The first two parts of this series dealt with warrants and justifications for traffic signals. Here the series concludes with a discussion of warrants and justifications for signs and other traffic control devices.
Justification for Stop Signs
Let’s go to the Minnesota MUTCD; section 2B.4:
.03 YIELD or STOP signs should be used at an intersection if one or more of the following conditions exist:
- An intersection of a less important road with a main road where application of the normal right-of-way rule would not be expected to provide reasonable compliance with the law;
- A street entering a designated through highway or street; and/or
- An unsignalized intersection in a signalized area.
.04 In addition, the use of YIELD or STOP signs should be considered at the intersection of two minor streets or local roads where the intersection has more than three approaches and where one or more of the following conditions exist:
- The combined vehicular, bicycle, and pedestrian volume entering the intersection from all approaches averages more than 2,000 units per day;
- The ability to see conflicting traffic on an approach is not sufficient to allow a road user to stop or yield in compliance with the normal right-of-way rule if such stopping or yielding is necessary; and/or
- Crash records indicate that five or more crashes that involve the failure to yield the right-of-way at the intersection under the normal right-of-way rule have been reported within a 3-year period, or that three or more such crashes have been reported within a 2-year period.
.05 YIELD or STOP signs should not be used for speed control.
The typical location is a minor street approach to a major arterial, like all the minor side streets at Lyndale Ave. I doubt many or most of the Minneapolis “basket-weave” stop signs meet MUTCD recommendations, and the discouragement of stop signs for speed control is pretty clear. It breeds disrespect for all traffic signs and is counterproductive and even dangerous as motorists can tell when their time is being wasted and impatiently stomp on the gas between signs, then start doing “rolling stops”, then finally just blow through them (including the final sign which is actually entering a street busy with cars and pedestrians). Meanwhile, pedestrians are lulled into a false sense of security. In over 70 technical papers, it’s been shown that “speed control” stop signs increase safety or decrease speeds only in the very limited situation of on-street parking causing a sight distance problem.
Engineers aren’t necessarily always against speed control in appropriate places, say, to discourage regional traffic from cutting through on local streets to avoid congestion on an arterial. This is a particular problems in older neighborhoods where the lack of a hierarchical road network and capacity problems at intersections enable and encourage motorists to cut through on local streets. But using geometrics like chokers and traffic circles or signal timing to accomplish it is a lot safer and more effective.
Justification for 4-Way Stop Signs
Once again lets see what the MUTCD has to say; Minnesota MUTCD 2B.7:
.01 Multi-way stop control can be useful as a safety measure at intersections if certain traffic conditions exist. Safety concerns associated with multi-way stops include pedestrians, bicyclists, and all road users expecting other road users to stop. Multi-way stop control is used where the volume of traffic on the intersecting roads is approximately equal.
.02 The restrictions on the use of STOP signs described in Section 2B.4 also apply to multi-way stop applications.
.03 The decision to install multi-way stop control should be based on an engineering study.
.04 The following criteria should be considered in the engineering study for a multi-way STOP sign installation:
A. Five or more reported crashes in a 12-month period that are susceptible to correction by a multi-way stop installation. Such crashes include right-turn and left-turn collisions as well as right-angle collisions.
B. Minimum volumes
1. The vehicular volume entering the intersection from the major street approaches (total of both approaches) averages at least 300 vehicles per hour for any 8 hours of an average day; and
2. The combined vehicular, pedestrian, and bicycle volume entering the intersection from the minor street approaches (total of both approaches) averages at least 200 units per hour for the same 8 hours, with an average delay to minor-street vehicular traffic of at least 30 seconds per vehicle during the highest hour; but
3. If the 85th-percentile approach speed of the major-street traffic exceeds 40 mph, the minimum vehicular volume warrants are 70 percent of the values provided in Items 1 and 2.
Perhaps a 4-way stop would be appropriate at 46th and Bloomington, but having them on Lyndale Ave and a minor street, say at 25th St where there’s been calls for a traffic signal, or where unwarranted signals should be removed clearly violates MUTCD recommendations and would preclude any attempt at platooning.
Justification for Warning Signs
In engineer speak, here is the justification for warning signs; Minnesota MUTCD 2C.1:
Warning signs call attention to unexpected conditions on or adjacent to a highway, street, or private roads open to public travel and to situations that might not be readily apparent to road users. Warning signs alert road users to conditions that might call for a reduction of speed or an action in the interest of safety and efficient traffic operations….The use of warning signs should be kept to a minimum as the unnecessary use of warning signs tends to breed disrespect for all signs.
Warning signs are effective when used to warn drivers of discrete, static, unexpected conditions, and preferably inform them of what action to take. Examples of effective warning signs are curve ahead signs, mid-block crosswalks, and uneven lanes. Here’s one from Nova Scotia, where they want you to get the idea that the expressway ends and if you don’t stop and turn you’ll have a close encounter with a balsam fir tree.

Nova Scotia Expressway Ending
Here’s another good installation of warning signs, this time a little closer to home near Worthington. Local county roads were used as a detour during completion of the MN 60 Expressway, and the road takes a sharp turn that is likely unexpected in flat farm country and difficult to see due to being on top of a small hill. The signs warn of an unusual, static, always present hazard and are clear what action to take (slow down to 40 mph and be prepared for the turn).

Appropriate Warning Signs
Warning signs are not effective when warning of ordinary conditions that may or may not be present and are unclear about what action to take. An example would be Slow-Children at Play signs.

Slow Children Sign
Besides contributing to motorist disrespect for all signs, they endorse the idea of children playing in the middle of the street, lull children and parents into a false sense of security, and give the impression that if the signs aren’t there, then it’s OK to drive fast and it’s not necessary to watch for children. And how slow is “slow” specifically? The speed limit? 25 mph? 10 mph slower than whatever you’re doing? When is it OK to speed up again?
Similarly, those “Deer” signs on rural roads are ineffective and are now being removed. Here’s a sign from New Brunswick. It’s dramatic, but is it effective?

New Brunswick Warning Sign
Signs to actively warn motorists if deer are present have been tested by Mn/DOT as far back as 2001, and were found to be effective, reducing car vs deer crashes by a third to a half, but the technology was just too expensive and cumbersome. Attempts to tweak it are continuing.
Inappropriate stop and warning signs (and signals) are of course extremely common; nothing gives the perception of safety like a nice shiny sign. A study in Stearns County showed one-third of the highway signs int he county were not needed, and maintaining unneeded signs is not a trivial expense. Some agencies have practiced a don’t ask / don’t tell policy on speed control stop signs, removing them when they’re called on it. Sometimes politicians like to play “traffic engineer” and order a sign installed with absolutely no engineering justification. Often an audit finds signs for which there are no records of who, when, and why it was installed, or not uncommonly signs that residents have bought and illegally installed themselves! Here’s an illegal sign in my neighborhood.
Although it’s unlikely the city has the wherewithal to find the responsible party and prosecute a crime like this, if, say, an illegal sign alters the intended right-of-way in an intersection or a child thinks it’s safe to play in the street and a crash occurs, a lawyer is a lot more likely to find out who put up the sign.
Crosswalk Warning Devices
If it’s desired to have something more than an uncontrolled crosswalk but less than a full-fledged signal (or a signal is not warranted) there are options. One is a Rectangular Rapid Flashing Beacon (RRFB).
These use a standard pedestrian sign with yellow strobe lights to warn motorists that pedestrians are present.
Here are LEDs embedded in the road in Wisconsin Dells
Another is a High-intensity Activated crossWalK beacon, better known by the catchy but contrived acronym “HAWK”.
These are substantially cheaper than conventional traffic signals (about $80,000), and have close to 100% compliance from motorists (as opposed to 25–45% on crosswalks even when equipped with a flashing yellow beacon). Unlike most of the other devices we’ve been talking about, there are specific warrants for these. The warrants are for any 1 hour of a day.
These tend to work best on isolated, wide, suburban-style roads where there’s no attempt at a green wave and platooning and there are a lot of pedestrians crossing, like near a park or school. I’ve seen several of these on the Wisconsin Dells strip where people who don’t want to pay to park in the waterpark parking lots are walking from their motels.
Another thing worth noting: the city of Minneapolis, with permission from the American Association of Highway and Transportation Officials (AASHTO), the agency that develops the MUTCD, has been experimenting with flashing red lights on an overhead sign, in order to see if they have the benefits of a HAWK at an even lower price. The thinking is that flashing red lights send a stronger message, since the legal meaning of a flashing yellow light is “Caution” or “Look at this Sign”, whereas flashing red means “Stop”.

Minneapolis Pedestrian Crossing
These omit the pedestrian signal present on regular traffic signals and HAWKs, and with the visors you can’t really see what indication motorists have. I found it disconcerting as a pedestrian not knowing if they had switched to flashing red yet or if they were even working. Since ADA compliant push-buttons will eventually be installed at all signalized intersections, I wonder if they could also be used here with a feedback message like “31st Street has red lights”.
Since the cost of a fatal crash is high ($10.6 million dollars) and all these various blinking light devices are fairly cheap in capital costs (an RRFB is about $15,000) and don’t create much in congestion costs, the question might be why don’t we use more of them? I see more and more going up, which is good. These are going up in Bloomington, including one crossing on Lyndale Ave near the VEAP social services building, and another where students cross from the Holiday gasoline station to Kennedy High School in the middle of the block rather than walking 200 feet to a signalized intersection.
But there is such a thing as too much of a good thing, so we need to be careful and use these warning systems with discrimination. Take the center-mounted brake lights on cars: when they were introduced rear-end crashes immediately went down 35%. But now that every car has them, they’re becoming part of the background noise of the road; effectiveness is now at 4.3% as drivers have tuned them out. Use too many signs, too many lights, and when everything screams at motorists from every direction, it all just gets tuned out as noise.
Very interesting! Two points of further interest:
1. Could you go over two-way stops? There are so many of them in my neighborhood in Minneapolis which was interesting to me because back in DC where I grew up, I remember there being hardly any two way stops. The issue I have with two-way stops is that many drivers assume they are four way stops and thusly give them the “rolling stop” treatment. There are many cases where I think drivers are completely unaware that it isn’t a four way stop as I have been honked/yelled at/almost hit when I have continued through intersections where I have no stop but the honker/yeller/quasi-stopper does. The issue is exacerbated by the fact that drivers park their cars right up to the crosswalk (sometimes amusingly, into the crosswalk), ignoring the 30 foot rule, so the lines of sight are very poor for those dutifully stopping so poor driving can result due to poor sight lines.
2. Is predictive modelling used when looking at signal changes? Are the traffic flow impacts and warrants/justifications, etc. modeled for prospective changes? Using the Lyndale and 25th St. example from above and completely made up but hopefully illustrative results, say in the model, traffic volume on 25th street at Lyndale Ave would be increase to be large enough if stop signs were removed and signals added to justify the signal installation. Further, the newly signalized 25th St. would suck up almost all of the traffic volume from 22nd St., resulting in a 22nd St traffic volume that only justifies a two way stop. Even further, the newly signalized 25th St. really creates new hyper-localized traffic demand so now traffic volume on 27th St. is now also large enough to justify a traffic signal.
The fact that there’s one line in the MUTCD which says…
“YIELD or STOP signs should not be used for speed control.”
…absolutely does not mean that yield or stop signs should not be used for speed control. It means the MUTCD reflects dated thinking and is largely irrelevant for streets in complex urban environments.
Huh? Did you not read the rest of that section?
“In over 70 technical papers, it’s been shown that ‘speed control’ stop signs increase safety or decrease speeds only in the very limited situation of on-street parking causing a sight distance problem.”
An aversion to stop signs isn’t a pro-car, drive-through-the-neighborhood-at-45 mentality. It’s simply a reflection of the fact that it doesn’t work.
I strongly suspect those studies were done on wider suburban residential streets without on-street parking, boulevard trees, or even a consistent sidewalk network (putting any potential peds in the street and at greater risk). Where the design of the street – both width, corner sightlines, and length between intersections – really do allow drivers to speed up to 35-45 mph between pesky stop signs.
On Minneapolis streets, where those design elements are all present, I find it hard to believe going above 30mph between stop signs is common. Like I said in a different comment section: I think >20mph is too fast for a side street. So I get frustrated when I see someone who is probably pushing 30 going down my street with 50-75% parked cars hiding what could be kids or dogs or adults coming out between them. But an engineer wouldn’t find fault in this at all since the posted speed limit is actually 30.
A stop sign in old neighborhoods on nearly every corner does enough to more than wash away any time savings from avoiding arterial congestion. At least, enough if you wanted to do a road diet or change/add signals on said arterials to slow vehicle traffic down and didn’t have the budget to add roundabouts or diverters in the neighborhood.
Context is important, but this sentence means something very different than what most people are interpreting it as meaning.
I read this sentence, and expand it to read thusly;
If a stop sign is placed when it does not fulfill any warrants, and is only a brazen attempt at traffic calming in a location where it is unsuited, it will not only be ineffective, but quite possibly to probably cause an effect counter to the intention. HOWEVER, this does not mean that a stop sign, placed where it increases safety, or where there would be a desire to control traffic at an intersection, increases speeds.
Most Minneapolis stop signs actually meet the warrants laid out above (see my comment below). And therefore I have no evidence to say if they have lowered or raised speeds in the neighborhood. Personally I think they have no effect, but the narrower streets, with 2-sided parking, lots of kids and pedestrians and bicyclists, creates the slower streets, and the stop signs simply “reset” every driver every few blocks.
Also, I’d bet good money that parking on your “old neighborhood street” as well as trees, fences and even houses, cause sight-distance issues which can only be resolved through stop signs, and thus, from Sean’s comment and Monte’s post, these signs may actually reduce speeds.
I don’t have anything more than anecdotal evidence to go on, but I have seen the “speed up quickly after a stop sign” phenomenon in Minneapolis. It started appearing in my old neighborhood shortly before I left for the Navy, and I’ve seen it occasionally on trips home since…
I’ve noticed that commuters will use a road if there isn’t a stop sign for two blocks. It’s an opportunity to gain acceleration since they won’t have to stop at the next block.
I’ve been trying to get the City of Minneapolis to install a 4-way stop intersection at 53rd and Girard Ave. S for several years now (It’s currently a 2-way). It’s a fairly active pedestrian and biker intersection.
I’ve tried to get the Lynnhurst neighborhood association on board, on the grounds that it’s safer for pedestrians, bikers, and children.
I’ve even tried to get the Minneapolis bicycle coordinator, Simon Blenski, on board. This was due to seeing one too many close calls between westbound bicyclist almost getting hit by commuting drivers. He didn’t think it was warranted because nobody has been hit or killed at that intersection yet.
We’ve even sent a signed petition with over half of the residents on the two affected blocks to our counsel member, Linea Palmisano.
It’s nearly impossible to convince the City of Minneapolis to install just two more stop signs.
Is there any case examples, or a better way to get a 4-way stop intersection installed without somebody dying? I’d be interested in finding out if this is a lost cause despite the tax paying residents (directly or indirectly) are all in favor of this intersection change.
“The neighborhood thinks someone will be hit some day” is not listed in the MUTCD as a valid warrant for a 4-way stop.
There have been case studies. Over 70 of them that all say that unwarranted stop signs do not increase safety or decrease motorist’s speed unless there’s an unusual sight distance problem.
Hundreds of studies supported Phrenology, does not mean it was a correct way to think.
I would suggest reading 2B.4.04 – 2 again (as well as 2B.4.04 – 1), as these are the justifications for most of the basket weaves.
ADT on most St Paul local streets ~ 900-1,000 ADT, thus an intersection of two local roads would approach the 2,000 movements per day threshold without any bike/ped movements.
Definitions of unsafe sightlines are not found in the MNMUTCD but actually in AASHTO. To leave an intersection safely uncontrolled, there needs to be a specific amount of sight distance, (if you have an AASHTO Green Book, “Intersection Control, Case A-Intersections with No Control” specifically Exhibit 9-51.) It is up to every engineer to interpret this guidance and the existing conditions carefully and to uphold the Oath of the Engineer as well as the ethics they agreed to by obtaining their P.E.
Also you forgot the Option warrants for all-way stop control…
Other criteria that may be considered in an engineering study include:
A.The need to control left-turn conflicts;
B.The need to control vehicle/pedestrian conflicts near locations that generate high pedestrian volumes;
C.Locations where a road user, after stopping, cannot see conflicting traffic and is not able to negotiate the intersection unless conflicting cross traffic is also
required to stop; and
D.An intersection of two residential neighborhood collector (through) streets of similar design and operating characteristics where multi-way stop
control would improve traffic operational characteristics of the intersection.
More like “trying to keep the article to a reasonable length” as opposed to “forgot”.
A new stop sign was put in at Wentworth and W 41st St in South Minneapolis. There used to only be one stop sign on 41st between Lyndale and Nicollet (at Grand). People would use it as a shortcut because every other street north of 50th has at least two stop signs or stoplights between Nicollet and Lyndale. And they would drive fast.
The new stop sign almost certainly doesn’t meet warrants, but it has slowed down car speed considerably, and driven more of the east-west traffic to 40th (which was designed for more cars). It seems like a great win for the neighborhood.
Joe, Do you know what the community’s action was in order to get the city of Minneapolis to install a stop sign?
Was there a death, injury, or car accident history?
Did neighbors reach out to their counsel member and they acted as an advocate?
What is the full case history on this? In DC, 4 way stops are everywhere. Even in progressive St. Louis Park, just on the other side of France Avenue from Minneapolis, south of Cedar Lake, 4-way stops are in abundance at every street corner.
I don’t see why the city feels that traffic should
be allowed the room to speed up to 30 mph on narrow residential streets with parallel parked cars on both sides of the street.