To Stop or Not to Stop

My friend, a recent bike-commuting transplant from the Midwest, was describing his 35-mile round trip commute during the misty, quiet of the early mornings.  He sets off in the dark, living his ideals, even on those days when he doesn’t really want to.  His impressive commitment put me in awe, but what he wanted to talk about was stop signs.
 

“What do people here in Portland do?” he asked, eager to know.  “At 4 AM, when no one is around, do you stop?” 

At first, I thought he was joking.  Is there any place in the US where cyclists stop at stop signs at 4 in the morning when no one is around?  But he seemed earnest, awaiting my sage response.  Because I am a Portland native and a 25+ year bike commuter, he must have thought I have some insider track.  But as I considered his question, I realized that I am but one single biker with my own opinions of right and wrong.  Had he asked my 13 year old daughter, an avoid rule-follower, he would have gotten a completely different response.   I know what I do.  I have a feel for what I see around me, but what is the official Portland ‘culture’ when it comes to stop signs?

The only way to find out was to dust off my college anthropology-class tools – a trusty pad of write-in-the-rain paper and some wool socks - and observe the natives.

I found a significant observation point from which to gather data for my Stop Sign Obedience Observation Study.  It was meaningful because the Portland police often choose Ladd Circle in SE Portland as a place to set up large-scale infraction crackdowns aimed at bicycle riders.  I had recently made a narrow escape from the stop-sign crackdown - I turned before the cop writing tickets could wave me over to the side of the road.   I figured my study might serve two purposes.  I could answer my friend’s question and determine whether the cops’ focus on stop sign-blowers at the Circle was well-placed.

I set up shop early one sunny afternoon, leaning my bike against a concrete utility housing and hopping up on top.  I sat cross-legged in my yellow bike slicker, facing SE Ladd as cyclists rode towards me from downtown on their way home for the evening.

I tracked all vehicles as they approached the intersection of Ladd with the Circle, where a red octagonal sign stood, clearly telling them that traffic law required them to stop.  There was no ambiguity here, but what I witnessed surprised me.  In the one hour I sat there, less than half of all vehicles actually stopped, and by ‘stop’ I mean wheels stopped turning, a standard I would later question.   

Thirty-six percent of cars failed to stop and 71% of bikes, which averaged 60% non-compliance among all vehicle types.  The difference between the behavior of people when they sat behind a steering wheel  and when they sat atop a bicycle seat was significant: 34% more bikes disobeyed the law than cars.

As I sat there, easily identifiable as a biker by my attire and accoutrements, and obviously taking note of my fellow cyclists, I got a variety of responses.  Many tried to behave.  A few were successful, others were not.  They put down a foot to imply ‘halt of forward progress,’ even as they continued to roll, or they squeezed on the brakes all of a sudden when they saw that I was tallying up their actions.  A few sneered and made unkind remarks from afar as if my presence was some sort of a threat or proof of an alliance with Big Brother or the fuzz.

 

I tut-tutted the high rate of disobedience as I watched, but all along, I knew that my own traffic behavior was no different from those I observed, both on 2 wheels and 4.   I felt no guilt for my own infractions, nor did I judge any of the vehicles harshly, because, despite the overwhelming number of tallies in the ‘no stop’ column, I did not witness a single dangerous traffic interaction.  In fact, traffic never slowed and not a single driver or biker seemed uncertain about what those around them were doing.  The lack of a full complete stop didn’t have much effect on the success of the intersection. 

Indeed, the traffic circle configuration itself is a serious tease.  Traffic circles are meant to speed traffic along without forcing a stop or the ‘look left-look right-look left’ that is required at four-way intersections.  In Bend, the traffic circle is meant to be entered and exited without stopping.   My data collection point, I decided, was not the typical intersection on which I could base conclusions about Portland bikers in general.   

As I wrapped up an hour of data collection, I knew that further study was required to fully answer the question about what Portland cyclists do when faced with a stop sign.  I found two additional, dissimilar observation points along the same bike corridor.  One has a flashing red stop light on a four-way intersection with decent bike traffic and relatively low cross-traffic.  It is situated on SE 34^th and Clinton.  The other data collection spot is relatively high cross-traffic intersection along the same bike corridor.  It is the four-way stop at SE 26^th and Clinton.  

 

I also figured I needed to measure behavior at different hours of the day to have a well-rounded definition of how Portland bikers, as a group, behave.  So I selected 15 minute increments in the morning rush hour and similar periods in the afternoon rush.  In all, I tallied up 16 sessions of 15 minutes each, a total of 240 minutes.

The Data

I tracked 2,517 vehicles; 1,523 cars and 994 bikes.  There was also one skateboarder, but I didn’t count him (he didn’t stop).  The busiest intersection was 26th and Clinton, with 1,474 vehicles, followed by Ladds Circle, with 701.  The intersection at 34th and Clinton had 342 vehicles.  (Note:  I measured vehicles from all directions at the four-way intersections, whereas I had only measured the one direction, the typical commute pattern, at Ladd’s Circle due to visibility issues.)

The statistics varied by time of day and by intersection type.  The most extreme sample was on a slow morning, during which 92% of all vehicles failed to stop.  It was at Ladd’s Circle: 95% of bikes did not stop and 83% of cars did not stop (I guess the cops had it right).  Of the people who did stop, it appeared that they did so only because there were pedestrians in or near the crosswalk.  Driving behavior and biking behavior was more similar in this sample than in any other.  Apparently, no one thinks it is important to obey the stop sign in the morning when there is very low traffic, regardless of what conveyance is used.  From my observations of the sparse, one-directional commuting traffic at that hour, it made perfect sense to me too.

Overall Traffic Counts at all Three Locations

Morning Rush

In the mornings, bikers did not stop much: 75% didn't stop at 26^th, with somewhat heavy four-way action; 85% didn't stop at 34^th, a quiet residential street that nevertheless has a blinking red light; and 95% didn't stop at Ladd's Circle.  Drivers followed the same pattern, with 26^th being the most stopped-at intersection and Ladd’s the least:  45% didn’t stop at 26^th; 56% didn’t stop at 34^th; and 83% didn’t stop at Ladd's (if the Portland police are looking to write a lot of tickets, Ladd’s Circle is indeed an area technically ripe with violators).  Conclusion:  drivers don’t stop much either.  Of the 938 morning vehicles at all locations, 33% came to a complete stop and 67% did not (Stoppers = 313, No stoppers = 625).  But on average, bikes glided through the stop signs 34% more than cars did, and surpassed cars in this measurement for every time period observed.

Intersection	Bikes – No Stop	Cars – No Stop	TOTAL NO STOP
26th x Clinton	75%	45%	53%
34th and Clinton	85%	56%	68%
Ladd’s Circle	95%	83%	92%
TOTALS	86%	52%	67%

 

Afternoon Rush

The afternoon commute brought slightly different results.  Out of 1,579 vehicles, 983 were cars and 596 were bikes, a surprising number because I always thought there were as many bikes as cars.  Two hundred and thirty bikes did not stop (61%), and 297 cars did not stop (30%).  On average, afternoon commuters did not stop at stop signs 42% of the time.  By location, cars did not stop at Ladds 36% of the time and bikes 71%; cars did not stop at 34th 46% of the time and bikes 71%; cars did not stop at 26th 28% of the time and bikes 48%.  Again, bikers rolled through more often than cars for every location and every period, on average 31% more.

Intersection	Bikes – No Stop	Cars – No Stop	TOTAL NO STOP
26th x Clinton	75%	45%	53%
34th and Clinton	85%	56%	68%
Ladd’s Circle	95%	83%	92%
TOTALS	61%	30%	42%

Even at the heaviest traffic of the afternoon at 26^th and Clinton, 32% of all vehicles did not stop.  But I saw no accidents, no near misses, not even that “you go – no, you go – no, you go” behavior that sometimes results from ambiguity among those who share the road.   

Morning Traffic Counts per Location                                                        

Afternoon Traffic Counts per Location 

 

 

Totals AM and PM Rush Hours 

Intersection	Bikes – No Stop	Cars – No Stop	TOTAL NO STOP
26th x Clinton	57%	33%	39%
Ladd’s Circle	79%	49%	71%
34th and Clinton	83%	53%	64%
TOTALS

Definition of “Stop”

I used the legal ‘complete stop’ threshold for determining whether behavior qualified as stopping or not.  As I tallied up hundreds of commuters, this standard revealed itself as less and less useful when comparing cars and bikes.  Because of their slower overall speed, bike riders have just as long, if not longer, to approach an intersection, observe the participants and determine whose turn comes next.  For the most part, they slow down sooner and move slowly for longer as they evaluate the timing and safety of crossing.  Rarely does this require a complete stop.  If a bike were to stop, it takes longer for them to get started again than it does for a car.  

As drivers, we approach intersections faster, decelerate later and faster, and spend less time gathering data about the likely traffic pattern.  We are also able to accelerate quicker when we start up again.  Because of these more dramatic differences in auto travel speed, acceleration and deceleration, when a car approaches a stop sign and rolls through without actually coming to a complete stop, it looks more like a stop than when a cyclist does the same.  About halfway through my study, I realized that I was guilty of this bias in perception, causing me to question my data.  But I persevered, consciously noting the movement of spokes and hubcaps as a way to focus more accurately on a halt of forward motion.  I wondered if police officers have a trick to overcome this bias.

As traffic participants, however, I don’t think we care much about stopping perfection in our own behavior or in that of others.  We do appear to understand and accommodate a slightly different set of physical and logistical factors when bicycles are part of the equation.  In fact – I think all bikers have experienced this on occasion - some drivers motion for bikes to go ahead even if it is not their turn.   

For this reason, I believe that experienced Portland drivers and bike riders probably agree that not stopping can be the most courteous thing to do at intersections: it allows the next intersection participant to make their move sooner.  The difference between people’s behavior when they are in cars versus on bikes is simply one of degree.  We are almost all guilty of not executing textbook stops as required by the law, but it is much easier to do so, or appear to do so, when in a car.

Results

The compiled data from all locations, mornings and afternoons, supports this.  Only 49% of all vehicles stopped: 62% of cars stopped and 29% of bikes.  (The only vehicle type with a perfect stop score was TriMet busses.)

The busiest intersection had the most stoppers in both cars and bikes.  At 26^th and Clinton, 61% of vehicles stopped – 67% cars and 43% bikes.  Curiously, I saw close similarities between people in cars and people on bikes in the extreme opposite situation as well.  During both the heaviest traffic and the lightest, bikes and cars behaved more similarly than in moderate traffic scenarios.  Across all times and vehicle types, the intersection at 34^th and Clinton, where cross-traffic was minimal and total vehicle counts were the lowest, saw the highest incidence of not stopping:  84% of bikes did not stop and 53% of cars did not stop.

The data tells me that a little over half the time, Portlanders technically obey the law at stop signs.  But functionally, most of the interactions I observed worked as the intersection designers intended.  Most of the people I saw took the following steps:  they approached the intersection, slowed considerably, looked both ways, allowed the waiting cross traffic to proceed if they were there first, and then continued to creep ahead until it was their turn, then accelerated through the crossing.  I marked all of those people as non-stoppers, even though they treated the other vehicles respectfully and, I like to think, made the entire interaction more efficient.  I did not have a special tally mark for rude intersection non-stoppers – people who zoomed through without waiting their turn or in such a manner that the smooth, expected  flow of traffic was interrupted.  Had I, there would only have been 10 hash marks in that column, and all of them would have been 2-wheeled travelers. 

Out of 2,517 vehicles, 1,229 stopped (51%) and 1,288 did not (49%).  Nine hundred forty five of those that stopped were cars and 284 were bikes.  Five hundred and seventy eight cars did not stop while 710 bikes did not stop.  Portlanders, when on their bikes, tend not to execute perfect stops at a significantly higher rate than when they are in their cars.    

Now I can answer my out-of-town friend.  Half of Portlanders stop, whether in a car or on a bike.  Most bikers don’t stop.  The bottom line is this:  like marijuana, ignoring stop signs is illegal, hence it carries the risk of a ticket.  Does that make it dangerous?  Let your conscience be your guide.