[ygra]

Usually I stand at a crosswalk and thus have little risk of walking off a bridge or into traffic. Also I see little point in always anticipating someone who drives into pedestrians. But if they do, I guess standing to the right of the traffic light gives me a slightly better chance of survival.

And while it routinely happens that motorists will ignore pedestrians or cyclists at corners or that they drive over red lights, I have yet to see someone accidentally driving on the sidewalk at a traffic light, endangering pedestrians there.

[tbirdz]

You should be glad it didn't happen to you, but unfortunately it happens to other people all the time.

http://news.ca.msn.com/local/montreal/pedestrian-killed-on-s...

http://www.streetsblog.org/2013/11/11/three-killed-by-curb-j...

http://www.myfoxny.com/story/19681198/car-drives-onto-sidewa...

http://gothamist.com/2013/11/11/two_pedestrians_killed_after...

http://www.sbsun.com/general-news/20131226/san-bernardino-ma...

And these are just the first few links that show up on googling for "pedestrian killed on sidewalk" there are many, many, many more articles just like these.

[ncallaway]

News is news because it is a rare or uncommon event. I'd be very wary of extrapolating "there are news articles about this event" to "it happens all the time".

I would much prefer a link that provided me actual data in order to believe that this is an event that occurs all the time (such as the number deaths of pedestrians on sidewalks per capita).

For example, the National Highway Traffic and Safety Administration put out a 2008 report (I chose 2008 as it was the top result on Google) listing the average pedestrian fatality rate per 100,000 people by state [1]. It seems to range from ~1 to ~3.

According to the CDC, there are 799.5 deaths per 100,000 population in the United States (2010 data) [2]. This means my chance of dying as a pedestrian, compared to other potential means of me dying are around 0.3%.

A few studies have found that between 80% and 90% of pedestrian fatalities are a result of the pedestrian being at fault (this comes from the Center for Problem Oriented Policing [3], which cites two sources for this information [4] [5]). This means, again as a percentage chance of all other forms of death, dying while standing on the sidewalk is somewhere in the ballpark of 0.06%.

[1] http://www-nrd.nhtsa.dot.gov/Pubs/810968.PDF (page 18, figure 2-4).

[2] http://www.cdc.gov/nchs/fastats/deaths.htm

[3] http://www.popcenter.org/problems/pedestrian_injuries/

[4] Lee and Abdel-Aty (2005)

[5] Teanby, Gorman, and Boot (1993)

[ygra]

I'm a bit hesitant to extrapolate a danger from anecdata. I might try finding actual data tonight. From experience so far I'd guess I'm much more likely to die when exercising my right of way as a cyclist.

[talmand]

If you're speaking of the traffic light pole providing protection, that might not be the case. I'm sure it's not the same everywhere, but for my area the poles are made to snap off at the base in case of a car hitting it. I would imagine the thought is that it saves the occupants of the car from a more severe impact, especially with the big poles.

[ygra]

I wasn't too serious about that, but from what I've seen they don't seem to snap easily; at least here. I'm also not sure whether it's better to have a traffic light pole falling on an intersection instead of just having a car stop at the pole (the former sounds like there could be more injuries involved).

[aestra]

Most are designed to snap off when hit. There is actually a real LOT that goes into roadway sign engineering and we have learned that it is the safest option. It's called a breakaway feature.

http://safety.fhwa.dot.gov/roadway_dept/policy_guide/road_ha...

>Roadside signs, utility poles, lighting structures, traffic signals, railroad warning devices, motorist-aid callboxes, mailboxes, and other rigid objects can become deadly roadside hazards if placed where run-off-the-road vehicles can strike them. While it is preferable to maintain obstacle-free roadside clear zones, this is not always a practical option. When rigid objects cannot be removed or relocated, potential crash impacts can be mitigated by specifying breakaway features, or by shielding the object with a longitudinal barrier or crash cushion.

And here's a video of a light pole breakaway feature being tested. :)

http://www.youtube.com/watch?v=bbg5NkbuN6I

[couchand]

I would call that "feature" a "bug" on a street corner where pedestrians are standing.