Friday, 23 August 2013

Safety on the Road

When I started blogging about urban cycling I decided that my focus would be on safety. Year after year I see more cyclists on the roads, and I have spoken with many who have wanted to ride. Not only that, but the range of reasons for cycling has increased as well. People are becoming more self-conscious about exercise, and cycle commuting is a “free lunch” so to speak, getting exercise and getting to work at the same time. There is also the economic benefit, no paying for parking or gas, or transit pass. There are initial sunk costs, but they are quickly made up. Then there are environmental benefits, less pollution and less reliance on a non-renewable resources.

So there are a lot of reasons to ride.

I believe that for the majority of NEW riders, the primary concern with cycling is safety. I will say more about some dissenting opinions later, but for now let’s take that as a given. I believe that cycling is poised to become a much larger part of everyone’s transit experience as economies change and technological networks change to accommodate environmental and energy realities, and as such, that a discussion of safety is crucially important. If people thought they could cycle safely they would be much more likely to do it.

With that in mind, how do you assess the safety of road riding?

Well, I have about 1200 hours of urban cycling under my belt, over 5 years. Toronto, as it happens, has one of the worst traffic commuting times in the world, so that suggests that my experience would be representative of other major urban centers.

I have also done some research, and my go to source for cycling statistics is John Forester, a US cycling advocate. I disagree with Forester about several things, but I respect his experience and his use of statistics and scientific evidence.

So what do the stats say about cycling safety?

First off, experience matters.


Forester compares “club” cyclists (cyclists that have joined a riding club where they can learn safe riding methods and compare experiences with others) to non-club cyclists, and shows that non-club cyclists get in more accidents. Or as he puts it, the number of miles you ride per accident (or the number of miles on average that you ride before you get in one) goes up with time on the road and time spent in cycling clubs, so experience matters to safety.

However, the novice driver, by definition, has little experience, so you need some other solution if you want safe riding for new drivers.

You need rules and suggested methods for riding.

What rules and methods work?

I will present two answers to this question.

The first answer is general, the other specific.

The specific answer is to consider various traffic situations and discuss what to do. I will do that in later posts.

Here instead I will start with some general principles, based on my understanding of the statistics.

Breaking it Down: The Numbers
According to Forester, 38% of bike-car collisions occur when the cyclist is following the rules of the road. I will ignore the fact that this leaves 62% of our cyclists not following the rules of the road when the accident happens. I know that breaking the rules can be bad, so I’m more interested in what happens when you follow them.

What Forester’s statistic tells me is that accidents can occur even if you are doing what you are supposed to do, so doing what you are supposed to do isn’t enough if you want a safe ride. Let that sink in for a moment, as it has some profound consequences. If 38% of car-bike collisions happen when the bike rider is obeying the law, then if you always obey the law on the road while cycling you have a non-trivial chance of getting into an accident.

Thus breaking the law must be necessary in certain contexts. More on that later.

Now just to give you a comparison, of bike-car collisions, 7% occur when riding on the sidewalk. So you are 31% more likely, almost a third more likely, to be hit when you are riding where you are supposed to be on the road than when you are on the sidewalk, where it is often illegal to ride.

That seem odd to you?

It should, because it flies in the face of what you are told about cycling.

89% of-car bike collisions occur at intersections, when passing through or turning. 7% happen when you are being overtaken by a car coming from behind. Again, the stats prove your instincts wrong, most of us feel more threatened by the overtaking traffic (as we can’t see it coming unless we turn or look in a mirror) but the intersection, which we can see ahead of time, is far deadlier.

So what do you figure is the difference, why are bike car collisions so much more likely to happen at intersections? Why do accidents happen when you are doing what you are supposed to be doing?
Leaving out the obvious candidates like drunk drivers, panicking drivers who make the wrong decision, etc., it strikes me that there are two obvious problems here: visibility and speed.

When cars are coming up behind you they can see you. Overtaking accidents occur when a cyclist is driving at night without proper reflectors or lights, when they rejoin traffic unexpectedly (say coming off a sidewalk), or when the driver makes a right turn across their bow as he thinks he can make the turn before the cyclist gets there but he cuts them off.

Visibility helps with the first two problems. Riding at night with reflectors and lights, and not moving in and out of traffic unexpectedly. If you bomb down the sidewalk and then through pedestrian cross walk cars come to the intersection to make a right turn, look for pedestrians, and start to make the turn. If you are coming up fast on the sidewalk you can easily arrive when the turn is being made but not be seen. Bikes are fast and silent. Making yourself visible means that most cars just go around you. I ride up Keele Street to work every day, Keele is quite busy and the cars are fast. However, unless the traffic is bumper to bumper the cars tend to leave the right lane for me and pass in the left. If the traffic is bumper to bumper I just ride up the side of the road. In both cases as I am visible they generally go around me, or I go around them.

Motorists don’t, for the most part, want the hassle of hitting you and dealing with the consequences, they just want to get on with their day and get past you. So visibility is key.

At intersections the issue becomes reacting to the changing traffic situation. This is where speed is added to visibility as an issue. When you are riding on the road you need to be seen as you can’t compete with the cars when they are going fast, and when they are going slow and you are passing them you want to be seen so no one opens a door in your path or cuts you off with a right turn. Being visible allows others to go around you.

At intersections decision making comes into it. Say you are at a busy four way intersection, cars are lined up at all four spots. You roll up in a bike lane and arrive at the front of the line of cars. When do you go? In the sequence you arrived? Perhaps, but you can’t always be seen by the other cars, particularly when the lead cars beside you are vans or large trucks. Personally, I wait until the lead car beside me drives straight through and I ride out with it, effectively protecting my exposed flank and allowing me to avoid having to negotiate my position with other cars.

At traffic lights I have encountered the opposite problem, I’m visible, but many motorists try to “jump the turn”. I’m going straight through an intersection and the car across from me in the oncoming lane is signaling a left turn across my path. When the light changes I have the right of way, but the car is faster and can make a quick turn across the intersection before I’m even half way out. So they will often make the turn despite me coming through.

I have also seen many cyclists jump the light, they wait on the red, and when the last car has gone through the intersection they start to enter even though the light hasn’t changed. It gives them at best a few seconds, but it happens a lot.

Well, if 89% of car bike collisions happen in intersections, and in 62% of car bike collisions the bike was not following the rules of the road, you can see why I would be skeptical of the safety of jumping the light. Cars move fast, and stragglers can arrive at the intersection late. All this suggests that cyclists need to obey the law at intersections, and that they should focus on being visible and driving at a speed that allows them to react to the changing traffic situation.

23% of bike-car collisions happen when a cyclist is entering a roadway from a sidewalk, driveway or other road. This happens, I suspect, as the cyclist is not noticed, and thus the driver either doesn’t give the space or reacts badly to the attempt to enter traffic. Once I came to understand the dangers of re-entering moving traffic I started to wait until the road was empty before entering, at least when I was not completing the turn in sequence with traffic at a light. “Empty” is a relative term, what this means is that I’m not rejoining the flow of existing traffic, I’m waiting for a break in that flow and entering the road then. This gives me time to adjust the bike, but more importantly it allows me to be visible to traffic approaching from behind me.

There is more to say, but that’s enough for today. Visibility and speed are important to safety. Riding fast gives you less time to react to traffic situations, and makes it more likely you will appear when a motorist isn’t expecting you. Riding without considering visibility means that you will frequently find yourself unnoticed when a car is moving into your space.



Here are a few more pics:

 The Humber Trail near the Weston Road exit

The East/West portion of the Finch Hydro Corridor Recreation Trail at York University

No comments:

Post a Comment