SFMTA Wants Stop Lights, Not Signs, To Keep Muni’s 5-Fulton Moving

An SFMTA board, displayed at a Wednesday community meeting, explained how adding traffic signals can speed up the 5-Fulton. Photo: Aaron Bialick

The latest of SFMTA’s efforts to speed up Muni lines to run into some neighborhood opposition involves its proposed replacement of stop signs with transit-priority traffic signals. Some Western Addition neighbors have protested a proposal to signalize five intersections on McAllister Street to speed up the 5-Fulton, one of the designated “Rapid” routes receiving upgrades under the Muni Forward program (also known as the Transit Effectiveness Project).

Initially, the complaints were driven by fears that signals would bring dangerous speeding to McAllister. Muni planners responded by holding more outreach meetings, and presented data showing that pedestrian injuries declined on similar streets after signals were added. They also say speeds won’t go up significantly, since signals will be synchronized for speeds below 20 mph.

A September hearing on the transit-priority signal plans for McAllister and Haight Street drew strong opposition from neighbors, leading the SFMTA to postpone the plans’ approval and drop a signal . D5 Supervisor London Breed asked the SFMTA to do the extra outreach, but is cautiously supportive of the agency’s efforts, said aide Conor Johnston.

“When it comes to transportation, her priority first and foremost is improving transit,” he said. “The only thing that trumps that is public safety.” Johnston said the data on injury reductions were “helpful, but not a complete answer.”

The stop sign at McAllister and Laguna Streets will remain, though five other intersections are proposed to get signals. Photo: Peter Ehrlich/Flickr

After an SFMTA study of 12 intersections that were signalized, “We found that the total traffic related pedestrian collisions at these intersections decreased from 30 to 8, indicating a safer pedestrian environment,” states a flyer [PDF] handed out at a packed community outreach meeting on Wednesday. As for concerns about drivers accelerating to beat yellow lights, SFMTA planners say they haven’t observed that behavior.

Sean Kennedy, the SFMTA’s Muni Forward program manager, said the data seems to quelled some neighbors’ fears, but that the complaints have shifted. “What we hear is that there’s a lot of concern over the neighborhood feel,” he said. “And that’s something we can’t really dispute with facts. It’s an individual preference if people do or don’t like signals.”

Natalie Burdick of Walk SF told Streetsblog in September that the Muni proposals “should not conflict with the SFMTA’s own stated priority for ensuring the safety of the city’s road users.” She noted that “signalized intersections can support safer walking environments, if they are designed effectively. For instance, signals can be timed to calm traffic with lower speeds, and provide regular phases for pedestrian crossings.”

The SFMTA’s broader plans for the 5 and other priority transit corridors include transit bulb-outs (a.k.a. sidewalk extensions) to provide roomier bus stops and improve safety by slowing cars at intersections, more visibly-marked “ladder” crosswalks, and some left-turn restrictions.

A snapshot of the SFMTA’s plans for McAllister [PDF], which include a new signal at Steiner Street (Fillmore already has one).
Along the 5, the SFMTA predicts that transit-priority signals, which stay green when a bus approaches, will save 1.5 minutes alone in each direction. An additional six minutes will be saved by speeding up boarding through bus bulb-outs, removing or relocating some stops, and adding right-turn lanes to keep turning cars out of the way.

Some of those improvements have already been made since the SFMTA made initial upgrades to launch the 5L-Fulton Limited service, which runs about 17 percent faster than the 5-local. The SFMTA says the 5L has attracted about 1,900 new daily riders (initially reported as 2,000), though some riders may have switched from other routes.

Bob Esfandiari, a 5-Fulton rider who lives at the west end of the line on La Playa Avenue, said he’s excited for a speedier commute.

“People are apprehensive, and I understand that,” he said. “But my commute right now is 55-ish minutes, and I definitely feel the slowdown once the bus turns on to McAllister. The bus is stuck behind cars, stuck at stop signs.”

Esfandiari also pointed out that the community meetings seem heavily attended by neighbors from the immediate area, but that Muni riders who travel through don’t seem to be heard from as much.

“I’m pretty sure anyone who lives past 30th Avenue would appreciate” the improvements, he said. “I probably should get the word out to my neighbors.”

  • david vartanoff

    Rescue Muni asked Muni to do this on the N,K,L, J and M. over a decade ago. Maybe someday…

  • BBnet3000

    Ah, the “letting the public do engineering” black hole.

  • Bruce

    The N, at least, is having stop signs replaced with “traffic-calming measures” at several intersections.

  • Greg Costikyan

    The 5 does run very slowly, and I can see how this would improve it. I used to bike McAllister every day, though, and my impression was that streetlights exist at every intersection with a high level of cross traffic, and that stop signs were reasonable, given the low level of cross-traffic on less frequented streets. What about two-way rather than four-way stops, with McAllister proceeding by default?

  • Michael Smith

    This posting completely missed reporting the problems with respect to what the SFMTA is trying to do. Speeding up buses of course is a good thing. Converting existing traffic signal to Transit Preferential Signals is of course a good thing and should have been done a long time ago. But replacing 4-way stops with traffic signals makes the situation much worse, especially for pedestrians. There are far better, more cost effective solutions.

    First of all, Muni is fudging numbers, like they previously did with schedule adherence (check out previous streetsblog articles on that one!). Stop signs don’t slow down vehicles by 18 seconds. If you go out to a few intersections and time the buses you will find that the a bus is typically slowed down by a stop for only a few seconds (certainly less than the claimed 18 seconds). And if your look at a traffic signal, such as the absurdly placed one at Haight and Ashbury, you will see that the signal actually significantly slows down the vehicles. Note that Transit Preferential Signals don’t always give the buses a green light. The SFMTA acknowledges that the lights will hold a signal green at most only for a few seconds. This means that buses will still often stop of those intersections with traffic signals. And note that the SFMTA never sites a real single study showing their numbers are valid.

    And they talk about “synchronizing” signals for buses?? You can’t synchronize signals for buses when buses stop for an indeterminate amount of time at the bus stops. You can synch them for cars, bikes, and sometimes even express buses. But you can’t do so effectively for even Limited buses such as the 5L and the 71.

    And if you talk to them about their “safety” study you will find that they are comparing apples to oranges. They were looking at very different types of intersections compared to the ones they are now proposing to change. Traffic engineering studies have long showed that traffic signals are significantly more dangerous than 4-way stops due to cars speeding through the signals. Just look at the stats. After the last hearing there was yet another serious collision, this time on Laguna. Was it at a 4-way stop? Nope, it was at Laguna and Fulton where there is a traffic signal (as opposed to Fulton and Haight where there is currently a 4-way stop).

    There are other solutions. For some intersections there should simply be no traffic signal nor stop for the buses. That will allow the buses to go right through the intersection without getting stopped at a red light. If pedestrians want to cross they will still be able to without having to wait. Both pedestrians and transit users win. And for the intersections where can’t remove the stop signs for the buses then it is simply not worthwhile to degrade the pedestrian environment for at the most a couple of seconds time saving for the buses. The SFMTA should concentrate on the cost effective improvements including Transit Preferential signaling where there are already signals, bus bulbs, stop consolidation, dedicated right of ways, turn restrictions, etc. Those are solutions that are known to be effective and worthwhile. Note that those improvements are expected to save MINUTES, yet the SFMTA is concentrating on spending funding on the signals which at most improve travel time by a few SECONDS.

    At the last hearing the consensus was universally to make these other cost effective improvements to the 5 and the 71 and to NOT replace the 4-way stops with signals.

    Wow, you read this far? I’m impressed.

  • hp2ena

    What about traffic signals that are off most of the time (i.e. they only flash red lights for all four-ways), and only turn solid red (for crossing traffic) and solid green (for the street with a bus route) when a bus triggers the priority signal sensor? Regular traffic would continue to stop, but buses would be able to breeze through the intersection.

  • EastBayer

    Great points. I’d add pedestrian delay to that equation as well. A pedestrian can move fairly continuously through a street with stop signs (realistically, bikes too) but signals will cause significant delay at most blocks. Really bad for joggers.

    And last, although I am often skeptical of giving too much a voice to residents about engineering issues, shouldn’t concerns about livability be worth something?

    I think traffic engineers like signals bevause they make it so easy to (over-)engineer, control and predict.

  • Thanks for the perspective, Mike. I’ll look into this.

  • Mario Tanev

    Your argument about fudging numbers is frankly itself fudging numbers and sounds exactly like NIMBY arguments I hear (what would be SFMTA’s motivation to fudge numbers in this case?). The graphic makes it clear that the cost of a stop sign should factor in the acceleration and deceleration and 18 seconds sounds more probable than just a few.

    That’s not to say that there aren’t other solutions (like hp2ena’s), and I’ve never heard a good argument from SFMTA against it, but there probably is one. Status quo is not a reasonable solution, just because you don’t like the particular solution proposed.

  • David Marcus

    According to this article ( http://www.dmgov.org/Departments/Engineering/PDF/FAQ15_Unwarranted_Traffic_Control_Device_.pdf ), the delay to left turns is ~7 seconds (for cars). I imagine it should be the same for going straight. 18 seconds does seem a bit high. That would imply an extra minute of delay every 3-4 blocks. I would guess a 5L not making stops can probably cover 3 blocks in a minute, including the time in motion.

  • djconnel

    I find it ironic that when signals are malfunctioning traffic tends to proceed more smoothly than when they are functional. On the peninsula, where cycles are much longer, typically with generous left-phases in addition to straight phases, vehicles drive quickly on the ubiquitous multi-lane roads, then spend minutes at a time waiting at red lights.
    The key in this case is a traffic light takes right-of-way from pedestrians, who otherwise cross as soon as they establish their intent.

  • Mario Tanev

    David, with all due respect these are pseudo-science arguments I expect from climate change deniers and NIMBYs.

    First, a bus is much heavier than a car, it has a longer stopping distance so you have to slow down earlier.

    Then, what you quoted was according to the Highway Capacity Manual, which has been used by traffic engineers for years to implement wrongheaded policy. The number you quote is implied to be the delay ACCEPTED (just like whether drivers accept a speed limit) by drivers for a turn. I just don’t know how relevant this figure is.

    Then comes your computation about total running times if there is a stop sign on every block. But there isn’t a stop sign on every block, so you have to diligently do the math on the entire run to see if it’s really consistent. But even then, things are not as simple. The 18 second delay is assuming free flow otherwise. But you don’t have free flow even if you eliminate stop signs – you have turning vehicles, you sometimes have a red light and so on. While a stop sign causes 18 second delays (assuming the number is correct), it doesn’t mean it can all be recovered.

    This was a superficial reading on your part, whereas if SFMTA has done it’s job, they took more care with yours. But this precisely is what NIMBYs do, they argue with experts and question parts of reports that they disagree with with arguments that on the surface sound plausible. You can question any piece of data and you will sometimes truly find a hole. If SFMTA is unprofessional, you will find many holes, even on items you agree with (data that justifies more bike or bus lanes), so you’re being a hypocrite. If SFMTA is professional, they keep their errors in check, but as a NIMBY you yourself are biased on a particular number so you will only argue with a particular number, and are statistically biased to unearth an error in that number.

    This is not to say we shouldn’t be skeptical if a number appears way off and is hard or convoluted to measure . Recently, someone told me that 40% of arterials in SF have bike lanes. I was very skeptical, and asked for the data. The data showed a few issues with that number: some arterial classifications (or lack of classification) were dubious, some of them didn’t have bike lanes for the entire duration, and for some, as a percentage of all lanes bike lanes were a bigger part (not really arterials) than others (really arterials), so if a weighting was applied the number surely wouldn’t be 40%. The person admitted that the number could be wrong so we decided not to argue further. But measuring average delay of stop signs is not that hard and perhaps SFMTA has done it empirically and I just don’t see a reason to doubt that number unless it really tells the wrong story for you. But if you want to approach this properly, just ask SFMTA for the data on that 18 seconds, how did they arrive at it, rather than quote the second law of thermodynamics.

  • Mario Tanev

    A lot of pedestrian and transit advocates like to give Europe as an example that has done all of the following well:

    1. Transit is prioritized
    2. Pedestrians are safe
    3. The pedestrian experience is encouraged.

    It is very clear (or at least, it’s in SFMTA’s interest, so that’s why they’re pushing it), that transit will be helped by this change, so check.

    It seems some people are arguing with the pedestrian safety aspect, but SFMTA data seems to show that pedestrian-involved accidents will be reduced (perhaps by discouraging pedestrians due to delay, so see below).

    The third argument is the one people are passionate about. But look at Europe – there are almost no stop signs there and while smaller streets have no traffic devices at all, major street gets free flow, the equivalent of a two way stop sign. That means that crossing a major streets requires a bit of a walk. Yet, a lot of Europe does indeed have an inviting and safe pedestrian experience. So we’re overplaying the importance of stop signs in this.

    Last, I should mention that I did question with my supervisor and SFMTA the third signal phase installation at 30th and Dolores. My observation was that pedestrians used to cross less on red before, that J ran smoother and so on. Couple that with the death that happened there recently so I probed the SFMTA again. It seems their data does show fewer left-turn accidents, so I just don’t want to quibble with that.

    The phenomenon of what’s happening here is really a very familiar phenomenon in the workplace: micromanagement. The populace is the ultimate manager (many levels removed), but also the one with the minimum visibility and very selective interest in what the report has done. A good manager sets up a good structure (by voting for establishment of SFMTA per say), asks occasional questions (say, how did you arrive at the 18 seconds number) to get better informed, but when a report says they’ve done their job, the manager needs to trust that until evidence to the contrary arrives or fire the report if the manager doesn’t trust them with the basic stuff.

  • p_chazz

    There is no intersection of Fulton and Haight. The streets run parallel. Did you mean Laguna and Haight?

  • Flubert

    Mario, it’s true that Europe doesn’t have 2-way and 4-way stops. That’s a peculiarly American thing as far as I know.

    But they have roundabouts and traffic circles in place of 4-way stops. And “Give Way” signs in place of 2-way stops.

    The advantage of both of those mechanisms is that, when traffic is light, you don’t have to stop at all. Easily the most frustrating thing about a Stop sign is when visibility is good and there is no cross-traffic, but you still have to stop and start.

    Wherever possible, let the traffic flow. I’d support giving traffic circles another shot.

  • Flubert

    The key to that traffic flow in the suburbs is phased signals. So while you might sit for two minutes at a red, and it’s annoying, when you do get a green and yu maintain a reasonable speed, you catch a whole string of green signals and can make good progress.

    SF only does that on a few major through roads like Pine/Bush, Fell/Oak, Franklin/Gough etc. And of course that cannot work on every street by definition.

  • David Marcus

    Hey, for the record I support the SFMTA’s plans! I was just trying to track down a second source for the 18 second figure.

  • Michael Smith

    I mistyped. I indeed meant Laguna and Haight as the intersection with the 4-way stop that the SFMTA is proposing replacing with a signal.

  • Michael Smith

    I like this idea. Unfortunately the SFMTA will argue that it is unusual and people often don’t do well with unusual intersections. Some might not appreciate the difference between a flashing red, to be treated the same way as a stop sign, and a solid red, which means that one must stop because cross traffic has a green light. If someone misinterprets the light it could cause a serious crash.

    But I think it is certainly worth studying.

  • Michael Smith

    Their report unfortunately shows that the SFMTA has not yet done their job.

    First, the 18 second number. It was originally 30. After mentioning to them that when I measured delays it was much less than that they clarified that the 30 seconds was for streetcars (which we aren’t talking about here!). For buses it was only 20. Well, great. Measuring things and questioning their numbers moved things in the right direction. After more prodding they have lowered the number to 18 seconds. But that number is a guestimate made on bad assumptions. It assumes that the buses will also be slowed by cars, which is often not the case at the intersections in question. That brings it down to about 10 seconds, which is still a bit higher than what I measured. Plus as I previously mentioned, their numbers don’t take into account the time that buses will have to stop for a traffic signal, which will still be quite significant.

    Key thing here is that they should measure the real world situation at the intersections in question and also take into account when buses get slowed by signals, even Transit Preferential ones. They have not done so.

    Their safety numbers are even worse. They looked at a few intersections where 4-way stops were replaced by a signals. Not only were these intersections not directly comparable but they had a shockingly small amount of data.. They could have easily compared many comparable intersections. Of the pedestrians who are killed or seriously injured, what percentage to you think are at 4-way stops instead of traffic signal intersections? Yes, almost none. Pedestrian advocate we have long advocated for stop signs instead of traffic signals for relatively lower volume roads because the numbers show they are safer.

  • Michael Smith

    Note that in my original comment I have proposed that the SFMTA follow other more cost effective and beneficial solutions that are not the status quo.

    And as I said in my other comment, the 18 second infographic is not based on measurements. Saying that you think it is “more probable” is not as useful as going at and measuring the delays.

  • gneiss

    One option the city could pursue, of course, is to restrict private vehicles along segments of the transit routes for the 6, 71, or 5 by creating forced turns before congested blocks.This has been a very effective strategy on Market, and would make the streets where the buses run safer by de-prioritizing private vehicles.Additionally, these forced turns could be restricted to certain hours, say between 8-10 AM and 4-6 PM to accommodate the need to increase transit speed during commuter hours.

    To those that say this would require too much traffic enforcement, posting cameras like those used on the Bay Bridge or Golden Gate bridge on these blocks could be used for automatic enforcement rather than posting police officers.

    The fact that SFMTA is only considering traffic signals rather than tabling the intersections and changing to 2 way stop signs, or having blinking reds which change to green for transit vehicles or pedestrian access shows how the long hand of the more conservative DPW influences traffic planners in the city. They are also the same people who like nothing better than to waste street space by adding expensive to maintain center medians in our roadways rather than parking protected bike lanes. Remember, each traffic signal costs the city some $400,000 to install. Then, there is the annual $10,000 per year to maintain them. All moneys of which flow into DPW’s coffers.

  • Flubert

    The problem with restricting private vehicles from certain blocks around there is that it merely pushes that extra traffic onto the surrounding streets. If you look at the east-west streets around this area, you have bus routes along Turk,Fulton, Hayes and Haight. So you’d push traffic onto the only other four east-west streets there: Page, Golden Gate, McAllister and Grove, which would expect a doubling of traffic. You also have more un-necessary traffic on the north-south streets that connect them and, since the Panhandle blocks many of them, you increase the mess on the already busy Masonic and Stanyan.

    Those east-west streets are two-way (except Fell and Oak) and so not wide enough to enable a bus lane or, in the case of Turk which is wide enough, it’s not necessary.

    Finally you would have to allow lots of exceptions for access: cabs, delivery trucks, senior transport etc., so enforcement by camera isn’t a good option as you would be sending out many invalid tickets.

  • gneiss

    You could make the same argument for Market Street. We’ve just pushed private traffic off Market onto other surrounding streets. How is this any different? After all, just like the traffic pushed off Market, where there is the Fulsom and Howard couplet, we have Oak and Fell, two blocks away from Haight or Hayes Street.

    As for lots of spurious tickets being sent, the enforcement would include someone looking at the violation before it got sent to a person, which would allow SFMTA to see if it was a taxi or not. And what I’m talking about it through traffic – people could still get onto the street, just not from the particular direction they wanted to, just like disallowing left turns during certain hours on Masonic.

  • Dark Soul

    Did SFMTA say Stoplight reduce delays when traffic lights present??? ..i dont think so.. Stop sign only requres few second to stop then to go.

    Drivers/Buses can stop at a STOPSIGN,which can take around 3-10 seconds compare to Traffic Light can take up to 60 seconds or more. (Guessing the numbers)

  • Thomas

    Traffic lights are faster if buses have priority, i.e. they automatically request and get a green light. With this system, buses don’t have to stop at intersections anymore because the stoplight is always green for them.

    It’s not only the stop that counts, but also the time lost at decelerating and accelerating, which can easily amount to SFMTA’s 18 seconds figure.

  • hp2ena

    I forgot to clarify that this treatment already exists at Powell and California. It’s a four-way flashing-red (stop) intersection until a cable car comes along. When a cable car comes along, the intersecting traffic is stopped (solid red) whereas traffic running parallel to it continues to operate (although I think to my understanding when a cable car crosses the intersection, traffic all four ways are held).

  • Michael Smith

    I didn’t know that. Thanks for the info.

  • Under California law it is illegal to cross the street midblock between intersections unless the intersections are signalized. That is this state’s legal definition of jaywalking. So I would say that this approach fails to encourage the pedestrian experience.

  • djconnel

    I’d like to see who spends more time waiting / light / car: Sunnyvale or San Francisco? I’m still guessing Sunnyvale, even with attempts at synchronization, which obviously can’t work for all directions at every intersection. The other issue with Peninsula lights is the ubiquity of left-turn phases, which reduce the fraction of time available for any given vehicle direction, not to mention pedestrians.

  • p_chazz

    It is different. Market Street is a commercial street east of 10th Street, while Fulton, McAllister and Haight are residential for much of their length. What if you lived on a street with a forced right? You would have to make a right and three lefts to get home. Seems like this would create more problems than it solves..

  • 94110

    I think you mean the opposite of what you typed there. Assuming that, I’m also concerned about making pedestrian movements illegal by installing signals.

    One obvious compromise would be installing signals at every other intersection so the legality of crossing would not be changed.

    Checking the law before contradicting you, I just learned the California definition of “alley” (which effects the presence of unmarked crosswalks) mentions San Francisco by name and allows the city leeway in designating roads that would not usually qualify as alleys. Wonder if there is a list?

    Reference: http://archive.desertsun.com/assets/pdf/J11680111215.PDF

    Also, I’m pretty concerned about transit priority signals on Mission. The pedestrian countdown runs down, people stop crossing, then notice the light is still green and resume crossing. This seems more dangerous.

  • Thanks for catching that, whoops.

    I have long wondered about the S.F. exemption for alleys, since it is rarely indicated on street signs. I know Balmy Alley is an alley because of its fame for murals, but for the most part we’re left guessing. (Of course, there’s widespread ignorance about pedestrian right-of-way in unmarked crosswalks anyhow.)

  • jonobate

    I’m with the SFMTA on this one. For traffic flow, signals are better than stop signs for intersections with higher traffic volume, as a queue of cars can clear the intersection in one light phase without having to keep stopping and starting. On this particular road, keeping the traffic moving means keeping the 5 moving.

    With traffic lights, you also have the option to program them to give a longer green to the “major” street (McAllister) to prioritize east-west traffic; you can’t do that with four-way stops. This is in addition to any gains that might be made through transit signal priority.

    It is the case that traffic lights are slightly worse for pedestrians than four-way stops, as pedestrians don’t queue up at an intersection and cross one at a time; they bunch up at the curb and all cross as soon as it’s safe. So traffic lights just keep pedestrians waiting longer with no real benefit to them.

    However, in this case I think the need to speed up the 5 is greater than the need to minimize pedestrian delay. I understand that you might feel differently if you live in the neighborhood, but I also think that the people who travel through the area on the 5 should also have their needs taken into account.

  • Gezellig

    Europe is not all roundabouts 😉 In fact, in a country with pretty good transit and hands-down the best bike infrastructure–the Netherlands–non-roundabout intersections are still far more common. Two-way stop signs do exist, though they’re smartly avoided when possible on bike routes. In fact, one use case for them there is indeed to force drivers to give way to through bike traffic when shark’s teeth/yield signs have unfortunately been disobeyed by too many drivers:





    To reduce car speeding on through routes there are various strategies, including:

    –frequent speed bumps
    –police enforcement
    –visual narrowing of roadway
    –one-way roads alternating flow every block (except for transit/bikes)
    –periodic dead-ends for cars–even on relatively main streets–but never for transit/bikes. Example below:

    To get from the NW corner of the screenshot (Jodenbreestraat) through the Muiderstraat and on to Plantage Middenlaan in a car is not straightforward, but transit/bikes/pedestrians can easily get through both directions:


  • BBnet3000

    Why would someone jog on the side of the street opposite the park when they can jog on the park side and avoid cross-streets almost altogether?

  • Sprague

    My two cents worth is these are great improvements to speed up Muni service and to improve pedestrian safety. Motorists tend to respect red lights whereas stop signs are often treated as mere suggestions. And since the 5 Fulton is an important crosstown route, the rather minor changes proposed here certainly serve the common good. Signalized intersections have benefits for cyclists, too, by allowing for a more consistent speed of travel. These proposed changes to McAllister seem to come up short at the Van Ness, Polk, and Larkin intersections – where McAllister will continue to be extremely wide. Those busy intersections sorely need pedestrian bulbouts on most or all of their corners.

  • Wanderer

    18 seconds strikes me as a low end estimate for delay to a bus at a stop sign. It assumes that there’s no queue at the stop sign. If there’s any queue the delay can get much longer. This means that service will tend to be worse during peak hour, exactly when it needs to be better. It also helps account for Muni’s terrible on time performance, though there are numerous contributors to that.

    MTA could prioritize the bus route by moving to 2 way stop signs. But a signal is a more pedestrian-friendly solution than a 2 way stop sign. The pedestrian gets some protection crossing the main street, at the cost of waiting a little while for the light to turn red.

    Muni’s a comprehensive, frequent system, but it definitely needs to speed up routes like the 5.

  • Dave Moore

    The bus must end up stopped by a light sometimes, right? I know they will have some control over holding the green, but still there will be occasions when the bus has to wait. I don’t see any estimate that shows negative impact on the bus travel time because of this. I just see the 18 seconds per stop sign gained.

  • Let’s just discuss it, and then discuss the discussion and discuss that and bring it to a panel to discuss the discussion of what was discussed. After that we’ll discuss.

  • Courtney

    Wow, I didn’t know that but this is a huge concern for me as I live at the corner of Steiner and McAllister. I cross both streets mid-block at least daily. I already didn’t want a light at the corner, as I think it changes the residential feel of the block; but I especially don’t want it if it is going to more or less criminalize walking out my front door!

  • Sharon P

    The tone of this piece suggests that the author supports the Muni proposal to add traffic lights along McAllister to “speed up” the Limited 5 Fulton. The Muni proposal does NOT speed up the Limited but slows it down and substantially decreases pedestrian safety.

    According to the California Manual on Uniform Traffic Control Devices for Streets and Highways Part 4 ( http://www.dot.ca.gov/hq/traffops/engineering/mutcd/ca_mutcd2014.htmt) there are 8 warrants that should be met to install traffic signals at an intersection. They involve volume oftraffic (vehicle and pedestrian), school crossing, coordinated signal system,crash experience, and roadway network. If we look at the intersection at McAllister and Baker and Muni’s own data, none of the warrants are met. This is a low congestion intersection. Muni bases their decision to install a traffic signal based on the perceived time savings of a traffic signal over a stop sign. This
    is not an accurate prediction.

    Muni claims they can save 18 seconds per intersection each way. Currently the 5L takes 8 seconds to pass through the stop sign at the intersection. If we deduct the 2 seconds it takes to pass through the intersection (which would exist for both stop sign and traffic signal scenarios), the stop sign delay is 6 seconds. I have measured the actual delay for over 500 Limited buses. Muni has NOT measured the delay, they are assuming a delay of 18 seconds based on their perception of the sum of the times to decelerate, stop and accelerate. Assuming a 20 second traffic signal cycle (Muni’s proposal) and a 50/50 probability of the signal being red, the delay each way caused by a traffic signal is 10 seconds, or 4 seconds more than the current stop sign system.

    Muni predicts that their transit signal priority system will save 1.5 minutes each way by holding the green light for approaching buses.
    This assumes that each of the 5 additional signals proposed for the 5 Fulton
    will save 18 seconds. They claim that this GPS-based signal priority system has saved them 5 minutes over 63 intersections on Mission Street which computes to only 4.8 seconds savings per intersection. Why such a difference? According to an actual study performed to determine the efficacy of signal priority systems, such as the one used in this Mission Street pilot study, these systems are NOT justified for low congestion routes. http://ntl.bts.gov/lib/jpodocs/repts_te/14299_files/chapter_3.htm).
    This report also states that a 3- second savings per intersection can be expected when properly applied. So even with a signal priority system the
    buses will be slowed down by 1 second. Muni should have conducted a pilot of this system on a low congestion route to firm up their numbers before proposing the expenditure of millions of dollars over a faulty prediction and the strong objection of the neighbors.

    The second major reason to oppose this action is safety. Let’s look at McAllister at Baker again. At 1750 McAllister Street sits a
    12- story, high –rise building that houses over 100 elderly citizens. In the neighborhood, there are also many half-way houses with disabled residents who cross the streets many times each day. I observe these folks with their helpers, walkers, canes and wheelchairs crossing the street every day to catch the bus. In order to make this a safe crossing for these folks, the traffic light would need to be held for at least 25 seconds, adding another 5 second delay. This would make the delay due to traffic signals greater by 6 seconds. The MUCSD states that adding a traffic signal where unjustified will degrade safety by increasing accidents due to red- light running. People are more likely to run a light when they are forced to wait an excessive amount of time with no traffic to justify such a wait. Currently Muni data show only 1 accident in this intersection over the past 8 years but adding an unwarranted traffic signal will certainly change that. Currently the disabled or their helpers are
    able to look up the street in both directions and usually see no cars (because the street traffic is very low), enter the intersection and if a car does approach it will be stopped by the stop sign allowing the pedestrian safe passage. Don’t change this. Many times even 25 seconds is not enough time
    for these folks to cross.

    In conclusion, in their desire to show change at any cost, Muni has made a poor decision with respect to efficiency and safety. . It is up to the people and the SFMTA Board ofDirectors to stop them.

  • Dark Soul

    Even with TSP for Traffic Lights,it does not speed the muni route up..Instead it slows it down most of the time. Stop signs are more reliable and safer than traffic lights,which can sometimes get power outage that traffic light get turned off.


Tomorrow: Hearing on Traffic Signals to Speed Muni on Haight, McAllister

On the agenda [PDF] for tomorrow’s SFMTA public engineering hearing are proposals to speed up Muni lines with transit-priority traffic signals and bus bulb-outs along Haight and McAllister Streets. These types of changes are central to the Muni Transit Effectiveness Project, but some residents have voiced concerns about replacing stop signs with traffic signals and […]

McAllister Street Set to Get Two Traffic Circles Instead of Signals

McAllister Street, a popular bike route where SFMTA’s Muni Forward planners want to speed up the 5-Fulton, would have stop signs replaced by traffic circles at two intersections under the agency’s latest proposal. Under the plan [PDF], which must be approved by the SFMTA Board of Directors, McAllister would become the first street to get traffic circles […]