Tuesday, November 3, 2009

Ft Collins analyzes bike accidents

It was about a week ago in conversation a friend told me she was 'hooked' when a car driver passed her then immediately made a right turn into her.

Right hook


"It was so weird," she told me. "I've never heard of anyone get hit like that before."

I, in my super knowledgeable bike advocate smarminess, informed her that this collision is, in fact, the most common type of bike vs car collision. It even has a name: The Right Hook.

It turns out I'm wrong, at least in Ft. Collins, Colorado. The city traffic department analyzed accident reports involving bicycles from the year 2000 to June 2009. The infamous Right Hook -- or what the city calls "Overtaking Turn Accidents at Intersections" -- is the second most common collision at 13% of the total in their count.

The most common collision is "The Broadside," at 60.5% of crashes. That's when a motorist goes straight through an intersection even when there's a bike right in front of him.

The Coloradoan article on this analysis highlights the fact that of the 214 "broadside" collisions, 123 of them involve a cyclist riding against traffic. Naturally, the public comments section focuses on the scofflaw cyclists as the cause of this traffic mayhem.

What the article failed to mention is that in 10% of the broadside collisions, the motorist failed to stop at a stop sign or even a red light. Two of the drivers were DUI. In 130 instances, the drivers were cited for "failure to yield right of way."

To be sure, cyclists should ride with traffic for safety, but that's not the only problem. There's a problem with the bull in the china shop, and that bull should be controlled (to use the methaphor from Copenhagenize).

Left Cross

The third most common collision type is the left cross at 9.3% of collisions. This is when a left turning motorist slams into a cyclist going straight through an intersection. Of the 33 left crosses, 3 involved a cyclist riding on the sidewalk, 2 were going the wrong way, and two failed to stop at a signal or sign. The overwhelming number of these were motorists who just kept going in spite of the presence of a bike in their path.

Hit from behind

After that, the next collision type is the dreaded "Hit From Behind." The 30 "sideswipes" recorded account for 8.5% of bike collisions. With the exception of a single head on, all fatalities are these types.

The report also counts 25 severe injury (including fatalities) collisions out of the 354 bicycle accidents in the analysis. About half of the severe injuries are from the "Broadside" collisions.

Many risks are controllable while cycling and the city report highlights some of the contributing factors that involve bicyclists -- you should generally ride with traffic, avoid sidewalks, and obey traffic control signs and signals. There's still work to do to reign in the bull as well -- Ft Collins with a population of 137,000 has a serious injury accident almost once a month.

But is it dangerous?

For the 9 year period that Ft Collins examined, the accident rate is 0.93 per 1,000 population. Compare that against an injury rate of 7.7 per 1,000 population for all people involved in car accidents. There were four bicycle fatalities in Ft Collins in nine years, compared against two to four traffic fatalities total each year.

While bicycling is generally a safe activity, there are risks in bicycling, and it's good to see Ft Collins quantifying some of those risks.

Download report PDF here.

16 comments:

  1. Do we know how Ft. Collins collects/keeps cycling accident stats? My own local gov't (fairly cycling friendly, as these things go) insists that it's not possible to do a similar evaluation for our city.

    ReplyDelete
  2. Leave it to me to post before RTFA. Data was compiled by looking through police reports. (Which I'd say skews the numbers low - lots of crashes go unreported.)

    ReplyDelete
  3. Thanks for the data. Broadside threats are the most common I experience. I see a lot of people push too far forward before they stop at lights/stop signs.

    I find that action is the one most likely to cause me to yell or call out to make sure the driver sees me. I can normally spot potential right hookers in my rear mirror and make sure they cannot pass me. Both types are drivers that I actively look out for at every intersection.

    ReplyDelete
  4. @MB: you're probably right. I've been about 50/50 in reporting my own bike vs car collisions, and in one instance the cop declined to even take the report. In Colorado, police are now required to file a report if a cyclist reports an accident, but this wasn't always the case.

    @Duncan: I admit that the "broadside" never really registered with me as a threat, but thinking back you're right - it's very common for me as well. I can almost count on a motorist running a stop sign along one residential street I ride every day.

    ReplyDelete
  5. "For the 9 year period that Ft Collins examined, the accident rate is 0.93 per 1,000 population. Compare that against an injury rate of 7.7 per 1,000 population for all people involved in car accidents. There were four bicycle fatalities in Ft Collins in nine years, compared against two to four traffic fatalities total each year."


    These can be misleading numbers. I would love to see these normalized for the percentage of the population that participates in riding and driving.

    .93 per 1000 looks like a very different number if 10 in 1000 rides a bike on or around the roads.

    ReplyDelete
  6. @SEan - I was going to explore that part but it was getting late and I went to bed ;-) But roughly, over that period something like 2% of commuters in F.C. biked to work. 2% of 7.7 is 0.15, so cyclists have roughly 6x the risk of an injury collision as car occupants (fudging also the fact that not all bike vs car accidents result in injury for the cyclist).

    For 2008, the bike modal share in F.C. is over 7%, which edges out Portland OR's 6% share.

    ReplyDelete
  7. I just can't understand how anyone could overtake a bike and then turn in front of them. I believe a right hook is done in only one of two ways:
    1) the car passes the bike and turns, with the intent on causing harm to the bicycle (Why else would you do that?)

    2) The car is turning, and a fast moving bicycle runs into the turning car (the cyclists fault)

    I just can't see how it would happen any other way.

    ReplyDelete
  8. "The Broadside ... That's when a motorist goes straight through an intersection even when there's a bike right in front of him"
    vs
    "After that, the next collision type is the dreaded Hit From Behind"
    What's the difference here?

    @Tony Bullard : I think it happens because the motorist doesn't judge the cyclist's speed well, and believes he/she can safely overtake the cyclist and turn out of the intersection before the cyclist gets to it.

    ReplyDelete
  9. @Tony: I was hooked a couple of years ago. I was cycling at close to 25 mph in a bike lane (Willow Road, Menlo Park, CA). The driver passed me at probably 35 to 40 mph, slammed the brakes and turned directly in my path pulling into a gas station. I grabbed brake but I only had maybe two feet of response time, hit the back of the car and flipped up and over.

    I'm pretty sure the driver either didn't see me or misjudged my speed, so it clearly wasn't intentional.

    So explain to me again how this is the cyclist's fault?

    ReplyDelete
  10. Yakota, I could see it being the cyclists fault if they're tearing down the road and the try and cut in front of a turning car. I'm not saying this happens often, as I imagine most cyclists that daring would just pass the car on the left.

    As for the case where you got hit, I wasn't there, but I'd have a hard time believing the car that crossed in front of you just misjudged. Well, that's not true, there are plenty of stupid drivers out there. I myself would have simply waited till you passed the gas station, but I'm a cyclist, so I think about other on the road more than most motorists.

    ReplyDelete
  11. "bull in a china shop" may not be the right line of thought or argument.
    Cars also very much have rights on the road, and are fundamental to our society.

    Besides, they outnumber us.

    That's not to "justify" or "coutenance" any of the behavoir - especially the really daffy stuff.

    By the way, as a driver, the cyclists who alarm me the most are the ones doing 'weird' things - riding on the wrong side of the road, ignoring red lights, appearing from odd places.
    It's worth thinking about what things you see that baffle you or surprize you while driving, and remember that while riding, and vice versa.

    ReplyDelete
  12. To echo Mr. Willman, I agree.

    We can talk all we want about improving driver behaviour, but the single best thing we can do for anyone who's found this site is to tell them what they _personally_ can do.

    I think the takeaway, from this, is that sidewalk riding, salmon riding, and running lights is more dangerous.

    Ride defensively. If we could put it to a vote today they'd ban bikes, not regulate motor vehicles: So let's not get too excited about that.

    ReplyDelete
  13. Of the 126 broadside collisions with the "Bike riding against traffic", 80 of them were "Bike on sidewalk/crosswalk". It may be a bad idea, but it's not quite the same as riding the wrong way on the street.

    ReplyDelete
  14. The data is interesting, yet it begs for questions it cannot answer. For instance, in a 'broadside' collision - who was at fault? Did the cyclist fail to yield or did the motorist? Prying out details is essential if we're to use this information to prevent future crashes or try to influence public policy. And they're more properly called CRASHES rather than accidents. Crashes have causes and they're preventable. Accidents are more in the realm of Acts of God.

    ReplyDelete
  15. What is most frightening about this data, to me, is that the most deadly type of accident (hit from behind/ sideswipe) is the thing that even riding defensively can do little to avoid. I can avoid a lot of potential right hooks by never coming up next to a moving car at at an intersection, I can stay out of the door zone, and avoid most broadsides and left hooks by being very careful not to go into an intersection until I see that it's clear and make eye contact with any oncoming traffic. But someone drunk or texting who veers out of his lane- not much I can do to see that coming or prevent it. Thats the kind of thing that scares me...

    ReplyDelete
  16. =v= These findings are interesting because they seem to be broadly consistent with the analyses done by Right of Way, on New York City data. Right of Way focused on fatalities, though, and the Ft. Collins study has a much smaller sample size of those (a total of 2, both sideswipes). The prominence of overtaking and cars running red lights and stop signs is about the same, as is that of bikes running stop signs and going the wrong way on the street.

    ReplyDelete