Using Data To Further Vision Zero


Earlier this month at the Seattle Bicycle Advisory Board, two employees of the Seattle Department of Transportation (SDOT), Monica DeWald and Chris Svolopoulos, presented the top-level of an intense study into the locations of pedestrian and bicycle collisions in Seattle. SDOT is analyzing collision data on a scale that it has not done prior to now, with the hope that the data obtained can inform key network planning and improvement projects. The team at SDOT looked at collision data over the years from 2007 to 2014 and will be releasing the full data set in the coming months.

The question is the degree to which SDOT will see the forest for the trees in terms of a systemic approach to intersection design with regard to furthering Vision Zero goals. Will spot improvements be the extent of the application of collision data, or will the findings be used more broadly to further all intersection improvements?

The pedestrian collision data showed that close to a third of pedestrian collisions were left hook collisions at controlled intersections, but the top type of serious and fatal collisions was one occurring at a mid-block crossing, likely because drivers are traveling at a higher speed and do not expect someone to be crossing. In other words, most pedestrian collisions are likely not “jaywalking”-related but many serious and fatal ones may be.

Top three factors in pedestrian collisions from this study. (City of Seattle)

Bicycle collisions were more broadly dispersed, but more than a fifth of serious and fatal collisions also involved left hooks.

Top four bike collision causes. (City of Seattle)
Top four bike collision causes. (City of Seattle)

Other findings showed a predictable lopsidedness in terms of the type of road that collisions occur on, compared with their overall prevalence in the city’s streetscape. Pedestrians and people on bikes want to use primary arterials.

Prevalence of street types amongst collisions. (City of Seattle)
Prevalence of street types amongst collisions. (City of Seattle)

During the meeting, Chris Svolopoulos discussed the current limitations of the data being used: things like the current signal configuration and the placement of streetcar tracks on the streets involved are not currently factored in, but can be loaded in at a later time with a more detailed set of geographical data.

Another big question is the degree to which this data will inform the implementation plans for the multimodal master plans that already have their wheels set in motion. Will we have to throw out what’s already been proposed, and is that a good thing? Almost every safe streets advocate in Seattle is greatly frustrated over the amount of master planning that has been occurring in contrast with the amount of changes taking place on the ground. But if a master plan that truly utilizes the data showing where people are least safe can be developed, that may be worth the price of delay.

We hope you loved this article. If so, please consider subscribing or donating. The Urbanist is a 501(c)(4) nonprofit that depends on donations from readers like you.

Ryan Packer lives in the Summit Slope neighborhood of Capitol Hill and has been writing for the blog since 2015. They report on multimodal transportation issues, #VisionZero, preservation, and local politics. They believe in using Seattle's history to help attain the vibrant, diverse city that we all wish to inhabit. In December 2020, Ryan started a three-month stint as editor of Seattle Bike Blog.

Inline Feedbacks
View all comments

Whose graphs are these? I’m hoping the n of fatalities is not high enough, that numbers like 33% v 20% of subdivided data are meaningful. How many pedestrians die in “left hook” collisions over what period of time? How many die in mid-road crossings?

Want vs have to

I’d echo that when I walk or bike I don’t want to use primary arterials, but that’s often the only place I can get across busy streets at a light. That’s where the businesses and grocery stores are located. That’s where the bus stops are. People walking and bicycling are not hit because they want to be hit or want to be traveling on busy roads — they are hit because those streets are not safe.

Al Dimond

No, we shouldn’t delay in order to re-master-plan everything.

1. People don’t inherently want to ride on “primary arterials”, people want to take direct, flat paths to their destinations. Many of these paths are currently occupied by major arterial roads, but not all — hence the popularity of some rail trails and greenways along particular corridors.

2. Today’s crashes occur on yesterday’s road designs. We know many ways to build better intersections today. The dumbest thing we could possibly do is to stop redesigning intersections based on today’s best-practices because we’re waiting for yet another comprehensive plan to come out.

3. Safety isn’t the only problem with our bike network today. Connectivity and utility are huge problems, possibly bigger than safety. Broader studies on the relationship between cycling and collision rates show that collision rates drop as cycling rates rise. This could be because having more people biking makes people drive better, or because safer roads entice more people to ride. Either way, fixing the obvious gaps in the connectivity and utility of our bike network couldn’t help but make key intersections safer, and will get more cyclists out there.

Here’s a poster-child for all three points: the intersection of Dearborn and Rainier. It sits between a major arterial and a greenway leading to an off-street (freeway lid) path, which together constitute one conceptual bike route. It’s a very outdated bike lane design, and a basic redesign along modern principles would improve safety considerably. And to the extent that it scares people off from riding the I-90 route, it’s a major regional connectivity gap. So here’s the question: would you delay the upcoming improvements to Dearborn/Rainier until a safety plan was complete? Of course you wouldn’t! That would be like delaying bike paths over environmental-impact statements…

Al Dimond

23rd is an example — what exactly to do with the street was a hard decision and I don’t think the result was great for biking. And there are a few other examples, like N 45th, NE 65th, and Madison, that are looking similar.

So should we re-plan them? Let’s look at 65th, where the arterial-vs.-greenway argument played out in the theater of Seattle Process. 65th was in draft versions of the standing BMP (even if only as an abstract corridor) and shouted down by local activists, who, to their credit, turned people out. So what happened then? A line in a PDF was moved a few pixels (up to 68th) and re-shaded (to the greenway color) — it hasn’t made any implementation plans yet.

If we opened that up again there’s no reason to believe the politics would be different, and the city is still limited by those politics. But maybe by the time we actually go to do something about 65th the politics will be different. Maybe we’ll be able to get bike lanes on some part of the street (i.e. not between 20th and 25th) as part of a paving project, and make the rest look like a gap to close — by changing relevant conditions. Just talking about bike safety isn’t going to change anyone’s mind. Just as many cyclists pay lip service to VZ because we want a more connected and better route network, the business owners there will pay lip service to any number of concerns, including pedestrian safety, to protect street parking.

What about 23rd, or Madison? Both were redesigned recently and I don’t think anyone believed that directing people to side streets was going to be best for bikes. But on both streets there was a strong argument for the importance of transit reliability and pedestrian safety over bike routes. I doubt that understanding this data changes those priorities or the conclusions they led to.