Saturday, 18 February 2012

How Wide Should the Lane Be?

Is this bike lane preferable? This cyclist doesn't seem to think so.
Is this wide outside lane preferable?
Why not this standard width lane?

Many times, I see discussions online about lane width. We are told that cyclists need bike lanes, or that, in the absence of bike lanes, cyclists need wide outside lanes so that we can operate safely alongside motorized traffic. Personally, I think these ideas are both based in misconceptions about what bicycles are and how they work. Both are based in the false notion that cyclists need only a small amount of space on the road, and that bicycles are somehow fundamentally different from other vehicles and ought to be kept out of the way. But in practice, narrow bike lanes tend to confuse motorists and increase conflicts at intersections. Conversely, wide outside lanes prevent cyclists from exerting the control over the lane that they may find necessary. The fact is, there is a precise width of lane that must exist for cyclists to operate their vehicles safely, and it's not the 4ft standard width bike lane, nor is it the 6ft standard width bike path, nor is it the 14ft wide outside lane. The proper width for a bike lane is the same as that of a standard or minimal width traffic lane - between 10ft and 12ft. Let me explain why.

In terms of bike lanes, they were originally designed for the convenience of motorists and they are almost always implemented in ways that were influenced more by what could be squeezed into the existing road width than by what cyclists actually need. Let's face it - bike lanes are as narrow as their designers can get away with. If transportation engineers figured that they could get away with making bike paths 2ft wide, they would do just that - and in some cases, they do. The US is better than Europe in this regard, and the standards here are (mostly) followed, but the fact is, that standard width of 4ft is not wide enough, nor is 6ft, which is the standard width for bike paths, a width that some cyclist advocates are requesting be made standard for bike lanes too. Even 8ft is too little. Such narrow lanes are ill-suited for use by bicycles, because bicycles, being two-wheeled vehicles with an inherent lack of stability and and proneness to wobble (especially at low speeds), need a lot more lateral room than their appearance would suggest. In my opinion, this is a big reason why cyclists on narrow bicycle facilities have so many collisions with static objects. It is also a big reason why bike lanes tend to increase intersection conflicts and why so many cyclists are killed by turning motorists at junctions or as a result of doorings mid-block. Cyclists (and indeed motorists) need a bike lane to be between 10ft and 12ft wide. 10-12ft allows motorists to fully merge before a right turn, and it allows cyclists to operate outside the door zone mid-block and with no fear of their vehicle's inherent instability putting them in conflict with other road users. 4ft is nowhere near enough, nor is 6ft.

As for wide outside lanes (14ft or more) vs. standard outside lanes (12ft), I prefer a standard width lane over a wide lane. The standard 12ft wide lane evolved to be the way it is for a number of reasons, and the design of both the bicycle and the car was influenced by this. The design elements that made up the bicycle, the car and the standard width lanes they travel on did not evolve in a vacuum, and although the full rationale behind these design elements is little-considered these days, there are good reasons for the standard width lane that go well beyond what is necessary for automobile traffic. Wider outside lanes would be good if many of us were driving tanks on the road - but they are not good for other vehicles, because lane control is essential if 'right of way' is to have any meaning, especially where cyclists are concerned. And unless we want chaos on the roads, right of way must have not just meaning, nor just force of law, but it must be supported by road design too. 12ft is good, because it allows cyclists to control the lane when necessary, preventing compact cars and larger vehicles from overtaking in the same lane. 14ft or more is not good, because of two reasons: 1 - in many states, lane control is illegal for cyclists on a 14ft lane (because cyclists are required to operate far-to-the-right in such a lane), and 2 - even where it is legal, when a cyclist finds it necessary to control the lane, the wide lane encourages motorists to squeeze past in the same lane as a cyclist, potentially on either side, at a time when such a maneuver could be extremely dangerous.

The best lane width for bikes is between 10ft and 12ft - no more, no less. That was the case in the 1890s and it's still the case today. It will continue to be the case as long as the bicycle is around, even after the last car disappears from the road. Bicycles are designed for relatively wide lanes (not too wide so that lane control becomes a problem, not too narrow so that bicycle control becomes a problem) - this is something that transportation engineers consistently fail to understand, because they are often not well versed in the mechanical and physical properties of the bicycle, so they do not understand how bikes move on the road or why their operation requires more than a narrow strip of asphalt.


  1. As a cycling commuter, I would fully agree with your synopsis in this article. I've found that I prefer 12 ft wide lanes or narrower for precisely the reason you provide, better lane control. Lucky for me, most of the roadways on which I ride are four lanes or more, so centering in the lane makes life easier for overtaking motorists.

    I've noted that your blog has an RSS feed for comments, but I've not been able to locate a feed link for the body of the blog. I've been lucky that other blogs to which I am subscribed have linked to yours. Can you enlighten me on a link for the main feed?


    1. Thanks Fred,

      To get updates from the blog, I think you have to join as a 'follower' - there should be a clickable button saying 'Join this site' on the right hand side of the website under 'followers'.

    2. On the issue of lane control, I think a lot of veteran commuters eventually reach the same conclusion about lane width as you (and I) have. It's ironic in a way, that wide outside lanes actually enforce gutter riding, because one would think that more space equals more freedom. But that's not how it works in practice, as the extra room only really gives freedom to the faster vehicles, because more width equals the loss of the ability to control the lane for narrower vehicle operators. And when you have no control over the lane, the only workable option is to stay out of the way of faster traffic.

      For me, it took many years to get confident enough to start to exert control over even a narrow lane, and for years I just assumed that staying out of the way of cars was logical. It took a lot of experience for me to figure out why I was having so many close calls on the road, and I see now that I rode quite unsafely for years, due to trying to stay out of the way. Some cyclists figure out very fast that gutter riding is dangerous - but for folks like me it takes a long time to sink in.

      Nowadays, I am much more confident on the road, and better at controlling the lane when necessary. On very busy roads, there's still a certain nervousness at the idea of riding assertively, but having seen the studies and statistics, I know it's safer to do so.

  2. I follow your logic here, except for the instability of a bicycle requiring 10- to 12-foot wide lanes. I think 6 feet is plenty wide for a single bike. Also consider that when a group of cyclists control a lane, they naturally move into a two-abreast arrangement. And they can ride this way safely.

  3. Firstly, bikes work on principles that demand that they do not travel perfectly straight - no cyclist can keep a perfect line and the average cyclist's wheels describe a wave pattern on the road about a foot wide, if not more. Secondly, bike lanes are not 6ft wide. The standard is 4ft. But even if they were 6ft, that only allows 3ft either side of the tires. And that doesn't take into account the bike's width. When you consider that most bikes+rider are about 2ft wide, that only gives the cyclist two feet of extra wobble room each side, before they're hitting the gutter debris or edging into the next lane. I don't know about you, but I've certainly had numerous experiences in which a wobble or an avoidance maneuver - the need to avoid a wet/slippery manhole cover for example - has resulted in me moving more than two feet from my line. When a bike lane is only 6ft wide, that's hazardous. When a bike lane is only 4ft wide, as most are, that only leaves 1ft wobble room on either side, before the cyclist is entering danger territory. And when you consider door zone bike lanes and the doorings that go along with them into the equation, that makes even riding in the middle of the bike lane a potentially deadly pursuit.

    With all the above factors, I have to conclude that cyclists need, at the very minimum, 6ft from the curb or from parked cars (just to avoid car doors or unexpected pedestrian conflicts), and at least another 4ft on the left, to cater for the usual road problems and bike control issues and to ensure that such issues don't put cyclists in conflict with the lane to the left. There's no getting around the fact that that demands a minimum of 10ft.

  4. Ten feet is better than twelve simply because it tends to make all the motorists a bit more cautious and inherently understanding about why the silly idiot on the bike is occupying the entire lane. An interesting adjunct to the wobble observation is that inexperienced cyclists (the one people claim bike lanes will attract) need MORE room.

    1. Yes, I tend to agree with you. I think ten feet is probably the optimal width for a lane.

    2. Long live the ten foot lane!!!

      To Steve's point, this is why it's easier to convince a novice to adopt lane control (so long as you teach them correctly). Novices are uneasy riding near the edge, whereas roadies consider it a measure of their skill.

  5. Long Live John Forrester and his Vehicular Cycling ideas which have kept the mode-share of bicycling at 1%!

    1. An interesting assertion, but I don't see much in the way of persuasive argument there.

      Since, during the last few decades, John Forester's philosophy of vehicular cycling has been completely ignored by the vast majority of bicycle advocates, transportation engineers and government (except in terms of being targeted as an object of ad hominem attacks, sarcasm and ridicule) I don't think anyone can seriously suggest that vehicular cycling is to blame for the failure of cycling advocacy to increase cycling mode-share. Nice try, but that dog won't hunt.

      I think it's long past time that so-called bicycle advocates admitted that their chosen philosophy of cycling inferiority has been an abject failure. Not surprising, since it has made its corner stone the idea that the road is scary and should be avoided. Is it any wonder that 99% of people still don't want to get on a bike?

      I think it's a great shame that there is a witch hunt mentality that is demonizing vehicular cycling and basically bringing a whole avenue of potential cycling advances to a standstill. I think it's very sad indeed that this situation is likely to cost many cyclists their lives over the next few decades.

  6. Fred Oswald wrote this, but Blogger seems to eat all his posts, so I'm reposting it:

    While I also much prefer a "real" travel lane, I think your argument that 6' is not enough is weak, except on a fast downhill run. A reasonably skilled cyclist should be able to stay in a 6' lane and to maneuver around obstacles as needed. But a real lane is much better for seeing and being seen.

    The main safety problem with typical segregated bike lanes (which, in my experience are usually 5' wide, not 4) is that often much of that width is not safe to use. When these BL's pass parked cars, typically half or 2/3 are in the door zone. Someone trying to teeter along the left stripe is inviting close passing. And this allows no margin for error nor a "flinch allowance".

    Perhaps I should acknowledge that in my travels, I've seen BL's in 3 places that were far enough from parked cars to be outside the door zone. One was in Santa Cruz, CA and the other two in Victoria, BC, Canada. However, I saw other BL's in these cities in the DZ.

    There are many other problems with segregated BL's, including requiring improper positioning at intersections, encouraging motorist mistakes, discouraging cycling education (indeed, providing "anti-education) and otherwise reinforcing nasty fallacies about bicycle operation (see

    Fred Oswald

  7. I didn't see the filth mentioned, but if readers want to go back to the pictures above, one will see a lot of grunge swept into the bike lane or wide outside lane by cars. That, as well as right hook or left cross problems, also puts me into the camp of preferring ten or eleven foot travel lanes safely shared in urban and many suburban settings.