1989Pre 

California and Florida have more people per square mile than any other state. That's the reason, I think. Up here in Maine, I can ride for miles and not see a house..., or a car.

Daniel4

"As far as I'm concerned, causes 2-5 are all rider error. My sister-in-law went down"

So just because your sister fell over a root, you conclude that being struck by a vehicle and being in a collision are rider errors?

Let's look at two real life examples:

US:low bicycle-motorvehicle ratio- high rate of fatalities.

Denmark:high bicycle motor-vehicle ratio - low rate of fatalities. Those cyclists aren't bumping into each other as often as you'd think.

DrIsotope

Reading is fundamental. Cause #1 is collision with a motor vehicle. Causes 2-5 have nothing to do with cars, and everything to do with avoidable factors. Pay attention. Don't crash into things. Don't get drunk and ride a bike.

Further, anyone who tries to use Denmark, the Netherlands, etc. as compare/contrast subjects needs to reevaluate their thinking. Most of the riders over there are riding single-speed citybikes on well developed cycling infrastructures... at about 9mph. The US will never have any sort of infrastructure like that, simply due to size, and this nation's continuing obsession with the automobile. Let's not ignore the fact that Denmark and the Netherlands will fit comfortably together... inside the state of Indiana (the 38th largest state.)

No one ever brings up Germany in these "Europe is a cycling paradise" examples. Because in statistically comparable categories, they run roughly parallel with the US.

KD5NRH

...and don't forget that their parents are driving there too.

jeichelberg87

I hope your sister-in-law is okay.

Following up, if you had to attribute the crash to just one factor, what would it be? You do not think the study would check off one crash into four different categories, do you?

DrIsotope

She's off the bike until 2017, and this happened in July-- been in one of those back brace things for almost 3 months. I would consider it rider error, because like I said before, it's a spot she (and everyone who has ridden that trail) has been over hundreds of times. She even said afterward, "I just wasn't thinking." It became a perfect storm of causes 2-5: a root lift she forgot about washed out her front wheel, so she bounced off of a chain link fence and fell off the bike.

This doesn't wholly forgive the poor maintenance level on the river trail-- those rifts are getting worse every day, and there's about half a dozen of them that I bunny-hop over in a ~3 mile stretch. At least they do bulldoze the muck off of it fairly quickly after the rains.

CliffordK

When looking at a per capita basis, the more bicycles on the road, the more that will die.

So, I would think Florida would be a cycling paradise. Flat roads. Sun. Good Weather. But, it also has a lot of older drivers. However, just basic per capita info is meaningless. Better to look at per cyclecommuter, avid cyclist,or estimated miles ridden.

Oregon also gets a lot of cycle commuting... and is thus ranked poorly due in a large part to the increased exposure of the bikes.

Does anybody in Maine actually own a bicycle?

revchuck 

It would be interesting to see the data further broken down by urban/suburban/rural occurrences. My guess, unsupported by data, is that the majority of the deaths occur in urban areas due to both traffic and population density.

jefnvk

And IMO, cause #1 can very well be the riders fault as well, as much as we like to pretend cyclists never are in the wrong.

As to the actual data, nothing too surprising. As others have mentioned, states with more population, more condensed population areas, and more traffic are deadlier. If I've got ten cars streaming past me a minute in Michigan, I'm just that much more at risk than ten an hour in Montana.

howsteepisit

Also, the more robust the cycling culture of an area the more cyclist miles are ridden, and a greater exposure and risk. The is an offset for more cyclists making more familiar drivers but it seems to me that in many cases familiarity breeds contempt.

Jarrett2

Check this out:

Texas:
Miles of roads: 675,580
Population: 26,448,193

More roads, less people. I guess that contributes to it being much farther down on the list.

Spoonrobot 

Where'd you find this?

I found the same stats at the site below but it's for injuries, not deaths. I'd be very surprised if 28 people (700 x 4%) were killed due to bicycle collisions with a dog in a one year span.

Pedestrian & Bicycle Information Center

Digging a little further and the stats are from this NHTSA survey:

2012 National Survey of Bicyclist and Pedestrian Attitudes and Behavior | National Highway Traffic Safety Administration (NHTSA)

What's interesting is that only half (51%) the respondents rode "at least once a week" in the past year. The rest are listed below:

At least once a month: 25%
At least once in the summer: 18%
Never: 6%

All above from Pg. 7 of the full report.

Length of ride is another telling stat but I won't spoil it for the curious. I highly recommend at least flipping through and looking at the charts and graphs.

wphamilton

Counter-intuitively, not necessarily! One third of the fatalities occur in rural areas, two thirds in urban (FARS data for 2014). Since there are a lot more cyclists in urban areas, it leads me to believe that cycling on rural roads has a greater risk of a fatal accident.

Auto accidents are much more likely to be fatal in rural areas, in part because an injury incurred there is twice as likely to be fatal. I suspect that some of the reasons why (which aren't completely nailed down yet) would also pertain to auto-bicycle crashes. Crashes might even be less likely, but when you have one it's more likely to be fatal.

Relating fatalities to safety makes me uneasy. For one thing, there are only around 700 or so fatalities per year, a number which is too small for reliable analysis, and rare enough that it begs the question of how much we should do to reduce that risk. Compare that to cycling-related ER visits: 480,000 per year. It seems to me that if we want to look at how safe or dangerous cycling is, looking at fatalities alone is certain to yield the wrong answers.

FBinNY 

I agree, the picture these statistics paint may be deceiving. Fatalities are only a tiny piece of the puzzle. Trying to draw generalized conclusions about overall statewide or rural vs. urban safety is like trying to say what the picture is in a puzzle while you're only holding a few pieces.

Seattle Forrest

Garmin and Strava have (and publish) some good data on this, but it doesn't represent the average cyclist.

CliffordK

Yes...
Urban problems like bumping into a rider on a sidewalk while pulling out of a driveway may be common, but happens at low speed.

On the other hand, rear-ending a person at 60 MPH is at a much greater risk.

One also has urban riders heading out to rural areas to ride, as well as rural riders heading into the city. Which is greater? Perhaps those long-distance riders are the ones putting themselves into harm's way. Certainly they spend a lot more time on the road than the urban bicycle shoppers.

Leebo

Umm, you have lots of weather, 72 F and sunny is still weather. Nice weather.

wphamilton

We really need to know about all of the cyclists, or a big enough cross section of them that we could extrapolate.

I'm pretty sure that Google knows, or would know if they cared to. Anyone with location services enabled on their Android phone (just about all of them) has their movements tracked and recorded by default. When I've looked at the history page, Google already has trips tagged as bike rides, with a fair degree of accuracy. For awhile Google was popping up an alert from my system tray for my commute home at about the time I'd leave, complete with how long Google expected it to take and a map of my route. Even though I've never told Google about any of that. I have no doubt that Google can figure out whether my phone is on a bike, in a car, or with me when I'm walking, and has a database record of every time I leave out the front gate. If it's accurate for me, they could do it for anyone who carries a phone.

They're just hanging onto that data experimentally, but it would be great to use it to get a better real world picture of cyclist movements and that information would be outstanding to normalize accident and injury data. I happen to have access to the processing power to analyze it, regardless of how much data it is. I bet I could pinpoint individual accidents. I'd do it for fun, and my bosses would let me. Anyone know an exec at Google?

Leebo

Yikes. Go the the proper Portland, Maine. Plenty of bikes. Have you ever been to FL? You must hear about in every state, some older( +75) driver hits the gas instead of the brake. All the time, at least here in MA. FL is where everyone just goes to die. Really. It's warmer so it takes longer. Everyone( OK east coast) goes there to retire. Would love to know how high the average driving age is in FL?

wphamilton

That's a couple of ideas, and for certain hit from behind is the most common crash in fatal bike accidents. There are some other potential factors that make sense, but we don't really know for sure. Not enough reliable detailed data.

Seattle Forrest

Yeah, and you can't predict how many kids ride bikes where based on Strava's data which you have to assume is mostly people riding for fitness. I was thinking that Garmin and Strava's data were the best available, though. You have a point about Google.

Google Now. It doesn't work as well for me but I think some of the reason is a hardware problem in my phone. (Sometimes it tells me I have 30+ minutes to go to get home after I arrive.)

I haven't seen it auto-tag my cycling. S Health tries to do that and does a bad job, so I'd be skeptical that my phone can do it. But it sounds like it's working for you.

I'd really love to see more of this kind of data find its way into planning decisions when it comes to infrastructure.

CliffordK

So, how many avid December/January cyclists are there in Portland Maine? As many as Florida?

I did make it to Florida once, but not Maine. But at that time I flew into Orlando, and rented a car, perhaps driving in a 50 or 100 mile range from Orlando. It was interesting seeing the stark difference between the coast and inland just a few miles. But, I have heard that Florida does represent a huge diversity of populations. Old people retiring at their beach houses is just a small part of it.

scott967

I went to the NHTSA FARS site to try to get a handle on how the data was collected, but couldn't really find anything. So I'm not too sure how much faith to put in the data.

But I would want to know more about the "type" of cycling engaged in at time of accident. My uninformed impression is that most fatals here are transportation cyclists after dark.

As far as Florida is concerned, I think the "old people who can't drive" is a myth. Yes there are some of these but in recent years seem to be vastly outnumbered by (Florida back-woods)-types in pickups. (And this is in traditionally "oldster" areas specifically gulf coast Tampa/St Pete/Clearwater). Mainly due I think to fact that after the housing crisis you could pick up a 2-bed detached house for $30k for example.

scott s.
.

vol

"5.7 cyclist deaths per million residents" makes no sense. Those among the "million" who didn't ride bikes should not factor in the stats. Different states have different % of people who don't ride bikes. A state where only 0.1% ride bikes, the death rate can't be high even if all the cyclists there died, compared to a state where 99% ride and only a small % died.

Maelochs

The City of Orlando's official motto was "Imagine of all the cyclists here died .... Let's make Orlando Great again!"

I met some Very civic-minded drivers on the roads of Greater Orlando.