MTA to increase G train service but won’t give free transfers

The MTA will shift G train stopping positions to better align with platform staircases.

The MTA will shift G train stopping positions to better align with platform staircases.

After conducting a full line review of the G train this year, the MTA will increase nine additional trains to the much-maligned ride during the afternoon rush hour but will not allow for free out-of-system transfers. The agency will also implement a series of changes to reduce the impact of the so-called G Train Sprint and will work to improve operations to ensure that wait times along the IND Crosstown line are consistent. Despite these victories, though, the G train’s chicken-and-egg problem remains.

“The G line is a vital connection for customers in fast-growing parts of our service area, and this review will be an important tool for making both short-term improvements and long-term additions to our service. We are pleased to be able to take these steps to improve service for all of our G train customers,” MTA Chairman and CEO Thomas F. Prendergast said in a statement.

In conducting the review — which is now available online — the MTA found facts that furthered their own talking points but also discovered ways to improve service. Notably, the agency continues to maintain that, despite significant growth in ridership over the past decade, overall ridership lags behind the rest of the system. This is my aforementioned chicken-and-egg problem. Ridership has grown despite infrequent service in relatively poorly maintained stations, and ridership hasn’t grown more because of the lack of connections to other lines and the long headways.

Yet, despite ridership that the MTA says falls within their load guidelines, the agency will add nine additional trains to the line between 3 p.m. and 9 p.m. each day, reducing headways from 10 minutes to a more manageable eight. So then why is the MTA adding this service? The report says doing so will help with “service irregularities resulting from the merge with the F.” Reducing headways to eight minutes will allow the G to better interact with the F — which has peak headways of four minutes. As an added benefit, the increased service will reduce travel times those riders who have to make transfers. The service increase is contingent upon additional funding totaling $700,000.

Additionally, to improve the customer experience, the MTA examined the G train in the context of its longer platforms. The builders of the IND Crosstown line were ambitious in their construction efforts as they built out giant stations for stations will low ridership. To save costs, the MTA runs shorter trains, and regular G train riders often have to sprint down long platforms to each trains. At nearly every station, the MTA will adjust where the G train stops to better allow for access to the more popular exits and entrances. The appendix materials detail the changes, and for many, the G Train Sprint will become a thing of the past. By the end of 2013, the MTA will also post signage indicating where along the platforms G trains will stop.

Finally, Transit will make some operational tweaks to better distribute passengers as well. Train doors will remain open for longer at Court Sq., and platform benches will be moved to align with train doors. Scheduled holds at Hoyt-Schermerhorn will allow for regular service too.

But despite these increases, the report is notable for what it rejects as much as for what it promises. The MTA will not lengthen G trains. Calling 600-foot trains a “misallocation of resources,” Transit says longer trains would lead to less frequent service. Transit will not increase A.M. peak hour service frequency or off-peak trains either. The biggest rejection though concerns out-of-system transfers.

Transit advocates have long asked for free transfers between the G at Broadway and the J/M at Lorimer St. and the G at Fulton St. and the rest of the subway system at Atlantic Ave./Barclays Center. Noting that such a walking transfer would take nearly seven minutes, the MTA cites operational and revenue concerns in rejecting the transfer:

Given current ridership patterns, an estimated 4,000 trips per weekday would be eligible for a MetroCard walking transfer, slightly under half of which use pay-per-ride MetroCards, which would result in an estimated annual revenue loss of $1.34 million with restrictions to reduce multiple trips for the price of one entry, and $7 million without these restrictions. Given the density of subway stations in Downtown Brooklyn and the heavy commercial activity in the area, restrictions would not clearly distinguish between transfers that improve connectivity and entries for a second trip that use the walking transfer to avoid paying a second fare. A significant portion of these “transfers” would likely be stop-overs by riders who travel to the area and then, within the MetroCard two-hour time limit, re-enter the subway at Fulton St once they are done with their activities.

The same analysis holds for Lorimer where MTA estimates revenue loss ranging from $770,000 to $1.1 million. On the flip side, though, the MTA has a budget of $13.8 billion, and the report pays short shrift to the fact that nearly 50 percent of riders — those without unlimited cards — are charged two fares to make these transfers. The debate remains ongoing.

Politicians and members of the Riders Alliance who have pushed the MTA to make improvements heralded the results of the study. “Now G train riders will be en route to much-needed relief that may one day lead to the G meaning great,” Senator Daniel Squadron said. “These recommendations will allow the G to keep pace with skyrocketing growth in Brooklyn and Queens – and make the notorious G Train Sprint a thing of the past. Increased frequency, shorter wait times, and better communication will go a long way for many riders.”

“Over time, the recommendations outlined in the MTA’s review of the G Line will greatly impact the quality of service for thousands of daily commuters. I applaud the MTA for a thorough assessment of the G and for putting a plan in action that will almost immediately alleviate some of the difficulties riders had pointed out,” Senator Martin Dilan said.

50 Responses to “MTA to increase G train service but won’t give free transfers”

  1. Eric Brasure says:

    Still digesting the report, but I am SO happy that they are aligning the southbound stopping position with the northbound stopping position at Metropolitan Avenue. I’ve been saying they should do that for years. That small change alone should help with crowding.

    • Bgriff says:

      It will be interesting to see how they execute this. The problem with the G train stopping positions is that with 4 75-foot cars, it is the same length as a 5-car train of 60-foot cars. The stopping markers are all based on 60-foot car lengths, so the G train stops at the 6-car marker during weekday service (and the OPTO marker during overnight and weekend service). The 6-car marker at Metropolitan theoretically can’t move up since should a 6-car train of 60-foot cars ever wander through that station, the back end of the train would not be in the station. So presumably the MTA will introduce some kind of new signage–maybe a 5-car marker? Or just a special marker that targets the G train specifically?

      (end dork rant)

  2. Seth R. says:

    Much as it makes sense to be able to transfer from the G to these other trains, I think most G riders would prefer that that kind of money was just applied to more frequent service.

    Anyone have an idea how much it would cost to combine Lorimer and Hewes street?

  3. Bolwerk says:

    I think they actually do the G right for the most part. There isn’t a chicken and egg problem. They actually have this many riders, and I don’t think more frequent service or longer trains will ever be able to override the fact that the G serves Brooklyn-LIC and not Manhattan. Still, we should all wish we had more G Train-like services around the parts of the city that don’t have anything better than a bus.

    So, the price of the Broadway Line-Crosstown transfer is estimated at $335/rider/year. But are they assuming they wouldn’t induce more ridership with such a transfer? And, if they do, I get the impression those who exploit the two hour limit would not be very disruptive, especially given that the G and J/M/Z both seem to have extra capacity.

    • Alon Levy says:

      In addition to induced ridership, there’s also a matter of ridership diverted from the L. Less L crowding means less need for rush-hour service, which is a large cost saver given the high marginal cost of peak service.

  4. llqbtt says:

    Funny, but I don’t see how people not being able to squeeze on their G meets light ridership level standards. What I am referring to specifically is the northbound weekday rush through Greenpoint from about 8:25 to 9:05. During this period, service is woefully inadequate, and at 1 point, there is a 9 minute headway between Gs with the following Gs then running every 8 minutes. This service is far too infrequent given the current 4 car train set make up of the G.

  5. John says:

    I rode the G regularly when I was a student at Pratt, many years ago. Back then, it wasn’t crowded. Even when you waited 25 min. for a train, it would show up, four cars, and had sitting room.

    I always wondered why, if they were never going to run more than four cars, why the TA wouldn’t just put railings up on the platforms, which would easily designate where the trains would stop.

    • Bgriff says:

      Other trains are sometimes rerouted through the G line when there is a problem elsewhere–in particular if there are delays on the F in Manhattan the MTA will often send a few (or more than a few) F trains through the G line between Brooklyn and Queens to keep things moving. Those full-length trains still need the full platform.

  6. Demetria says:

    Was the possibility of routing the G to Forest Hills when the M train is not running even considered? A three seat ride from Astoria/Queens Boulevard local stop on weekends/late nights certainly stymies potential ridership growth. Many times I’ve routed through Manhattan just to avoid having to wait R/E/G, or N/7/G. Seems ridiculous considering the proximity of Greenpoint/Williamsburg to Astoria/Woodside.

    • Dan says:

      From what I’ve read, the demand really isn’t there for it. People on QBL generally want Manhattan and the relative handful who need Greenpoint/Williamsburg can transfer at Court Square (or from the G to the 7 or QBL if they need Queens coming from Brooklyn).

      If the MTA does anything going forward, it will most likely be increased R service on weekends or maybe the E running local in Queens.

  7. Jerrold says:

    Transit terminology frequently has a weird-sounding nature. Sometimes it’s the acronyms, but this time it’s a word.
    What does JUSTIFY a train mean?

  8. Patrick says:

    If the current way out-of-system transfers are done makes it undesirable to expand, maybe it is time the MTA should overhaul the way they handle subway-subway out-of-system transfers.

    Perhaps they can devise some sort of out of system transfer box and place them by all the exit turnstiles at the Broadway and Hewes Street stations. If you want to take advantage of the out-of-system transfer between those two stations you would have to swipe your MetroCard at the box before exiting the fare control area. After swiping your MetroCard at the Transfer Box you would then have 18 minutes to walk to the other station (either Broadway or Hewes Street) and swipe your MetroCard at the turnstiles there.

    If you did not first swipe your Metro-Card at the transfer box before leaving the fare control area at the first station or you did not get to the second station within the allotted 18 minutes you will be charged another fare. That way the out-of-system transfer can’t really be taken advantage of by anybody other than the people exiting station A and directly entering station B.

    If a concept like this works then it can be expanded to other station pairs that might benefit from out-of-system transfers.

    Just an idea,
    ~Patrick @ The LIRR Today

    • Eric Brasure says:

      This is intensely cynical of me, but the MTA has zero interest or intention of ever letting the free out-of-system transfer (not replacing any service pattern) happen. It would drive demand with no increase in revenue and would drive up the cost of operating the train lines in question.

    • Bgriff says:

      I agree that this kind of system would be helpful — the current system is too open to abuse, since there’s not even the slightest logic applied as to when you get a free transfer. For example you can take the 6 from 77th St to 59th St, get off, run errands, and board an uptown 6 at 59th St, and receive a free transfer — even though that free transfer is intended for people coming from the F train, and it is fairly obvious that no one starting that journey at 77th St would have been coming from a transfer off the F train.

      But if you read the report carefully, the MTA seems to be making a more nuanced point: they are saying, basically, “our policy is that we only institute out-of-system transfers because a service change takes away something that had already existed.” (F move from 53 to 63 took away transfer to 6 at 53/Lex; G cutback to Court Sq took away transfer to 7 at Roosevelt.) They also say that they would need to revisit this policy entirely if they were going to establish these G train transfers. So basically, they’re afraid to do this because it could open the floodgates to other similar requests, like between the 3 and L in East New York, and who knows where else.

      • Alon Levy says:

        The problem with these ostensibly neutral policies is that sometimes the policy itself is not optimal. But once a policy is set, the MTA justifies the policy in terms of itself, instead of openly discussing what works best for the riders.

        For a different MTA example, consider crowding guidelines. They are currently based on average crowding at the most crowded point of each train. But it’s not the only way to do things. In one direction, since trains have variable amounts of crowding, it’s worth looking into guidelines that are based on maximum crowding instead of average crowding. In the other direction, having to stand for two minutes is not the same as having to stand for twenty, so it’s worth looking into guidelines that are based on crowding over the worst n minutes (say, n = 5) rather than worst instant).

        • Bgriff says:

          I agree that the MTA’s rationale for making policies may be flawed (to put it mildly). But, if one is going to argue with them on this specific point, it is important to address the reasoning they give for their decision: they did not say that they dislike out of system transfers on principle, or that either of these two possible transfers are unacceptable on their merits (though they hint that they may feel that way about Fulton-Atlantic/Barclays), but instead merely point out that thus far they’ve only instituted OOSTs under a specific set of circumstances (essentially as a last resort, not a customer convenience), and a new generalized principle for when to offer OOSTs would have to be established if they were going to do one of the proposed OOSTs on the G. So any further discussion of OOSTs on the G would have to start with such a principle.

          • Alon Levy says:

            The problem is that they (or at least their defenders here, since I’ve never talked with anyone in an official capacity) refuse to entertain other principles, such as “provide unlimited OOSTs within a 2-hour period” or “provide OOSTs when two lines cross without an in-system transfer station nearby.” With the latter principle, we would not need an OOST between 53rd/7th and either the 1 or the N/Q/R since there are nearby in-system transfers at Columbus Circle, Times Square, and Herald Square, but we would need one between the G and the J/M/Z, as well as between Atlantic-Pacific and Fulton.

            • Andrew says:

              Other principles sound great, but they also cost money.

              The two proposed OOST’s involving the G would alone result in a revenue loss of over $2 million annually (much of that due to “accidental” transfers), and that’s without considering other OOST’s elsewhere.

        • Andrew says:

          In one direction, since trains have variable amounts of crowding, it’s worth looking into guidelines that are based on maximum crowding instead of average crowding.

          In any given hour, the single most crowded train is presumably the one that was delayed a bit (or whose leader was running a bit ahead of schedule).

          Are you seriously proposing basing service levels on a brief blip in service rather than on the overall hour?

    • Henry says:

      The problem with that is that any such transfer box could cause crowding due to queues, since a free transfer would attract a lot of people. It probably wouldn’t work within the existing footprint of certain stations – Lex/59th is a station that is pretty much at capacity.

      The smartcard system that they are trying to develop (or whatever it is at this point, because CTA’s similar system is an absolute horror) should theoretically be able to handle exit taps, making it possible to regulate transfers more strictly. This is how Shanghai manages its free out of system transfers.

      • Patrick says:

        The perhaps a system could be developed where you could mount readers on the exit halves of turnstiles at the affected stations…if you want to take advantage of the transfer swipe you MetroCard before going through the turnstile, if you don’t just keep walking through.

        ~ Patrick @ The LIRR Today

    • pete says:

      The cubic turnstyles at St George on SI, that are part identical to NYC subway turnstyles, are swipe on entry AND exit. The parts are made. It is just political. The only thing missing would be a bypass button if you didn’t enter the system with a MC (senior/disabled half fare coupon for cash entry). Then again, the MTA could just say, forget to swipe on exit, too bad.

      • Andrew says:

        There wouldn’t be any need to swipe on exit for anybody not making an out-of-system transfer.

        But the issue isn’t that there isn’t a way to turn a turnstile around and put the swipe head on the inside. It’s that there’s no way to encode a MetroCard with a special extra transfer for a brief period.

    • Andrew says:

      Devise? It’s a good idea, but any such device would have to first be designed and built by Cubic, since it doesn’t currently exist. And the MetroCard system is on its way out, so would it really be worth the expense?

      I do hope that the smartcard system is being designed with this capability in mind, so that the policy can be revisited in a few years.

  9. Bgriff says:

    Also, I noticed recently some new signage at Clinton-Washington station giving somewhat more guidance as to where the train will stop, though I can’t say it was great. The signs were quite easy to miss if you weren’t looking for them. Hopefully the MTA comes up with an improved variant before the full roll-out.

    • Patrick says:

      Perhaps they can take a page out of the commuter railroad’s book and paint a big thick green line where the trains will stop. At most of the stations (especially along the LIRR Babylon Branch) the platforms have a green line painted right next to the yellow one and a sign saying “stand by the green line during these times” to alert you where to stand.

      • SEAN says:

        Sene the green line in Minneola. MNR doesn’t do that, rather the train stops close to the front end of the platform.

        Would it make more sence to extend the G to either Jamaica or Coney Island? Easy way to solve the transfer problem. I know, I know, it makes no sence but at least I tried.

        • Dan says:

          Can’t really extend the G to Coney Island because there’s no room for it.

          For Jamaica, it depends on whether or not G trains could switch to the express tracks at Forest Hills without disrupting the E and F, and when the R did that to run local to/from Jamaica in the past it didn’t really fare that well so we don’t know if the ridership levels need that.

          • Andrew says:

            The demand for the G south of Church and north of Court Square is quite small.

            The only reason the G ran north of Court Square until 2001 is that Queens Blvd. would have otherwise had only one local.

    • Kai B says:

      Yes, Clinton-Washington and Bedford-Nostrand have the new “test of signage” already. The signs are the same size as the station name signs, are white-on-black, with some sort outline of a train car and the text “trains stop here”. At first I thought it was an unauthorized project of stickers. They are not too noticeable. Hopefully the full rollout brings along better ideas.

  10. Kai B says:

    Interesting that was no mention of a review of entrances and exits and their closed/open status.

    Also, fascinating that Greenpoint Avenue will becoming ADA compliant in the next six years (first I’ve read about this). The only question becomes, where will someone who needs an ADA compliant station take the train to? The only option the G right now is Church Avenue.

    • Bgriff says:

      Yeah, Greenpoint Ave being on the 100 key stations for ADA accessibility list is very weird. In theory those passengers could transfer to A/C trains at Hoyt-Schermerhorn and get to a few more places, but not many, and the train-to-platform gaps at H-S are not especially designed for wheelchair transfers. Maybe the key stations list was drawn up back when the G still ran on Queens Blvd, at which point there would have been more potential destination points?

      BTW, the remaining list of planned ADA conversions is here, which does not give any further insight as to why Greenpoint is special:

    • Doug says:

      There are also stations where ‘same platform transfers’ are possible:
      (I haven’t confirmed how feasible this is in reality however.)

      Rush-hour map of ADA-accessible stations, including those ‘same platform transfer’ stations:

      Suddenly, the system feels a lot smaller.

  11. Eric Brasure says:

    And of course the day after this report was released, the G blew its schedule this morning and I waited 15 minutes for a southbound train.

  12. John Doe says:

    The solution is so simple! Just add extra cars! Gee whiz it’s common sense….pesky humans.

    • Andrew says:

      Pages 18-19:

      As noted above, G trains are currently 300 feet long, consisting of four 75-foot-long R68/R68A cars, and are always shorter than the platforms where they stop. This can lead to customers waiting at the wrong section of the platform and to extended station dwell times, as train operators and conductors sometimes delay the train’s departure as a courtesy to customers coming from platforms areas outside of the train’s stopping position. These inconveniences have led to occasional requests for full length, 600-foot G trains. However, longer trains would require more cars, cost more to operate, and most likely trigger reduced service frequency per guidelines, with adverse impacts on transfer flows. Longer trains would also be loaded far below guideline capacity at all times of day.

      The R68 cars are linked into four-car units, so operation of five-, six-, or seven-car trains is not feasible. Expanding to 600-foot trains would require the acquisition of new cars at an approximate cost of $10 million per train (or $130 million for the full G fleet), and cars for longer G trains are not included in the capital program. Once purchased, each car also increases operating costs due to maintenance and power consumption.

      Longer trains run at current frequencies would lead to a considerable reduction in loads carried per car. If trains were lengthened to eight cars at the current frequency, the average car at the morning peak load point of Clinton-Washington Avs would hold only 64 riders, less than a seated load and only 44% of guideline load. By NYC Transit guidelines, this light load would trigger frequency reductions in the morning rush hour from 9 tph to 6 tph, for an average headway of 10 minutes or a 33% reduction in service.

      Given that increasing the length of G trains to 600 feet at current ridership levels would be a misallocation of NYC Transit resources and could lead to reduced service frequency and crowded transfers, other means were examined to address concerns associated with short trains, as discussed in the Next Steps section.

      Doesn’t look like common sense to me.

  13. John Doe says:

    Andrew – Do you believe everything you read?? That’s what the MTA wants you to think. They should run a pilot program with extra cars and let the people decide.

    • Andrew says:

      Which of the statements in those four paragraphs do you dispute, and how would a pilot program help to resolve them?

      What do you want the people to decide, whether they prefer 9 trains per hour or 6 trains per hour? I can tell you the answer right now, with no need for a pilot program.

  14. John Doe says:

    Andrew – Add the extra cars AND increase frequency the trains are run, it’s as simple as that.

  15. John Doe says:

    Eric – We’ve spent over one TRILLION dollars on wars in distant lands since 2001, I’m sure we can find the money to purchase new rolling stock.

  16. John Doe says:

    I’d rather spend money on rebuilding our infrastructure than on countless wars, wouldn’t you agree?

Leave a Reply

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>