T to bring back late-night service in test next year

The Globe reports the state and corporate sponsors are kicking in the bucks to keep subways and 15 bus lines running until 3 a.m. on Friday and Saturday mornings in a pilot project.



    Free tagging: 


      Night Owl Failure

      The Night Owl was designed to fail and I don't think overtime for the drivers was the reason why. The service was horrible in every way. They replaced trains with busses, only the busses had strange bus stops which where confusing. The buses ran infrequently and almost never kept to schedule. The fares were much higher which made people less likely to use the service or become angry when finding out they needed to pay triple the normal fare. (Or not have enough cash.)

      Many of the people who used the service were drunk which only exacerbated the other problems. I'd rather have the drunks on a bus then in a car but I did feel bad for the drivers who seemed to be getting the worst part of horrible plan.

      These problems wouldn't have happened had they just kept the trains running on a normal schedule a few hours later.


      I remember

      Trying to pick one up in Government center around 2am once, and the driver was less than helpful in answering my questions of which line they were running and if they were stopping at the usual stops. I turned around and waited for a cab.

      At least you know the trains, schedule and stops and don't have to deal with people who hate everything about their job but the benefits and paycheck.


      They replaced trains with

      They replaced trains with busses, only the busses had strange bus stops which where confusing. The buses ran infrequently and almost never kept to schedule.

      Even when they replaced buses with buses, they fucked it up by not replacing the entire bus route. I used to live on the 77 bus route in Arlington Heights, but the night owl version only went as far as Arlington Center. So even though that route was "important" enough to warrant night owl "service", it wasn't important enough to actually cover the entire route. One more way the T helped turn me from a T rider to a car driver.


      I agree. Years ago, I

      I agree. Years ago, I remember talking with a guy who was part of a business group who was trying to work with the T to get late-night service. He said that the T had a reason why every suggestion the group had wouldn't work. Subway service? Nope, need to work on the tracks (work on the tracks my @ss, but whatever,) So, they said, post signs and the T can run the trains on the opposite direction's tracks late night when they need to work on a section of track or close a platform (this is done in NYC pretty frequently.) Nope, nope, nope.

      Sounds like this time there is the political will from up high to make this happen.

      I don't even remember how long the Night Owl Service lasted, but I never quite figured it out, and I never took it.


      "Many of the people who used the service were drunk which only exacerbated the other problems."

      Indeed. I never took the night owl service, but part of the reason why is that I found the whole schedule/stop layout very confusing. And I figure, if I was confused by it when I was sober, this must have been REALLY confusing for someone who was drunk and not able to think clearly.

      i think it's funny how the

      i think it's funny how the article is hardcore pushing an agenda that this is ONLY being done for the 'people who live and work in the innovation district, because they're different.' i guess the rest of us who have been yelling and screaming about this for years don't actually matter.
      -someone who actually used the night owl bus


      very silly indeed...this

      very silly indeed...this struck me as out of place right off the bat, though I do wonder if there's any validity to growing "innovation" economy lending to the critical mass of getting this done. At the very least, it seems anecdotal...

      Shirley Leung

      That's because it was written by Shirl the Hurl, who never saw a small child that she didn't want to sell the organs of. She is one of the worse writers on the Globe's staff and is normally confined to the Business pages where her bullshit more or less fits (or at least people don't notice because it's all about the money, so who cares if it's cellphone apps or baby spleens). I just find it ironic that the entire city MUST bend to the will of the Innovative Master Race to which we owe all our existence and to which we must pay tribute, when it seems clear that they didn't even turn out in substantial numbers to vote in this last election (otherwise we'd be talking about the other white meat as Mayor elect).

      Hopefully they can make the late night service work, because it's needed. Just the same, how come they magically found $20 million for carting around puke-spewing 20-somethings on the weekends, but for the elderly and handicapped it was "pay up bitch"? Y'know, I'm beginning to think that politicians aren't being honest when they talk about whether or not we have the money for different programs/projects....


      Great question...

      Great question...

      Check this out:

      What about the Green Line?
      Unfortunately, the Green Line does not have the same type of train tracking technology as other lines. As Green Line riders ourselves, we feel your pain! Our teams are currently working to design a tracking system but expect it to be a few years before we can make something public.

      Who knows when, but I at least appreciate the acknowledgement...


      Needs line of sight to the heavens, so it doesn't work in the tunnels.

      My guess is the volume of trolleys on a number of close lines is also a problem for radio tracking and triangulation in the tunnels themselves. I'd image (guessing) it's much easier with the lines that don't split, and where trains run further apart and at a more regular schedule.

      Doesn't a real time information

      system require what you're tracking to run on a predictable schedule to be truly effective? With all the short-turns, headway adjustments, and other "musical trains" gimmicks the Green Line dispatchers are famous for, it seems to me that providing real time information is impractical at this point.

      Serious question. I've

      Serious question. I've noticed in many of your MBTA related posts, you always make a point about short turns, schedule adjustments etc. Are you implying that these moves are made at the random whim of dispatchers without some sort of operational philosophy or strategy behind them?

      re:Roadman's posts

      I think he has a right to wonder. For instance, the Oak Grove platform fixer-upper project appears to have been finished for several weeks; there does not appear to be anyone working on the platform at your right as you are facing it, which is the second platform to be fixed. Yet they are still running one train in and out of Oak Grove. Why? Who knows.

      Maybe if the MBTA were ever to respond to a question it would solve a lot of life's mysteries.

      Well... sure!

      Are you implying that these moves are made at the random whim of dispatchers without some sort of operational philosophy or strategy behind them?

      Haven't you ever been to Brigham Circle? Of course they have no idea what they're doing!

      Sidenote: With real-time tracking, the MBTA could actually make better decisions regarding short turns. Also, if they finally turned that switch at Brigham into an automatic (they use it enough!) then maybe short turns on the E would be a lot faster and more efficient.

      Based on 30+ years

      of riding the Green Line, plus my experiences monitoring (until recently) the Green Line dispatcher's frequency, it became all too apparent to me some time ago that the dispatchers are making many of their decisions, perhaps not "on a whim", but in a very inconsistent and upredictable manner. Recent examples - Why else would one day you hold both a Riverside and Cleveland Circle train at Park Street, yet on another day at the same time you hold the Riverside train but let the Cleveland Circle train pass it on the opposite track. Why else would one afternoon you run three trains through to Government Center, yet on another afternoon at the same time, with the same three trains, you re-route one or two of them into Park Street.

      The truth is that the Green Line is scheduled and dispatched as four separate lines that happen to share part of the same track (although the T would never admit this in public for obvious reasons). The fact they have a signal system and operating rules set up on purpose to deliberatly restrict capacity, plus the unwillingness of management to actually put enough cars in service, only excerbates this philosophy.

      And if I sound cycnical, it's because I am. Why? Becaus I'm sick and tired of a management that constantly ignores the real problems facing the system (overcrowding being the biggest one), yet seems to have no problem with spending time with web sites and smartphone apps and "let's redesign a map that's perfectly good" contests. Almost as bad is a media and public that blindly accepts such "feel good" non-improvements in "customer service" that do nothing to provide the customers with better service.


      So I'll ask you the same

      So I'll ask you the same question I asked you in a previous comment thread. You believe that the problem with the green line is there aren't enough trains in the central subway?

      And what would be their motivation for "deliberately" restricting capacity?

      If they had an extra $150 million or more laying around, I could fault them for using antiquated signals, but we both know that isn't the case.

      Not "motivation"

      Try "excuses" that intentionally reduce capacity.

      As in "safety dictates we force all trains to arbitrarily come to a full stop at certain signals, even though the operator is running at or below the speed limit and has a clear view of a particular section of track."
      As in "safety dictates we only allow one train to berth on a platform at a time (Park Street westbound being the sole exception) even though most platforms are long enough to accommodate at least a pair of two-car trains."
      As in "because insuring good schedule numbers is more important than serving customers, we hold loaded westbound E or C trains outside of Government Center so we can turn empty B or D trains (which delays the loaded trains trying to berth at Government Center), and then hold said B or D trains at their berth at Park Street - further delaying the E or C trains."
      As in "because insuring good schedule numbers is more important than serving customers, we short turn overcrowded North Station trains at Government Center and/or overcrowded Lechmere trains at North Station."
      As in "because remotely throwing switches based on the power draw of the approaching streetcar (power throws the switch one way, power with braking throws the switch the opposite way) is considered "old fashioned", even though it worked very well for several decades until they "upgraded" to the Boeing LRVs, so we'll install an expensive overbuilt "AVI" system that's far less reliable.

      Need I go on?

      So I'll ask again what is

      So I'll ask again what is their motivation for intentionally reducing capacity?

      And again, you believe there are too few trains in the central subway?

      Frankly, taken in a vacuum, none of the scenarios you describe really signify anything. In order to make a judgement, I would need to see all the information available to a dispatcher at that moment. Frankly, if I had a platform full of people at Gov't Center and a full C and E train, I would certainly put the empty b and d trains in front to reduce boardings on the C & E.

      The real answer - which the MBTA

      will never admit to in public - is simple. Fewer trains means fewer operators required - makes budget look better. Reduced capacity is not so much a motivator for this action, but rather a consequence of it. And, as I've stated in other posts, the truly logical thing to do would be to run all eastbound service to Lechmere, and dispatch westbound service from there. Instead, they persist in this nonsense of "in from Riverside (or wherever), so it must go back out to Riverside." When trains start to bunch, that's what encourages the short turning. With this "In from X, out to Y (with Y being the next destination in the schedule)" scheme, bunching doesn't matter because you have a self correcting system.

      And, with respect, there's a term for forcing people off crowded trains and onto other ones that are equally crowded - and become more so with the influx of additional passengers. It's called reduction in capacity.

      You think the MBTA doesn't

      You think the MBTA doesn't freely admit that fewer people operating trains = lesser costs? Why do you think the Red and Orange Line went to one man operations over the past 3 years? I think the T has been pretty open about the fact that they have no money and that service often reflects this.
      And frankly, do you really think the new countdown signs and maps are coming in lieu of improvements to service/capacity? You realize that the countdown signs were done mostly with IT staff and cost less than $1 million? The maps were done in a contest. I'm not sure it will cost the T anything. It isn't like they were pulling green line operators and dispatchers off trains to do these things.

      And why do you keep dodging my question that you think there are too few trains in the central subway??

      The Green Line signal system

      was rebuilt after the 1987 flooding, but only "in-kind". So it's not the fault of the people who managed the project, but the fault of the higher-ups at the MBTA who decided doing things on the cheap was more important than taking advantage of the opportunity to make improvements to the signal system was.

      Just another example of what happens when you let accountants dictate how you run and maintain a transportation system.

      Helpful, But Still Not Late Enough

      With the last trains leaving the city at 2:30 am, the extended service will help some people, but it's still not late enough for many others. When bars or restaurants close at 2:00 am, it leaves only thirty minutes to make connections and get to your final station. Employees who have to close up will have no chance of catching a train.



      This is what most concerned me about the plan when I first read it. It's a D.C. WMATA-style late night service, but at least in D.C. the last trains depart Metro Center just after 3 a.m. on the weekends, not 2:30 a.m.

      Having said that, it's likely that the T will continue their policy of guaranteed connection on the last train, so if you make it onboard then you'll get to your final station.

      But for later hours workers, this still won't help. But, the T already runs special very early morning bus routes to help workers get to the airport and other early jobs. I don't see why that can't be extended and expanded to help other night shift workers get home. It would require study and some funding. Guess where you can find funding?

      In an interview with me last month, Walsh envisioned a late-night Boston with downtown bars serving until 2:30 a.m. and dancing until 3:30 a.m.

      Charge those bars an annual fee to stay open later! Use that money to pay for cleanup, emergency services and late night transportation. Done and done.


      Envious Of Montreal

      Along with their subway system, the overnight bus service in Montreal is something to envy. When the Metro shuts down each night for maintenance, frequently running circuits of busses overlap their routes and those of important daytime busses. Most important destinations have service 24 hours a day; if not door-to-door, at least door-to-walking distance.

      BTW - If you're a fan of subway systems, this website by Matt McLaughlin is the nicest "worship site" about any transit system I've ever seen. Check out how nice his interactive system map and station navigation works. I especially like the architectural descriptions and station ratings!

      Montreal, SF, and us

      I wouldn't mind seeing the T emulate STM in that way. But one distinction: Montreal has a grid street network, which makes it a lot easier to replace trains with buses and not confuse riders. Same with SF, for the most part (except for the Twin Peaks tunnel, but you'll notice they only run one of the routes over the mountain at night).

      The more I looked at comparable bus and MBTA subway routes, the more I noticed just how awkward it is. Leave aside the rapid transit routes which cut across the city: even the Green Line routes, which follow streets, are a pain because of the crazy one way streets near Back Bay, around Boston Common and further into the city. Especially if you plan to run all the night buses to a central meeting point such as Gov't Center, or Haymarket (as was done in the old days).

      In terms of getting people to understand and generate ridership, just extending hours is the easiest. But that can't be done 24/7, so some more thinking is required to create reasonable routes for late night workers that don't just fail like the Night Owl did.


      There exists a network of

      There exists a network of early morning bus routes that are centered on Haymarket. They connect outlying neighborhoods with downtown and currently only make one round trip per day. It would be great to see this network expanded: additional radial routes with some crosstown lines, a full overnight span of service (12:30A to 5:15A), and seven day service. Operating an all bus network could certainly reduce operating costs, plus a full seven day schedule avoids/minimizes the problem of premium pay for bus operators. (In short, the premium pay for operators for 2001-2005 era Night Owl was not just because the runs were scheduled very late, but also because it was just for two nights per week.)

      Certainly, to make that work the T must have good branding. The early morning routes have distinct route numbers (191, 192, etc.) but the destination signs on the buses are based on the routes they parallel (i.e. "39 HAYMARKET"). I've noticed a few newcomers getting confused by this. Many of the downtown stops also lack signage (who knew you can catch a bus to JP a little before 5:00a on Chauncy Street?), so there'd have to be a small bit of capital invested into that as well.

      That would be nice

      I've noticed that, for example, the 57 bus route is drawn up as Watertown Yard-Haymarket in the NextBus API. Curiously, the lines on the map show that the early morning version of the bus goes through the Downtown Crossing pedestrianized streets. I haven't ridden it so I don't know if that's actually the case. I also don't know what the impact would be of having all buses tracing all four remaining Green Line branches, plus the 57, going through there at regular intervals. And even with that particular route choice, it's still pretty awkward, crossing from St James to Newbury around Copley. And that's just for the Green Line.

      Do you know more about this? Some historical sources I've consulted say that pre-1960 the buses met up at Haymarket, and for 2001 they met up at Gov't Center. It seems logical to run a low-frequency service like night buses with a "pulse" where all the buses connect at a central location periodically. However, that requires space for a dozen or more buses to meet up and stage, and then exit in waves. Presumably Haymarket and Gov't Center have enough curb space for that.

      Also, I'm curious about more info regarding the rules providing premium pay for certain types of shifts, if you have that available.

      Funny you ask...

      Haha, I use these routes with some regularity (they all happen to stop near places where I party late). The downtown routing for the 193 ("57") is Copley, Boylston, Essex, Chauncy, Arch, Milk, Pearl, Congress to Haymarket; returning via New Chardon, Cambridge Street, Scollay, Tremont, Stuart, Saint James, Huntington, et seq. The same is true for the 192 ("39"). The 191 ("15"), coming from Mattapan, Ashmont, and Dudley, follows the Silver Line up Washington, then picks up the downtown loop from Essex onwards. The 194 ("89/93") nominally takes a left from Causeway onto Canal to reach Haymarket - think about that. In my hypothetical all-night system, a timed transfer at Haymarket would be vital. The existing runtimes for current early AM routes are about the same, so coordinating them shouldn't be an issue. There's room for at least 6-7 buses to queue curbside at Haymarket (if you use the full block) and using the busway could be an option.

      The premium pay info I've only been able to cobble together here and there from different sources. Basically, I originally wanted to know why running the 171 and tacking it onto regular work isn't such a big deal as the Night Owl work.

      As for the Green Line bustitution: is late night/early morning ridership along the D enough to warrant an all night bus route? Or the C? That's a sincere question. If some late night revelers could be accommodated by an extra late-night trolley trip, just do that. That last trolley deadheading from Heath could maybe run an extra trip out via the D or C to cover the post-bar crowd. It's heading out to the yard anyway.

      Ok, this is good info.

      Ok, this is good info. I don't know why the online data has the 193 drawn through DTX. Must be an error. Do you know when the 193 leaves Haymarket? The schedule only shows when it leaves/arrives at Kenmore.

      Currently they only extend some of the bus lines at night, AFAIK, there is no subway replacement. If the program is expanded, then you would imagine some kind of subway-tracing bus routes would be added. And then they'd need more than 6-7 buses at Haymarket.

      I agree that specific ridership needs should be looked at late night hours, rather than assuming it's the same as daytime. But there's also equity issues to contend with. If you decide to save money by only night-busing some lines and not others, or to shorten the trips, then someone will complain.

      Probably the best way to handle drunken revelers is to do what they're doing and just extend hours. That won't work for the bar staff, though, since they have to close and that'll finish after the last train departs.

      So the question here is, how to provide transportation for late-night / early-morning workers? It seems that the equity issue hasn't raised its head so far with the early morning routes, so expanding that program might be a way to duck it. Or, maybe it's just because few people know about them?

      Anyway, lots of things to think about. And a pilot program to observe in the spring.


      The 193 departs Haymarket at 5:04A sharp, Monday through Saturday. No service on Sundays. As for the geographic coverage of a late-night/all-night bus network; put the service where the people are. Yeah, that may mean Braintree or Riverside don't get a bus. People will bitch, but that is what twitter is for. If there is enough potential demand in an area not being served, I think a trial route is certainly worth exploring. It will be great to see the (hopefully, very detailed) data from this 2014 pilot. Alas, it would've have been great to see how a revived Route 09N would play out in a post-yuppie Southie; but it isn't one of the key routes.

      On weekends, yes

      DC last call at 3 a.m.

      Of course, if the T ends service at 3 a.m. the last trains won't get to Riverside/Braintree until 3:30 or so. First trains inbound depart around 5, on Saturdays at least. Just run a train every half hour all night; you're not going to get any track work in anyway, and you can use operators on double-pay shifts through those early trips.


      Needs to adjust then, I'd imagine.

      Hell, welcome to the life of a train commuter. When ever I've needed to catch a train, or wanted to catch an express, I've found them to be scheduled to leave 5-15 min before I had any hope of reaching them. Some lines have really weird scheduled that seem to try to beat the commuters themselves, and stick them on the every stop local.