Aug
14

Thought Experiment: Closing 25 subway stations

By

Franklin Street on the 1 train on a particularly busy day. (Photo by flickr user NYCUrbanScape)

Dotted throughout the subway system are a series of shuttered subway stations that pop up now and then as ghosts. They are reminders of an era when subways were shorter and stations even more plentiful than they are today. Many people don’t know these stations exist, and as trains fly past 91st St. and Broadway or Park Ave. and 18th St., astute straphangers can espy glimpses of a past when the city closed subway stations.

Today, we are set in our ways with 468 subway stations, and in a few years, we’ll see that total creep up by four to 472. Closing a subway station, even temporarily or overnight, is pretty much a non-starter as community groups and politicians who otherwise look down on public transit or ignore its needs entirely throw a fuss whenever the MTA threatens service. Still, it’s probably possible to argue that, in some spots, we have too many subway stations.

Yesterday, in the comments to my post on the city’s ever-growing wishlist for transit maintenance prioritization and improvements, a few SAS regulars and I got to talking about station closures. The challenge: Identify 25 stations to close that would cause minimal disruption to those who rely on transit. Challenge accepted.

In essence, it’s a thought piece. With staffing levels on the decline, shuttering 25 stations wouldn’t make much of a dent where it counts in the MTA’s budget. Sure, the MTA wouldn’t have to rehab these stations every three or four decades or paint and clean them more regularly. But the jobs saved would add up to maybe a few million dollars while a lot of people would be both upset and angry at the MTA for failing to provide an adequate level of service.

Still, thought pieces can be fun, and as I toyed with the idea, a few obvious stations came to mind, many of them along the same few routes. First, Rector St. and Franklin St. on the 1 are ripe for closures. Neither provide transfers to other lines, and entrances to both are around 0.3 miles away from the next closet station. A five- or six-minute walk is hardly a major disruption. With fewer than 6000 riders per day, Franlkin is the 240th most popular stop in the system.

Other 1 train stops were prime culprits as well. Up in northern Manhattan, barely 2000 riders per day use 215th St., and 18th St., a relatively popular station, is not even a quarter of a mile away from 14th St. In Brooklyn, Cortelyou Road and Beverly Road are 0.2 miles away from each other, and in Queens, 21st St. on the G averages 1123 riders per weekday. While those folks would argue for their stations, it’s easy to see a rational argument against keeping these stops open. Subway rides would be faster, and expenditures on stations less.

Of course, we can’t just cut willy nilly. We can’t leave neighborhoods without subway stations, and we can’t discourage system-wide use. We need to maintain regular service while allowing for it to be fast, reliable and comprehensive at the same time. So here’s my own challenge should you choose to accept it: Help me develop those criteria for assessing stations. I’ve looked at exit coordinates, distance to the next nearest subway and ridership as potential indicators of a station’s value. What do you think? I’ll develop a full list with some explanations on the stations we pick for a post early next week.



215 Responses to “Thought Experiment: Closing 25 subway stations”

  1. Alex C says:

    Off the top of my head:
    Bay Parkway or Avenue I on the Culver. They’re close together (Avenue I is about a block away from 18 Ave) and neither gets much ridership. Bay Parkway is a ghost town there (one could say literally with the cemetery there). Similarly, the Courtelyou and Beverly on the Brighton line. It’s ridiculous that the two stations at a block apart. Should have just one station in the space between the two.

  2. Moses Gates says:

    I’ve already got my 25, but they’re heavily geared toward ridership numbers (they are all amongst the 50 least-used stations). Almost all are stops well out in the boroughs on branch lines.

    Some considerations:

    #1 – check ridership trends, not just 2011 ridership. A station becoming more and more used should probably stay, even if it’s still very low ridership. Also make sure the ridership isn’t an anomaly – some stations have low ridership in 2011 because they were closed for rehab most of the year.

    #2 – take into account future development. 21st street on the G, for instance, has very low ridership, but that area is slated for a lot of development. As such, eliminating the stop would be a bad choice.

    #3 – don’t eliminate transfer points, even with low ridership (Broad Channel is the best example).

    #4 – the tipping point for eliminating Z service is very close. If you eliminate as few as 2 or 3 stops on the J (maybe even one), it’ll make Z service not worth it. This could, I suppose, be a plus or minus.

    #5 – I say counting Aqueduct as a station is cheating.

    • Bill Reese says:

      The major development in Long Island City is on the Waterfront. The 21st Street G station may be close to the development, but the Vernon Jackson and Hunterspoint Avenue stops on the 7 are just as close, if not closer. They are also direct routes to Manhattan. It is only 2 blocks from 21 Street to the 7 and G stops at Court Square, and an extra block underground to the E/M stop at 23 St.

      There’s also the fact that 21st Street is one of the most poorly maintained stations in the entire system. When I suggested my girlfriend and I get a place in that neighborhood, she said she didn’t want to have to take the subway every day at a station that looked like “Silent Hill.”

    • Matthias says:

      These are good considerations and probably rule out most of the stations suggested here. Low-ridership represents an opportunity for growth (how many people rode the Flushing line when it still ran through farmland?). Running time saved would be minimal, and shuttering one station may simply increase crowding and dwell time at adjacent stations.

      I hadn’t considered the J/Z service before, but skip-stop only saves 5 minutes on the entire run. The 9 train was eliminated partly due to growing ridership at skipped stations; I wonder how long before riders at Z-only stops complain about 10-minute headways (to the J’s 5) and the J takes over the entire route.

      If the purpose of this exercise is to promote thought and discussion, it’s certainly succeeded in distracting me!

    • Henry says:

      Does J/Z skip-stop provide a meaningful reduction in travel times? I know the E is much faster than the J/Z if one wants to go from Jamaica to Lower Manhattan

      • ajedrez says:

        Actually, I think the J is still a little faster, even without skip-stop.

      • Andrew says:

        Per the schedule, the E takes 49-53 minutes from Jamaica Center to WTC, and the J/Z takes 44 minutes from Jamaica Center to Chambers. (For comparison, the non-skip-stop J takes 52 minutes.)

        The E is faster, but it has a much longer route, so it takes longer.

        The J/Z is probably more reliable and is definitely less crowded.

        • Bolwerk says:

          Whether you take the J or E from Jamaica probably should depend on whether you want to go to City Hall or the WTC or World Financial Center. Any time you’d otherwise save taking the J would be lost in the walk if you use it to get to the WFC.

  3. Henry says:

    I’m not sure if this counts as a complete closure, but I remember reading that at one point, the MTA was considering closing two stations on the J/Z and building one brand new one as a transfer to the G.

    I would also recommend having the 7 stop at Willets Point only during sporting events – as of right now, the only things surrounding that station are Corona Park and the auto shops at Willets Point that are supposed to be redeveloped at some point in time.

    • al says:

      The adjacent Bus and Subway depot has workers coming in by 7 train. They are a loud constituency. There is also a park and ride. Last but not least, Willets Point station, like 111th st for short run local trains, is also a turnaround point for short run express trains.

      • Matthias says:

        They can’t all be turned at Main St on 3 tracks? How do they turn them all at Times Sq on 2 tracks?

        • Quinn says:

          There’s more congestion when there are three tracks, strangely enough. A lot of trains usually end their runs at Willets Point and reverse into 111 St because they can do so without interfering with traffic unlike Main.

        • Henry says:

          I know that in the morning, one platform is designated for express services while the other is designated for locals. Having a train shuffle across three tracks to get to the furthest platform is inevitably going to cause a lot of conflicts.

        • Andrew says:

          A three-track interlocking takes up more space, which increases the time it takes for a train to cross the interlocking, which reduces capacity.

          (The trains that turn at Willets Point are all locals, and most people going to Flushing are on expresses, so I don’t think this is a big problem.)

      • Henry says:

        The short-run evening rush trains are annoying – at that point, mostly everyone on the train is heading to Main St, and all the people remaining on the train are inconvenienced by the short run.

        I know it’s useful from an operations standpoint, but usually short-runs shouldn’t end short of a major destination.

        • Bolwerk says:

          Given the crush loads at Flushing-Main, the 7 should probably just be extended from Flushing-Main. Maybe even extended in two directions (for two modern terminals), giving locals and expresses alike different passengers to collect in relatively low-density neighborhoods.

          • Henry says:

            That would be nice, but Main St was redone with a brand new entrance in the 90s that blocks any extension. It’s sort of stupid, because prior to the building of the entrance there were tail tracks that went past Main.

            • Bolwerk says:

              That’s something that shouldn’t be hard to fix.

              I’d be more worried about the NIMBYs.

            • Andrew says:

              Main Street had tail tracks? I don’t think that’s correct.

              In any case, the narrow staircases to the narrow platforms couldn’t handle the crowds. Building a new entrance past the end of the platforms was a no-brainer, especially for ADA purposes.

      • Henry says:

        Also, out of curiosity, is this park and ride heavily used? The surrounding areas are fairly dense, and quick & easy access to the mornings using the Grand Central, the LIE or the Van Wyck is never guaranteed…

    • Kevin Walsh says:

      What about people who want to use the park and don’t want to go in at the 111th street end?

  4. Caelestor says:

    In general, it’s better to close lightly used stations in the middle of the line to speed up the train than at the end of the line. Interestingly, very few IND stations need to be closed because they were spaced further apart.

    First, the obvious:
    – Intervale Ave (2/5), too close to Prospect Ave and Simpson St.
    – Zerega Ave (6), too close to Castle Hill Ave and E. Tremont Ave (though time savings would be minimal).
    – Bay Pkwy (F) is in the middle of a cemetery. Ave I and Ave U can easily replace it.
    – Cortelyou Road and Beverly Road should be merged together, but if one of them had to go, it’d be the former.
    – Liberty Ave is close to Broadway Junction, a far busier complex, and Van Siclen Ave.
    – Same with Alabama Ave.
    – Same with Bushwick Ave – Aberdeen St (L).

    Stations that can be closed to speed up service (but have decent ridership):
    – Norwood Ave (J).
    – Seneca Ave (M).
    – Hewes St (J/M).
    – Franklin St (1).
    – 18th St (1).
    – Either Elder Ave or Morrison Ave (6), probably the latter.
    – 69th St (7).
    – 39th Ave (N/Q).
    – Either President Street (2/5) or Nostrand Ave (3).
    – Either Knickerbocker Ave or Central Ave (M).
    – 20th Ave (N).
    – 20th Ave (D).

    Stations with low ridership that could be closed but would leave geographic gaps in coverage:
    – 215th St (1).
    – E. 143rd St – St. Mary’s St (6).
    – Cypress Hills.
    – 21st St (G).
    – Beach 105th St (A/S).

    Lastly, Aqueduct Racetrack just to make it an even 25.

    • Caelestor says:

      Well Aqueduct Racetrack isn’t ever open anyway, so add Neptune Ave (F) to the third category of stations.

    • Stratford says:

      “Cortelyou Road and Beverly Road should be merged together, but if one of them had to go, it’d be the former.”

      WHAT? You would remove the subway station on the commercial street with all of the shops, restaurants and bars, and high capacity housing stock… but you would keep open Beverly Road Station — the quiet residential street with single-family detached homes?!

      I get that this is a thought experiment, but just looking at the subway map isn’t how you choose which stations to close.

      • BrooklynBus says:

        But Beverley gets much more ridership. This idea of making it one station has been debated for years and rejected probably because if the stations were joined and both exits maintained, walks to the exit woud be increased for riders of both stations and only one minute would be saved which would be negligible.

        • Stratford says:

          Umm, BrooklynBus. But that’s not actually true. Daily ridership for Cortelyou is almost TWICE that of Beverly Road. 6,812 vs. 3,567 daily midweek riders.

          That being said, you’re right. Both exits would still have to be maintained. If, say, Beverly was closed Coretlyou would probably gain 2,000 more riders (the rest would go to Church). The single staircase bottleneck is already slow going at current levels.

        • cortelyou says:

          Beverly gets much more ridership? Really? Have you visited that station in the last 5 years? Cortelyou is on the avenue, it gets 3x the ridership of Beverly anytime of day or night. Check your facts.

          • MaximusNYC says:

            Cortelyou Road is an up-and-coming commercial street. It’s developing a “restaurant row” that is a destination for people all over the city. A couple of foodie friends of mine visiting from out of town just took the Q out to Cortelyou this weekend to meet me for lunch, instead of going somewhere in Manhattan or Park Slope; they read some of the rave reviews of Cortelyou Road restaurants that have appeared in the NY Times and elsewhere. There is also a farmers’ market on weekends. Closing this station would kneecap a vibrant and growing, but still fragile, district.

            As for the idea of combining stations: How would that work? A station in the middle of that residential block would be tricky to access; homeowners wouldn’t want entrances amid the houses (which would probably require eminent domain). Entrances at Beverley and Cortelyou would require some sort of long walkways to the station… which would probably be awfully desolate and creepy at night.

        • Andrew says:

          Average weekday ridership is 3,567 at Beverley vs. 6,812 at Cortelyou.

          Average weekend ridership is 3,393 at Beverley vs. 7,284 at Cortelyou.

          Annual ridership in 2011 was 1,095,502 at Beverley vs. 2,135,444 at Cortelyou.

          These numbers are all posted right here. Now do you understand why I don’t trust your numbers and analyses elsewhere?

    • Bronxmet says:

      I would not close Intervale Av (2/5) station for two reasons:

      1. There is a slow timer at Simpson St (thus not saving that much time).

      2. Residents around there would not be thrilled to walk up a hill to reach Simpson Street. They fought against the closure back in the late 1980s when a fire closed Intervale station.

    • Caelestor says:

      I messed up regarding Cortelyou and Beverly Roads. Closing Cortelyou Road would leave a smaller geographic gap, but Beverly Rd does have significantly less ridership. It’s a tricky situation.

      Also, note that I don’t advocate closing 25 stations. In fact, only the first 7 I mentioned are the ones that can be closed for little to no downside, and even Cortelyou/Beverly probably should be moved to the second category of stations. The next 12 are decently-frequented stations that can be closed if you really wanted to speed up service (but I personally don’t recommend closing them right now). The last few stations have low ridership, but I believe they serve important geographic locations and are close enough to the end of the line that their closures wouldn’t speed up travel times significantly. It’s also the reason why I didn’t list half the Rockaway stations.

    • Bronxite says:

      Soundview-Morrison is vital to the communities served by the BX 27.

    • ajedrez says:

      Liberty Avenue is over a 1/2 mile walk from Broadway Junction.

  5. David says:

    Never really understood the need for the 28th St station on the #6 line.

    Just 5 short blocks to either 33 st or 28 st.

    No idea what the numbers are.

    -David

  6. Alexjonlin says:

    How about the entire Rockaway Park Branch? It would leave that area without any subway service, but that level of ridership can honestly be much better served by buses.

  7. aestrivex says:

    Challenge accepted! Mostly I focus on stations with low ridership close to neighboring stations, with majorly inconveniently located stations a second priority. This makes my list somewhat different in emphasis than Caelestor’s above who nukes some very high ridership stations (like 39 Av on the N, that station certainly doesnt need to go). The 1 train really needs the cleanup in lower manhattan as it takes forever to trudge through stations every four blocks everywhere, so I make some exceptions there.

    Franklin St – 1
    18 St – 1
    28 St – 1 (the 6 can keep its stop because its the 6)
    215 St – 1
    145 St – 3
    Park Place – franklin av shuttle (too inconvenient and not even very far from the 2345 station. low ridership backs me up.)
    Neptune Av – F
    Avenue I – F
    Beach 105 St – A
    Beach 36 St – A
    aqueduct racetrack – A (this one doesnt count towards the total, okay)
    aqueduct n conduit – A
    163 St – C
    Bushwick Av – L
    Whitlock Av – 6 (this station is positioned very strange so as to be fairly inaccessible and extremely close to Simpson St on the 2. It does have a huge ridership increase in 2011, I’m curious why. Even so its a very low ridership station thats tucked inconveniently out of the way.)
    Elder Av – 6
    Zerega Av – 6
    Intervale Av – 2
    55 St – D
    Alabama Av – J
    Cypress Hills – J
    Norwood Av – J
    21 St – G
    Seneca Av – M
    Central Av – M
    Cortelyou Rd – Q

    • Billy G says:

      YoY ridership stats may not be fully accurate if only measuring 2010 vs 2011. There was work done on the Elevated 6, including Whitlock and Elder in that timeframe. Whitlock is the last station before going underground. I also believe that it’s the station that’s most raised from the ground on the Elevated 6.

      • aestrivex says:

        ah, that makes sense. that’s probably why in 2010 the station had an extremely low ridership, but it skyrocketed in 2011. still, in 2011 ridership was quite a bit higher than before it plummeted in 2009, and even so it is still an extremely low ridership station. it would still be fairly uneventful to axe it because it is very close to the 2, serves an extremely small area relative to other stations, and is not very far from huntspoint av on the 6 either.

        you’re right about elder av, though the ridership figures suffered in 2011 but were much better before. it shouldn’t go.

    • Jesse says:

      Wouldn’t it make more sense to keep Cortelyou Rd. and close Beverly, since Cortelyou is the main commercial strip in the area?

      • aestrivex says:

        it may, as people have commented on above. i’m not at all familiar with the area. all i have to work off of is ridership statistics and maps. both stations have high ridership despite being three blocks away from each other — in terms of not leaving large gaps, cortelyou road is the better choice, but it may be that beverley road is a less valuable area for a station.

  8. Alon Levy says:

    One criterion I’d use, in addition to ridership and distance to the nearest station, is nearby population density, or intensity of activity. In a CBD, it’s okay to have a stop every 400 meters, since it reduces walking distances to a crucial destination. In the suburbs, speed is more important. What this means is that on general principle it’s good to prune stations in the Rockaways, while in Lower Manhattan Rector Street should stay, and of course nothing in Midtown should go.

    Another criterion is location on the line. The closer a station is to the end, the less justification there is to close it. More technically you should look at how crowded the trains are with people passing through the station. On the north-south lines, which thin out continuously as they go Uptown (especially the IRT), it’s not going to be very helpful to close stations close to the north ends. In contrast, on the lines going into Queens, the Jamaica and Flushing ends are major anchors, so if there’s a station that’s too close to another station and too sparsely used, e.g. on the J/Z, it’s more justifiable to close it.

    A third criterion is scheduling. It doesn’t matter on most lines because trains that share tracks make the same stops, but on the J/Z it can be useful to offer express service with timed overtakes. In that case, stop spacing can be adjusted to better-serve the overtake. Within an overtake segment, if it’s longer than just one stop, stops should be retained, because they substitute for a long dwell time while the local train waits to be overtaken. Outside an overtake segment, it’s more useful to close stations if they cause too much of a speed difference between local and express trains.

  9. Phantom says:

    I’m the one who said 25 stations, but it makes my brain hurt to get that high in a reasonable timeframe.

    We should always be open to the idea of pruning back stations, in order to allocate resources where it can be better used.

    Here’s eighteen stations I propose for the chopping block.

    Manhattan

    1- 18th St
    R- 28th St
    1- 28th St
    B- 103rd St
    2- 116 St

    Queens
    7- 69th St
    Q- 30th Ave
    J 111 St
    M- Seneca Ave
    A- four Rockaways stations- Beach 105 St, Beach 96 St, Beach 60 St, Beach 25 St ( The Rockaways are wildly overserved. The remaining stations would serve this area well )

    Brooklyn
    Q- Neck Road
    D- 50 St
    D 79 St

    Bronx
    4- 176 St
    2- Simpson St

    • Bolwerk says:

      That’s three mentions of Seneca. What’s this boner people here have for closing Seneca of all things? It’s hardly an unused station, it’s in a fairly mixed use neighborhood, and closing it would open up a gap in service the size of the one between David Letterman’s front teeth.

      • aestrivex says:

        actually it’s very close — roughly three or four blocks — to the neighboring stations on either side, and its weekday ridership was 384th in the system, compared to much better ridership at myrtle-willoughby and forest. i’m not personally familiar with the area, and perhaps its a growing community — ridership has increased in that station over several years, slightly — but it would not leave a large gap in service.

        • Bolwerk says:

          That’s not really accurate. It’s three rather long blocks walking from Myrtle-Wyckoff, and probably a bit further as the crow flies from Forest. However, the el turns off Palmetto at Onderdonk and enters a private ROW where you can’t walk directly with it, so dispensing with Seneca would mean significantly longer walks to the subway for anyone geographically due north of the station.

          …but it would not leave a large gap in service

          To be clear, I was referring to a geographic gap. But I wouldn’t say the ridership it attracts is non-trivial either.

          • aestrivex says:

            As the crow flies, slightly further, perhaps 3.5 blocks instead of 3 blocks. The ridership isn’t nontrivial, I agree — after all, at many of the stations we’re considering cutting some 2000 people ride the train each day (excepting some rockaway stations). That being said, with the goal to identify 25 stations that could be eliminated with minimal impact in order to speed up the routes, eliminating Seneca Avenue basically asks riders to walk an additional 3 or fewer long blocks to get to the stations at Willoughby or Forest.

            Maybe there are other reasons to not axe this station, or maybe we simply don’t need to speed up ride times that much and some quantity of tightly clustered stations is okay. But I don’t agree that there is a large geographic gap here.

            • Bolwerk says:

              On a stub line like the M, I can’t imagine it would even meaningfully speed train traffic up – the major hurdle there is the grade junction at Broadway. Basically, it inconveniences ~2000 daily users and costs some percentage of ~$4000/day in revenue in lost riders. An extra three blocks actually seems like a lot to ask to me, considering some people might be walking from further than that already.

              Though, really, I’m of the opinion that 25 stations worth closing completely simply can’t be identified. I can see maybe identifying ~350 token booths to shutter entirely, but I can’t see identifying 25 stations.

    • al says:

      (7) 69th st closure would be fine as long as its partial and is open during rush hr. Long term, when the TriboroRX opens, it should reopen full time for transfer. There is a grade timer coming down from 61st St, so the time saved isn’t going to be as high as it used to be.

    • DanDAgostino says:

      30th Av in Queens on the N/Q? That would be very bad. 30th avenue is one of the busiest commercial areas in the neighborhood. Close down 39th Av-Beebe Av. Nobody ever gets on or off there. If any changes should be made, the Q should stop at QBP, Astoria Blvd and Ditmars and give Astoria some sweet express service.

      • JMB says:

        I was going to say the same thing. In the AM, 30th avenue is packed and in the PM, getting out of there takes some time.

        Also, agreed on express service on the el.

        • Matthias says:

          The last time Astoria had express service, the trains were nearly empty while the locals were packed to the gills. Have ridership patterns changed?

          • Eric says:

            Good question. I tend to doubt it, but I haven’t lived in Astoria for over two years now. Ditmars is Ditmars (i.e. busy) and Astoria Blvd has probably shown increased ridership (although I haven’t checked) but the ridership at those two stations doesn’t justify running even 1 out of every 10 trains as a express. And the ride from Ditmars to QBP is 10 minutes–hardly any time is saved by running express.

            That actually raises the question of why the Astoria line was built with express capacity at all. I assume because there were plans to eventually extend it, but I’m just guessing.

            • DanDAgostino says:

              Wishful thinking on my part! Although, the ridership on 39th, 36th and Bway has been going down while 30th, Astoria Blvd and Ditmars has been going up. I still think 39th could go though.

              • Bill Reese says:

                There have been several hotels built on or near 39th Avenue in LIC. The hotel lobby would fight tooth and nail to keep 39 Av. open, and they would win.

            • Henry says:

              There have been plans for an extension of the Astoria Line to LaGuardia for quite some time – although, I don’t think it will happen, #1 because NIMBY opposition, and #2 because there’s no good place to put it. The Grand Central already uses extremely low streetlights in that area because a landing path goes straight across the Grand Central, so I’d assume there’s no clearance for an elevated train line.

              They should extend the Astoria Line to Flushing though, if only to relieve the 7 and to provide a “redundant” option for 7 line closures.

              • Eric says:

                Right, but the Astoria line was finished over 20 years before LaGuardia officially opened as a commercial airport.

                • Bolwerk says:

                  It’s three blocks to the water from the end of the el. I always wondered if it could do a sudden dip there, and then bore east toward the airport. Would be helluva expensive, but it would provide a quick straight shot to LGA.

                  Maybe another option would be some spur from a Bronx IRT line to LGA. Again, helluva expensive, and Riker’s Island is in the way.

                  • Henry says:

                    I’d prefer it if the el was extended through Ditmars and Astoria Blvds before dipping underground to serve the Willets Point and Main St stations, with an “Airtrain LaGuardia” connecting LGA with the Astoria and Flushing lines through 94th St & Junction Blvd.

                    Due to lack of money and loud NIMBYs, this will probably never happen.

    • Alon Levy says:

      116th is a major street in Central Harlem; it should be kept.

  10. I used to thing that Neck Rd (on the Q line) should have been removed instead of being renovated.

    http://whatyourdonotknowbecaus.....lfway.html

    But, someone pointed out, I think on this forum, that when it was built it was surrounded by farmland and that people build homes and people moved there so they can live near a local stop.

  11. Jeff says:

    This would never happen, but they need to shut down one or two lines in Lower Manhattan and Downtown Brooklyn. There are so many stations located literally right on top of each other and too much redundancy. Lower Manhattan does not need a subway line on every block.

    Maybe just shut down all of the stations and leave the tunnels there for capacity purposes. The Montague line and Nassau Street lines would have all their stations shut down south of City Hall/Chambers and north of Dekalb and trains bypassing all of those stations (in the case of the Nassau line, just close it). Get rid of some of the stations on the 2/3 north of Atlantic Av while we’re at it.

    • aestrivex says:

      redundancy is by no means an entirely bad thing. it allows for multiple paths and routes to elsewhere in the system, and it allows for travel through a very crowded area if one line breaks down or has maintenance (as with the fast track system).

      • Matthias says:

        Exactly. Where would the trains go if these lines were shut down? People would have to make all kinds of extra transfers, causing nightmarish overcrowding.

    • Henry says:

      Redundancy is actually the strength of the subway system. Whenever something happens (like say, the signal blowout at DeKalb today), trains can be rerouted around problem areas. It came in handy during 9/11., and it’ll certainly be handy in the near future.

      • As a former New Yorker currently marking time in San Francisco, where exactly zero of the subway tunnels (BART or Muni) have any sort of redundant service, let me just say: dear lord, what Henry said. System-wide shutdowns of BART and Muni subway service are a commonplace occurrence, because they have basically no means of routing around certain classes of problems. It’s maddening, and the MTA’s resiliency is one of the things I most miss about NYC.

        • Alon Levy says:

          Not only that, but BART and Muni are parallel downtown, and the tunnels were constructed at the same time. But instead of making it flexible (not terribly important) and allowing cross-platform transfers, they decided to put Muni above BART and require people to go up to a mezzanine level and then down again to transfer.

    • Andrew says:

      The lines serve different corridors. The point isn’t to serve different streets downtown, it’s to serve different parts of the city.

      Many of the downtown stations are also quite small, already struggling to meet their current passenger loads.

  12. Larry Littlefield says:

    The MTA will be facing a big nut when the time comes to replace the signals on the Rockwaway line, the least used line in the system. I suppose one option would be a Rockaway terminal, and a busway serving the east-west line (the Rockaway Park side pretty much only gets a shuttle now).

    The advantage of a busway is it could be expanded west past Rockaway Park in some form of on-street BRT, if a ramp is built down to street level, and perhaps on to the subway in Brooklyn.

    Another option would be to extend the busway over the Rockaway line in mainland Queens as well, and terminate the A at JFK. Busway riders would not only be able to transfer to the A at JFK, but also to the Queens Boulevard line for a faster trip to Midtown.

    • Andrew says:

      Haven’t the signals in the Rockaways been replaced in the past 15-20 years?

    • Justin Samuels says:

      The challenge was not to leave neighborhoods without train service, which getting rid of the branch to Rockaway Park would do. And actually, that side is served by the A train during rush hours. More importantly, lately the Rockaways are experiencing a big construction drive and gentrification. Tearing down the train there would destroy that. Ben’s challenge is to get rid of redundant subway stations, like 18th street on the 1, not to leave sections of the city without trains, which would ultimately discourage public transportation use all around (as the train wouldn’t go everywhere).

      And don’t even mention the horribly crowded bus going to Rockaway Park, the Q53 . The train ride there is much nicer.

  13. Phantom says:

    It was the ” old ” M train that ran via Fourth Avenue to the West End up to Bay Parkway

    Which made it easier for the MTA to put the trains to bed in the yards, but which didn’t really provide much of a needed service. I’ve long thought that the way to go would have been to restore the long lost ” Chambers St Special ” from Bay Ridge to Chambers St as a rush hour only service.

  14. David Brown says:

    I might choose Bowery (J), 21st (G) and 18th (1) (Not that they would actually close that station).

  15. sajh says:

    A lot of people are making arguments from their perspective while ignoring the effect on others or the capacity of other lines to pickup the slack. Rector St 1 is actually needed and WELL used by the residents of BPC/downtown. Maybe when the Cortland WTC stop is reopened…. The 18th St 1 however does not make sense since there is no 18th St stop on the Broadway or 4th Ave/Lex lines. I do agree that the Willets should only be opened during the day or game time. When it is late, the park is not used and therefore the line doesnt need to stop there… All in all however, it is hard for us to comment on which stations are useless unless we’ve used them daily and at varying times, esp during rush hour. It’s wrong however to assume anything about these stops due to a ‘passing knowledge’ of using the line once and a while.

    • al says:

      They should had waited until Cortlandt St reopened to build the new South Ferry Station. That way the station could be much deeper and 1 train have a shot at going under NY Harbor to Governors and Staten Island.

      The entrances to the 7th ave/18th st station are near the 8th ave/16th st and 6th Ave 16th st entrances of the 8th ave/14th st and 6th ave/14th st stations.

      • Phantom says:

        Service to Staten Island straight into Manhattan? Not gonna happen any century soon.

        Not sure that many Staten Islanders will want it either – you’d lose the island’s low rise character pretty damned fast if you had subways straight into Manhattan.

        • Bolwerk says:

          Bah. Low-rise character like this? The scene is near Courtelyou and Beverly, probably a 30m ride to Manhattan on the Q.

          Nothing about mass transit means things become all high-rises, as the vast majority of NYC’s subway network outside Manhattan ought to convincingly show.

      • BoerumBum says:

        Wasn’t there chatter a couple years back about extending the J train to Governor’s Island and from there to Red Hook?

    • John-2 says:

      The reason 18th Street on the 1 stayed open is because of the placement of 14th Street on the Seventh Avenue line. On the 4/5/6, N/Q/R, F/M and the A/C/E stops at 14th, the station orients north, so that 14th is at the south end and there are exits as far north at 17th St., in the case of the Union Square stop. But 14th on the 1/2/3 orients to the south, so that it’s other exit is at 12th Street. That creates a four-block gap to 18th — not huge, but enough to justify the station’s existence and explain its relatively high passenger usage numbers.

  16. Ivan says:

    I’ve always wondered why there is a stop on 86th St and Central Park West (B/C), when there is another one just five blocks away on 81st St. I’d get rid of the one on 86th St.

    There’s also the 135th stop (B/C too), which while 10 blocks from the nearest stops (125th and 145th), hardly anyone uses. I say that because, ironically, I use it, but the station often seems deserted.

    • Tower18 says:

      81st St stretches 79-81 and 86th St stretches 86-88, so they’re not really as close as they seem. Same as 79th and 86th on the 1. Besides, that is a very densely populated area, with large buildings particularly on 86th. 81st has the museum, 86th has the crosstown buses, which do you cut?

    • Alon Levy says:

      86th is a key bus connection point.

  17. MRB says:

    I would eliminate the 4/5 stop at Nevins Street; the 2/3 stop at St. Marks (it’s two blocks from the Atlantic Ave stop); Bergen St on the F/G…

  18. Ian says:

    Maybe the solution is not to identify stations that are underused and could be shuttered, but identify stations that are underused and rezone the areas around them for higher densities.

    • Quinn Raymond says:

      I like that idea much better.

    • al says:

      TOD, or build giant mixed use building with next to it to drive ridership.

    • Jeff says:

      The idea behind this exercise is that many of the stations in the NYC subway are redundant or plain not useful, and all of these combine to add up to 460+ stations, which is BY FAR the highest station count for a subway station in the world, even though the total area coverage of this system is not that big compared to some other subways.

      Stations require routine maintenance, utility costs, staffing, and rehabilitation every few decades, and all of this adds to the inefficiencies in the system.

      • Quinn Raymond says:

        Closing virtually any station would be an unwise move– think back to how hard it has been to build any new stations/lines at all. It’s a lot easier to do higher density development around existing stations than it is to add new stations.

        Also keep in mind that even the least used station has better ridership than many entire bus lines.

      • Ian says:

        Stations are only redundant if the redundancy is not the station nearest to you. Even the lowest ridership stations are likely useful to the riders who use it.

    • TP says:

      Absolutely. Closing stations is the wrong solution.

  19. Mark L says:

    I don’t think anyone has mentioned this one yet: the 3rd Ave stop on the L train. It’s very close to both Union Square and the 1st Ave L stop. Anecdotally, when I used to take the L train frequently from 1st Ave, I rarely saw people get on or off at this stop.

    • al says:

      The 2nd Ave end of the station is going to be the transfer to 2nd Ave subway.

    • aestrivex says:

      Yes, I agree, when I had to take the L train to brooklyn I would often walk to 14th street, but the 3rd avenue stop was so pointlessly close to union square that I would walk to union square instead where I had a much better likelihood of finding a free newspaper.

      • Matthias says:

        The ridiculous thing about these stations is that the staircases face each other. There should be entrances toward University Pl and 2 Av. There should also be an Avenue A entrance for 1 Av.

    • Bolwerk says:

      That’s a terrible idea. It will simply crowd Union Square and First Ave., which are already prone to crowding. Especially Union Square.

    • Andrew says:

      Union Square and 1st Ave. aren’t crowded enough already?

  20. Stefanie G says:

    I nominate the L at 3rd Ave. It’s not that difficult to just walk to Union Square.

    • Lady Feliz says:

      Unless you’re a senior and/or disabled and need to transfer to the Third Ave buses. It’s always easy to close someone else’s station and make their commute harder, ain’t it?

  21. Bolwerk says:

    Of all the candidates mentioned here that I am at least casually familiar with (probably about 60% of them), I’m not sure I can count even five good proposals. I proposed Courtelyou myself, but maybe Stratford is right about Beverly making more sense. I’d buy 18th in Manhattan if it weren’t a local line anyway. Hewes on the J/M/Z would make a lot of sense if it weren’t for the unpleasant walk to Marcy from the area around Hewes.

    Anyway, I think we can dispense with the supposed advantages:

    Financial. This whole “thought experiment” only even makes sense if we keep the status quo, where stations are needlessly expensive to operate because of token booth staff overhead. Eliminate even a fraction of that overhead and the financial savings would be much grander than the benefits of closing 25 stations and making the system less useable. And for the finances to make sense, you have to carefully pick stations that don’t have a positive contribution margin.

    Saved time. How much time does eliminating 25 stations really save? At 45 seconds/stop, about 19 minutes of travel time? That would be great on a single train line, but we’re talking about saving 19 minutes of travel averaged over the entirity of NYCTA’s operations. Most riders would literally be lucky to save a few seconds.

    Maybe 4 or 5 stations make sense to eliminate, but finding 25 is pure fantasy.

    • Justin Samuels says:

      I agree with you Bolwerk, because if you close down 25 stations, you’re discouraging public transportation as you make it a lot less convenient to take trains.

      I could see the MTA closing 18th street on the 1 train, as 14th street is only 4 blocks away . I’m not saying this should be done, but something like that isn’t a great loss. But closing a bunch of stations around the city goes along way towards dismantling public transportation, which has been done nationwide since the 60s. Even NYC lost all of its trolleys, and a lot of elevated lines town down were never replaced.

  22. Streetsman says:

    Here are some additional criteria that come to mind:

    -whether there is a solid bus alternative on the route
    -whether or not the station is handicapped accessible
    -state of repair/cost to maintain
    -crime problem stations

    Didn’t read all the comments to see if these were mentioned already but some others that come to mind:
    -Bowery on the J
    -18th St on the 1
    -28th St on the 6
    -20th Ave (Bk) on the D
    -Eastern Parkway/Brooklyn Museum on the 2
    -East 105th St on the L

    • aestrivex says:

      This is my bias, since I live around the area and know what’s going on, but Bowery on the J does not strike me as a good station to remove because the area is gentrifying so rapidly. The J still has the problem that it doesn’t cover much ground in manhattan, but it may be that the immediate area around Bowery and Delancey will be massively commercialized in 10 years from now. Another thing that should be implemented and should have been implemented years ago is a free out of system transfer between bowery and grand street on the D, 1 block away. while less of a requirement now with the M, it would still have its uses.

    • Bolwerk says:

      There is no such thing as “solid bus alternative” to a subway. Start throwing buses into the mix and say goodbye to much of the ridership. And then expect it to cost a hell of a lot more to transport whoever is unlucky enough to not be able to vote with their feet.

    • Henry says:

      It would probably be MORE expensive to operate buses on the same route, considering you’re paying more for drivers and fuel per person.

  23. and I think we should consider how shutting down a station effects people. Some people may have moved their business or homes to a certain location because of its proximity to a subway station. It might be more than a matter of convince. This kind of stuff can effect the livelihoods and lives of many people.

    One should look beyond the numbers.

    • Nick says:

      I agree.

      Even if you look deeper into the numbers, there’s greater effects at hand. Property values around the closed subway stop tumble. Tax revenue disappears. Economic activity vanishes. Strain on other parts of the system and alternate forms of transit increases.

    • Spendmore Wastemore says:

      How about a trial scheduled shutdown, say 3 months. For 18th st on the 1, does anyone really care about the 3(?) block walk to the nearest entrance of the 14th st station?

      With the trains themselves being a touch over 3 city blocks long, do we really want stations 4 blocks apart?

      • Andrew says:

        A trial of what? What are you trying to accomplish?

        It’s just over 4 blocks between the nearest entrances at the two stations – the northernmost entrance to 14th is on the south side of 14th. That end of the station is already congested, since the transfer from the L feeds in there. And the first two cars on northbound trains pick up large loads at 34th and 42nd – do you really want to push even more people toward the front of the train?

        And, for the record, an IRT train is 514 feet long – shorter than two city blocks. (An IND/BMT train is 600 feet long – about two blocks and a quarter.)

  24. Kid Twist says:

    Chicago used to have el stations a quarter-mile apart or less. When the Chicago Transit Authority took over the els in the 1940s, it closed tons of stations to speed up service into the Loop.

    I think that has to be the goal here. Speed up service by removing redundant stations in the middle of long routes. Whether it’s 25 stations or more, I don’t know.

    So because of that, I wouldn’t close stations in Manhattan (though I would shutter 145th and Lenox since it’s only five cars long and just three blocks from Lenox Terminal). I’d leave the Flushing Line alone since it’s busy and not all that slow. And Astoria is too short to bother with.

    Instead, I’d look at cutting down the number of stops along the White Plains Road Line, Jerome Avenue (close to the Concourse anyway), the Broadway-Jamaica El, the Culver Line and maybe the West End Line.

    This doesn’t just save money for the MTA. If you can shorten ride times into Manhattan, then you bring down housing costs because you make a larger of number of units available within any given travel time.

    (Rebuilding a bad bottleneck or two, like the Rogers Avenue junction, would also help.)

  25. Matt Kroll says:

    Here are some other stops no one has mentioned yet:

    2/3: Central Park North/110th St
    N/R/Q: 5th Ave/59th Street

    And I second the note on the 3rd ave stop on the L train…

    • Lady Feliz says:

      You close 5th and 59th on N/Q/R and all those tourists and park users will flock to 59th/Lex, which is already dangerously overcrowded, especially the stairs when transferring to/from the IRT. The BMT platform is like the 9th Circle of Hell, especially on a hot summer day.

  26. TP says:

    You can’t kill Franklin St! It has gorgeous floor tile work and stellar Yelp reviews: http://www.yelp.com/biz/the-1-.....n-new-york

    • Matthias says:

      I do like the floor, although it’s hard to see in the photo.

      What’s that big MetroCard on the ceiling?

      • Lady Feliz says:

        Maybe a newsstand that sells MetroCards is under the sign? Not sure, i rarely use that station. If memory serves me, Franklin was one of the first IRT stations to get a complete rehab about 20+ years ago, and it’s held up very well since.

  27. Nick says:

    I don’t think this is a good idea for the public now or in the future. There’s 2 arguments backing my opinion:

    1. The transit system does not need to be effective in part in order to be justified as a whole. Just because there is low ridership at a particular subway stop, does not mean that it isn’t contributing to the health of the city in a substantial way.

    If a station has only 2000 riders per day, that’s 2000 people who are not otherwise adding to traffic congestion. Traffic congestion has a cost as well.

    2. People need places to live. By decreasing the areas accessible by public transit, you push more people into already heavily populated areas. More demand for housing in those areas means rent increases.

    Furthermore, with many transit connected areas already having high rents and density that means that those places with low ridership are ripe for development.

    Does each station have it’s own situation? Sure, but before we start hacking away at our own transit system to ‘save some money’, let’s just continue putting pressure on MTA to figure out how to manage it. Try to put yourself in the shoes of someone who lives or owns near one of these subway stops. It’s really easy to write off the importance of a particular area of the city when you don’t live or work there.

    • Justin Samuels says:

      I agree with you, Nick. Public transportation is not a profitable enterprise in the since that say restaurants are . You don’t just close a station because it has the lowest ridership. Using that logic, the city should have torn down the J and M trains, and closed the L line. Only oops, gentrification has now caused stops in Williamsburg, Bushwick, and Bedstuy on these lines to be popular. Said gentrification never would have happened where those lines closed due to low ridership. For that matter, Park Slope and Keningston becoming more popular because of the G train never would have happened had the G been dismantled due to low ridership.

      • Bolwerk says:

        I always wonder, perhaps if you shave out management patronage and TWU welfare, and account for whatever (minor) cost increases occur as a result, the transit system probably is profitable. Much of the reason it doesn’t make money is, as I think Larry Littlefield said, the point of the MTA is to administer payrolls and pensions, not do transit.

        At least, it should be able to cover its own operating costs. In theory.

  28. John T says:

    I don’t think anyone mentioned these two already, but they have stops just 2 blocks away and NO bus service

    – West End 20th Avenue Station
    – Sea Beach 20th Avenue Status

    No Bus service is why the Brighton line’s Beverley Road station would be the right one to close too.

  29. Kevin Walsh says:

    I wouldn’t close any stations. There are people who can’t get around too well who would have to walk extra blocks out of their way to get to a station. And the IND stations are too far apart as is.

    • Spendmore Wastemore says:

      I can’t get around too well and I don’t want a station every other block. The same condition that makes it hard to get around also makes frequent stops and long trips a deal breaker.

      Disabilities work in odd ways that are not visible. Glad you are thinking about helping us, but don’t go by eye; “helpful” is not the same as actually helping.

      In general, having the city work well for healthy folks means it can afford to help us. I don’t want, say, a 25mph speed limit to make it easier for me to cross the street, that would drive up the cost of car service and reduce the distance I can travel (ppl with disabilities often have time limits). It also means everyone else is effectively poorer, meaning that people will want to cut the few disabled services that actually work (most disability-related agencies serve only the agency employees and managers.)

  30. ARC says:

    I think that several factors need to be taken into account. While ridership numbers may be a helpful starting point and distance to the nearest station would help as well, one would also need to take into account current development trends and the effect that that would have on future ridership. In addition, it would be very important to consider the number of people who rely on mass transit in the vicinity around the station to be closed.

    That said, I would add Park Place to the list on the 2/3. While the station ranks pretty high, it is also combined with figures from the Chambers St. A/C and WTC E stations. Furthermore, it is very close to Chambers St. and once the new Fulton Transit Center opens, will only be three blocks from an exit. Before anybody attacks this suggestion, I acknowledge that the location of the station downtown could cause bottlenecks. This is just a suggestion.

    I would also close 155th St. on the C. It’s roughly 1000 ft. from 155th St. on the B/D and less than half a mile from 157th St. on the 1.

    Beach 98th St. is a prime target for closure. It ranks 416th on the ridership list and is less than 10 blocks from the two stations on either side of it. I think you need to keep Beach 90th because it is the first stop on the Rockaway Park Shuttle and Beach 105th because you risk creating too large of a gap. I realize that this ranking is thrown off by the seasonal nature of the station’s use but I think it is pretty safe to say that the surrounding stations could absorb the additional riders.

    The Broad St. J/Z stop could be cut. It ranks 291st in the system and is close enough to the Wall St. stations on the 2/3 and 4/5. Furthermore, the out of system transfer to the Wall St. 4/5 station could be used as an entrance. Furthermore, it is only about five blocks from Fulton St. The station could remain in use for turning trains but it is not necessarily justified in terms of ridership.

    Beyond these, I agree with closing 18th St., Rector St. and Franklin St. on the 1, and Beverly Road on the Q. I think this is a great though experiment and hope to create a full list of 25 by the end of the day.

    • Spendmore Wastemore says:

      Leave Park Place open and charge rent to whomever lands there.

      “Welcome to New York”

    • Andrew says:

      Absurd.

      Park Place is a busy station, and it’s a long walk to Fulton on the 2/3. (The nearest entrance to the transit center is irrelevant – that’s at the other end of the complex!) It’s also a transfer point to the A/C/E (the only transfer in the area to the E, and an easier transfer to the A/C than at Fulton) and to PATH.

      Broad St. is the terminal. Closing it wouldn’t save anybody any travel time. The nearby stations are on other lines that don’t go anywhere near where the J/Z goes. It’s closed on weekends but quite busy during rush hours.

      Closing 155th on the C would save time for the very small number of C riders that close to the end of the line. It’s a pretty long walk, involving significant elevation change, to 155th on the B/D (which is actually a block or two north of 155th).

      I ask again, what’s the point?

      • Justin Samuels says:

        Its absurd and surprising this would be suggested on the second avenue sagas blog. Once a train station or line is dismantled, it likely will never be replaced, or if it is replaced, its going to take a lot of time and money.

        Take the tearing down of many of the city’s els Have the second and third avenue els been replaced? No, after billions of dollars, a second avenue stubway will open up in 2016. Its going to take a fantastic amount of money to open the whole thing, particularly if a Bronx extension is included (which will have to be elevated to save money)

    • aestrivex says:

      closing beach 98 and leaving beach 105 makes little sense to me. beach 98 is the 416th used station, well beach 105 is the 419th; only aqueduct racetrack is less used. moreover, the average weekday ridership at beach 105 is 255 people compared to more than twice that at beach 98.

      ridership is not necessarily everything, but look at the placement of these two stations on a map: beach 98 is a small residential area, and beach 105 is next to lots of warehouses and parking lots.

  31. Lady Feliz says:

    Funny how nobody seeks to close stations that THEY actually use. It’s all about closing some other station, and let people on that line walk “a few extra blocks” to get to their destination. I thought the whole point of mass transit was to be convenient as possible so we don’t have to drive to get around?

    • Spendmore Wastemore says:

      No, _we want to close stations _they use ;-)

      /sarcasm off
      I’m between 2 stations, so np to me to close 1 of them. Trains are 2 blocks long, no need for stations 5 blocks apart. Trains now are double the length they were when the IRT was laid out.

    • Alon Levy says:

      I use 18th Street a fair amount when I visit, because a close friend discovered a good restaurant (Legend, between 15th and 16th). Said friend’s boyfriend lives next to the stop, and so she uses the stop frequently.

  32. Henry says:

    On a different note, there should be a conversation about adding stops (express stops, that is). Major transfer stops should be made express where possible: 74th St on the 7, Woodhaven on Queens Blvd, 59th St/Columbus Circle on the 1, and Bleecker on the 6.

    • Spendmore Wastemore says:

      And how about having the express not dragging its brakes every 1/2 mile. Those timers were not there 30 years ago and the trains didn’t fly off the tracks*

      *unless the $100K/year train operator had 6 drinks before work and nips on the job.

      • Andrew says:

        I guess you’ve never heard of the Williamsburg Bridge.

        And there were plenty of timers 30 years ago.

        • Bolwerk says:

          I don’t think he’s saying there should be no timers, or that timers are automatically bad, but the MTA seems to take speed precautions to a rather absurd level.

          It takes pretty small increases in speed to get another few trains per hour through, and that’s a lot of money saved in labor, equipment, and maintenance – a savings that can be invested in safety, if anything.

          • Andrew says:

            Timers are installed as necessary to meet safety standards – that is, to ensure that the signal system does the basic task that it’s supposed to do. I don’t think that’s absurd.

            Part of the problem is that train operators are often afraid to go as fast as the timers allow. I don’t know if that’s how they’re trained or disciplinary procedures are overly strict.

            • Bolwerk says:

              It is conceivable that safety standards can go too far.

              • Andrew says:

                The safety standard in question is “keep trains from running into each other.” I don’t think that goes too far. It’s why the subway has a signal system, and I’d like the signal system to do what it’s supposed to do.

                • Steve says:

                  I’ve long wondered about this: is it the timers that make the B/C faster than the A/D between 125th and 145th in spite of the B/C stopping at 135th in between?

                • Bolwerk says:

                  What we’re talking about are ad nauseam measures to keep trains from running into each other. Even at the system’s nadir, so far as I know/remember, trains weren’t running into each other because of the signals.

                  I’m all for safety, but the difference in safety we’re talking about is probably something akin a modicum of safety traded for real world inconvenience for lots of people. Bringing us back to speeds that were possible in the 1980s still leaves the subway far and away the safest local transportation mode.

                  • Andrew says:

                    I’m not sure why you’re bringing up the system’s nadir – that had nothing to do with it. The problem that was uncovered in 1995 was that the signal system’s design assumptions no longer applied in practice.

                    As I said, the point of a signal system is to keep trains from running into each other. It’s not a matter of degree. Either it meets that goal or it doesn’t. And in 1995, it became plainly apparent that it didn’t.

                    If NYCT becomes aware of a design flaw in the signal system and doesn’t take action to correct it – or, worse yet, begins to take action and then reverses course – they’re asking for a major lawsuit (that they will undoubtedly lose, and rightly so) the next time one train runs into another and passengers are injured or killed.

                    Perhaps you prefer WMATA’s lack of a safety culture. I’d rather ride a system where safety is taken seriously.

                    As I pointed out yesterday, train operators often go a lot slower than they need to. Training them to run their trains at the posted speed would save time without jeopardizing safety.

                    • Bolwerk says:

                      I don’t know why you’re bringing up WMATA. The types of problems there seem quite analogous to ones we’ve had in the past in NYC, and generally ultimately came down to irresponsible human behavior. Regardless, I was not speaking out against tweaking the signal system to improve safety. I was only saying there seems to be an over-emphasis on limiting speed in the name of safety. I don’t know what assumption no longer applied by 1995, but that NYCTA can’t achieve approximate historical running times in 2012 seems a little incredulous. (Plus, it doesn’t necessarily improve safety if it means lower throughput and more platform crowding.)

                      And how can TOs unilaterally decide to slow down? Are they literally disregarding schedules?

                    • Andrew says:

                      I’m bringing up WMATA because WMATA has earned the reputation of lacking a safety culture. NYCT has a safety culture, and as a rider and taxpayer I’d like to see it stay that way. NYCT takes the integrity of the signal system seriously.

                      I wouldn’t put too much faith in printed schedules as an indication of how long it actually took a typical train to get from one place to another. In the absence of accurate record-keeping, a schedule may show faster and/or more frequent service than is achievable in order to impress politicians or the public. On the flip side, with accurate record-keeping, a schedule may be excessively padded in order to artificially boost on-time performance.

                      Train operators aren’t supposed to use schedules as a guide to how fast or slow to operate. They’re supposed to operate as fast as they safely can. But grade timers – the usual mechanism by which the signal system enforces speed limits – often scare train operators into operating the train slower than they need to, for fear of tripping the train’s emergency brakes. (If you’re not familiar with how grade timers work or what I’m talking about, there’s a good description here.)

      • petey says:

        “the $100K/year train operator”

        and worth every cent

      • Alon Levy says:

        *$63k/year train operator.

    • Alon Levy says:

      There were plans once to make 72nd local and instead convert 59th to express. I forget why they were shelved – could be cost, could be the realization that more people use 72nd than 96th.

      74th on the 7 would be precious to convert, of course. The other two I’m not so sure about. The busiest local 6 stops are 51st and Canal, and although Woodhaven is the busiest local QB stop, it’s not an important connection point.

      • Bolwerk says:

        It may have been logistics. The configuration of the 1 there is a pretty dense P-L-E-E-L-P. IIRC, it was Moses of all people who wanted an express stop there.

      • Henry says:

        I’d argue that Woodhaven has a big enough anchor to justify its conversion – Queens Center is the most profitable mall per square foot in the States, and a lot of buses either end or short-turn there.

        This is my bias because I used to go to Chinatown frequently, but Canal St is a labyrinth, and making it express would probably make it more of a disorienting maze than it already is.

        • Alon Levy says:

          Well, to be fair, Woodhaven does become a connection point on one of my fantasy maps, so it should be interesting to see what the cost of an express conversion would be. (Another point taken from fantasy maps: making commuter rail workable as urban transit means there would be much more express frequency between Jamaica Station and Penn Station or Grand Central, and this would make it more palatable to add stops to the E/F.)

    • aestrivex says:

      other than woodhaven which was designed with a potential future upgrade to express in mind, those would all be nightmarish construction projects. the bleecker street transfer to the F is long enough overtime; can you imagine the chaos of redoing the station and adding at least one island platform? it would probably mean no east side IRT express service between brooklyn bridge and union square for months, at the very least.

  33. Duke says:

    In terms of maximum functionality, often the best way to eliminate stations is to close two consecutive stations and build a new one between them, rather than just picking stations to close. This keeps the spacing more even, and is less disruptive albeit more expensive.

    Probably every example would be on the IRT or BMT. The IND was built later and the stations were already further apart to begin with. Compare in The Bronx where the D stops with where the 4 stops right next door. Also, dig up an old map and look at where the A/C stops compared to where the old Fulton el stopped.

    Some examples I would propose for cutting stations:
    – kill 238th and 231st on the 1 and replace with 234th
    – kill Junius, Pennsylvania, and Van Siclen on the 3 and replace with Van Sinderen and Vermont. Build a transfer to the L at Van Sinderen.
    – kill Norwood and Cleveland on the J/Z and replace with Shepherd
    – kill Hewes and Lorimer on the J/Z and replace with Union
    – just kill 69th on the 7, period
    – kill Elder and Morrison on the 6 and replace with Ward

    I could probably come up with a few more, but 25 is too many, IMO

    • Spendmore Wastemore says:

      That’s smart.
      Do those lines also have buses running along the subway route?

    • Andrew says:

      Let me see if I have this straight. You want to close 231st, a very busy bus transfer point, and spend scarce capital dollars to build a new station three blocks to its north in order to speed travel to the last stop on the line (which is much less busy than 231st) by a few seconds?

    • Alon Levy says:

      Not sure about the rest, but word on Hewes and Lorimer on the J/Z.

    • Henry says:

      Killing 69th probably would not save much time on the 7.

      Would the Hewes/Lorimer merger be combined with a transfer to the G? Because that would be a very valuable opportunity, indeed.

    • aestrivex says:

      if we were redesigning the subways from scratch, there are many, many changes one could think of. however, in the real world, we can with minimal expense close unnecessary stations, but we cannot simply tear down stations and move them several blocks away with minimal expense.

      yes, these are both thought experiments, but quite different ones.

  34. Alon Levy says:

    Okay, let’s try to rewrite the comment I wrote last night, which got eaten.

    I think in addition to your existing criteria for closure, you should think of three additional criteria:

    1. Ambient density/intensity. In higher-density neighborhoods, especially in the CBD, you’ll want shorter stop spacing. The upshot is that it’s okay to widen stop spacing in the Rockaways on general principle. In contrast, Rector should probably stay.

    2. Distance from the edge. The north-south lines in Manhattan and the Bronx thin out uniformly as you go north, which means closing stations close to the end is less useful. The point here is that a station stop imposes a cost to people who ride through, and if there aren’t too many of those, then closure is less justified. For lines with strong outer anchors this is reversed, so station closures on the Nostrand Line (Brooklyn College is one of Brooklyn’s top stations) and the J/Z (Jamaica has tons of demand, the other stations on the line not so much) are warranted.

    3. Scheduling. Currently, all stations that use the same tracks run at the same speed – they either make the same stops or run skip-stop. But there’s an argument for running local and express trains on the Jamaica Line with timed overtakes. In that case, it makes sense to adjust stop spacing based on feasible overtake locations.

  35. Andrew says:

    I’m sorry, this is just plain absurd.

    What’s your goal in closing stations? Keep in mind that (a) skipping a station saves about half a minute, or less in many cases, (b) if there’s express service, many of the through riders are on the express already, and (c) there’s a good chance that significant capital expenditures would be required to add capacity to the adjacent stations and/or to nearby stations on other lines.

    Let’s look at your specific suggestions.

    Rector is the next-to-last stop on the line, so only people going to or from South Ferry would save any time. Rector is the closest station to much of the residential part of Battery Park City, already a significant walk. Rector is also the closest West Side IRT station to Broadway in the Wall-Bowling Green area; closing it would divert many riders to Wall on the 2/3, which already has significant overcrowding problems and would be simply unable to handle the diverted riders. Ridership at Rector grew by 9.0% from 2010 to 2011, and in my experience it’s fairly busy during rush hours, although relatively quiet at other times.

    Your photo at Franklin is misleading. It’s probably of the southbound platform, which doesn’t attract many waiting riders, because it’s the fourth-to-last stop on the line. Take a look at the northbound platform in the PM rush, or go back to the southbound platform in the AM rush 15 seconds after a 1 train has pulled in, and tell me if it’s that empty. As with Rector, trains aren’t crowded at this point, since most through riders are on the express – so closing Franklin wouldn’t save many people time.

    215th is, again, near the end of the line, so there isn’t much savings for through riders. It’s also the last stop south of the river.

    18th is another local stop on a line where most through riders are on the express. Unlike on the Lex, the north end of the station at 14th is south of 14th, and diverting most riders from 18th to that entrance would probably overload it.

    Others have addressed Cortelyou and 21st.

    In all of these cases, the benefit to through riders would be very small in comparison to the increased walking time for people who currently use the station. Unless the former significantly outnumber the latter, closing the station is not worth considering.

    If a busy line with no express service happens to have one or two unusually quiet stations, then perhaps they could be closed. I can’t think of any stations that fit the bill. Can you?

    • I think you’re taking this too seriously. It’s a thought experiment, not a proposal to close stations. And one of the problems with closing just about any of the 468 stations is that you can make a very valid and compelling argument against it.

      • Andrew says:

        Pardon me for concluding that, by proposing stations for closure, you were proposing stations for closure.

        From reading the many comments, it appears that many of your readers had the same thought I did.

      • Justin Samuels says:

        I think its because of the thought of what could happen if these stations are closed. Ben, you’re a well known proponent of public transportation. If you sponsor a contest to declare what stations should be closed due to low ridership, who is to stay some budget cutter in the government won’t run with your ideas/suggestions? Do keep in mind NYC has already destroyed much of its public transportation network, from the 40s-70s and a lot of it will never be replaced. Considering the history of the system, overall I’d say any station closing is pretty much a bad idea, as it would make getting to any particularly place more inconvenient and make driving more desirable (even worse traffic).

        This blog, as I previously noted, is the Second Avenue Sagas and look how long it took for them to get just a stubway of the second avenue subway to be underway. Because the east side els were torn down ,the Lex line was left horribly overcrowded plus some people have to walk pretty far in bad weather to get to the train. Its very ironic you even suggest this.

        • Bolwerk says:

          If you sponsor a contest to declare what stations should be closed due to low ridership, who is to stay some budget cutter in the government won’t run with your ideas/suggestions?

          That would take one stupid budget cutter. I would guess ridership would literally have to be in the low hundreds before it makes financial sense to stop using a station. Even with three shifts of a booth and employment costs of $30/hr, it only takes 500 or so riders to cover the cost. Maybe add a bit more for electricity, turnstile maintenance, TVM maintenance, etc..

          Either way, it’s probably safe to say, if it gets 1000/riders a day, it’s probably well worth keeping.

  36. Someone says:

    If you close these subway stations, then the few people who actually use those stations will be angry that they have to walk longer. Just saying.

  37. Nyland8 says:

    It’s interesting to revisit this link in a post-Sandy world. How many riders were forced to use the “under-used” nearby alternates when their preferred subway wasn’t running? I can’t help but suspect that the ridership data was heavily skewed in the weeks following the storm surge.

    And imagine what would have happened if Rector Street had been chosen for closure on the 1 Line – because of it’s proximity to the N,R Rector and the 4,5 Wall? The loss of South Ferry would have been far more catastrophic than it already is.

    I can’t help but wonder if perhaps an ideal commuter “solution” to this exercise would be to close some inbound stations M-F during the morning rush – and reopen them for the ride home. Or skip-stop some inbounds during those times. But there will always be pluses and minuses for large segments of the population, no matter which stations close.

  38. Ed says:

    To some extent this already happened, with the discontinuation of the J,M, or Z service between Broad Street and Bay Ridge (I forget which of these trains actually went to Bay Ridge). Interestingly, this had the effect that the only direct train service between Lower Manhattan and Bay Ridge and a number of other Brooklyn neighborhoods is the really slow R.

    Otherwise, I don’t really understand the point of this proposal. Nearly all those train lines “congesting” downtown Brooklyn funnel through the Borough Hall -Court Street station. Unless you get rid of the Borough Hall station its generally not worth bothering about. Geographically, if you run a line anywhere between most of Brooklyn and Manhattan, it will have either go through Brooklyn Heights or up Flatbush Avenue. Which is what they do.

    Somewhat similarly, Lower Manhattan is where the tunnels to Brooklyn are. The only other ways to get trains to Brooklyn is over the Manhattan and Williamsburg bridges, and through the tunnel the L uses. What survives of the J service, incidentally, is the only train service connecting Lower Manhattan and the Lower East Side. During the end of my years working in Lower Manhattan I found the rush hour crowding and delays on the 4 and 5 unbearable. I discovered it was much better using a combination of the J and N (transfer at Canal Street) or or sometimes the N-R-Q and 2-3. In a weird way Lower Manhattan is actually underserved in terms of subways, no 6 below City Hall which means the 4 and 5 get ridiculously overcrowded between City Hall and Wall Street during rush hours, no B,D,F,N or Q and with the recent construction and the influx of tourists the narrow sidewalks in the area area also overcrowded, you really don’t want to do much walking there during weekday business hours.

  39. Jeff says:

    The Nassau Street line, R Line, and 1 Line are all fairly underutilized in Lower Manhattan, with the R a bit less so. Even during rush hours some of those stations can be deserted.

    I work in Lower Manhattan and basically use all of these lines depending on where I go after work, and its definitely convenient to have all of these different options, but that’s a luxury, not a necessity.

    And 4&5 will always be overcrowded because they are the only line serving the east side… It doesn’t mean that Lower Manhattan is underserved.

  40. Jeff says:

    Also, Lower Manhattan, specifically the south of City Hall area is simply not a very big place… You can basically get to anywhere from the Fulton Street station within 10 minutes. There’s no need to have so many stations packed in there.

  41. Phantom says:

    I take the R train every day – none of the lower Manhattan stations are deserted during rush hours, in either direction.

    While the train is not as jam-packed as say the 4/5/7 are, it’s crowded enough during the rush hours.

  42. Andrew says:

    The Nassau Street line feeds into the Williamsburg Bridge – none of the other lines go anywhere nearby.

    The R, again, feeds into a different line from anything else. There isn’t enough capacity on the bridge to handle the R in addition to the N and Q.

    The 1 essentially provides the ability to turn West Side local service in Manhattan. Remember the service mess after 9/11? That’s what would happen without the South Ferry line.

  43. Lady Feliz says:

    You guys are always bitching about folks in outer boros driving into Manhattan, and then you propose getting rid of the R and 1 lines downtown. How exactly would the 70,000 Staten Island Ferry riders get to Whitehall? Should they all cram on to the 4/5 at Bowling Green, or just walk 15 mins from Fulton St in the dead of winter? The R and 1 lines are a “luxury” only if YOU don’t depend on them. For thousands of ferry riders and downtown office workers and residents, it’s far from a luxury.

  44. Jeff says:

    Like I said, R trains aren’t as deserted as the other two lines, but part of that is from the low frequency of service there I think.

    But nonetheless I’ve seen both the Cortlandt St and City Hall stations look relatively empty during rush hour times.

  45. Matthias says:

    Let’s say you closed all stations below Fulton St. Can you imagine the bedlam that would result? Fulton St would turn into a deathtrap.

    The density of stations is a function of the density of lines serving a high-demand area. Reducing that number would require more people to transfer, resulting in overcrowding at the remaining stations.

  46. Andrew says:

    It’s 14 minutes walking (according to Google) from Fulton and Broadway to South Ferry.

  47. Henry says:

    Let’s keep in mind that even if Fulton has that giant egg shaped building as a station house, it will never have the capacity to discharge people at a rate of say, Chatalet-Les Halles or Shinjuku.

    The point of the Lower Manhattan stations is to discharge passengers more effectively instead of cramming them into one undersized station like Main St in Flushing.

  48. Justin Samuels says:

    You quite clearly haven’t worked in lower Manhattan. If I have to get off a train and walk 15 minutes to work (a walk which may take long if there’s a lot of snow on the ground, or a lot of traffic or crowds), that’s 15 minutes or so I could have been at work. Ever heard of time is money. Or if there’s a lot of snow on the ground and my walk to work takes 30 minutes, maybe my boss won’t like it and I’ll get FIRED!!!

  49. Lady Feliz says:

    After tons of service interruptions and Fastrack work, I can personally vouch to that timing. It takes 15 mins, even walking at a fast pace, to get from Fulton St/City Hall area to the SI Ferry terminal. I’ve done that walk at midnight, just in time to get to the terminal and watch the ferry pull out, forcing me to wait 30-60 mins for the next boat. It totally sucks, especially since the drive from Midtown to SI takes 30 mins total at that hour of the night.

  50. Andrew says:

    And that’s without hordes of tourists clogging the sidewalks, as is the case for most of the day.

Trackbacks/Pingbacks

  1. […] What Are NYC’s Least Necessary Subway Stations? (2nd Ave Sagas) […]

  2. […] « Thought Experiment: Closing 25 subway stations Aug […]

  3. […] neighborhoods, instead there’s a lot of hand wringing about licensing and curb pickups. Closing subway stations, though? Not a problem. […]

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=""> <strike> <strong>