NY Times article - Cycling will kill you!

On Friday, November 15, 2013 1:57:26 PM UTC-5, Jay Beattie wrote:
On Friday, November 15, 2013 9:05:45 AM UTC-8, Frank Krygowski wrote:

On Thursday, November 14, 2013 6:31:13 PM UTC-5, Jay Beattie wrote:

I would have just said that the markings did not comply with existing federal standards -- which were then changed. Illegal makes it sound so sinister.

The existing federal standards were not changed.

Experimentation was allowed. How ever you want to cut it, PDX was given the green light for the green boxes.

PDX was given the green light to perform an experiment. They were given the OK after the fact, due to political string-pulling. The engineers in charge are lucky to retain their licenses.

The green boxes are entirely self-explanatory. Federal regulators may be in a tizzy, but from an actual use standpoint, this is a tempest in teapot.

Understand, Jay, the installation was an experiment. The objective was to prevent (or at least, greatly reduce) right hook problems for cyclists.

Did it work? No, it failed. You can call it a "tempest in a teapot," but eight years worth of data (four before, four after the installation) shows that right hook conflicts increased greatly. No, it doesn't mean you're likely to get killed riding through one of these things. But it _does_ mean the design was a failure!

And this whole "tempest" thing illustrates just how illogical the "facilities uber alles" crowd really is. As with some other fashionable, "innovative" facilities, it doesn't matter if they actually make things measurably more dangerous. They still have dogged fans!

So they did the formal "experiment" with disciplined data collection, and found that the bike boxes made things worse.

That's not true. What they found was there were more right hooks involving moving cyclists.

??? The bike boxes were intended to prevent, or greatly reduce, right hooks, period! The right hook count roughly tripled! That isn't making things worse? That isn't failure to meet design objectives? Sheesh!

Also note that in the same time frame, it became legal to pass on the right. We also had an increase in cyclists. This could all be statistical noise. There is no proven causal connection or even intuitive causal connection between the green boxes and right hooks, at least not from the cyclists standpoint. I don't know what affect they have on driver behavior.

Seems to me the causative connection is pretty clear. I've explained it already.

... The box is (as you know) to get cyclists in front of STOPPED traffic to avoid hooking at the start of the green light cycle.

Yes, I know. I also know that cyclist behavior has been shown to poorly match the theory; most cyclists stay at far right. I don't know how much of that is because it's not necessarily easy to tell when the green light is imminent. I imagine few cyclists want to be 45 degrees to the flow of traffic and just coming to a stop when the light turns green. Again, in Europe the bike boxes tend to have separate signals for bikes, usually an advanced green phase. That solves that incipient green problem - at a cost of worsening traffic congestion by reducing intersection throughput, of course.

....

And there may be other problems. Left turning cyclists are being told to ride all the way up on the right, then move all the way to the center line (or into a left turn only lane) within a few feet of the stop line. ISTM this must be difficult if there are several cyclists already there, plus a light that's about to go from red to green!

Who is telling them this? Not the UVC or common sense. As you know, bicyclists may leave the bike lane to execute left turns.

Please report, what percentage of left turning cyclists do that at bike box intersections?

Green boxes may be a waste of paint and public funds, but they're harmless except to the extent they are slippery when wet.

They're a failed experiment. They did not achieve their objective, by the standards of measurement agreed upon. In fact, they did the opposite, and produced a greater count of the problem they were intended to cure.

- Frank Krygowski

Frank Krygowski writes:

Yes, I know. I also know that cyclist behavior has been shown to
poorly match the theory; most cyclists stay at far right. I don't
know how much of that is because it's not necessarily easy to tell
when the green light is imminent. I imagine few cyclists want to be
45 degrees to the flow of traffic and just coming to a stop when the
light turns green. Again, in Europe the bike boxes tend to have
separate signals for bikes, usually an advanced green phase. That
solves that incipient green problem - at a cost of worsening traffic
congestion by reducing intersection throughput, of course.

Does an advanced green for cyclists-only really solve the problem?
There can still be some cyclists not yet in the box and passing to the
right when the regular green goes active. Now they are in the
hook-zone.

It might work if, as is done in California, the right-turning cars are
required to merge into the bike lane (not that they always do, but they
are supposed to). Of course, once you do that, the rationale for a
green box mostly goes away.


--
Joe Riel

On Friday, November 15, 2013 3:40:03 PM UTC-8, JoeRiel wrote:
Frank Krygowski writes:



Yes, I know. I also know that cyclist behavior has been shown to

poorly match the theory; most cyclists stay at far right. I don't

know how much of that is because it's not necessarily easy to tell

when the green light is imminent. I imagine few cyclists want to be

45 degrees to the flow of traffic and just coming to a stop when the

light turns green. Again, in Europe the bike boxes tend to have

separate signals for bikes, usually an advanced green phase. That

solves that incipient green problem - at a cost of worsening traffic

congestion by reducing intersection throughput, of course.



Does an advanced green for cyclists-only really solve the problem?

There can still be some cyclists not yet in the box and passing to the

right when the regular green goes active. Now they are in the

hook-zone.



It might work if, as is done in California, the right-turning cars are

required to merge into the bike lane (not that they always do, but they

are supposed to). Of course, once you do that, the rationale for a

green box mostly goes away.


Exactly! Cyclists were getting right-hooked while riding through intersections during stale green lights. And to answer Frank's question, yes, left turning cyclists in bicycle boxes just get to the left of the box, wait for the light to change and go when traffic clears. The box has no effect on usual bicycle behavior. Really, except for being slippery when wet, the bike boxes are harmless reminders for cyclists to be out in front of traffic and to avoid right-turners. They were a publicity stunt after Tracey Sparling got whacked. Californication of our bike lane laws would solve most of the problems, as would education of drivers and cyclists.

-- Jay Beattie.

On Friday, November 15, 2013 6:40:03 PM UTC-5, JoeRiel wrote:
Frank Krygowski writes:

Yes, I know. I also know that cyclist behavior has been shown to
poorly match the theory; most cyclists stay at far right. I don't
know how much of that is because it's not necessarily easy to tell
when the green light is imminent. I imagine few cyclists want to be
45 degrees to the flow of traffic and just coming to a stop when the
light turns green. Again, in Europe the bike boxes tend to have
separate signals for bikes, usually an advanced green phase. That
solves that incipient green problem - at a cost of worsening traffic
congestion by reducing intersection throughput, of course.

Does an advanced green for cyclists-only really solve the problem?
There can still be some cyclists not yet in the box and passing to the
right when the regular green goes active. Now they are in the
hook-zone.

Good point. I'm trying to remember if European ones also include some warning that the motorists are about to get a green.

But to be clear, I don't remember ever encountering a European bike box. I did see some bike lanes with advanced green phases for cyclists, in Stockholm, I think, but no bike boxes that I recall. My remark about the separate signal phases was based on what I remember from reading articles, etc.

Speaking of reading, here are John Allen's comments on bike boxes:
http://john-s-allen.com/blog/?p=60

- Frank Krygowski

On Friday, November 15, 2013 8:50:05 PM UTC-5, Jay Beattie wrote:

Exactly! Cyclists were getting right-hooked while riding through intersections during stale green lights. And to answer Frank's question, yes, left turning cyclists in bicycle boxes just get to the left of the box, wait for the light to change and go when traffic clears. The box has no effect on usual bicycle behavior. Really, except for being slippery when wet, the bike boxes are harmless reminders for cyclists to be out in front of traffic and to avoid right-turners.

They might be harmless if they didn't increase the incidence of the problem they were trying to prevent!

They were a publicity stunt after Tracey Sparling got whacked.

I agree. And that's a terrible justification for a weird bike facility design.

Californication of our bike lane laws would solve most of the problems, as would education of drivers and cyclists.

I agree again. But green paint gets SO many more pats on the politician's back than calls for education!

- Frank Krygowski

Frank Krygowski writes:

On Friday, November 15, 2013 6:40:03 PM UTC-5, JoeRiel wrote:
Frank Krygowski writes:

Yes, I know. I also know that cyclist behavior has been shown to
poorly match the theory; most cyclists stay at far right. I don't
know how much of that is because it's not necessarily easy to tell
when the green light is imminent. I imagine few cyclists want to be
45 degrees to the flow of traffic and just coming to a stop when the
light turns green. Again, in Europe the bike boxes tend to have
separate signals for bikes, usually an advanced green phase. That
solves that incipient green problem - at a cost of worsening traffic
congestion by reducing intersection throughput, of course.

Does an advanced green for cyclists-only really solve the problem?
There can still be some cyclists not yet in the box and passing to the
right when the regular green goes active. Now they are in the
hook-zone.

Good point. I'm trying to remember if European ones also include some warning that the motorists are about to get a green.

But to be clear, I don't remember ever encountering a European bike box. I did see some bike lanes with advanced green phases for cyclists, in Stockholm, I think, but no bike boxes that I recall. My remark about the separate signal phases was based on what I remember from reading articles, etc.

Speaking of reading, here are John Allen's comments on bike boxes:
http://john-s-allen.com/blog/?p=60

Hmm. From that it appears the box is being used by cyclists
to make a left-turn. Is that the intent? I assumed it was
mainly for straight-ahead bike traffic, to avoid a right-turning
vehicle.

The following silly video illustrates a toy bike turning
left from the bike box.

http://www.edmonton.ca/transportatio.../bike-box.aspx

That seems insane. He's actually turning from the right lane,
not the center lane. How is that a good idea?

--
Joe Riel

On 11-16-2013, 00:10, Joe Riel wrote:
The following silly video illustrates a toy bike turning
left from the bike box.

http://www.edmonton.ca/transportatio.../bike-box.aspx

That seems insane. He's actually turning from the right lane,
not the center lane. How is that a good idea?

You have to look at _both_ videos to get it. I think the Edmonton style
in the second video makes a LOT more sense than any of the other styles
I've seen. Not thrilled about the first one, but since NO ONE can go
straight at that intersection, it's not bad.

On Friday, November 15, 2013 9:27:07 PM UTC-8, Wes Groleau wrote:
On 11-16-2013, 00:10, Joe Riel wrote:

The following silly video illustrates a toy bike turning

left from the bike box.



http://www.edmonton.ca/transportatio.../bike-box.aspx



That seems insane. He's actually turning from the right lane,

not the center lane. How is that a good idea?



You have to look at _both_ videos to get it. I think the Edmonton style

in the second video makes a LOT more sense than any of the other styles

I've seen. Not thrilled about the first one, but since NO ONE can go

straight at that intersection, it's not bad.

The second video is SOP around here -- and what is required by the UVC.

-- Jay Beattie.

Wes Groleau writes:

On 11-16-2013, 00:10, Joe Riel wrote:
The following silly video illustrates a toy bike turning
left from the bike box.

http://www.edmonton.ca/transportatio.../bike-box.aspx

That seems insane. He's actually turning from the right lane,
not the center lane. How is that a good idea?

You have to look at _both_ videos to get it. I think the Edmonton
style in the second video makes a LOT more sense than any of the other
styles I've seen. Not thrilled about the first one, but since NO ONE
can go straight at that intersection, it's not bad.

Thanks, I didn't realize that first video was a three-way intersection.
The pictures below the videos are confusing; they depict the three-way
intersection, but don't show enough detail to clearly show it is a
three-way. If smarter I could have deduced that from the arrows.

--
Joe Riel

On Fri, 15 Nov 2013 14:10:10 -0800, Frank Krygowski wrote:

snip



PDX was given the green light to perform an experiment. They were given
the OK after the fact, due to political string-pulling. The engineers
in charge are lucky to retain their licenses.

The green boxes are entirely self-explanatory. Federal regulators may
be in a tizzy, but from an actual use standpoint, this is a tempest in
teapot.

Understand, Jay, the installation was an experiment. The objective was
to prevent (or at least, greatly reduce) right hook problems for
cyclists.

Did it work? No, it failed. You can call it a "tempest in a teapot,"
but eight years worth of data (four before, four after the installation)
shows that right hook conflicts increased greatly. No, it doesn't mean
you're likely to get killed riding through one of these things. But it
_does_ mean the design was a failure!


What's the source of your information?

Here's a link to a report evaluating Portland's bike box "experiment."

http://nacto.org/wp-content/uploads/...Bike-Boxes-at-
Signalized-Intersections-2010.pdf

Your attention is called to page 2 of the Executive Summary to wit:

Do the bike boxes improve safety? After controlling for volumes, the
number of conflicts decreased and yielding behavior increased. In
addition, user perceptions of safety improved.
• Conflicts: Overall, the number of observed conflicts decreased from 29
to 20 while the total number of cyclists increased 32% and motor vehicle
right-turn volumes increased by 7%. Controlling for differences in
volumes of bicycles and right-turning vehicles, fewer bicycle-motor
vehicle conflicts are expected for an intersection with a bike box.
• Yielding: Our analysis of the three additional intersections for
yielding behavior found an increase in the key behavior of right-turning
cars yielding to cyclists at the treatment locations and a decrease at
the control location. The increase is partially driven by additional
interactions as a result of increased volumes; however, the increase in
yielding is proportionally more than the volume increase.
• Perception of Safety: The surveys found that both motorists and
cyclists perceived the intersections to be safer after the installation
of the bike boxes. In particular, 42% of motorists who are not cyclists
felt driving through the intersections was safer with the bike boxes
(compared to 14% who felt it was more dangerous). Moreover, 77% of
cyclists felt bicycling through the intersections was safer with the bike
boxes (compared to 2% who felt it was more dangerous).


I'm aware that this shows conflict reduction and not actual accident
data. Section 1.4.1 of the report describes the difficulty of using
actual accident data, to wit:


1.4.1 Accident Records Analysis
In theory, bicycle crash data offer concrete evidence of changes in
intersection safety; however, in practice drawing valid conclusions from
available crash data has proven very difficult. Three of the six bike box
evaluation studies analyzed bicycle and motor vehicle accident records
(Allen et al. 2005; Newman 2002; Rodgers 2005). However, none of the
studies found that there was an adequate amount of crash data to draw
firm conclusions. Newman (2002) found an overall trend toward accident
reduction after installation, while Allen et al (2005) found some sites
with increased casualties and other with decreased casualties. Neither
study had statistically significant findings. Rodgers (2005) analyzed
accident records for the period prior to the bike box installation, but
concluded there was not enough data in the short period of time after the
installation to merit including after-installation data. Atkins (2005)
utilized accident records to identify which intersections were
appropriate for study, but did not include any analysis of accident data
in the study.
The difficulty presented by incomplete accident data in assessing
intersection safety is not unique to these bike box studies. A meta-
analysis of accident reporting across 13 countries, including the United
States, and found that single-vehicle bicycle accident reporting was the
lowest of all categories – at less than 10% (Elvik and Mysen 1999).
Another study found that fewer than 10% of emergency room cases that
resulted from bicycle accidents were duplicated in state accident files
(Stutts and Hunter 1999).

snip



They're a failed experiment. They did not achieve their objective, by
the standards of measurement agreed upon. In fact, they did the
opposite, and produced a greater count of the problem they were intended
to cure.


Several advantages are attributed to bike boxes by the NACTO

http://nacto.org/cities-for-cycling/...on-treatments-
old/bike-box/

to wit:

Bike Box Benefits

Increases visibility of bicyclists.

Reduces signal delay for bicyclists.

Facilitates bicyclist left turn positioning at intersections during red
signal indication. This only applies to bike boxes that extend across the
entire intersection.

Facilitates the transition from a right-side bike lane to a left-side
bike lane during red signal indication. This only applies to bike boxes
that extend across the entire intersection.

Helps prevent ‘right-hook’ conflicts with turning vehicles at the start
of the green indication.

Provides priority for bicyclists at signalized bicycle boulevard
crossings of major streets.

Groups bicyclists together to clear an intersection quickly, minimizing
impediment to transit or other traffic.

Bicyclists can avoid breathing exhaust while queued at the signal.

Contributes to the perception of safety among users of the bicycle
network.

Pedestrians benefit from reduced vehicle encroachment into the crosswalk.


I don't wish to leave the impression that bike boxes are the greatest
thing since sliced bread. As noted in my previous thread, no single
strategy fits all situations. The need to move all traffic efficiently is
important in dense urban environments. Let me quantify what that means in
NYC. Here's the incoming cordon count for cars and bicycles crossing the
60th St corridor during the 3 hour morning rush hour (Oct 2011)

Street,vehicles/bicycles
York Ave,2713,0
Second Ave,8242,307
Lexington Ave,5079,132
Park Ave,5545,32
Fifth Ave,8032,231
Seventh Ave,3076,157
Broadway,4644,247
Columbus/9th Ave,5608,182
West End/11th Ave,3924,68

FDR Drive,21411,0 (cars only limited access hwy)
West Side Highway,14480,0 (cars only limited access hwy)
Hudson River Greenway,0,956 (class 1 bike lane adjacent to West Side
Highway)

Bike Share was introduced at the end of May of this year. So far, the
number of daily trips within the CBD has averaged about 30,000 trips
daily.

These numbers should give you a measure of the problem to be solved. I
don't expect these numbers are duplicated in places like Palo Alto CA.
However, I would take a deep breath before suggesting what somebody
believes is adequate for Palo Alto should be the model for Midtown
Manhattan.

Stephen Bauman