortable space, sometimes equivalent to a full travel lane.

In a wide, buffered bike lane, you can ride double with a friend. Intersection conflicts are mostly resolved. It’s easy to move left to improve sight lines. This approach offers a wide mixing zone that is intuitive for car drivers to merge into. You still have to watch for the screened left cross, but a large clear area helps you see and be seen. It’s not much different from controlling the right lane on a multi-lane road, except that you are not sharing the lane with cars outside the mixing zones.

I’d prefer a buffer on the right. Unfortunately, in a world where people have been repeatedly told to fear overtaking traffic above all the far more common threats, they’re going to want the buffer on the left. Whatever. It’s fine as long as I can use that space for defensive driving purposes, and cross it to prepare for a left turn.

And yet, a buffered bike lane is not good enough for some advocates.

Imagine my horror at seeing a movement to convert these bike lanes to “separated” bikeways by adding barriers such as flex posts, bollards, curbs, and a host of other innovations.

I get the desire to feel protected from cars, but at what cost? First of all, “feel protected” is all you get. Posts and curbs will not stop a moving car. They will, however, cause a bicyclist to crash. This is a known hazard which causes actual casualties, including serious injuries. Yet, these crashes don’t show up in national crash data, because it counts bicycle crashes only if they involve a moving motor vehicle.

Barriers create additional operational problems for bicyclists.

Barriers decrease operating space and utility. A bike lane with a painted buffer allows for operating space to pass a slower user, avoid a hazard, improve sight lines when approaching an obscured intersection, and change lanes to prepare for a left turn. All those movements are restricted or prevented by a barrier.

A barrier in the middle of the buffer reduces the buffered area by more than half, because the barrier itself requires shy clearance.

Oh, and surface hazards we need to avoid? Those accumulate because the barrier creates a requirement for special street sweepers to clean the bike lane. So you not only have the expense of installing the barrier, you have extra expense to keep the bike lane usable. Flex posts that are knocked down have to be replaced. In the meantime, the broken pieces are an additional crash hazard. They’re bad for the bicyclists, and bad for the taxpayers.

In cold climates, snow can’t all be plowed to the curb, and the bikeway becomes a sheet of black ice when snow melts across it and refreezes.

They may even be an ADA issue.

In my 15 years of teaching, I have taught numerous students who have good enough eyesight to ride a bike, but can’t get a license to drive a car. The bicycle plus training gives these students independent mobility. In areas where bus headways are often an hour with multiple transfers to get to many destinations, being able to use a bicycle for transportation is freedom! The catch is, many people who can’t pass the vision test for a driver’s license don’t have the visual acuity to see some hazards. As seen in the story above, people with perfect vision have also failed to detect posts, and crashed. 

That is a lot of downsides.

We should ask, what problem are we solving?

The research behind the FHWA’s claim didn’t include junctions, only mid-block segments.

The only relevant crash type is a mid-block overtaking crash, around 5% of total crashes for all roads, including ones with no bike lane. The majority of overtaking crashes are actually sideswipes in narrow lanes (the motorist misjudges the space). We have a robust dataset from Mighk Wilson’s crash analysis in Orlando. In it, overtaking crashes on streets with bike lanes were 1.5% of crashes. The majority of bike lanes in the area are narrow and non-buffered. Paul Schimek’s study in Boston came to a similar conclusion.

The FHWA document states that 1/3 of Killed/Serious Injury (KSI) crashes are caused by overtaking motorists. In the data from the Orlando Metro area, overtaking crashes accounted for 10% of KSIs and overtaking in bike lanes was 2.1% of KSIs (14 crashes in 7 years). The percentages for overtaking KSIs will certainly be higher when adding rural roads to the dataset. Overtaking crashes are worse on rural roads. But that’s not where they’re wanting to put separated lanes.

Vehicle speeds and time of day are important factors, too. The risk on urban streets is minimal. In 7 years, out of 657 total KSI crashes, there were zero KSI bike lane overtaking crashes on 25-30mph streets, two on 35mph streets, and both of those were at night. On 40-50mph roads there were 38 KSI overtakings, and 60% were at night. If you’re looking for truly effective crash mitigation factors, reduce traffic speeds and improve lighting. A little traffic justice would go a long way, too. We’re talking about crashes caused by careless driving, not innocent mistakes. Why is the proposed solution to put the bicyclists behind bars?

For a deeper dive, see Mighk’s 3-part series on his research here, here, and here.

Research result, or advertisement?

The FHWA social-media post and linked documentation read and look like a commercial advertisement. An “up to” claim is an advertising trick to mention only a single most favorable result without incurring legal jeopardy. Such claims are common for weight-loss programs, car ads promoting gas mileage or EV range, and the like. From a government agency purporting to show the results of research that affects public safety, it’s… troubling.

If the finding being promoted is a comparison between a bike lane with flex posts and one without, then the only honest comparison is between a bike lane with a painted buffer and one with a painted buffer and flex posts. Preferably, it should be before and after on the same roads. Otherwise, there are likely to be confounders distorting the data.

Can there even be a statistically significant number of overtaking crashes in buffered bike lanes? I have not seen the comparison streets or crash data behind this finding. We’re trying to obtain it. But looking at long-term metro area crash data, I feel confident saying the bold claim of 53% crash reduction is deceptive. The real exposure can’t be high enough for that to be meaningful.

If it were, we’d have a larger problem in society than flex posts could solve!

Engineers have an ethical responsibility to do no harm.

Converting a useful, comfortable facility to one that increases cost, reduces mobility, and introduces crash hazards should be supported with significant, irrefutable data that it will do more good than harm. It’s not OK to ignore the existence of the uncollected data for bicycle crashes not involving motor vehicles. Public policy that affects people’s safety should not be made on dubious findings, or aimed at people’s feelings.

And that brings me to my next point. There has long been an end-justifies-means attitude in bicycling advocacy. At times, it borders on immoral (see door zone bike lanes). People who want other people to ride bikes want facilities that make them feel safe. Making people feel safe without actually making them safe, while at the same time decreasing the utility of a perfectly good facility, is counterproductive. To make matters worse, it sends a damaging message that even wide, buffered bike lanes are not safe enough unless they have some sort of barrier. 

Separation, or separation theater?

Even the most wishful thinker has to realize that separated bikeways will never be available on every road. They are certainly not appropriate on every road. That’s not to say that separated bikeways are never appropriate. They are warranted in some high-speed corridors, but the separation should be more robust than flex posts if it is actually to protect bicyclists from reckless drivers. Flex posts are separation theater, not actual protection. Overuse of this type of facility in inappropriate contexts sends a message which is far more likely to discourage bicycling than encourage it.

---

Admins of the Encinitas Facebook group have been tracking injuries in the separated buffered bikeways on Highway 101 in Encinitas. This stretch of road had no crashes prior to the installation. They are at 32 and counting now, including one fatality.