Thursday, January 19, 2012

What Is A Sharrow And What Does It Want From Us?

Not a bike lane, but a bike-friendly street
Okay folks - we've heard there is some confusion about a new bicycling symbol that is popping up with increasing frequency on Philadelphia's streets. They have been spotted in Manayunk, in Center City, and in South Philly. Here's the lowdown on this new addition to Philly bike paint:

What Is That Funny Marking?
If painted on a road, it is a sharrow (a mix of the words 'share' and 'arrow'). If displayed on a sleeve, you are addressing a Corporal in the Bicycle Army.

Is It The Same As A Bike Lane?
No. It marks a lane shared by vehicles and bikes.

So What's The Point?
A sharrow is a visual reminder that bicyclists share the road, and offers an ideal 'line' for bicyclists to take on the street. They have many uses, such as bridging a gap in a bike lane network (example: on Pine between 23rd and 22nd St).

Is a bike lane
Where Are They Placed?
They are placed 11 feet from the curb if there is on-street parking, to guide cyclists away from the "door zone." Without on-street parking, they are placed 4 feet from the curb. Riding away from the curb keeps bicyclists away from debris and discourages unsafe passing by motor vehicles.

Did San Francisco Invent This Crap?
Sorry, yes. The concept was developed in San Fran as an improvement upon an existing, more confusing symbol.

When Were Sharrows Included In The Manual Of Uniform Traffic Control Devices?
We get this question a lot. It was added in 2009.

Are They As Good As A Bike Lane?
Nope. They aren't recommended for streets with speed limits of 35 mph or higher. If a street can support a bike lane, a bike lane make the road safer for everyone.

So, Sounds Like A Bicycle-Friendly Addition To A Street, But Not As Great As A Bike Lane.
Yup. Ride safe!


Kelly said...

I think sharrows are pretty great on streets that don't have room for a bike lane (Grays Ferry for example). Is there any way for the biking public to suggest what streets these should go on?
Hint: It would be really really great to get one on 16th street in Center City.

Ian Brett Cooper said...

Bike lanes are not safer than sharrows. In fact studies show they're more dangerous than an unmarked road. Bike lanes exist to encourage timid cyclists to use the road. They are not for experienced cyclists and, for the sake of the rider's safety, their use should be abandoned once the rider is more comfortable on the road.

Sharrows should be used in any lane which is too narrow to share. The 30mph upper limit is nonsense. Cyclists need to be urged to ride farther out into the lane, as gutter riding encourages drivers to try to squeeze by when there isn't enough room for a safe pass within the lane.

Ian Brett Cooper said...

"shared by vehicles and bikes"

A bike is a vehicle.

A sharrow is really to notify motorists that cyclists may use the full lane. This is, of course, the case on any road that is not restricted for use by cyclists, but sometimes, drivers need a reminder and sharrows perform that function.

andrewlevitt said...

Sharrows = "Bicycles Have Full Use Of Lane."

On normal streets, people might forget that bicycles have full use of lane everywhere.

And without the "Full Use Of Lane" part, people in cars might think that "sharing the road" means "bikes outta my way".

Rich F said...

I'm wondering how much money the Philadelphia area plans to spend educating cyclists and motorists about proper lane positioning for cyclists? My worries are: 1)that communities typically paint sharrows on the road before they publicize what they are for, and 2) their use creates the false impression that cyclists may only ride where there are sharrows or bike lanes.

What might happen regarding the advancement of cycling (and a better level of vehicle operation in general), and true "sharing" of roads, if proper cycling were taught in the school phys ed program, driver's ed, and driver certification testing included cycling questions? Will we ever know the answer? Not if we keep putting the cart before the horse - or the sharrow before the education.

Anonymous said...

Kellly - You can e-mail Aaron Ritz, the City's Active Transportation Coordinator ( and suggest those sharrows!

Ian - experienced riders can and should take the lane and use the road. But we're in the business of encouraging all types of bicycling, and towards that end, bike lanes are invaluable.

And on safety: our research on Spruce/Pine found a 44% decrease in serious crashes after the installation of the bike lanes.

Rich F - City Paper just wrote an article addressing this issue:

In it, Jeannette Brugger of the City Planning Commission acknowledges that the educational component of the sharrow program is missing.

Anonymous said...

Thank you for your honesty! I believe that sharrows have questionable efficacy, but are perhaps a helpful compromise until we can garner further support for proper infrastructure.

As for the lone John Forester disciple above, I will continue to ignore the continual trolling.

Anonymous said...

A note from Charles Carmalt, the City's Pedestrian & Bicycle Coordinator, on why the sharrows are sometimes in the left-hand lane (originally posted to the BCGP Philly yahoo group):

"On 15th Street, the sharrows were marked on the left side of the street because that is the side of the street where parking is not permitted.

On one-way streets, the Penna. Vehicle Code permits bicyclists to ride on either side of the street. Parked cars are threats to bicyclists and bicyclists will be safer if they ride on the side of the street without the parked cars. In addition to dooring, drivers may pull out from a space unexpectedly or they may suddenly stop in order to park. Both situations happen fairly frequently in areas where parking turnover is high. (Riding against the curb also eases the transition from street to sidewalk when you reach your ddestination).

Riding on the left side of the street can also be advantageous on one-way streets that have high frequency bus traffic."

Ian Brett Cooper said...

"...we're in the business of encouraging all types of bicycling, and towards that end, bike lanes are invaluable."

They are useful in getting people to believe they are safe while on such facilities, despite the fact that numerous studies show that they are less safe.

"And on safety: our research on Spruce/Pine found a 44% decrease in serious crashes after the installation of the bike lanes."

Is this study available? I'd like to examine it, because it goes against almost every previous study I've read on the subject. Are you certain there was no selection bias? Does the study account for how the situation changed during the study period? I see references to cities doing studies, but very rarely do I see the methodology. And let's face it - cities are under enormous pressure to claim that bicycle facilities are safe.

Ian Brett Cooper said...

By the way, I'm all for sharrows. I do not believe that they will lead drivers to believe that bicycles can't use the full lane on streets without them, and I think they will get drivers used to cyclists using the full lane.

My issue is with bike lanes, which study after study has shown to increase right hooks. Right hooks on bike lanes are killing lots of cyclists on 'London's Cycle Superhighway' bike lane - so many have died, in fact, that the British government has advised cyclists to avoid certain areas on the 'Superhighway' as they ride to the Olympic Games this summer.

In Britain, the Cycle Superhighway is starting to be widely regarded as an abject failure. I wish we could learn the lessons that the Europeans have learned, before they get implemented here, but I guess we need to do the same dance over and over. I guess we should take comfort in the fact that at least it will only kill a few cyclists before they scrap the idea here. Glad it's not going to be any of my loved ones, because I teach them to avoid bike lanes like the plague.

John Boyle said...

The 44% is a comparison of 2010 and 2008 PENNDOT Crash data for Spruce and Pine St. You can contact the Mayors Office of Transportation and Utilities for their methodology. To obtain actual crash data you will need to submit a formal request to PENNDOT.

To look at the pre and post bike lane data go here -

Reported bike crashes in Phila. have declined steadily since 1998 from about 1100 to 500 (PENNDOT), while the rate of bicycling has tripled in the same time period. (US Census). In 1998 there were 30 miles of bike lanes. In 2010 there were more than 200 miles of lanes.

So Philadelphia has more bike lanes, more cyclists and fewer reported bike crashes. That doesn't prove bike lane design safety but I believe it would be difficult to counter those facts with studies that categorically dismisses all bike lanes as dangerous.

Ian Brett Cooper said...

A comparison of crash data is not necessarily a 'study'. A study has to eliminate confounding factors. Has this comparison been reviewed? Published in a journal?

Crashes and fatalities can decline despite factors that make facilities more dangerous. Confounding factors such as increases in cyclist populations and mass changes in route can make nonsense of a mere comparison. This is why studies need to be rigorous, reviewed by disinterested peers and widely disseminated so that potential flaws can be examined.

The recent Lusk study concluded that Montreal streets with bicycle facilities were safer than what the study called 'reference streets'. It was only after the study became available online that flaws in the study - primarily blatant and cynical selection bias - were pointed out. Lusk has a lot to gain by promoting bicycle infrastructure - she has made her career in bikeway promotion. Whether she and her colleagues tweaked the study so that the results fit their pre-existing requirements subconsciously or not is open to debate, but the fact is, this sort of thing is not unusual - in fact, where money is to be made, studies often support the financial welfare of the people doing the study.

Lusk's study is still very highly regarded in the media and often referenced. Reid's more recent review is ignored - partially because it concluded that the overwhelming weight of literature comes out against bicycle infrastructure. Reid's conclusions are not politically correct in a climate where bikeways are to be pushed no matter what their safety record is, because cyclists feel comfortable on them, and our crumbling infrastructure needs some pressure taken off it, so getting bums out of cars and onto saddles is where it's at.

Andrew J. Besold said...


There are good bike lanes and there are bad bike lanes. Just because there are really horrible bike lanes in London and elsewhere doesn't mean that they are all bad.

I do have a big issue with bikes lanes that place cyclists to the right of right turning traffic and the use of bike boxes where there is a right turning conflict between cars and cyclists (a la
Portland). I think we both agree here.

Most if not all the bike lanes in Philly are pretty damned good because they replicate proper vehicular cycling lane position. I like this design approach because it trains novice cyclists as to where to best place themselves when there are no bicycle lanes. I can't see how one could argue against the Spruce and Pine lanes. A lane taken away from crazed motorists made into a protected space for cyclists with mixing zones when approaching the intersections.

Lanes work! Don't underestimate the value of a painted line. Imagine the chaos if 6 lane interstates had all the lines removed.

BTW - I've been riding for 20+ years, I'm an LCI and a professional in the bike/ped planning field who is not afraid to criticize bad designs. I agree with you that there is plenty of bad stuff being built but there is good stuff as well. Philly is lucky because they have great people working there who are actual experienced cyclists and understand cycling from an experienced and novice end-user perspective. Unfortunately this is all too often not always the case.

My best!

Andy B

Ian Brett Cooper said...

Andy, all bike lanes are by their nature bad because they ALL create unnecessary conflicts at intersections. This is a feature of the design - it has nothing to do with how they are implemented.

The Spruce and Pine bike lanes, as far as I can see, are segregated 'far to the right' lanes. They feature the exact same problems at intersections that every other kind of bike lane has - they invite right hooks and left crosses. They are segregated and not in the door zone, but that only makes them less dangerous - it does not make them safe.

If only transportation engineers would adopt sharrows and bike boulevards, which actually do help cyclists learn to operate in a safe and integrated manner, we would have far fewer cyclists getting injured and killed on these supposedly 'safe' bike lanes.

Lanes work to separate traffic that moves at similar speeds and when all road users understand that turns can only be made from the outside lane. But that is not the case with bike lanes. When a street is equipped with a bike lane, speed differentials are high between two very close lanes of traffic. More importantly, turns are made from the two outside lanes. This is inherently dangerous. As an LCI, you should appreciate this more than anyone.

Ian Brett Cooper said...
This comment has been removed by the author.
Ian Brett Cooper said...
This comment has been removed by the author.
Ian Brett Cooper said...
This comment has been removed by the author.
Ian Brett Cooper said...

I've expanded on my comments and put them on my blog at