Thursday, November 07, 2013

In praise of speed bumps

The reason we have roundabouts is that they let traffic flow without stopping through intersections. The problem with roundabouts is that having traffic flow without stopping is extremely hazardous to pedestrians. Roundabouts seemed like a great idea when when we were building them on roads that don't have pedestrians, but then we started putting roundabouts in front of high schools and other high-pedestrian areas, and now we have one hell of a mess. Since the Homer Watson/Block Line roundabout was built, pedestrian accidents and injuries have doubled.*

It was folly to build so many roundabouts so quickly, but now we have them - at great expense - and we have to find a way to make them safe for pedestrians. The only solution I can see is speed bumps. For new roundabouts we could use temporary metal speedbumps until drivers get used to them. Problem roundabouts like the one on Homer Watson should have permanent speed bumps.

Jeff Outhit has argued that speedbumps are bone rattling, snowplough-wrecking menaces (I may be exaggerating here) that fire departments loathe. But well-built speedbumps are shallow and smooth - it's mostly parking lot speedbumps that are the bone rattlers. Plus, nobody, not even a fire truck or ambulance, should be speeding through a roundabout.

Most of our roundabouts were built on the edge of town where there are few pedestrians, but as we increase density there will be more people travelling on foot. We've got to fix our roundabouts so pedestrians, even children, can be safe on them.

*According to Jeff Outhit in The Record, during the 509 days before the Homer Watson roundabout was built seven people were hurt, none seriously, in five injury-causing collisions. In the 509 days after the roundabout was built 14 people were hurt, including two seriously, in 10 injury-causing collisions.


Anonymous said...

Your whole argument that there is a problem with roundabouts in general is based on almost completely useless stats.

First, you are only looking at data for only one roundabout. Therefore, you can't generalize past that one roundabout.

Second, you are only looking at the data for two time intervals, one before and one after. What if the 509 days before the roundabout had a particularly low number of accidents? That hypothesis may not be true, but the data you give would fit that hypothesis just as well and your hypothesis that there is an increase in accidents as a result of the roundabout. Therefore, you can't even make a general statement that accidents have increased at that one roundabout.

Furthermore, you haven't provided any evidence that there has not been an increase in accidents at other intersections where roundabouts have not been installed (maybe for some reason unrelated to roundabouts there has been an increase in accidents). I'd say you are committing the classic correlation = causation error here, but you haven't even provided enough data to remotely show a correlation.

Yappa said...

To Anonymous at 10:54.

Hi and thanks for your comment. I'm not arguing against roundaboauts; I'm arguing for using speedbumps on dangerous roundabouts.