Advocacy & Safety - Rear-end collisions

Does anyone know a reliable source for data on bicycle collisions? Specifically, I'm interested in statistics about collisions occurring after dark. According to Cyclelicious, rear-end collisions are exceedingly rare and cyclists are irrational to worry about being hit from behind:

http://www.cyclelicio.us/2010/bicycle-lights/#more-5446

I have found some statistics from local jurisdictions that cuts both ways. For example, Burlington, Vt. amassed 3 years of data and found that 47% of cyclists were hit from behind, while stats from Austin showed 3%. This data seems too limited to draw any conclusions from.

You only need one to have a bad day;

http://www.bikeforums.net/showthread.php?670912-It-finally-happened-to-me-hit-by-a-car

Does anyone know a reliable source for data on bicycle collisions? Specifically, I'm interested in statistics about collisions occurring after dark.

I don't think there is any good compilation of this data.

Note that the frequency of collision types changes with location. That is, rear end collisions might be uncommon (or less deadly) in urban areas. Front collisions might be less frequent in more rural areas (where there are fewer intersections and where the cyclist can see other drivers better).


According to Cyclelicious, rear-end collisions are exceedingly rare and cyclists are irrational to worry about being hit from behind:
30% isn't "exceedingly rare" (nor is it "rare"). Also, rear end collisions (especially, on high speed roadways) can be more deadly.


70% of bicycle crashes are from the front. If I had to choose between front and back lights, I’d pick the front.

===============


That should be "exceedingly rare," not "exceedingly rear"!
You can edit your posts (to fix these minor mistakes).

===============

Anyway, why do you care?

If you are riding at night, have a front and rear light.

30% isn't "exceedingly rare" (nor is it "rare"). Where did 30% come from? Sounds like one of those "86% of all quoted statistics are made up on the spot...."

Sorry, I do not know any reliable source of such statistics.

I will point out not taking reasonable precautions after dark is just stupid. Unlit, unreflectored and wearing a black shirt is asking for trouble.

I have seen things like that, and once out in the Palmdale Lancaster area I came across a cyclist riding the wrong way in the fast lane of a 2 lanes each way highway. Sooner or later such become statistics. Just pointing out that those cyclists who when added up are the statistics include a lot of people doing things I sure hope no one here does.

My personal opinion is that many of those preaching avout the dangers or lackof dangers of rear end collisions are doing just that preaching a preheld opinion. I have had only one accident involving contact with a car and in that I was rear ended. So I am evidence it is not a vanishingly small risk. But I'll be 100 time more concerned about right hooks and I'll also bet that in some areas those get included as rear enders.

Come to think of it I'll bet that is a major contributer to the difference between 3% and 47% that the OP cites.

Where did 30% come from? Sounds like one of those "86% of all quoted statistics are made up on the spot...."

Look at the link: "From Behind 30%" in the graph below the video.

=======================


According to Cyclelicious, rear-end collisions are exceedingly rear and cyclists are irrational to worry about being hit from behind:

http://www.cyclelicio.us/2010/bicycle-lights/#more-5446


The link indicates that Cyclelicious is not saying anything much like what Freewheeling is saying it does.

Look at the link: "From Behind 30%" in the graph below the video.

I did and I advise anyone to totally discount their numbers. They have forced everything into from the front and from behind. Seems to me I have had far more from the side close calls than either front or back. That they have zero is unbelievable.

BTW at teh time when I did a few doubles, including the 2 back to back for the Land Rush I had my bike setup with reflective tape that meant I'd be visible from any angle as long as headlights hit me. The article does NOT seem comprehensive and is addressing one issue, what is more important front or back lighting and so falls into some very sloppy work.

BTW on that they forgot one very important thing. If you have a rear light and it stops functioning yuo may well be unaware of this for some time. If the one light was all you had and it does not work as a reflector yuo are dark in back. Front lights fail also, but at leat the rider will know right away. (Thinking back on it the lighting system I used back then has 2 headlights, the lower powered one, which was always on, had clear plastic screws on hte sides that meant I was lighted from the side. Not a huge light, but something).

As I see it one does not have to become a lighting and safety Nazi to realize riding in stealth mode after dark is not a good idea.

I did and I advise anyone to totally discount their numbers. They have forced everything into from the front and from behind. Seems to me I have had far more from the side close calls than either front or back. That they have zero is unbelievable.

As I said earlier.


I don't think there is any good compilation of this data.

===============

It appeared you (Keith99) were asking where I got it from.

The only reason I mentioned the "30%" was to refute "Freewheeling Sp's" assertion "According to Cyclelicious, rear-end collisions are exceedingly rear and cyclists are irrational to worry about being hit from behind."

(I have no idea whether it's right or whether it was made up.)

The original post doesn't go anywhere and it isn't clear what "Freewheeling Sp's" is getting at (it doesn't appear that he read the link he provided very carefully).

Does anyone know a reliable source for data on bicycle collisions? Specifically, I'm interested in statistics about collisions occurring after dark ...
I do not like to re-post information that I recently posted in another A&S thread, so here is a synopsis for your browsing convenience.

The often cited major US study is Bicycle and Pedestrian Crash Types of the Early 1990's (FHWA-RD-95-163 (http://drusilla.hsrc.unc.edu/cms/downloads/PedBikeCrashTypes.pdf), all 9 MB of it available online). Of their 85 crash types, 5 (6 including "Type unknown") are in the "Class D: Motorist overtaking bicyclist". They compared their data to the 1977 Cross and Fisher study and for this class of crashes:

1977 study, Total Fatal=166 of which 37.8% were motorist overtaking, Total Nonfatal=753 of which 10.5% were overtaking
1990's study, Total Fatal=41 of which 29.3% were motorist overtaking, Total Nonfatal=2453 of which 9.8% were overtaking.

The North Carolina Dept of Transportation, Division of Bicycle and Pedestrian Transportation, maintains an on-line 1997-2007 database which you can do your own queries on bicycle crashes by location statewide/county/city/region, year(s), and two other fields of your choosing to produce rXc tables of any of the two fields you chose. In your case, the fields of most interest would be "Crash Type" vs. "Time of Day". It would be of most interest if you did this search 11 times, one for each year from 1997 through 2007.

I have done the query on "Crash Type" vs. "Rural / Urban Location" for each year and graphed the results (note this is all crashes, not just fatal ones):
164544

Totals of all crash types were stable to slightly decreasing over these 11 years. While there is no statistically significant change in the fraction of motorist overtaking fatalities or occurrences between the 1977 and 1990's studies, you really do not need to do any statistical analysis to see the fraction of motorist overtaking crashes in N. Carolina is increasing from 1997 through 2007. In rural areas, it becomes a substantial fraction of all crashes.

For further discussion of the above findings, see my posts #67, #81, and #84 (http://www.bikeforums.net/showthread.php?667384-Best-safety-investments-for-cyclist/page3) for other links, additional N. Carolina graphs (including vs. cell phone use), statistical package output, and brief discussions of relative risk, exposure risk, and the statistics of rates and proportions.

I only know of my own experience, and that is that the only collision I've ever been in was a rear-ender, but it was in broad daylight. Luckily the guy behind me spotted me in time, so I just got a little bump - didn't even bend my mudguard.

Of course, I've had a few near misses - a couple of right crosses and a couple of right hooks.

njkayaker - you're absolutely correct that the Cyclicious post does not use the phrase "exceedingly rare," and I would agree that (assuming the numbers are correct) 30% is not "exceedingly rare" or "rare." I posted here only to see whether someone could provide better statistics on this issue (thank you Giro!).

For the record, here is what Cyclelicious wrote:


70% of bicycle crashes are from the front. If I had to choose between front and back lights, I’d pick the front.
Think about this: a motorist approaching from behind has his headlights shining directly at you. Unless you’re a complete bike ninja with black clothes on a black bike and a black backpack, the motorist at least has a chance of seeing you. He has headlights to ensure he doesn’t run over stationary hazards on the road, so hopefully he’ll see moving objects as well. Even cars with broken taillights aren’t invisible to following traffic.

Note that the frequency of collision types changes with location. That is, rear end collisions might be uncommon (or less deadly) in urban areas. Front collisions might be less frequent in more rural areas (where there are fewer intersections and where the cyclist can see other drivers better).

+1

Rural rider here. I believe my chance of getting rear-ended is higher due to winding, hilly roads with few cyclists on them. I choose my routes with safety in mind.