Muni Switchbacks Stink, But What’s the Real Root of the Problem?

Photo: ##http://www.flickr.com/photos/sf2006/5938257382/##Confetti/Flickr##

The San Francisco Civil Grand Jury released a report [PDF] today blasting Muni’s regular practice of dumping riders and turning vehicles around early.

Known as a “switchback,” the practice is used by Muni management as way to alleviate delays when buses and trains are bunched together by redirecting a vehicle to another point in the system where it’s more needed. The practice was scrutinized by the Board of Supervisors last spring, and SFMTA Transit Director John Haley says the agency has made progress in reducing them and warning riders of them in advance. Members of the SF Transit Riders Union say focusing on switchbacks distracts from the root of Muni’s problems, like getting stuck in traffic and poorly-maintained vehicles, which make the measure necessary in the first place.

The Grand Jury said Muni officials’ use of the practice “shows a callous disregard for the welfare of riders,” claiming that few other major transit systems practice it regularly except in cases of breakdowns and emergencies.

At a press conference called by the SFMTA today, officials presented a document (summary [PDF], detailed [PDF]) responding to the Grand Jury’s claims, saying that the report ignores evidence and defending the use of switchbacks when necessary to alleviate problems.

“They don’t suggest an alternative,” said Haley, adding that Muni intends to propose scheduled switchbacks on some lines within the next six months, similar to regular practices on many other systems, including BART. Still, he said he doesn’t think unscheduled switchbacks “will ever be at zero.”

“If you look at the unevenness of where the demand is,” said SFMTA Director Ed Reiskin, “and people getting on and off the buses, it just doesn’t make sense to run every bus to the end of every line on every run.”

Ben Kaufman of the SF Transit Riders Union said switchbacks are just one symptom of Muni’s greater structural problems, and that a holistic approach is needed to improve the system. “The only way to minimize the amount of switchbacks is through a network of lines that don’t have to deal with external factors, like traffic congestion, getting stopped at stop signs and red lights,” he said. “That’s what we should be focusing on, not condemning the MTA for making switchbacks.”

  • Anonymous

    The SF Civil Grand Jury has a history of getting involved in issues it doesn’t understand and recommending nonsensical solutions.  They even got involved in that neighborhood schools push a few years ago, again with little supporting evidence and no credible conclusions.  I wish they could make more useful contributions, but they seem to be constrained by a lack of data (eg Muni operations records) and a lack of expertise.  

  • Anonymous

    Instead of switchbacks why don’t they express buses through to a certain point? That way some of the riders can get where they need to go and the others can wait for the next bus (and don’t have as far to travel).

  • jwb

    Switchback and midpoint relief at the twin middle fingers from Muni to riders.  It’s totally disingenuous to compare switchbacks with BART’s short runs.  When you get on the train, it says right on the sign that the destination is MONTGOMERY, in all capital letters, as plain as it can be.  

  • ch

     Speaking as a Muni inspector who supervises bus service, on the diesel bus lines we do that. It’s called “deadheading” – sending a bus out of service, non-stop up the line either to put it back on time or to fill a gap in service. But the switchback issue involves the LRV/streetcar system and the trolley coach system. Neither can accommodate express or deadhead moves – you can’t pass another streetcar or trolley coach. So, if there’s a disruption that delays a few vehicles, you have a gap in service. That gap will ricochet back and forth along the line. To fill it, you switch a car or coach back. It’s annoying and disruptive to the unfortunate passengers on that particular vehicle, but it plugs a hole in the line that would otherwise delay many more passengers, possibly for hours. The only way to seriously reduce the number of switchbacks is to reduce the interaction between Muni vehicles and unpredictable surface traffic. That means taking street space from autos. While I fully support doing that, I’m not very optimistic that we can…

  • david vartanoff

    legitimate short routes/turnbacks have been employed in transit for over a century to better utilise equipment and operators.  These are properly signed so that a boarding rider knows how far the car/bus will go.   In contrast, Muni has a dismal history of surprise short turns which have often dumped a full 2 car train in the fog with the empty promise of another train right behind.   At one point the switch irons used to shift the rails were confiscated.   All of this is driven by a stubborn refusal by Muni to properly manage evening rush hour service.   Aside from the chronic driver shortage (deliberate understaffing to make the fraudulent budget numbers) Muni does little or nothing to maintain the scheduled sequence of outbound trains from Embarcadero.  If the inbound L is delayed, potential L riders crowd the platforms asmultiples of other routes cycle through some nearly empty.  If Muni chose to dynamically reassign trains to maintain service to each route, there would be less crowding and fewer riders delayed..   Once the errant cars come in, they too should be reassigned to rebalance the fleet.   This has been recommended by various rider advocacy groups for over a decade with Muni  always complaining that work rules make such plans impossible.  No less than three union contract cycles have passed with no effort by Muni to change the alleged rules.       

  • david vartanoff

    @ch,  IINM the 40ft  ETI trolley coaches have battery/EPU equipment designed to allow them to briefly operate off wire.  This should be enough to “pass up”  other coaches a couple times.  As to the LRVs,  the bogus turnbacks would be less anger inducing if it were the phantom “next car” which was emptied onto the leader and turned rather than leaving riders standing around on the pavement waiting.   As I said in another comment dynaqmic resequencing outbound from Embarcadero would go a long way.

  • jwb

    I was on a crush-loaded 21-Hayes line outbound once which short-turned and disgorged everyone at Fillmore.  The next bus, while actually right behind us, was also crush-loaded and didn’t even stop to pick up anyone.  I walked home.  Dozens of foreign tourists in some kind of group looked confused.  There is no excuse for turning a fully loaded vehicle.

  • Anonymous

     ch: In a word, *wrong*.  In three words: wrong, wrong, wrong.  Did I mention that you’re wrong?  You’re wrong.

    Unscheduled short runs are not only disruptive to the riders on the vehicle being turned back, they’re extremely disruptive to people who live on the parts of the line being denied service.  Unscheduled short runs don’t plug the gaps, they simply shift them to the portion of the lines you’d prefer to ignore.  Think of it this way.  If there’s so little need for service at the ends of the lines, why do we have an N Express bus whose sole purpose is to serve Ocean Beach?

    Likewise I realize that it’s en vogue to lay blame exclusively on private vehicles, but guess what?  Private vehicles make up the minority of the delays on the metro lines.  When you, the supervisor, let *non communicating trains* into the subway you create gigantic gaps in service that are tempting to fill by gutting surface service.  Trains that won’t/can’t operate in auto mode (of which there are plenty, check out all of the delays listed on the SFMTA’s twitter account) are limited to lower speeds and increased minimum following distances.  In addition, subsequent trains must also follow at a greater minimum distance… delaying many more passengers, possibly for hours.

    I watched earlier this year as a supervisor turned three of five outbound Ls in a row back at Sunset.  There were about ten people waiting for the inbound L at the Sunset stop across the intersection.  Not one of the four MTA employees it took to turn felt compelled to walk across the intersection, shout, wave, etc at the prospective passengers.  Their excuse?  There was an inbound L due in a few minutes.  Think about that for a second.  There’s already inbound service, but the proper thing to do was to dump most (60%) of the passengers trying to go towards the end of the line.  Right-o.  Thing is, if you were to look at the automatic people counter data, you’d see miniscule ridership that far west… but you’d be missing the fact that most people just gave up at Sunset or 22nd and started walking.

    MUNI’s core routes are built around the use of feeder lines.  This idea of gutting service outside of a few key corridors is absolutely detrimental to public transit in San Francisco as a whole.  Yes, the downtown corridors are far busier than the outer reaches.  But how do you get to the downtown corridors in the first place?  If you live nearby you’ve got plenty of alternate routes within stumbling distance.  If you live on Taraval, east of 19th you’re within walking distance of the subway.  If you live west of Sunset, you’ve got far fewer options and are more likely to drive downtown and create this “unpredictable surface traffic” that’s purportedly so problematic for MUNI.

    That person who might be stuck for hours at Montgomery waiting for an outbound L is far more likely to disembark somewhere before Forest Hill.  Instead of spending two hours lamenting the lack of Ls, they could take a K, M, or F (J, N, or any number of buses if they’re not going all the way to Castro).  Hell, they could even walk that distance without any large hills.  That person you just gave the finger to that was trying to get to 44th and Taraval?  You’ve left them with zero options beyond walking.  You can chant the TEP mantra of “starve the feeder routes” all you’d like, that doesn’t actually make it constructive.

    It’s far easier to stick it to the riders, than to try and extract any sort of meaningful reform of things like work rules.  We get that.  But short runs as a go-to solution hurt riders.  Full stop.  That you, Ed, and Tom can only justify the status quo goes to show that MTA employees do not ride or depend on the metro routes for anything other than a paycheck.  As David points out, dynamic outbound reassignment would go a LONG way towards mitigating this supposed need for unscheduled short runs.

  • djconnel

    The real problem is the institutional arrogance which this reveals.  If there’s two characteristics which drive people nuts, it’s arrogance and hypocrisy, and MUNI reeks of each. MUNI in its present form will never be a solution for the city, only a black hole for precious resources.

  • nt

    You are kidding, right? It looks like popular opinion (based on rider’s
    evidence) is against you. Look at all of the other comments on this
    article, and comments to the sfgate article. Switchbacks happen all the
    time. I question MUNI’s data on switchbacks, just like I question their
    “fudging” of on-time performance.

  • Anonymous

     @zippy_monster:disqus : It does seem like switchbacks can help reduce bunching in a line by evening out service, even in a situation where three of five cars are turned back (that’s 60%, or a little more than half). But if Muni is turning back cars to serve an overcrowded downtown core, it sounds like they may be doing something wrong. More shuttles would make sense in that case—I was happy to find an unscheduled one yesterday after getting off at the downtown Caltrain station. And as others have pointed out, it makes the turnback more predictable (this one was signed to turn back at Church St., though not everyone knew how to read the sign). Automatic service is also more reliable.

    As a rider of the J, one of the lines with the worst on-time record, I’m all for minimizing surface traffic. Dedicated lanes seem to keep the N moving pretty zippily along the end of the line. But I dunno if Church street could realistically close to vehicular traffic. A man can dream …

    Undergrounding of lines would be even awesomer, but an even bigger fantasy.

  • Anonymous

     @fea5a07e4c2ec6da7c1533cf1db75048:disqus : It also just occurred to me you can absolutely express the head car in a streetcar line when there is bunching. That one doesn’t have to turn back. Most often when I see bunching on the J there are only two cars bunched together.

  • Mario

    I think there is a problem of communication between both sides.

    For example, Ed Reiskin was quoted as saying “The notion that we should run every bus and every train to the end of the line on every run is not a good practice, let alone a best practice”. This is not what the report contends however. A scheduled announced switchback is called a short-run and short runs are one of the recommendations of the civil grand jury. So it doesn’t seem like the civil grand jury objects to short runs. Similarly, the civil grand jury doesn’t seem to object to switchbacks used when there is a breakdown as it only suggests to “eliminate switchbacks 
    except in cases of equipment breakdowns, accidents, and unavoidable emergencies”.So really, the issue at hand are switchbacks that are the result of regular bunching. The report contends that even if they are announced properly, they are still unacceptable. Muni contends that it tries to avoid them by holding in headways and deadheading.So the question is, do other agencies who implement holding in headways and deadheading still rely on unscheduled switchbacks in non-breakdown situations? It’s unclear from either party’s research.

  • lasertag

    I wish the MUNI management could experience the frustration of being kicked off the L or N while it’s raining. Please publish their email addresses so I can send them a message congratulating them on pissing off a very high number of people on a daily basis. MUNI = FAIL

  • Anonymous

    I have no idea why Muni does not propose transit only lanes for the J from Duboce to 18th St. The street is wide enough here that there will still be a lane for auto traffic an no (or very little) removal of parking, and it would also benefit the 22.

  • Anonymous

    Well that’s easy.  First.Last@sfmta.com.  John Haley’s been known to respond to e-mails from time to time.

  • Anonymous

    tinyurl.com/cyk9xz2

ALSO ON STREETSBLOG

Supervisor Wiener Calls Hearings to Assess the Cost of Muni Delays

|
Just how bad is Muni? And is it getting any better? Supervisor Scott Wiener has called for monthly reports from the SF Municipal Transportation Agency and the City Controller’s Office to tally up the true cost of transit delays and track progress on Muni’s reliability. The regular reports to the Land Use and Economic Development Committee would […]

Today’s Headlines

|
SFMTA Looks to Simplify and Streamline the Residential Parking Permit Program (SF Examiner) Better Market Street Planners Aim to Bring “a Simplicity of Place and Movement” to the Corridor (SFGate) Muni’s Operator Shortage Continues to Drive “Shadow” Service Cuts (SFGate) SF Examiner: “Muni Has Made Big Inroads” on Reducing Switchbacks Beyond Chron: SF Civil Grand […]

What’s Next for the TEP?

|
Can the TEP transform Muni, dramatically improving its reliability and on-time performance? Bashing Muni, an agency that has been historically underfunded, is a San Francisco pastime. Riders are used to break downs, delays and general unreliability. Could all that be about to change? After an arduous process of gathering ridership data, community input, and considering […]