A Glimpse of Biking Network Effects

Recently I’ve become the steward of an amazing and rich set of non-motorized transportation data through work with Bike Walk Twin Cities (BWTC), a local non-profit administering the federal Nonmotorized Transportation Pilot Program (NTPP), since 2007 (part of the SAFETEA_LU legislation).

This $28M pilot program, awarded to 4 communities around the country and luckily our fair city of Minneapolis was chosen (Sheboygan, WI, Marin County, and Columbia, MO are the others). Among other things, this funding is intended to do two main things: provide funding for bike and pedestrian infrastructure projects, and measure the effect of any infrastructure projects on biking and walking traffic.

Suffice to say, I love my job.

To measure bike and pedestrian traffic patterns, each September the NTPP communities deploy volunteers to dozens of locations, counting observed bicyclists and pedestrians between 4pm and 6pm per the established National Bike and Pedestrian Documentation Project methodology.  Since 2007, Minneapolis has seen great increases in both bike and pedestrian traffic. To be specific, across 40 measurement locations in the Metro Area, bicycle traffic is up 51%, pedestrians are up 24%, with a combined non-motorized transportation change of 37%.

Whatever is happening, it seems to be working.

However, there are some locations that are decreasing in absolute traffic numbers. In fact, one of these declining locations is on Lyndale Ave, just north of Franklin. Isn’t that pretty close to hipster-central? Why would such an important urban arterial see a 24% decrease in bike traffic since 2007? The answer is network effect. That is, traffic patterns are shifting.

A quick look at the context in this location gives it away. Just two blocks to the east, the Bryant Ave Bike Boulevard exists, connecting the Loring Bikeway and downtown, south as far as 50th, through South Minneapolis.

Luckily, we have a measure of that as well. One of the BWTC count locations is on the Loring Bikeway Bridge, just above the Lyndale location. Not surprisingly, we have seen a similar increase in observed cyclists; 23% since 2007.

Now, those numbers are a little misleading.  That -24% change on Lyndale Ave constitutes 34 observed riders from 2007 to 2012, whereas the +23% change on the Loring Bikeway Bridge constitutes 114 bicyclists.  That is, the bike facility is changing traffic patterns, and it is attracting more and more riders!

Here is a neat graphic to demonstrate the point:

There you have it, one line goes up, while another line goes down: network effects.

There you have it, one line goes up, while another line goes down: network effects.

If you look closely, you can see the light and dark green boxes which represent the locations of screenlines where each count is located.

A similar effect was observed before and after the Cedar Lake Trail was connected to the West River Parkway.  Before 2009, the trail and connecting facilities extended from St. Louis Park and beyond to the west, up to about 12th St on the edge of downtown Minneapolis.  The trail saw good ridership before the extension project, but once that missing link was mended, there were massive observed increases in ridership.

Now, this isn’t to say that we want cyclists to essentially get out of the way of motorists.  Indeed, there should be many more and much better bike and pedestrian facilities on many more of our streets.  Additionally, we should be looking to increase the non-motorized network beyond the Twin Cities into the suburbs and beyond.

This, however, is just one interesting location which hints at much larger and many more implications for non-motorized transportation.  The whole 2012 Count Report from BWTC can be found at their website, and keep your eyes peeled for more interesting analysis, including monthly count information, and an address of the impacts of weather.

Keep on riding-


Streets.mn is a non-profit and is volunteer run. We rely on your support to keep the servers running. If you value what you read, please consider becoming a member.

6 Responses to A Glimpse of Biking Network Effects

  1. Bill Lindeke
    Bill Lindeke March 8, 2013 at 11:00 am #

    This reminds me of work that shows how running red lights or stop signs decreases when you actually build infrastructures that make sense for bicyclists (e.g. separate signal phases, leading intervals, etc.). Sidewalk riding drops when you build safe and comfortable bike spaces on the street, etc.

    If cities really want bicyclists to “obey the law”, the way to accomplish that is to make spaces that make sense for how people actually ride, not just browbeat them.

    • Prescott Morrill
      Prescott Morrill March 11, 2013 at 3:41 pm #

      Totally. This neat statistic was cited in the results of a 2006 study looking at a bike scramble tested in Portland (of course):

      “From PDOT’s cyclist counts before the intersection was installed, 71.8% of all cyclists passed through the intersection illegally. This number has fallen to 4.2% for November 2006.”

      -http://web.cecs.pdx.edu/~monserec/courses/urbantrans/projects/ce454_bikescram_rpt.pdf

  2. Matt Brillhart March 8, 2013 at 11:50 am #

    I actually performed the 2011 counts on Lyndale at Franklin. As you can see, that stretch of road does still attract plenty of bicyclists. Anyone that lives on the east side of Lyndale Avenue is not going to be compelled to use the Bryant Ave bikeway and bridge. That stretch of Lyndale, between Franklin and the cycle path to all points north, is in desperate need of some kind of improvement. I observed cyclists riding on the sidewalk, riding on the wrong side of the street (southbound next to Rudolph’s, against vehicular traffic flow), etc. People are coming and going to the cycle path that terminates at eastern end of the bike bridge and a proper connection to Franklin Avenue, on Lyndale, is crucial. It would likely result in the loss of a few parking spaces on Lyndale, but it would make things safer for bikes, peds, and possibly afford Rudolph’s a little more room for their cramped patio seating in warmer months. I can’t imagine that their business lives or dies on ~4 on-street parking spaces, especially considering they own a large surface lot on the other side of Lyndale.

    • Prescott Morrill
      Prescott Morrill March 11, 2013 at 3:42 pm #

      Yes, there is always room for improvement, especially when a bike facility terminates at a sidewalk. As noted, this is just a quick vignette.

    • David March 12, 2013 at 5:15 pm #

      Bingo. Lyndale at Franklin is a biking disaster. What other place has a bike path that naturally suggests that the bicyclist ride in an auto lane opposing traffic?

      I’m not the least bit surprised about the traffic pattern shift but I think it is a sign of a problem as well as a sign of better infrastructure via bike boulevard and bridge.

      Janne is spot on. Often I have wanted to bike to somewhere on Lyndale after spending time downtown. Then I get to that path and just give up. It’s easier to just ride home to the Wedge.

  3. Janne March 9, 2013 at 10:34 pm #

    Matt, thanks for taking over my spot when I couldn’t count there any longer!

    You’re comments are exactly what I saw.

    The other specific concern I have about the network effect of bikes moving from Lyndale to Bryant is that I want to ride to destinations ON Lyndale, and the more press is given to bikes belonging (and being) on Bryant rather than on Lyndale, the less likely it is I’ll get reasonable accommodations that get me closer than two blocks from my destination. For some reason, people forget that people on bikes aren’t usually out just to get fresh air, they’re going somewhere. And that random house on Bryant just isn’t my destination.