Right-hook risk drops with flashing “Yield to Bikes” sign on NE Couch

New safety signal at Couch and Grand-5-8

The right-turn warning at NE Couch and Grand is the only such sign in the country.
(Photos: J.Maus/BikePortland)

As Portland edges closer to possibly adding protected bike lanes to its downtown, a new study has found that one of its most unusual bike-lane intersection treatments seems to be working.

The LED sign above the intersection at NE Couch and Grand that flashes “Turning Vehicle Yield to Bikes” seems to have reduced right-turn conflicts by more than 60 percent since its 2011 installation.

However, the design hasn’t eliminated injuries or conflicts — an engineering term that refers to braking or changes of direction to avoid a collision — in the way that a dedicated bike signal phase might be expected to.

Here’s how the sign, the only one of its kind in the United States, works: there’s an inductive bike detector loop in the pavement in the bike lane just east of the intersection…

loop detectors

…and when it’s triggered, it causes the LED sign across the intersection to flash, hopefully reminding people in cars that they need to yield to a bike overtaking them on the right before making their right turn.

Many Portlanders weren’t big fans of the sign when it was installed. One of them was Kirk Paulsen, an analyst for Lancaster Engineering who worked on the new city-funded study.

“Before going into the data, I was skeptical of the effectiveness and thought it would be viewed as more of a neon advertising sign, and was surprised that it was shown to be as effective as it was,” Paulsen said in an interview Wednesday.

Advertisement

But after analyzing 24 hours of daytime video footage from before and after the sign was installed — 7 a.m. to 7 p.m. on two days in late September 2011, compared to the same times on two days in early October 2013 — Lancaster’s analysis found that the number of turning conflicts had fallen 66.7 percent:

conflict stats

After adjusting for different numbers of conflicts per right turn and per bicycle, the results came out even better. “Major” conflicts, categorized as those that required substantial braking or course adjustments by one or both vehicles, plummeted more than 80 percent after the sign’s installation. Total conflicts fell 68.5 percent per right turn and 61.4 percent per bicycle.

conflict stats per turn

“Based on the research it’s a really strong benefit,” said Peter Koonce, manager of the city’s signs and signals division.

Based on these findings, Koonce said the sign — which isn’t in any official federal manual of traffic signs — is on the table as an option for the coming upgrade to downtown biking and walking facilities.

“One of the next steps that we’ve contemplated was studying Broadway and Hoyt, which is a notoriously bad location,” Koonce said. “That sign is a viable option for Broadway and Hoyt.”

But Koonce also sounded notes of caution, saying that he too isn’t sure the LED sign will work elsewhere.

“I worry that it’s just noise, and that this is just one study,” he said. “It’s like the Food and Drug Administration, right? Your results may vary. … The takeaway that I hope we can share is that we’re going to be looking at these things.”

Several other factors are at play. At the time of the 2011 video footage, the bike box there was only one year old, and its effects on behavior, if any, might still have been developing. Also, there’s no scholarly consensus about whether reducing the number of conflicts between cars and bikes reduces the number of crashes.

However, Lancaster’s analysis also looked at the number of serious collisions at the location that were reported to authorities and/or BikePortland in the year before the sign’s installation and the two years after. Then they used their video statistics and Hawthorne Bridge bike traffic trends to estimate the number of bikes that came through the intersection during those periods. Using that method, the number of known serious collisions fell from three per year to two per year.

Those figures are far from definitive, though, because bike-related collisions go unreported.

Moreover, the estimated number of people observed on video biking through the intersection declined slightly from 2011 to 2013, a change that might be weather-related but suggests that despite the drop in conflicts, users didn’t find the new design significantly more comfortable or otherwise appealing.


Footage of a December 2013 right-hook collision, captured by a TriMet bus. The victim received upper and lower back injuries that required months of physical therapy and prevented use of his bike for transportation.

And Koonce noted that according to these findings, the LED sign failed to eliminate crashes. The city says it aims to completely eliminate traffic fatalities and serious injuries.

“Does a sign do that?” Koonce said. “No, a sign wouldn’t give you zero. But a signal probably would.”

On the other hand, Koonce said, using a dedicated bike signal to reduce right-turn conflicts can also increase traffic delay for all users. In some situations, like the one at Broadway and Hoyt, it might also require a general travel lane to be converted to a dedicated right-turn lane, because the rightmost lane with car traffic couldn’t simply display a green light.

Another downside of a dedicated bike signal phase, Koonce said: if it doesn’t offer a clear safety benefit, some people might not comply with it while riding bikes.

— Portland has tried many different approaches to dealing with right-hooks over the years. Learn more by browsing our right-hook story archives.

Correction 4/3: An earlier version of this post incorrectly described the loop detectors.

Michael Andersen (Contributor)

Michael Andersen (Contributor)

Michael Andersen was news editor of BikePortland.org from 2013 to 2016 and still pops up occasionally.

Subscribe
Notify of
guest

44 Comments
oldest
newest most voted
Inline Feedbacks
View all comments
Zaphod
8 years ago

I’m glad this sign exists and it is indeed helpful. I’m on this road quite a bit. Given light timing and it’s a downhill, cyclists can easily keep up with traffic. I always take the lane here and wait behind right turning vehicles. This creates the safest and most predictable traffic flow. It moves the most vehicles through. I never get honked at or otherwise hassled for this approach. While the sign is far better than no sign, my approach is a far lower risk option even though you’re more in the auto traffic mix.

MaxD
MaxD
8 years ago

I bike through here as part of my daily commute, and I am surprised that the sign is being credited with success. IMO, the bike box and ever-growing numbers of bikes on the road has a bigger effect. I regularly see cars cruising through this on red because they are no looking up at signs; they know the signal is red so they start looking left for an opening in traffic. Very dangerous place to be a cyclist or ped moving against traffic flow on the sidewalk. It would be cool to test the sign without a bike box, though, to get a better sense of efficacy.

Kyle
Kyle
8 years ago
Reply to  MaxD

I bike through this intersection daily as well and I always get in front of the vehicle waiting at the bike box before the light turns green, so they see me clearly, because I’ve seen so many potential right-hooks. Unfortunately it also opens me up to the occasional rude cyclist who will fly through to my right, often before the light has even turned green.

On a side note, I’ve recently stopped using eastbound Burnside up the hill from Grand due to the frequent and daily threat of right hooks at each of the side streets. Not only have I nearly been the victim of right hooks there on a weekly basis, but I’ve witnessed a right hook actually occurring about a block ahead of me.

No amount of signs, striping, or laws will change the fact that a significant percentage of drivers are distracted and incompetent. The only solutions are rigorous education or massive amounts of protected bike infrastructure, neither of which seem likely in the current crappy political climate.

Eric
Eric
8 years ago
Reply to  Kyle

Protected infrastructure would require closing intersections. How about a concrete bollard with a right-yield-to-bike sign sitting in the traffic lane just before each corner? Lanes are far too wide and unobstructed for city speeds.

paikiala
paikiala
8 years ago
Reply to  Eric

Standard lane widths are 10 feet. Have you seen a bus try to use a 10 ft lane with a UPS truck parked at the side of the road? Couch is a truck street as well, with several nearby truck based businesses.
One size only fits one size.

Spiffy
8 years ago
Reply to  Kyle

right hooks are just as bad on Hawthorne east… every week I’m nearly run over…

it’s just not fun commuting on the main bike routes any more…

invisiblebikes
invisiblebikes
8 years ago

But Koonce also sounded notes of caution, saying that he too isn’t sure the LED sign will work elsewhere.

“I worry that it’s just noise, and that this is just one study,” he said. “It’s like the food and drug administration, right? Your results may vary. … The takeaway that I hope we can share is that we’re going to be looking at these things.” Koonce

But the problem with Broadway and Hoyt is its a naked intersection. the lines are faded and markings are faded. If they went in and put in green painted path across hoyt, and green box along with this sign it would absolutely solve the issue.
But they can’t just drop the sign in and leave that intersection as naked as it is now, it would be noise then.

Also they should paint “right turns yield to bikes” about 50 feet before the intersection on the road in the right lane on Broadway.

hat
hat
8 years ago

This design should never exist. The right hand lane for this couplet section of couch would be best as a right turn lane only with Williams style diverters every block. Instead of spending money on a sign like this, make a safe street. 6 right hooks compared to a probable zero with diverters.

John Liu
John Liu
8 years ago
Reply to  hat

Are you suggesting the left lane be the only available through-lane for cars trying to go straight down NE Couch to the S-curve and Burnside Bridge? With the right lane being forced to turn right at every intersection?

That would be a disaster.

At morning rush hour, NE Couch is already full of cars, sometimes so much so that traffic cannot keep up with the traffic light timing. Almost all of those drivers are trying to go straight through to the bridge. If they were all forced into a single lane, the street would be total gridlock.

Also, almost all of the cyclists on NE Couch are also trying to go straight through to the bridge. Are you suggesting cyclists be jammed into the left lane with the cars if they are trying to go straight through? Or are you suggesting cyclists should proceed straight in the bike lane with cars being forced to turn right across that bike lane? That would be a disaster too.

NE Couch works fine as it is. I ride it every morning.

paikiala
paikiala
8 years ago
Reply to  John Liu

February 2015 at 8th: 11,000 cars per day. AM peak 1028 cars, PM peak 755. That’s one count on one day, but 1,000 cars per day per lane is the rule of thumb to enable a road diet, or not. It’s just over, but it is also a more significant street we would prefer auto traffic use since it became a couplet. The Davis greenway needs to be developed.

Adron @ Transit Sleuth
Reply to  John Liu

Another option, make it no turns!

Adam H.
Adam H.
8 years ago

Signage Is not a viable substitution for good bike infrastructure. Install more signs, sure, but not without adding dedicated bicycle infra —i.e. protected bike lanes

Alexis
Alexis
8 years ago

I commute into downtown on NE Broadway every day, and the bike signals at the NE Victoria and N Williams intersections make me feel very safe. Especially at N Williams, where there’s a very prominent sign warning motorists not to turn right on red. Both motorists and cyclists seem to know what’s expected of them.

I don’t know the statistics, but I would LOVE to see this design pattern instituted other places that have a high risk of right-hook collisions.

Reza
Reza
8 years ago
Reply to  Alexis

You need dedicated right turn only lanes and protected right turn signals for that design to work. Not an option here, nor at Broadway and Hoyt or 3rd and Madison downtown.

paikiala
paikiala
8 years ago
Reply to  Reza

I would say maybe. The Couch design should have anticipated heavy right turns at this intersection. MLK/Grand is a couplet. Anyone should expect heavier turns onto a couplet in the direction of each half, particularly where a higher level street intersects.

Hebo
Hebo
8 years ago

Improved signage, no matter how good, cannot make bad drivers into good drivers. I had a close call just this week at this intersection when a delivery truck made an unsignaled right turn from the left lane – so across one lane of traffic and the bike lane.
I do think the sign helps, but part of the improvement is by making cyclists (especially unsuspecting cyclists new to the intersection) of the risk. I also recall that the timing of this light and the preceding light changed around the same time. I think the intersection is safer if I am stopped to the right of stopped auto traffic and we both get a green than if traffic is already moving when I approach the intersection.

Daniel Costantino
Daniel Costantino
8 years ago

Why do we need a whole dedicated bike signal phase? Why can’t there just be a bike signal for the right lane that turns green 5 seconds before the vehicle signal does?

Once there’s a steady stream of bikes going through, drivers could wait the stream out before turning. A few more seconds of delay, but much less than what would result for everyone with a full dedicated phase.

Reza
Reza
8 years ago

To my knowledge, leading bicycle intervals (that you are describing) are generally not allowed under MUTCD because FHWA doesn’t want a green bike signal and green ball on at the same time. The idea is that a green bicycle signal implies an exclusive/protected phase, whereas in your scenario a rider that’s entering the intersection on a “stale” green may be unknowingly putting himself in harm’s way (right hook risk).

You could have a 5-second leading pedestrian interval and have signage instructing bicycle riders to follow the walk signals, IIRC.

Reza
Reza
8 years ago
Reply to  Reza

To clarify, this is the case at intersections where motorists CAN turn right, inviting conflict with bicycle riders going through the intersection.

J_R
J_R
8 years ago

Your solution would prevent only the right-hook for a motorist and cyclist waiting together on the red at the traffic signal. My closest near disaster from a right-hook motorist was when she caught up to and passed me about 30 feet from the intersection. (I was eastbound downhill on Hawthorne at 12th and going nearly auto speeds.) She came right up next to me, got her rear bumper about 5 feet ahead and simply began her turn. We both had a green. An advance green for bikes would have had no effect.

Chris I
Chris I
8 years ago
Reply to  J_R

Some motorists behave like dogs (as in, they don’t know that they have a huge mass behind their head that can hit or get in the way of things).

Hello, Kitty
Chris
8 years ago
Reply to  J_R

On Hawthorne heading East, the only safe solution is to take the lane. Alas, it is illegal. But safe!

PNP
PNP
8 years ago

I hope these numbers are true, but given my own observations of how seldom drivers actually look at signs, I have my doubts. For instance, I’ve made a habit of never being to the left of any driver who is in a lane that is marked “exit only.” They can drive right under two or three big orange signs with “exit only” in nice big letters and still slam on the brakes and make an abrupt, usually unsignalled lane change to the left just before the lane splits off. I see it just about every time I’m on a street or freeway with that configuration. Sure, this is anecdotal, but to my mind, it’s evidence of a lack of attention to the task at hand.

hat
hat
8 years ago
John Liu
John Liu
8 years ago
Reply to  hat

What is Google maps supposedly showing there? I’m missing the point.

adam
adam
8 years ago
Reply to  John Liu

Ifyiu look at the satellite view of the intersection Google took, it shows a giant car crash!

spencer
spencer
8 years ago

that intersection is insane, the only way to tame it is to do bike only signals, which carves out time from the rest of the grid, slowing all users down. I’m all for protected infra’ but it comes at a cost, slower riding and driving. i suspect that many would attempt to use a “car” signal to avoid waiting.

hat
hat
8 years ago
Reply to  spencer

This might be the only way to ensure safety if a massive redesign is not feasible. I’d wait. People on bikes are much more apt to obey traffic signals if it is meant exclusively for them.

While I love them often enough, I think it may have been a mistake to build the pedestrian extensions. PBOT cornered themselves into a design that cannot be fun for both cars or bikes. When the streetcar goes in sometime in the next decade or so it’ll be interesting to see how they change bike infrastructure on Couch.

paikiala
paikiala
8 years ago
Reply to  spencer

I would remove the parking, maybe change the corner and put the through bike lane (buffered) in the middle with a right turn only lane to Grand, ala Madison at Grand. The turn counts there probably justify a right only auto lane.

hat
hat
8 years ago
Reply to  paikiala

This is exactly what I was thinking. Would one or both sidewalk extensions be too narrow for standard width lanes? Do you remember what the impetus for the current design was (i.e. narrowing Couch)? Thanks paikiala. You’re a superstar.

hat
hat
8 years ago

I suppose another solution for this is simply move the bike lane to the South side of the street. Where Couch and Burnside Meet, bikes could cross with the ped. signal.

John Liu
John Liu
8 years ago
Reply to  hat

That would simply replace right hooks with left hooks on Couch, and delay cyclists by forcing them to wait at the Couch/Burnside crosswalk for a signal.

All a cyclist has to do, if he is really worried about right hooks, is to take the lane, which is easy because Couch is downhill and the lights are timed at 20 mph or so.

e2pii
e2pii
8 years ago

It would be nice to see more data: particularly the standard deviation in major and minor conflicts per day in the pre- and post- treatment period.

With just the numbers presented, and with the extremely small sample size, it’s really hard (mathematically) to say the sign having an effect and that the decrease isn’t just part of the random fluctuation. Is this a case of the Strong Law of Small Numbers?

Spiffy
8 years ago

I don’t think the sign works… it just adds to the urban clutter… it looks like a store sign… non-compliant colors and lighting… as a driver I’m looking for white or yellow signs to tell me what to do… blue and green signs help me navigate… a black sign with lit up letters? I’m not even going to read it…

it also doesn’t help that there’s no bike lane here… there’s a shared bike/parking lane… paint the inside stripe on the bike lane so it actually is a bike lane… I hate when they do the shared bike/parking lanes… cars park wide, and it’s not clear what it is…

adam
adam
8 years ago

There is a flashing “no turn on red” sign at a busy intersection along the new multi use path that parallels the new Orange MAX line / freight line.

But it only activates when there is a freight train approaching or crossing. It is a total waste of a signal – I have almost been right hooked several times at that intersection. Why can’t the City tweak it so that it triggers when a bike is lined up on the green bike loop detector rectangle?

SW
SW
8 years ago

At the Springwater crossing of Bell , there is also a large LED “NO RIGHT TURN” sign that comes on just before the diagonal bike crossing green light illuminates.
I’ve watched many drivers look up at it and attempt to sneak a right turn anyway. Signage doesn’t stop all poor behavior . A good rule is to verify the cars movements before you move.

OBTW: the green box to trip the sensor loop is almost completely worn away , needs to be repainted. Many “new to the trail” cyclists miss it.

eli bishop
eli bishop
8 years ago
Reply to  SW

Contact SAFE to report the need for repainting (safe@portlandoregon.gov) or use the Portland reporter app.

Keith
Keith
8 years ago

Nice article. A point of clarification about how inground vehicle detection works. It uses induction, not magnetics. If it was magnetic it would only work on ferrous metals. Induction works like a metal detector that’s placed in the pavement. It will work with any metal or metal alloy that conducts electricity. With bikes, alloy wheels are actually easier for the equipment to detect than steel wheels.

The sign was made in Woodland Washington by Wells Sign Mfg. The owner has worked with the City on many LED signs including the ones on the Light Rail Lines and the signs at N Interstate and Oregon.

Keith
Keith
8 years ago

Michael, For the purposes of this discussion…probably not. We’re talking inductive loop detection and I don’t think that if you zoom in that far you’re talking the same thing, but hey, I’m just a retired electrician, not a physicist. If you want to discuss minutiae over a cup of coffee, I’ve got plenty of time and opinions. On a serious note, I think it’s important that people understand how things that literally affect their lives work. Vehicle detection has many different forms and people need to understand how it works. It’s not weight, it’s not magnetism. In this instance it’s an inductive vehicle detection loop. Most detection in Portland is this type, but there are some other locations where it’s video detection and some places where it’s a form of microwave radar, but hopefully the end result is the same; all modes of transport are being detected and given a green light in a timely fashion.

John Schubert
8 years ago

Everything you folks are considering has some or all of the following drawbacks:
— Less than 100% effective. There will still be right hooks.
— Increases congestion and pollution.
— Slows bicyclist trip time.
— Slows motorists trip time
— Will not be done at every intersection. The city has targeted a very few intersections with the highest collision rates, but not even considered a behavior change that would be safe at all intersections.
— Costs lots of taxpayer money. (I’m fond of pointing out that someone worked third shift emptying bedpans for that money, and it should be spent carefully.)
So. . . . you all know where this is going:
Why not: recognize that roadway designs and traffic control devices that put bicyclists and motorists in this position are inherently dangerous?
Why not embrace “Bicycles May Use Full Lane?”
— Cheap. Leaves more money for patching potholes.
— Guarantees there will be no right hooks.
— Doesn’t decrease city trip times by much — certainly not as much as a separate signal phase.
— Not dependent on a complicated nonstandard gadget, or on motorists’ understanding of that gadget.
— Makes the safest behavior uniform throughout the city, instead of varying from one intersection to the next.

e.
e.
8 years ago

I was hit by a right hook while biking through a green light at this intersection just last week & had to be taken away in an ambulance because I wasn’t (and am still not) able to walk as a result of my injuries. No police report, no citation, absolutely no official documentation that a collision even occurred there, which completely angers me & makes me wonder if there’s a reluctance to report incidents that happen at this intersection because it contradicts this rose-colored glasses impression that the sign is improving the situation here.

Erinne
Erinne
8 years ago
Reply to  e.

I think I saw you in the aftermath. Unfortunately, I came too late to be a witness to what happened. Sorry, that’s awful.

Also unfortunately, I don’t think the “reluctance to report” has anything to do with this particular intersection or sign. It happens in incidents throughout the city. We really shouldn’t put the onus of making sure collisions that cause injury are reported on the victims, but it seems like that’s often what happens.

Hope you heal up quickly.

e.
e.
8 years ago
Reply to  Erinne

Thanks for the kind words. I was lucky that a few people did seem to see what happened & stopped to call for help & make sure that I was doing okay, so it could have been much worse.

And I totally agree about the city-wide issues with reporting collisions – it’s just frustrating to know that this intersection is a particularly known problem spot & that under- or mis-reporting of incidents there might further contribute to other people having the same misfortune. But as you pointed out, you could say the same thing about dozens of other places in Portland.