NYTIP – enhancing the nyc subway: broadway and queens boulevard

UPDATE (09.07.2022): Post substantially revised to reflect v1.0.0 v0.8.0 of the enhanced NYC subway on 09.05.2022. This update adds commentary on the transfer passage between 51st Street and Lexington Avenue – 53rd Street stations.

In my last post, I discussed the South Brooklyn redesign. So far, the redesigns contemplated by NYTIP only involve operational changes. In this post, we’ll explore the Broadway (N, Q, R, and W) and Queens Boulevard (E, F, M, and R) trunk lines. While redesigning the former is trivial, the latter is much more challenging.

Note: Click any image to enlarge.

I. Broadway

The Broadway trunk line in Manhattan runs from 57th Street and 7th Avenue to Canal Street in Manhattan. The N and Q trains run express and cross the Manhattan Bridge, while the R and W trains run local via Lower Manhattan. Conflicts occur north of 34th Street – Herald Square station when the N joins the R and W on the local tracks; the three services run together until reaching Queens, where the N and W serve Astoria and the R serves Queens Boulevard. Meanwhile, the Q serves the Second Avenue Subway (SAS) via 63rd Street.

Track map of area around 34th Street - Herald Square station.

[Fig. 1] Snippet of vanshnookenraggen’s track map showing Broadway line conflicts.

This pattern causes delays and restricts N, R, and W service since they share tracks from the 60th Street tunnel to Times Square. However, this conflict is trivially addressed:

Reroute N trains via the SAS.

Modified subway map showing N train reroute.

[Fig. 2] Overview of the N train reroute, via Brand New Subway.

Rather than switching to the local tracks north of 34th Street, the N would run with the Q to 96th Street and 2nd Avenue. This leaves only the R and W trains on the 60th Street corridor. To make up for the loss of the N through the 60th Street tunnel and Astoria, the service plan for v0.8.0 of the enhanced NYC subway would increase W service to 14 trains per hour (TPH) during peak hours; of these, 10 TPH continue to Brooklyn per the South Brooklyn redesign and 4 TPH turn at Canal Street. The R would run 10 TPH all day, ensuring combined service at least every 3 minutes all day through the 60th Street tunnel.

In a previous version of this post, I presented the “Canal Street Flip” as an alternative option:

Canal Street Flip track map, image 1 of 2.

Canal Street Flip track map, image 2 of 2.

[Figs. 3, 4] Overview of the Canal Street Flip.

First conceived decades ago in the MESA study, which incidentally addresses the SAS, the Canal Street Flip is a capital investment that would send Broadway express trains via Lower Manhattan and Broadway local trains over the Manhattan Bridge. In this scenario, the N and R would run local and the Q would run express in Manhattan; this would establish the Q as an alternative to the Lexington Avenue (4/5/6) lines. However, in Brooklyn, the N and R would run express and the Q would run local due to track configurations.

While there is still a fair concentration of jobs and attractions in Lower Manhattan, Midtown has superseded Lower Manhattan as the top jobs destination within NYC. Therefore, the Canal Street Flip is no longer under consideration for NYTIP.

II. Queens Boulevard

NYC Subway map snippet.

[Fig. 5] Snippet of the NYC Subway map showing the Queens Boulevard trunk line and branches.

The Queens Boulevard (QB) trunk line serves three distinct Manhattan trunk lines and connects to many other lines, making it one of the busiest corridors outside of Manhattan. It serves 8th Avenue (E train), 6th Avenue (F and M trains), and Broadway (R train). E and F trains run express in Queens, with E trains entering Manhattan via 53rd Street and F trains entering Manhattan via 63rd Street. M and R trains run local, with M trains entering Manhattan via 53rd Street and R trains entering Manhattan via 60th Street. This pattern poses several conflicts:

  • The E express and M local merge near Queens Plaza.
  • The F express diverges west of 36th Street station to serve the 63rd Street line; this switch induces delays on the QB express.
  • Forest Hills – 71st Avenue, due to its nature as a relay terminal for the M and R trains, induces delays on the QB local.
Track map of Queens Plaza station area.

[Fig. 6] Illustration of existing conflicts on the QB and Astoria lines.

Unfortunately, addressing these conflicts is not trivial. In 2018, vanshnookenraggen explored some of these difficulties at length; he proposed sending the R to Astoria and rerouting the N via 63rd Street to Forest Hills to replace the R. While it’s not a bad plan, I think service increases on the Broadway-SAS line – especially for SAS Phase 2 – could relieve overcrowded Bronx subways, so I’m not considering this plan.

In a previous version of this post, I considered sending the R to Astoria – Ditmars Boulevard station and building a new storage yard to the north:

Satellite image of potential Astoria Yard site north of Ditmars Boulevard station.

[Fig. 7] Potential yard location within the Con Edison property with a provision for a future LaGuardia Airport extension.

This would’ve served as a “down payment” for the oft-discussed Astoria line extension to LaGuardia Airport. Longtime readers know I am a strong proponent of that extension. However, since the W would have direct access to Coney Island yard in v0.8.0 of the enhanced NYC subway, the Astoria yard could be deferred to a future phase of NYTIP.

So how can we address the many issues plaguing QB? Let us consider several options.

Partial De-Interlining Options

Option 1a: Partial de-interlining with G extension

Image 1 of 2 showing Option 1a, which is described below Figure 9.

[Fig. 8] Overview of Option 1a.

Image 2 of 2 showing Option 1a.

[Fig. 9] Option 1a track map showing eliminated conflicts.

Under Option 1a, the G runs with full-length (600-foot) trains instead of 300-foot trains and extends to Forest Hills to replace the R; the R would serve Astoria under this option, replacing the W. The F and M switch alignments west of 36th Street station, with the F running via 53rd Street and the M running via 63rd Street; this swap removes conflicts with the E. As an optional enhancement, the G and/or M can extend to Jamaica – 179th Street, allowing the F to run express east of Forest Hills. This option requires additional railcars to allow longer G trains.

Option 1b: Partial de-interlining with R extension to Jamaica – 179th Street

Image 1 of 2 showing Option 1b, which is described below Figure 11.

Image 2 of 2 showing Option 1b.

[Figs. 10, 11] Overview of Option 1b.

Option 1b, like Option 1a, sends the F via 53rd Street and the M via 63rd Street to remove conflicts with the E. Rather than extending the G train, Option 1b would extend the R train to Jamaica – 179th Street instead. The purpose of the R extension is relieving congestion at Forest Hills – 71st Avenue station; under Option 1b, only M trains turn at Forest Hills, mitigating “bunching/gapping” delays. In conjunction with the South Brooklyn redesign, this option truncates R service to Whitehall Street, while the W serves South Brooklyn in its place.

Full De-Interlining Options

So long as the QB line serves more than two trunks or branches, full de-interlining is not possible. Furthermore, full de-interlining requires some level of capital investment. The Regional Plan Association, in their Save our Subways publication, tries to get around this by rerouting the M via the J line to Broad Street. This would leave QB with only the E express via 53rd Street and the F local via 63rd Street. RPA suggests doubling both E and F service to preserve service levels. Since such a drastic change isn’t necessary to achieve full de-interlining, NYTIP does not contemplate this option.

Before exploring full de-interlining options, let’s discuss the tunnels to Manhattan. The 53rd Street tunnel connects to 6th Avenue and 8th Avenue, while the 63rd Street tunnel connects to 6th Avenue and Broadway; the latter includes a provision for a future 2nd Avenue connection. Since the 53rd Street tunnel is the only tunnel connecting Queens to 8th Avenue, all 8th Avenue service should serve 53rd Street and all 6th Avenue service should serve 63rd Street. In addition, all Broadway local service runs via Astoria, so all W trains become R trains in this scenario. (This change requires a switch reconfiguration at Astoria – Ditmars Boulevard to increase capacity and reactivating the lower level of City Hall station to allow short turns due to the 21 TPH restriction south of there. It would also require an in-system transfer connecting Queens Plaza to Queensboro Plaza – otherwise, QB would be cut off from Broadway.) This yields two options.

Option 2: 6th Avenue service express, 8th Avenue service local

Image 1 of 2 showing Option 2, which is described below Figure 13.

[Fig. 12] Overview of Option 2.

Image 2 of 2 showing Option 2.

[Fig. 13] Option 2 track map showing eliminated conflicts.

Under Option 2, the E runs local while the F and M run express. This would eliminate all merging conflicts on the QB line. Since the E becomes the sole 53rd Street service under this option, service would need to increase to 24 TPH to preserve current service levels and to 30 TPH if the desired service level is the same as current QB express service. As there is limited short-turning capacity on the 8th Avenue line in Manhattan, Option 2 requires additional investments – such as signaling improvements or using the Worth Street subway provision – to accommodate service increases.

There are other issues with Option 2, including reduced service at Queens Plaza station and the M’s short length. Since the M uses 480-foot trains as opposed to the 600-foot trains found on the other QB lines, Option 2 would result in a net loss in capacity on the QB express – an untenable situation. (Running full-length M trains requires platform extensions at every station from Essex Street to Middle Village.)

Another issue is ease of access to 6th Avenue from the local stops west of Jackson Heights. Since the F and M diverge east of the next express stop (Queens Plaza), these passengers lose direct access to 6th Avenue. Such passengers would have to make a cross-platform transfer at 7th Avenue – 53rd Street to the B or D train to access 6th Avenue.

Option 3: 6th Avenue service local, 8th Avenue service express

Image 1 of 2 showing Option 3, which is described below Figure 15.

[Fig. 14] Overview of Option 3.

Image 2 of 2 showing Option 3.

[Fig. 15] Option 3 track map showing eliminated conflicts.

Under Option 3, the F and M run local and the E runs express; this option doubles E service to 30 TPH peak to preserve service levels on the QB express. This option also eliminates all merging conflicts on the QB line. The E branches east of Kew Gardens – Union Turnpike station, with one branch running full time to Jamaica Center and the other part time to Jamaica – 179th Street.

Option 3 also presents some issues:

  • As with Option 2, additional investments are required due to limited short-turning capacity on the 8th Avenue line in Manhattan.
  • The F, as a local train, becomes the longest local train in the system with 54 stops end-to-end. (Most passengers don’t ride end-to-end, so this may not be too big an issue.)
  • Unlike Option 2, local riders west of Jackson Heights have no opportunity to transfer to 8th Avenue service unless they ride back to Jackson Heights.

A capital solution to the last issue exists. With the R rerouted to Astoria and only 6th Avenue and 8th Avenue services on the QB line, there is now room on the 11th Street Cut west of Queens Plaza to connect the local tracks to the 63rd Street line:

Modified track map of Queens Plaza station area under Option 3.

[Fig. 16] Potential track connection from the 11th Street Cut to the 63rd Street line.

With this track connection, 6th Avenue trains can serve Queens Plaza directly before accessing the 63rd Street line, mitigating transfer issues and increasing service at Queens Plaza.


The Broadway line in Manhattan requires a simple fix – rerouting N trains via SAS – to mitigate delays and allow service increases. However, no simple fix exists for the QB corridor.

Given the difficulties presented by the full QB de-interlining options, I am selecting Option 1b for v0.8.0 of the enhanced NYC subway. While Option 1b does not eliminate all conflicts, it would still result in delay mitigation and service increases throughout the QB corridor. Importantly, Option 1b does not preclude the proposed QueensLink.

Complementary Enhancements

G Train Enhancement

Option 1b would include service increases on the G line. While the G would still run 300-foot trains initially due to rolling stock availability, the R211 order should yield enough cars for full-length (600-foot) G trains in the future.

<F> Culver Express Enhancement

At present, two <F> trains per direction per weekday spaced about 30 minutes apart run express from Jay Street – MetroTech to Church Avenue in the reverse peak direction. In conjunction with the G train service increase, the service plan for v1.0.0 of the enhanced NYC subway includes <F> express service in both directions every 20 minutes during peak hours. This results in at least 24 <F> train departures per direction per weekday; it also allows even spacing between (F) local and G trains in Brooklyn, which would run every 5 minutes each during peak hours.

Optional Transfer Passage Improvement – 51st Street/Lexington Avenue – 53rd Street

The 51st Street and Lexington Avenue – 53rd Street stations are connected by a transfer passage connecting the north end of the former to the west end of the latter. The transfer passage is somewhat long and requires a steep ascent or descent, depending on direction. The station complex is among the busiest in the city; nearly 19 million riders used it in 2019.

Before the 63rd Street Connector opened, ridership was so high and the transfer passage and platforms so congested that select peak-hour E and F trains bypassed Lexington Avenue – 53rd Street station. This is part of the reason why the F was rerouted via 63rd Street in 2001 and the V local (now M) replaced it on 53rd Street. MTA also built a connecting mezzanine in 2003 to ease crowding on the transfer passage.

With v0.8.0 of the enhanced NYC subway rerouting F trains via 53rd Street once more, an optional capital investment that could mitigate heavy crowding is a secondary transfer passage. The figure below shows what such a passage could look like:

Modified Lexington Avenue - 53rd Street station neighborhood map.

[Fig. 17] Conceptual diagram of secondary transfer passage between 51st Street and Lexington Avenue – 53rd Street stations.

The Bottom Line

The Broadway and Queens Boulevard redesigns would affect E, F, G, M, N, Q, R, and W service. The table below shows current frequencies and proposed frequencies under NYTIP (in minutes) for the F, G, M, and R lines; I previously addressed the E in my CPW redesign post and the N, Q, and W in my South Brooklyn redesign post.

ServiceCurrent Service
(Weekday Peak/Weekday Off-Peak/Weekend)
Proposed NYTIP Service
F4 / 8 / 124 / 6
G7 / 10 / 105 / 6
M7 / 10 / 10-126 / 6
R6 / 10 / 126 / 6

10 thoughts on “NYTIP – enhancing the nyc subway: broadway and queens boulevard

  1. Unrelated to deinterlining but I believe that option 2 and option 3 would be more effective if the BMT eastern division gets expanded to 10 cars that way the M train can pair up more smoothly with the F train in Manhattan and Queens but that’s a proposal for another time.

  2. I personally like deinterlining the queen blvd line.🙂 I believe that Queens Blvd line would benefit the most from deinterlining especially since the queen blvd line is getting CBTC which will allow extra trains to run per hour. Deinterlining would also allow extra trains to run per hour😊. Think of it as the best of both worlds😉. I believe that if we really want to reach the full potential of CBTC we must also deinterline queen blvd. What I’m trying to say is since CBTC being installed on the Queen blvd line will allow extra trains to run on the queen blvd line. Just think how effective queen blvd would be with CBTC and deinterlining. CBTC+ deinterlining=Much more reliable and effective service

  3. I was intending to use the K train label as a 2nd ave line but if this would be even better I believe the W train should be rerouted and instead used as a 2nd ave express train. My 2nd ave plan combined with this deinterlocking plan would include (T) (V) (W)

    • I forget to label the W train anyways if the K train is going used for 8th ave. So the 2nd ave plan would include the (T) (V) (W)
      Labels I don’t recommend using.
      H is the internal route designator for the Rockaway Park Shuttle. It was used publicly until 1993, when the public route emblem was changed to S. Since then, the shuttle has only been referred to as H in internal documents. It was last used publicly for a fare-free shuttle service in the Rockaways, started in November 2012 after damage caused by Hurricane Sandy rendered normal S shuttle and A train services in the area inoperable. This temporary H service lasted until May 2013, when full service on both routes to the Rockaways was restored.
      I has never been used due to its visual similarity to the number 1.
      O has never been used due to its visual similarity to the number 0.
      P has never actually been used, as it sounds like the word “pea” or the slang term “pee” (a colloquialism for urine).
      U has never been used, as it sounds like the word “you”.
      X was originally proposed as a route from 21st Street to the World Trade Center (via the IND Sixth Avenue Line), but the route never surfaced.[citation needed] X is used as a placeholder for routes under construction.
      Y has never been used, as it sounds like the word “why”.
      The JFK Express’s bullet, featuring an airplane symbol inside a turquoise circle, was used from 1978 until 1990.

  4. Let’s see what would happen if the F no longer served Culver and the V was brought back. Though I did it a different way. I used the E to go via Houston Street extension to South 4th Street, the K to go whee you wanted it to go, and both the F and V train via Worth Street. Oh. And there was an unused upper terminal for Winfield Spur. I would pick out a train to use it. But, It would look a little different if I had a Brown M to go there. It would go on 78th Street, Garfield Avenue, 65th Place, Fresh Pond Rd, Central Avenue, and go to where the LIRR Rockaway Beach line was.

  5. Thanks for your thoughtful explanation. It makes sense.

    With regard to commuter rail, fare integration and higher frequency would be an absolute game changer. I don’t know how practical it is on the main lines, but there are certainly some branches of the LIRR that come to mind as being really practical as adjuncts to the subway system. Particularly the PW branch, Atlantic branch, and the Lower Montauk branch. If you board at a station within city limits, pay a subway fare. That fare should allow you a free transfer to a bus and certain nearby subways.

    One problem with the out of system transfer at Lex/63 to Lex/59 is that the transfer takes the place of a transfer to a bus. There are many people in Queens whose trip starts with a bus and then to a subway and then probably at least another transfer of a subway to get to their destination. This is one of the key reasons why the transfer is not as popular as it otherwise would be to make a transfer from (F) to 4/5/6 and N/R. An out of system subway to subway transfer like Lex/59 to Lex/63 (and perhaps a LIRR to subway transfer like at Flusing-LIRR to Main St Flushing on (7)) should not count against your ability to take a bus, the same way that a true transfer between 6Av and Broadway trains at Herald Square does not count against your ability to take a bus.

    • If MTA rolls out OMNY correctly, it may mitigate the issue with out-of-system transfers you mentioned. Seems the solution is to allow unlimited rides on a single fare for a preset amount of time (say, 3 hours). Since there are a lot of extant free transfers, such shouldn’t erode MTA’s farebox revenue.

      I totally agree that rationalizing fares on commuter rail would be a game-changer. There would certainly be challenges (LIRR’s main line in NYC handling almost all branches, MNR’s Park Avenue tunnels, and at-grade merges between Amtrak and MNR south of New Rochelle and south of Riverdale being chief examples), but I think it’s doable even with current infrastructure.

  6. Good recommendations for Broadway. It seems that all who analyze the problem of interlining know that this is one of the key problems with the N switching from local to express.

    Queens Blvd is a much harder nut to crack. There is no easy solution here. You may still need a few Broadway trains to serve Queens Blvd. The G train won’t sufficiently draw traffic away from the other lines since it doesn’t serve Manhattan. (You correctly note the difficulties with needed capital improvements for options 2 and 3).

    The key with Broadway deinterlining is that no express train should cross in front of the local trains and vice versa. All Broadway express trains must either go to 2 Ave or reach Queens Blvd via 63rd street. All Broadway locals must either go to Astoria or reach Queens Blvd via the 60 st tunnel and 11 St cut.

    I think the best overall solution then would be to send all Broadway locals to Astoria. Send most Broadway expresses to 2 Ave. Some Broadway expresses (not necessarily all N trains, maybe just some N trains) get sent via 63rd street to merge with the M to form the Queens Blvd local. This would also mean that all Queens expresses (E and F) get sent via 53rd street.

    A key piece of the analysis is how may trains per hour can each track hold and how many trains per hour can each terminal turn back.

    It would also be nice to see how you would modify the schedules for late nights and weekends. Per MTA, during off-peak certain trains may not run at all or run to an alternate terminal. I’m curious to see if you can explain how those possibilities come together under NYTIP.

    • Indeed, Queens Boulevard is a tough nut to crack. Replacing the R with the G may be a hard sell (even though it makes SAS expansion much easier). If it proves too tough to overcome, then the N via 63rd Street/QB local is a good alternative. In this case I’d pursue the 11th Street Cut – 63rd Street connection to ensure the M/N serve Queens Plaza before hitting 63rd.

      If the N serves QB, then one could argue, as the Regional Plan Association did, that fare rationing and service increases on Metro-North’s Harlem Line could serve the 3rd Avenue corridor while SAS takes the cross-Harlem alignment. Another option is the Q via cross-Harlem and Concourse to give Bronxites an alternate to the overcrowded 4. However, this option requires capital investment to minimize interlining and could complicate the service pattern on Concourse.

      Concerning overnights/weekends, click here for a copy of the enhanced NYCS guide (Excel): https://nerdynel.files.wordpress.com/2019/05/enhanced-nycs.xlsx

      I hope this answers your question!

      • Agree that replacing the R with G will not go down well with Queens Blvd riders. The G wasn’t a popular service with QB riders when it ran there. I can’t really see what’s changed in the past 20 years that would make the G a more popular service. But the R (or another Broadway-bound service) is, although not nearly as popular as the E or F. The repurposing of the 11th St cut as a connection to the 63rd St Tunnel instead of the 60th is an interesting one, but I have to wonder if the connecting tracks will easily clear the web of underground tracks already in that area.

        All this, plus the RPA’s suggestion to run the N to QB via 63rd, got me to thinking that maybe there is a way to de-interline QB while keeping a Broadway service. Perhaps we can run the E and M to/from 71st Ave and the F and N express. The N would replace the E to/from Parsons/Archer. The N would likely have to run the same 12 tph to/from Parsons as the E due to the poorly placed crossover switch, but I think it would be less of an issue because the N is unable to turn more than that at Stillwell Ave (or Brighton Beach if it runs via Brighton Express). Though an Queens Blvd express N via 63rd does have an issue of merging with the F northbound and the Q southbound at Lex-63rd, which could potentially be a bottleneck. But all of the merging in the Queens Plaza/36th Street area would be removed. One question would be whether or not Archer riders would prefer Broadway over 8th Ave.

discuss respectfully!