invisiblehand

An idea that gets thrown around a lot is that as the proportion of traffic that is cycling increases, cycling becomes safer. Generally there are argument is that motorized traffic learns to expect and what to expect from cyclists such that there are proportionately less accidents.

Anecdotally, I think that this is the case. Recently, the Arlington PD released the most troublesome intersections -- sorry I don't have a cite ... it was during a presentation -- and I was pretty surprised by the results. Essentially the crazy intersections with lots of cyclists did not appear on this list.

Here is the primary example ...

https://maps.google.com/maps?ie=UTF8&...01179&t=k&z=20

(1) Two right turning lanes from an I-66 exit ramp heading onto Key Bridge (north side of the picture)
(2) Heavily used path arriving from the east at a different grade from the exit ramp from the highway.
(3) Another heavily used path and commuter line arriving from the west along the sidewalk. Note that cyclists traveling along the sidewalk from the west are heading downhill.
(4) Motorized, cycling, and pedestrian traffic from the south is also very heavy.

This is an extremely busy intersection during rush hour. And a very fast intersection during off-peak hours. So one would expect people to be getting killed all of the time here. Turns out that despite the crazy design, no one seems to recall a cyclist-auto accident there for years ... reported or unreported. My thought is that everyone -- drivers and many cyclists -- realize that the intersection is a death trap and to expect cyclists zipping down the hill, popping up from the nearby park, cars aggressively trying to catch the light across the bridge, and so on.

I assume that there are similar observations elsewhere.

noisebeam

People do what has worked previously. They keep doing it until something happens significant enough to them to cause them reason to adjust previous behavior. This previous behavior was not necessarily risk free just that the risk was so low that the behavior is perceived as safe. If one increases the risk (say more cyclists) then there will be more encounters and more learning.

Al

invisiblehand

I had similar thoughts. I don't have historical data for this intersection; but I suspect that there were a few accidents at some point after which people adjusted to some new equilibrium.

coldfeet

Interesting, would you say this has some connection to the places in Europe where they are taking out the road signs?
What they are finding, is when there are "no rules" everyone starts getting cautious. This junction looks so much of a free for all, that I, for one, would drive it with my head on a swivel, and maximum alertness.

invisiblehand

That is interesting. I had no idea that this was happening.

I think that it makes examining a local area for trouble spots difficult. Suppose the reason why there are a lot of accidents at one intersection is due to another intersection being dangerous.

Roughly outlining the idea, a dangerous intersection that requires extra care has everyone slow down and raise their level of awareness. If you think that after passing the high-risk area that people (1) increase their speed to make up lost time and (2) lower their awareness to recuperate from the effort then accidents would rise.

Alternatively, consider that people in the area learn that a particular intersection is high-risk. So they avoid that intersection and travel though an alternative increasing traffic past its "safe" capacity.