tallard

From what I've read from the links in this forum, my impression of the stats comparing rural cycling versus city cycling is that they are in a sense misleading. If my memory serves me right, it's something like 40% vs 60%, more or less...

I wonder if a more revealing statistic would be number of motorists encountered. If we were able to account for that factor, I believe we would find that rural cycling is multitudes more dangerous than urban cycling. As in a city, an hour ride may have thousands of "car encounters" whereas in a rural area, on hour cycle my only reach a dozen "car encounters". I would bet that reassessing all collisions in relation to number of "car encounters" would give us an entirely different outlook on rural cycling.

The statistical approach of any question is always debatable, I think it's important to explore the different options for assessing risk.

Have any of you encountered this type of assessment?

-=(8)=-

I might agree with that.
I have found that although one might only encounter a few
motorists on rural roads, the ones I have experience on are
usually very narrow and a few of those encounters might be
a caromming bigfoot truck going way too fast for the width and
blind corners on that road.

Bekologist

I suspect the overwhelming majority of bicycle/automobile collisions occur in urban areas.

Why would the OP think rural cycling more dangerous, if cyclists encounter exponentially more traffic in urban conditions?

-=(8)=-

Here is something mildly relevant that came up after the most basic
Google-age : https://www.brake.org.uk/index.php?p=631

The parts that are even remotely related :
Cyclists
Rural roads have more bends than urban roads and have fewer cycle lanes/paths to keep people on bikes away from the rest of the traffic.

Similarly to incidents involving people on foot, people on bikes involved in collisions on rural roads are more likely to be killed or seriously injured than those on urban roads.

In 2004, 4,542 children were hurt while on a bike on a built-up road. Of these, 543 (12%) were killed or seriously inured. 20 were killed. While the number of children hurt on bikes on non-built up roads – 137 – was much fewer, 34 of them (25%) were killed or seriously injured. Five of these were killed. [20]

In 2004, there were 10,347 adult cyclists hurt on built-up roads, of which 1,390 (13%) were killed or seriously injured. 66 of them died. On non built-up roads there were 1,011 adult cyclists hurt, of which 306 (30%) were killed or seriously injured. 43 of these people were killed. [21]

Case study: four cyclists killed in crash
Four cyclists, including a 14-year-old boy, were killed after a driver lost control of his car on the A547 in north Wales. Thomas Harland, 14, Maurice Broadbent, 61, Dave Horrocks, 55, and Wayne Wilkes, 42, died when a car spun sideways into their path near Abergele. The riders were in a group of 12 from Rhyl Cycling Club. Chief Inspector Lyn Adams from North Wales Police said the driver appeared to have lost control in icy conditions on a gentle left-hand bend, striking a wall and rebounding into the road.

Allister

Granted

To be fair, the number of cars isn't the only criteria. But as for the relative danger of rural/urban roads - it's not something that I think is a worthwhile line of contemplation. Once you start thinking a particular road is safe, or at least safer than another one, there's the danger of falling into complacency, and realistically, I think all roads deserve the same amount of vigilance.

tallard

No disagreement on behalf of the cyclists behavior, but I'd like to consider the educational aspect for motorists and policy makers. If on a per car basis, the stats demonstrate that rural motorists have much more reckless behavior than urban motorists, maybe we need to change the focus of education towards motorists in these areas...? As I'm presently in Whitehorse, Yukon, I notice motorists weave all over the road, over lane lines, just because they can basically. Also, multi-use paths here are falsely considered the safe option as at most of the shared intersections, the path is actually invisible from the stop sign location, either because of trees or dirt piles or 3D configuration in general... City planners have cut out huge swatches of nature in order to "give safety to other users" and yet, the place is more dangerous than before...

Robert C

In the US I live in a rural area. I can barely compare the feeling of safety between the "urban" and "rural" environment.

I really feel safer riding in traffic in China than I do riding on California's Hwys' 263 and 96. If I get bumped in traffic in China, that is it, I get bumped. Being hit on a rural highway in the US is another matter altogether.

The rural highways are high speed roads with no shoulder and a long time to any medical facilities. People can talk about "taking the lane" however, I see that practice as stupid and unsafe when you take into account continual sharp bends in the road that limit visibility.

Rural towns are a different case altogether. I would say that Yreka is one of the best towns I have loved in for bicycling. I liked Chico; however, I really did not like Sacramento for riding at all. While there are traffic rules, I would consider Sacramento worse than the near similar, population sized, city that I live in here in China (of course, the US has bike racks and less anti bicycle laws).

dobber

More often then not I encounter the slow moving farm tractor with the hay wagon slowing oscillating back and forth, the one lame warning light blinking dimly.

Bekologist

I agree, Allister, both rural and urban riding requires vigilance and awareness, I simply dispute the OP's assertion that 'rural' riding is more hazardous that urban riding.

Additionally, tallard, any area with 'city planners', bike paths, and stop signs would be considered an urban environment......

Whitehorse, your provincial capitol, with a population of 24,000, is definetly 'urban'. I've lived in cities of 24,000 and they are distictly 'urban'....you just don't have large city suburbia. but I'll bet you a loonie Whitehorse DOES have it's version of 'suburbs'....

new MUP's aside, do you feel at greater risk of an accident in the CITY of Whitehorse, or out riding to Haines Junction or east towards the 6?

sggoodri

Speaking objectively, average motor vehicle speeds on urban streets tend to be much lower than speeds on rural roads. This provides more time to be seen, and more time for reacting/braking, as well as less severe collisions when they occur. Those urban streets where traffic does move much faster than bicyclists are more likely to have additional pavement width for drivers to be able to pass safely with less delay and without moving into the oncoming lane compared to rural roads.

But subjectively, it seems to me that urban drivers on average are more competent at dealing with traffic, and it seems there are more rural drivers who are reckless lead-foots and hesitant to brake. The frequent overdriving of one's sight distance that I witness in rural areas would result in a car-car crash pretty quickly in the city.

A bill for a graduated licensing program for teenage drivers that was introduced in our legislature a couple of years ago was staunchly opposed by a legislator from a rural area, who suggested that it be limited to urban drivers. The bill supporters subsequently did some research (with NCDOT's help I believe) and determined that by far the highest rates of serious car collisions among teen drivers happened in rural areas. The bill ultimately passed.

Bekologist

do more accidents and fender benders occur in cities or in the countryside, steve?... not 'serious car collisions'.

I'm sticking with my assertion that many more riders get tagged in urban areas (from driver INattention, etc) than on rural roads. And I doubt there is much hard data collected on rural/urban comparisions of bicycle accidents.

sggoodri

Having spent a few years on a planning and zoning board where municipal expansion into unincorporated areas was a constant source of political controversy, it's my experience that arguments over what is rural versus urban or suburban can spiral out of control as people attach different personal meanings to each.

For instance, there is the land use perspective, where a primarily agricultural land use makes an area rural, while a primarily residential bedroom community would be considered suburban or urban. But people who live in formerly agricultural areas on one-acre or smaller lots and commute into incorporated areas for work tend to cling to the "rural" designation when the city planners show up to try to fix their failing wells, septic, systems, and roads in conjunction with annexation requests from their neighbors.

From a transportation perspective, I believe a more useful set of criteria for classifying a road as rural or urban would consider traffic volume, 85th percentile speed, intersection density, driveway density, pavement width, curb/gutter, and sidewalk.

Low-traffic 2-lane roads with few junctions can be great fun to ride even with narrow pavement and high posted speed limits. These I would call rural roads.

Higher-traffic roads with lots of driveways and intersections, ample pavement width, curb and gutter with sidewalk, and lower speeds tend to have lots of pedestrians and cyclists around here. I call these urban or suburban streets.

What I dislike most are very narrow two-lane roads with high posted speed limits and long distances between junctions, but heavy traffic. I call these rural-design roads with urban traffic, and they are most common on the urban fringe, where commuters living in their "rural" bedroom communities drive into the urban areas or suburban office parks to work.

sggoodri

Setting aside freeways, which are hard to classify and irrelevant to the discussion, more non-fatal car-car collisions occur in urban surface streets compared to roads in unincorporated areas. These are typically intersection-related, where traveling more slowly usually reduces the probability of collision.

But the car-car collisions in the unincorporated areas are much more catastrophic, with a much higher fatality rate. The number of fatal collisions in the unincorporated areas is highly disproportionate to the population there. Also, a great many of these collisions happen in non-intersection locations. When looking at pedestrian and cyclist fatalities, these tend to happen on the higher speed roads, most of which are in unincorporated areas, so even though there are fewer pedestrians in those rural areas, their fatality rate is higher, likely due to a combination of high speeds, darkness, and lack of sidewalks. There are too few cyclist fatalities in my area of NC to make many useful generalizations about contributing factors (some are intersection related, some are in darkness, and one or two have been daylight overtaking on high speed rural roads).

So, the tradeoff of urban cycling is likely more potential conflicts with cars at a lower probability of collision and severity of collision with each. I read a study once that showed that high-intersection-density urban cyclists were somewhat more likely to get into a collision with a car, normalized according to total cycling distance, but were less likely to receive serious injuries, normalized according to total cycling distance. As a utility cyclist, my destinations are urban, so that is where I cycle, and I appreciate the lower traffic speeds. But when I ride as a recreational cyclist, I like fewer interactions with cars, and so I choose out-of-the way rural routes and shrug off the occasional redneck I see careening around blind corners and over hills in his pickup.

tallard

Your assessment of rural/urban is completely relevant to Whitehorse, The City of Whitehorse is merely an agglomerate of over a dozen smaller communities or neighborhoods, each with their own grocery stores and schools. In other parts of the world, they'd be called separate little villages. The downtown of this huge 24,000 city is only 2000 people, and is as geographically distinct and separate as the other neighborhoods. In addition to these neighborhoods, there are many residences along the highway, the only only way, so it's in effect a country road.

Yukon roads are 6 times as dangerous (road deaths/injuries per 10,000 people) as any other place in Canada. People here drive like absolute morons as a general rule, EVEN THOUGH the speed limits are all very slow and most roads are wide with plenty of shoulder. The shoulders in Yukon are out of this world. The original Alaska highway was narrow and windey, but that has now been completely rebuilt and although mostly it's still an undivided 2 lane road, the total shoulder (striped) width is often nearly that of the roadway and the trees are cut down five times as wide as the pavement. Deaths here don't happen because of lack visibility and shoulder widths or high speed limits, these policies were enacted to decrease fatality rates but still the rate will not drop. Bottom line, drivers here are just completely moronic. Also there's a law here that encourages pedestrians to walk out on the street without looking because of certain "let pedestrians cross" bylaws, but people get hit a lot because they've stopped watching out for themselves.
In all it's a crying shame

Mr. Underbridge

To me, the only relevant statistic is curses/mile. By which I mean, the number of times that someone scares/angers me enough to get me to curse, per mile traveled. In fairness, the statistic is well under 1.0 curses/mile. I'd say I average about, maybe 0.08 curses/mile.

As for rural vs. urban, there are tons of confounding issues, so many that it makes trying to isolate single variables almost pointless.

SonataInFSharp

After riding in the suburbs for the better part of 20 years and now having ridden in the big city for 2 years, I must say from personal experiences that urban riding is much, much safer.

I don't care what statistics say as they can be interpreted for the benefit of whoever wants to use them.

San Rensho 

I think its the speed difference between cars and bikes that makes riding safe or unsafe.

Purely anecdotal, but I think city driving is probably safer as long as the cars are going at the near the same speed as bicycles. I feel very safe, even in fairly heavy traffic, when I can take the lane and flow with car traffic. If there is car traffic ahead and in back of me in my lane, its very unlikely a left turner will cut me off. Also, a right turner can't get to the left of me to hook me if I am taking the lane. And cars don't honk as much when they see I am going as fast as they are.

Rural riding, even when traffic is very sparse, often has a car approaching you from behind at 70mph, who doesn't expect to see a cyclist. If there is a breakdown lane to ride on then its usually pretty safe, but if you have to be in the lane, no matter how wide it is, it just feels dangerous to me.

Helmet Head

To the extent that a cyclist who doesn't employ good situational awareness and best practices is safe, it's probably mostly from relatively light traffic in rural areas and relatively good motorist driving skills and relatively low speeds in urban areas.

But a cyclist who employs good situational awareness and best practices is safe in either environment.

Either way, what matters the most with respect to how safe any cyclist is in any environment is, by far, the cyclist's behavior.

Roughstuff

I can understand adjusting accident/danger data for # of cars encountered. But this may still be misleading. In a rural area, where one encounters one car at a time, you do not have the risky interplay of numerous vehicles vs bicycles fighting for space, as you would in an urban area.

Whether driving my car or riding a bike, I have always found motorists in rural areas to be more courteous and professional than those in cities. I don't even bother to drive into cities like Boston anymore, the level of anger and road rage borders on the pathetic.

roughstuff

tallard

OP is not to question cycling practices but to challenge rural policy makers on motorists issues. If, as most posters seem to agree, rural collisions PER MOTORIST are way more common, then should we not be doing something about rural motorist's behaviors?

noisebeam

Here is a breakdown of cyclist fatality rate by county for Arizona:
https://www-nrd.nhtsa.dot.gov/departm...ata_2005_9.PDF

Pima (Tuscon) and Maricopa (metro-Phx) are suburban/urban, the rest basically rural.

Coconino includes Flagstaff, Yavapai includes Sedona.

edit: here is NY data: https://www-nrd.nhtsa.dot.gov/departm...ata_2005_9.PDF

Find other states here:
https://www-nrd.nhtsa.dot.gov/departm...B%20REPORT.HTM

Al

Giro

Thanks for the NHTSA link. As in many risk evaluations (health care, bicycling, etc.) there is often the problem of knowing the numerator (number of incidents) but not knowing the denominator (relevant numbers at risk).

The NHTSA link shows they use incidents per 100,000 population for pedalcyclists. It might be interesting to also look at fatalities per number of motor vehicles registered per county, the motor vehicle density per county (# cars per square mile) and the population density in the counties. My guess is the number of cyclists, hours cycling, and miles cycled per capita is higher in higher density areas so neither of these may be perfect denominators.

However, this might start getting at some factual rural vs urban fatality risks. Since the number of fatalities per individual county per year is low, the data would be needed over several years in all counties in a state or countrywide. Both the motor vehicle density and population density could be used as surrogate measures of how rural or urban a county is. You could then graph fatalities per these densities (the risk, on Y-axis) versus the vehicle or population density (the rural-ness to urban-ness) on the X-axis. Then we could see if the risk goes up, down, or shows a more complex behavior (e.g. U-shaped).

Thats the 1% inspiration. Now someone or agency needs to do the 99% perspiration of gathering and graphing the data. US Census data may be available online for the population density per county. Anyone have the spare time or able to make this into a school project?

tallard

Well those are indeed interesting numbers, however as generally happens, statistics regarding cycling fatalities overall give us little to truly compare between cycling groupings as the numbers may be overstated because of "per 100,000 population".

Lets take the Arizona numbers for example. Those numbers at first glance give quite a jolt. However, the numbers do not take into account the actual cycling population. If the actual cycling population "per 100,000 population" is 10 times higher in Maricopa, then the fatality rate, ALL OTHER THINGS EQUAL, will be 10 times as high in Maricopa. That then brings the numbers more like 1s and 2s, dealing with such small numbers, we must be careful to take into consideration "statistical significance". Hypothetically, the fatalities reassessed "per 100 cyclists" would give an entirely different assessment of risk. It is in this context where my "per car" collision rate gets driven up.

It is nearly impossible to assess cycling risk if cycling ridership/miles is not properly assessed. The error created is mostly unidirectional.

noisebeam

Exactly, which is why I pointed to the data, but gave no analysis of my own.
Al

Giro

We could try to create our own small data set to get a rough estimate of number of motor vehicles and number of bicycles on the roads. This bike fourm's participants could count the #cars and the #bikes encountered during our rides. Unfortunately, in dense traffic of a major city counting the cars would be unfeasible while paying attention to avoiding a collision. An estimate might be enough.

I have never started a poll on BikeForums. Can they record numeric data or do they only allow selecting from a multiple choice? Some relevant database fields might be what US county were you bikeing, time of day, weather, urban/suburban/rural area, bike lane vs path vs open road, and most importantly # cyclist ecountered and # motor vehicles encountered (or estimates of the numbers). For bike paths or trails the motor vehicle count could be the # encountered at road crossings.

I will leave it up to someone else to set up the poll(s) as a measure of interest in doing this. I would be quite willing to help in the data analysis.