Home View from Underground Guest Post: Transit’s subway frequency guidelines are the wrong approach

Guest Post: Transit’s subway frequency guidelines are the wrong approach

by Benjamin Kabak
As ridership peaks, a typical morning ride on the Q train involves lots of hair and armpits and very little space. (Photo: Benjamin Kabak)

Could subway service be improved if the MTA adopted better operations practices? (Photo: Benjamin Kabak)

Today’s post comes to us from long-time SAS reader and commenter Alon Levy. Alon originally published this on his own site this past weekend and has graciously allowed me to run it in its entirety here. With subway service trending toward unreliability and uneven headways becoming more pronounced, Levy has tackled the interrelated issues of the MTA’s load guidelines, service frequency, interlining and general frustration with NYC’s increasingly crowded subway system. You can read more of Levy’s work on his site Pedestrian Observations and follow him on Twitter.

In New York, the MTA has consistent guidelines for how frequently to run each subway route, based on crowding levels. The standards are based on crowding levels at the point of maximum crowding on each numbered or lettered route. Each line is designed to have the same maximum crowding, with different systemwide levels for peak and off-peak crowding. While this approach is fair, and on the surface reasonable, it is a poor fit for New York’s highly branched system, and in my view contributes to some of the common failings of the subway.

Today, the off-peak guidelines call for matching frequency to demand, so that at the most crowded, the average train on each route has 25% more passengers than seats. Before the 2010 service cuts, the guidelines had the average train occupied to exact seating capacity. At the peak, the peak crowding guidelines are denser: 110 passengers on cars on the numbered lines, 145 on shorter (60’/18 m) cars on the lettered lines, 175 on longer (75’/23 m) cars on the lettered lines. There’s a minimum frequency of a train every 10 minutes during the day, and a maximum frequency at the peak depending on track capacity. When the MTA says certain lines, such as the 4/5/6, are operating above capacity, what it means is that at maximum track capacity, trains are still more crowded than the guideline.

In reality, guideline loads are frequently exceeded. Before the 2010 service cuts, many off-peak trains still had standees, often many standees. Today, some off-peak trains are considerably fuller than 25% above seated capacity. In this post, I’d like to give an explanation, and tie this into a common hazard of riding the subway in New York: trains sitting in the tunnels, as the conductor plays the announcement, “we are delayed because of train traffic ahead of us.”

The key takeaway from the system is that frequency at each time of day is calculated separately for each numbered or lettered route. Even when routes spend extensive distance interlined, as the 2/3 and 4/5 do, their frequencies are calculated separately. As of December 2014, we have the following headways, in minutes:

Line AM peak Noon off-peak PM peak
1 3 6 4
2 6:30 7:30 6:45
3 6 8:30 6:45
4 4:30 7:30 4:24
5 5 8:30 5:45
6 2:30 4 3:18
7 2:30 5 2:30
A 4:45 10 4:45
B 8:45 10 9:15
C 9:15 10 10
D 6:15 10 6:45
E 4 7:30 4
F 4:45 7:30 5
G 6:30 10 10
J/Z 5 10 5
L 4:30 6 4
M 8:45 10 9:25
N 7:15 10 7:30
Q 7:15 10 7:45
R 7:30 10 7:30

Consider now the shared segments between the various lines. The 4 comes every 4.5 minutes in the morning peak, and the 5 every 5 minutes. There is no way to maintain even spacing on both lines with these headways: they share tracks for an extensive portion of their trip. Instead, the dispatchers move trains around to make sure that headways are as even as possible on both the shared trunk segments and the branches, but something has to give. In 45 minutes, there are ten 4s and nine 5s. Usually, on trunk lines with two branches, trains alternate, but here, it’s not possible to have a perfect alternation in which each 4 is followed by a 5 and each 5 is followed by a 4. There is bound to be a succession of two 4s: the second 4 is going to be less crowded than the guideline, and the following 5 is going to be more crowded.

It gets worse when we consider the extensive reverse-branching, especially on the lettered lines. For example, on its northbound journey, the Q initially does not share tracks with any line; then it shares tracks with the B, into Downtown Brooklyn; then it crosses into Manhattan sharing tracks with the N; then it again shares tracks with no other route, running express in Manhattan while the N runs local; then it shares tracks with the N and R into Queens; and then finally it shares tracks with the N in Queens. It is difficult to impossible to plan a schedule that ensures smooth operations like this, even off-peak, especially when the frequency is so variable.

Concretely, consider what happens when the Q enters Manhattan behind an N. Adequate separation between trains is usually 2 minutes – occasionally less, but the schedule is not robust to even slight changes then. To be able to go to Queens ahead of the N, the Q has to gain 4 minutes running express in Manhattan while the N runs local. Unfortunately, the Q’s express jaunt only skips 4 stations in Manhattan, and usually the off-peak stop penalty is only about 45 seconds, so the Q only gains 3 minutes on the N. Thus, the N has to be delayed at Herald Square for a minute, possibly delaying an R behind it, or the Q has to be delayed 3 minutes to stay behind the N.

In practice, it’s possible to schedule around this problem when schedules are robust. Off-peak, the N, Q, and R all come every 10 minutes, which makes it possible to schedule the northbound Q to always enter Manhattan ahead of the N rather than right behind it. Off-peak, the services they share tracks with – the B, D, and M – all come every 10 minutes as well. The extensive reverse branching still makes the schedule less robust than it can be, but it is at least possible to schedule non-conflicting moves. (That said, the M shares tracks with the much more frequent F.) At the peak, things are much harder: while the N, Q, and R have very similar headways, the D is considerably more frequent, and the B and M considerably less frequent.

I believe that this system is one of the factors contributing to uneven frequency in New York, with all of the problems it entails: crowding levels in excess of guidelines, trains held in the tunnel, unpredictable wait times at stations. Although the principle underlying the crowding guidelines is sound, and I would recommend it in cities without much subway branching, in New York it fails to maintain predictable crowding levels, and introduces unnecessary problems elsewhere.

Instead of planning schedules around consistent maximum crowding, the MTA should consider planning schedules around predictable alternation of services on shared trunk lines. This means that, as far as practical, all lettered lines except the J/Z and the L should have the same frequency, and in addition the 2/3/4/5 should also have the same frequency. The 7 and L, which do not share their track or route with anything else, would maintain the present system. The J/Z, which have limited track sharing with other lines (only the M), could do so as well. The 1 and 6 do not share tracks with other lines, but run local alongside the express 2/3 and 4/5. Potentially, they could run at exactly twice the frequency of the 2/3/4/5, with scheduled timed local/express transfers; however, while this may work for the 6, it would give the 1 too much service, as there is much more demand for express than local service on the line.

To deal with demand mismatches, for example between the E/F and the other lettered lines, there are several approaches, each with its own positives and negatives:

– When the mismatch in demand is not large, the frequencies could be made the same, without too much trouble. The N/Q/R could all run the same frequency. More controversially, so could the 2/3/4/5: there would be more peak crowding on the East Side than on the West Side, but, to be honest, at the peak the 4 and 5 are beyond capacity anyway, so they already are more crowded.

– Some services could run at exactly twice the frequency of other services. This leads to uneven headways on the trunks, but maintains even headways on branches. For example, the A’s peak frequency is very close to exactly twice that of the C, so as they share tracks through Lower Manhattan and Downtown Brooklyn, they could alternate A-C-A-empty slot.

– Services that share tracks extensively could have drastic changes in frequency to each route, preserving trunk frequency. This should be investigated for the E/F on Queens Boulevard: current off-peak frequency is 8 trains per hour each, so cutting the E to 6 and beefing the F to 12 is a possibility.

– Service patterns could be changed, starting from the assumption that every lettered service runs every 10 minutes off-peak and (say) 6-7 minutes at the peak. If some corridors are underserved with just two services with such frequency, then those corridors could be beefed with a third route: for example, the Queens Boulevard express tracks could be supplanted with a service that runs the F route in Jamaica but then enters Manhattan via 53rd Street, like the E, and then continues either via 8th Avenue like the E or 6th Avenue like the M. Already, some peak E trains originate at Jamaica-179th like the F, rather than the usual terminus of Jamaica Center, which is limited to a capacity of 12 trains per hour.

– The service patterns could be drastically redrawn to remove reverse branching. I worked this out with Threestationsquare in comments on this post, leading to a more elegant local/express pattern but eliminating or complicating several important transfers. In particular, the Broadway Line’s N/Q/R trains could be made independent of the Sixth Avenue trains in both Queens and Brooklyn, allowing their frequencies to be tailored to demand without holding trains in tunnels to make frequencies even.

For the lettered lines, I have some affinity for the fourth solution, which at least in principle is based on a service plan from start to finish, rather than on first drawing a map and then figuring out frequency. But it has two glaring drawbacks: it involves more branching than is practiced today, since busy lines would get three services rather than two, making the schedule less robust to delays; and it is so intertwined with crowding levels that every major service change is likely to lead to complete overhaul of the subway map, as entire routes are added and removed based on demand. The second drawback has a silver lining; the first one does not.

I emphasize that this is more a problem of reverse branching than of conventional branching. The peak crowding on all lines in New York, with the exception of the non-branched 7 and 1, occurs in the Manhattan core. Thus, if routes with different colors never shared tracks, it would not be hard to designate a frequency for each trunk route at each time of day, without leading to large mismatches between service and demand. In contrast, reverse branching imposes schedule dependencies between many routes, to the point that all lettered routes except the L have to have the same frequency, up to integer multiples, to avoid conflicts between trains.

The highly branched service pattern in New York leads to a situation in which there is no perfect solution to train scheduling. But the MTA’s current approach is the wrong one, certainly on the details but probably also in its core. It comes from a good place, but it does not work for the system New York has, and the planners should at least consider alternatives, and discuss them publicly. If the right way turns out to add or remove routes in a way that makes it easier to schedule trains, then this should involve extensive public discussion of proposed service maps and plans, with costs and benefits to each community openly acknowledged. It is not good transit to maintain the current scheduling system just because it’s how things have always worked.

You may also like


mister December 15, 2015 - 2:03 am

I rather enjoy reading Alon’s posts, and thanks Ben for including this. I have a number of thoughts about what is presented here.

For starters, what Alon presents as the peak frequencies of various lines is an oversimplification of what the reality is. For example, the A line. Alon lists it has having a peak frequency of 4:45 during the AM hours. However, while that is pretty accurate for the northern end of the line’s peak frequency, it actually understates what the Brooklyn peak frequency is – it’s actually closer to 3:30. A number of other lines have similar issues, the B is a good example of this. This introduces an additional complexity during the peak periods that isn’t really captured here.

Another piece of data missing here is the shortcomings of the physical layout of the existing infrastructure. For instance, in Alon’s previous post that he references where he goes over the potential of reducing the amount of reverse branching present in the NYC subway, he states that the only place where three routes share tracks in the subway system is the 60th street tube (N/Q/R). But while it’s a lot less obvious, the Rogers junction just east of Franklin forces the 2, 3 and 5 to all share a track, briefly. This is one major reason why some 5 trains don’t serve Flatbush, but instead terminate at Utica or New Lots. Trying to have the 2/3/4/5 all operate at the same headway would result in a reduction of service on all four lines under this proposal, or result in the same kind of schedule oddities as exist now, with a need to add more short turned trains.

Getting beyond those issues, I’m still not sure if the attempt here is to address an inadequacy in off-peak or peak services, which are very different. The post starts off disagreeing with the methodology for determining off-peak service levels but then addresses reverse branching issues that affect the lines primarily during peak hours (your 4/5 line example). During off-peak hours, with additional capacity available for each line to use, and more of an ability to build holds into the schedule, this is really not an issue.

But if this is an issue with peak services, it’s a different ball game. For instance, does operating the 4 at the same frequency as the 5 make sense? The post points out that, under the present system, if two 4s are followed by a 5, then the second 4 will be below guideline load, while the following 5 will be above it. If the 4 runs at the same frequency as the 5, will we end up in a situation where EVERY 4 is above guideline load while EVERY 5 is below it? It’s possible.

Many of Alon’s proposed fixes to the system introduce new challenges, or leave some of the same ones in place. Matching frequencies on every line that shares tracks will lead to some corridors in the outerboroughs being overserved while others deal with crowded trains that would exceed the discarded guideline loads. Ditto for the mandatory frequency solution that is discussed. Going to a system where frequencies work on multiples still creates scenarios that generate the exact same problem that we’re trying to get away from (i.e. – Alon’s solution proposed for the A/C lines, where the first A train will always be more crowded than the C and A that follow it). Trying to eliminate reverse branching is the most interesting solution that I see here, but it eliminates a number of conveniences that riders might balk at. As we’ve seen with the M, operating through service to a specific trunk can impact ridership significantly. Also, it would create other interesting capacity issues that would be difficult (though not impossible) to address.

I like the thought process to attempt to address the issue with guideline frequencies for individual lines that then share tracks, but calling the existing approach the wrong one in this case may be a failure to consider the totality of the suggested fixes.

Alon Levy December 15, 2015 - 9:13 am

I cribbed the frequencies from the Straphangers’ Campaign. I know they’re not 100% correct (for example, I’ve seen different peak tph counts cited for the 2/3 and the 6), and the example you give of different peak A tph counts in different parts of the system is another. In comments at my place, I got a link to more precise schedules, and the numbers I posted are largely correct; the off-peak 2/3/4/5 is sort of what I advocate already, with 8-minute service on each route, but the lettered lines have the 10 vs. 7.5 minutes alternation, and the headways between trains are not consistent, but vary by 1-2 minutes in each direction.

On to what you say about the alternate 4/5 situation: if the 4 and 5 have equal amounts of service, then we’ll see a bit more crowding on the 4 than on the 5, but not by a large margin. The reason is that the bulk of the crowding is in Manhattan, when the two lines run together. Something like two-thirds of boardings on the IRT are on the shared trunk segments; in the Bronx, the crowding is not very bad. Between that and the already not too large differences in frequency, it should not be an issue.

The reason the current situation is more problematic is that any schedule under which the trains on the Manhattan trunk are evenly spaced leads to an occasional 4-4-5 sequence. This means the 5 gets a much longer service gap than intended – around 7 minutes, against an average of 5 in the morning peak and 4.5 for the 4. In contrast, equalizing service leas to service gaps of about 4:45 on both branches, much closer to the crowding-based headways.

Finally, re Rogers: yes, three services share track there, but for a very short stretch, without any stations. So it’s less critical than the N/Q/R. There’s also a possible, though drastic, disentanglement: send the 3 to Flatbush and the 5 to New Lots.

Fbfree December 15, 2015 - 10:29 am

re: Rogers junction
Sending the 5 to New Lots still required merging with the 2/3 if the local stations at Nostrand and Kingston are to be served.

Brooklynite December 15, 2015 - 12:38 pm

That’s nothing two new switches just east of the junction can’t eliminate. A little bit of structural work will probably be required with moving columns, but it’s totally doable.

mister December 15, 2015 - 1:07 pm

‘A little bit of structural work’ = a lot of cost and service disruption. Structurally modifying a tunnel while it continues to host train service is not easy.

NYCT has looked into it.

Brooklynite December 15, 2015 - 4:27 pm

The Japanese managed to convert a rail station into a tunnel portal (with station) in four HOURS (see video). You’re telling me we can’t open up the street, take out a couple of columns, put in some beams, and close it up again over Thanksgiving weekend, for instance? Most of the proposals I’ve heard involve converting Rogers into something resembling 59th/CC, which would take much longer and be much more complex.

Brooklynite December 15, 2015 - 4:32 pm

Didn’t mean for it to embed like that… half of it is cropped out. I recommend watching it on YT.

mister December 15, 2015 - 6:06 pm

Super cool video.

I’m not saying it’s not possible. But how much did that project cost? Only figures I could find online was that it was 159.3 billion yen, but it seems that this included some additional work as well. It also looks like there were a lot of temporary works done underneath in preparation, not really an option at Rogers. Also, it’s kind of a different ball game trying to change overhead structure with soil, utilities and street overhead. Should we do it? Absolutely. Someone has to identify funding for it though.

Miles Bader December 20, 2015 - 12:58 am

Of course those four hours were preceded by years of track construction done during the nightly non-operating periods (roughly 1am-5am). They basically undermined the old tracks, putting them on supports and and building new gradually descending tracks underneath them to the tunnel entrance for the new underground station, and restoring the upper tracks to operating condition each morning.

It’s an interesting question how’d you do this sort of thing with 24-hour operation…

Brooklynite December 22, 2015 - 8:18 pm

A couple of long-weekend shutdowns should do the trick. Start over Thanksgiving and finish sometime over the holidays perhaps?

DF December 15, 2015 - 1:04 pm

It’s kind of puzzling, isn’t it, that during PM peak the MTA can equalize service on the 2 & 3 – even though (I assume) practically everyone taking the 3 would take the 2 if that came first, whereas the converse is not true – but there is such a large gap between service on the 4 & 5, implying that there is so much more demand for the 4 that the differing headways are considered necessary despite the kind of timing problems they create. Am I missing something that makes that seem more logical?

mister December 15, 2015 - 1:05 pm

As mentioned in a previous comment, that solution doesn’t fix Rogers.

If the vast majority of the crowding is on the shared segment of the 4/5 corridor, then why is is such a big deal if the 5 has one headway in a 30 min period that is 30 sec behind it’s desired headway when it could be right behind a 4 train through the most crowded segment of the line? Consider what 30 minutes of the schedule could look like with the 27 tph currently operated at the peak:

4- 8:00:00
5- 8:02:30
4- 8:04:30
5- 8:07:00
4- 8:09:00
5- 8:12:00
4- 8:14:00
4- 8:16:30
5- 8:18:30
4- 8:21:30
5- 8:23:30
4- 8:26:00
5- 8:28:30

Under this schedule, the one 5 that operates right behind the double 4s is only 30 seconds behind the desired headway, and also has a shorter headway between the 4 that leads it, which itself is likely able to handle additional passengers. How much additional crowding do you think will exist on this particular 5 train?

This post really feels like it’s creating a solution to a relatively minor problem, and in some cases, creating a bigger one. MTA has a lot of shortcomings, but it’s tough for one person to say that a whole department of people is wrong in their approach to something, unless you have a really good case to back up your claim. I just don’t see that here.

Alon Levy December 15, 2015 - 6:17 pm

A 1-minute difference is a big deal when it’s 2 vs. 3 minutes.

Nathanael December 21, 2015 - 11:46 pm

Put it this way: London does it the way Alon advises. The MTA has an entire department which is Doing It Wrong, and it’s for the reason documented in the next post: they didn’t think of it, and they simply refuse to look at best practices in other systems.

Nathanael December 21, 2015 - 11:45 pm

For what it’s worth, London has been aggressively redesigning service frequencies and routes for years to eliminate complex branching patterns and get even patterns of “trains per hour”, even at peak.

So Alon’s ideas are being implemented somewhere. London has a particularly complex network because the lines were mostly originally designed as “mainline” railroad lines with mixed freight traffic, intercity traffic, etc.

Marc Shepherd December 15, 2015 - 7:28 am

As far as I can tell, Levy’s approach is: Less service overall, and in particular, less along the East Side. I can just imagine the outrage on Ben Kabak’s twitter feed if this were adopted. Ben would need a new keyboard shortcut for the word “unacceptable” (which he uses too much already).

I do think it takes considerable chutzpah to write a post describing the current system as “the wrong approach,” when evidence that Levy’s is the right one, is so wafer-thin to non-existent.

There might be some decent ideas buried there, but as “mister” has documented (above) peanut-butter spreading along the trunk lines would lead to woefully inadequate service elsewhere, and the Rogers Junction problem throws a wrench in his plan, at least for the 2/3/5.

I haven’t thought hard about the other places it would break, but if Levy ignores the system’s best known bottleneck, what others might he have missed?

Alon Levy December 15, 2015 - 8:50 am

Where do I call for less service on the East Side? Yes, I call for equalizing East and West Side IRT service – via more West Side service. (The 2/3 is already the second most crowded track pair at rush hour, after the 4/5. We’re not talking empty trains here.)

Larry Littlefield December 15, 2015 - 9:50 am

It gets back to your other issue — the cost of capital improvements. The IRT is constrained by the capacity of the interlocking east of Franklin Ave. Plans to fix it are always put off due to cost and disruption.

I saw it get dropped from the NYCT proposal for one MTA capital plan with the budget director of the NYCT capital plan arguing for it.

Stephen Bauman December 15, 2015 - 4:36 pm

You can’t blame the bean counters for not fixing Rogers. They were told that CBTC would allow for drastically increased service levels regardless of track configurations. They believed it.

mister December 15, 2015 - 5:48 pm

Maybe you’re familiar with something from days past, but recent NYCT planning has suggested that they still want to fix Rogers.

You still haven’t answered how the Brooklyn IRT was ever operated without having a merge of services.

Stephen Bauman December 15, 2015 - 6:32 pm

NYCT Planning aren’t the bean counters.

I thought how they used to operate the Bklyn without merges was obvious from the 1954 service level diagram.

The August 1949 service level diagram is even more obvious. They operated 30 tph on both the West and East Side expresses. Only West Side trains operated to Flatbush and New Lots which was split 17 tph, 13 tph in favor of Flatbush. They turned 30 tph East Side expresses at Utica.

mister December 15, 2015 - 6:41 pm

Do you think that turning 30tph at Utica is realistic?

Stephen Bauman December 15, 2015 - 7:08 pm

The BOT reported that 87.47% of IRT trains operated on time between Jan 1 and Jun 30 in 1949. The percentages for the BMT and IND were 99.19% and 99.24%, respectively.

Brooklynite December 15, 2015 - 7:21 pm

For the record, anyone know what those percentages are now? And since they operated 11-car E/F trains in the ’50s they clearly had crowding issues back then as well.

Stephen Bauman December 15, 2015 - 7:37 pm

The MTA reports that this year’s on time performance is 70% and their target is 75%.

Nathanael December 21, 2015 - 11:49 pm


mister December 15, 2015 - 9:00 pm

That doesn’t answer the question that was asked here. It’s clear that current rules, regardless of your disdain for them, would not permit 30 trains per hour to be turned at Utica.

Stephen Bauman December 15, 2015 - 9:45 pm

Which rules are they and why are they important?

mister December 15, 2015 - 11:23 pm

Specifically we are referring to interpretations of rules requiring trains to be cleared out before relaying. I’ll not comment on why they’re important; I don’t think they are. But they exist and cannot be ignored.

Nathanael December 21, 2015 - 11:50 pm

So repeal those rules. They’re just internal NYC Transit rules, not federal law.

Eric December 20, 2015 - 6:51 pm

You could turn 20-25 trains at Utica and the remainder at South Ferry.

Henry December 15, 2015 - 8:29 pm

Demand is stronger on Flatbush than it was in the past, to the point where the “peak” East Side Flatbush service has become a full, weekday service. Running that kind of service pattern on Flatbush today would probably screw them over.

mister December 15, 2015 - 10:45 am

You can’t increase 2/3 West Side service without decreasing 4/5 East Side service. Rogers is already handling the maximum number of trains it can take.

Stephen Bauman December 15, 2015 - 11:41 am

In the old days they solved the Rogers capacity problem by removing it. Only West Side trains operated on the Nostrand Ave Line. That way both East and West Side lines could operate at greater service levels. Nostrand Ave passengers wanting to use the Lex would switch at Franklin Ave.

al December 15, 2015 - 11:46 am

Throw in short turn trains at places like Myrtle-Wyckoff. You get more tph with the same equipment count.

Stephen Bauman December 15, 2015 - 12:39 pm

They are already short turn at M-W on the L. However, it’s not balanced. This puts 6 minute gaps on trains from Canarsie.

Station turnstile counts are approximately equal east and west of M-W. That would indicate a service pattern like the old days would be appropriate. A balanced merge at M-W with trains from Canarsie running express between M-W and Lorimer.

I wonder if the CBTC system could handle the operational complexity without $$$ for reprogramming.

Alon Levy December 15, 2015 - 12:55 pm

But you can’t really run express trains on a two-track line…

Stephen Bauman December 15, 2015 - 1:30 pm

The BMT, BOT and TA managed to do it from 1936 to 1956. Riders complained, when they eliminated the 14th St Express.

Brooklynite December 15, 2015 - 4:20 pm

I believe frequencies were lower then.

Stephen Bauman December 15, 2015 - 4:49 pm

No. The 14th St Line currently operates at 20 tph. They operated 24 tph with the 14th St Express. Locals out of M-W operated at 12 tph. Expresses operated out of Canarsie and Lefferts at 6 tph each for a combined 12 tph. There were additional services out of both Lefferts and Canarsie that did not use the 14th St Line.

Robj December 15, 2015 - 10:59 pm

Can you please explain how this worked? Were there overtakes?

Brooklynite December 15, 2015 - 11:08 pm

There were no overtakes. The local left M/W right after the express before it and got to the all-stops section (Lorimer?) before the next express caught up.

Eric December 17, 2015 - 3:55 am

For example, if you have three trains with the following gaps between them:

Local – 6min – Express – 2min – Local

The express train can skip a few stops and save 4 minutes, then you will have:

Local – 2min – Express – 6min – Local

The problem with this is that if your system is busy, there should be 2 minutes between every train, so the express cannot skip stops without running into the train ahead of it.

You can run skip-stop service where one train stops at “odd” stops and the other at “even” stops, so the average speed is the same. The J/Z do this. But this doesn’t increase capacity either.

Tower18 December 15, 2015 - 12:53 pm

Seriously, no other city in the world has this obsession with one-seat rides as New York. And that’s because people don’t care as much about transfers if the maximum time they need to wait for a transfer is like 2-3 minutes.

So, if you detangled Rogers Jct such that Nostrand passengers wanting the East Side change at Franklin, that SHOULD be acceptable, provided that the waiting time on Nostrand was like 2-3 minutes, and the waiting time at Franklin is likely often nil, but no more than 2-3 minutes.

mister December 15, 2015 - 1:44 pm

If you look at commuter rail systems all throughout the world, many of them aim to do exactly that: provide one-seat rides. It’s kind of similar that that kind of system. No other Metro system in the world really has the kind of 4 track express service offered in NYCT either, except Chicago (there’s the ugly one-seat ride monster…) and Philadelphia.

The London Underground doesn’t do it on the same scale as NYCT, but it’s present there. Shucks, they mix commuter and metro lines there when interlining. Even trying to bring up that possibility for NYC on here causes people to get up in arms.

Brooklynite December 15, 2015 - 4:21 pm

Commuter trains do it because they have headways on the order of every 30 minutes. The subway has an order of magnitude more service, so the penalty of transferring decreases significantly.

mister December 15, 2015 - 5:45 pm

They have those headways during off-peak times. The Headway on many of LIRR’s branches is close to 10 minutes. Other commuter operations are similar.

Alon Levy December 18, 2015 - 2:37 am

No, it isn’t. One shared trunk line does (Main Line to Mineola and Hicksville, but not the intermediate stations), but branches run every 30-60 minutes off-peak.

Alon Levy December 16, 2015 - 11:44 pm

Seoul Metro Line 1 has four tracks.

Nathanael December 21, 2015 - 11:59 pm

London Underground has been spending decades trying to detangle their very complicated interlining patterns.

And for what it’s worth they have a four-track mainline from Finchley Road to Harrow-on-the-Hill — but they rearranged services when they built the Jubilee Line, so that it’s operated as two separate lines.

It’s also four tracks from Barons Court to Northfields, and again they rearranged services to operate it as two independent lines (Picadilly and District). Despite the flat junctions.

Earl’s Court is actually *six* tracks, largely designed to separate the “Wimbledon-Edgware” service from the main District Line service from the Picadilly Line.

They even detangled the DLR, which now runs as three functionally independent lines.

Sydney, Australia, has a metro system which is completely mixed up with commuter trains, intercity trains, and *freight trains*. They have also been spending a lot of money (more recently) to detangle the lines in the so-called “Clearways” program. The program is designed to isolate the lines into groups in order to prevent delays in one group of lines from causing knock-on delays in another group.

Henry December 15, 2015 - 8:34 pm

People don’t care as much about transfers, but the system in general is not designed to host large amounts of transfers in places where they might otherwise be logical. Where they do occur, like at Jackson Heights-Roosevelt, platforms are ridiculously overcrowded, to the point where it’s not uncommon for people to wait on stairwells.

Franklin has an issue similar to Camden Town on the Northern Line in London. The MTA has looked into separating East Side and West Side services on the Brooklyn IRT, but concluded that it would lead to too much transfer activity at Franklin.

Nathanael December 22, 2015 - 12:02 am

If this were London, what they’d do would be to massively expand Franklin Ave. station. They built a new station under the Houses of Parliament, not that long ago.

Imagine, building infrastructure specifically to allow improved operations. Shocking.

mister December 15, 2015 - 1:09 pm

Under this proposal, what serves Kingston and Nostrand stations on the 2/3?

Stephen Bauman December 15, 2015 - 2:03 pm

I did not suggest that they ran all West Side trains on the Nostrand Ave Line. West Side service was split between Flatbush Ave and New Lots.

They currently operate 18 trains peak hour trains out of Flatbush, evenly split between West and East Side lines.

Here’s a scenario by which they could increase the number of trains operating through Rogers without masive capital expenses. Let’s assume they could run 30 tph on each track, as they used to.

They could send the maximum number of West Side trains that the Flatbush Ave terminal could handle. The remaining West Side trains would continue and terminate at Utica Ave. The New Lots Branch would be handled by East Side trains. There are 15 of them at Utica and only 10 trains on the New Lots Branch.

This does mean unbalanced merges. There will need to be scheduled station wait times at the merges to create uniform headways on the trunk.

Fbfree December 15, 2015 - 3:07 pm

You can’t terminate the Eastern Blvd. locals at Utica. See here.

Stephen Bauman December 15, 2015 - 5:17 pm

These track maps indicate that reversing the locals at Utica is quite possible. Implementing it efficiently would require operational dexterity that NYCT has not exhibited in at least half a century.

mister December 15, 2015 - 5:43 pm

Those track maps show that it’s possible only by having the services cross in front of each other. Which means merges. Which is no better than the current system.

Stephen Bauman December 15, 2015 - 5:53 pm

It’s an improvement over the current system because there are fewer trains at Utica than at Rogers.

mister December 15, 2015 - 6:15 pm

On the other hand, it’s a downgrade from the current system because it would result in lower frequencies for Kingston and Nostrand Ave stations, and would put the merge at a terminal station where, regardless of your feelings about the practice, trains must be thoroughly checked so as to be certain they are clear of passengers.

Just doing the math, if they operated 18 peak hour West Side IRT trains to Flatbush, that would leave 5 to serve Kingston and Nostrand during peak hours. FIVE. IF they increase service, each train that goes towards Utica must merge with Lex trains south of the station.

Also, I don’t have the timetables handy anymore, but there’s no way that the 7th avenue line only has 9 tph coming from Flatbush.

mister December 15, 2015 - 3:30 pm

As explained above, you can’t terminate Local service at Utica while running expresses through. Unless… you want to merge West and East side IRT trains.

You can’t terminate all express service at Utica, because there’s just not enough terminal capacity. There also would likely not be enough local service on New Lots or Flatbush if you did that. Finally, if you wanted to send some east side service local but terminate the rest of it at Utica, 3 services still need to merge at some point.

I know it’s cool to think that everyone at MTA is a bumbling idiot, but they actually do know what they’re doing sometimes.

Brooklynite December 15, 2015 - 4:24 pm

If all 2/3 trains go to Flatbush there are two options:

1) add new switches, one per direction, to let the 4 stop at Nostrand and Kingston without interfering with the 2/3.

2) Disconnect and fill in the local track to let trains on the current express track platform at those two stops.

Larry Littlefield December 15, 2015 - 7:43 am

The problem is, crush loaded trains get delayed at stations. So if service has been cut so that trains on different branches arrive in the right sequence, the service cut on the more crowded line to achieve this would be self-defeating.

Right now you get crush loading off peak at some times.

Alon Levy December 15, 2015 - 11:11 am

Yes, but. I don’t think the off-peak trains are crush-loaded on average. The problem is that if half the trains have 50% more passengers than the guideline (87.5% more than seated capacity, quite full though not crush-loaded) and half have 50% fewer (37.5% below seated capacity), then three-quarters of the passengers see the more crowded train, and so the perceived loads are higher than the average load. Of course, for passenger comfort, what matters is the perceived load, while NYCT pays the operating costs of the lower average load levels. This is why it’s so important to equalize train loads.

22r December 15, 2015 - 10:28 am

As an aside: Simply put, if I’m waiting 10+ minutes for a train at 10pm, and the train is super packed once it finally comes (and this happens very frequently to me), then you guys need to run more trains…

Stephen Bauman December 15, 2015 - 10:58 am

Thank you for emphasizing the importance of balanced merges and uniform headways. It’s the secret that the grandfathers of today’s NYCT schedule makers used to reliably operate service levels in excess of 30 tph.

Their grandfathers did use one tool to simultaneously equalize passenger load levels that is currently lacking. They altered train length so that trains on relatively lightly used lines were shorter than trains on heavily used lines.

Brooklynite December 15, 2015 - 12:57 pm

If we’re talking about off-peak service, then the answer is simply to add more trains. We’ve all waited 10+ minutes for an off-peak train that then leaves people behind on the platform because it’s too crowded.

Regarding the peak hours, eventually subway planners are going to realize how inefficiently the system is being run today:
-First, capacity has decreased significantly over the years. Here’s a capacity map from 1954. Have a look: you’ll be surprised. http://transitmaps.tumblr.com/.....-flow-1954

-Second, interlining is a mess. Just look at the numerical acrobatics that Alon had to engage in to balance service on the various branches, or look at how many times the A/C report mentions how difficult it is to schedule everything around interlining services. Suppose the D train is found to be over capacity and needs a service increase. That means the schedules on the A, B, and N have to be changed, which in turn affect the C, Q, and R, which may propagate to the other services. It’s simply not a sustainable system, and de-interlining will (aside from preventing delays from propagating throughout the entire network) make scheduling many times easier.

Rogers Junction is not the chokepoint it’s made to be, it’s just operated in one of the least efficient ways imaginable. In 1959 96th Street on the West Side IRT was de-interlined, with all expresses going to Lenox and all locals to Upper Broadway. I propose the same thing in Brooklyn – all 2/3 trains run to Flatbush (which will presumably be rebuilt to improve terminal capacity) and all 4/5 trains will run to Utica/New Lots. A new switch on each level will need to be installed somewhere between Nostrand and the junction to let expresses cross over without interfering with the 2/3, but that’s doable.

mister December 15, 2015 - 1:18 pm

Thanks for sharing that link! That’s an awesome map.

I’m curious, what would your de-interlined service look like for Queens Boulevard?

You can’t say Rogers junction is not a chokepoint and then say that the way to de-choke it involves a major capital reconfiguration project. It’s operated the way it is operated because that’s the best way to operate it as it stands.

Brooklynite December 15, 2015 - 4:02 pm

Queens Boulevard is a tough one. In an “ideal world” there would be an 8th Av express service running QBL Local to Continental, and a 6th Av local service on QBL express to both Jamaicas. That completely cuts off Broadway (but more importantly, the in-station transfer to the 4/5) though.

Therefore we can realistically assume that the 60th St Cut must be used, and by a QBL local service (since the tracks are set up that way). That leaves ~40tph of current service to fill on QBL, and 60tph of capacity (53rd & 63rd St tunnels) to do it. You’d think that would be easy…

If we assume we don’t want two different services on the QBL express, that leaves 6th Av or 8th Av as the path for 30tph from Queens. 6th Av doesn’t work because the M isn’t a full length train, so it shouldn’t be express on such a crowded corridor. That would mean that the E would run 30tph, which would not work because of capacity issues at WTC terminal. Alternatively, the E could become the sole 8th Avenue-Cranberry St-Brooklyn/Queens route, replacing the A south of 42nd (and C in Brooklyn). Far Rockaway to Jamaica would be a massive, hugely delay-prone route though. Imagine delays on QBL because the South Channel drawbridge got stuck as it often does?

That leaves us with the need to have two different services running express in Queens, which is exactly what we have now. It’s the least-bad option I guess.

About Rogers Junction: most proposals to rebuild Rogers Junction have involved turning it into a 59th/CC-style interlocking, preserving the current service pattern. Compared to that, my idea (two new switches) is very minor. The removal of a dozen columns might require a Franklin-Utica bustitution for a long weekend or two, but that isn’t the end of the world. Something will have to be done about that junction eventually.

mister December 15, 2015 - 6:33 pm

For what it’s worth, I was told that they looked into adding the switches you’re proposing and they said it would be disruptive and expensive. Consider how long it takes right now to just rebuild the track work at a switch. Add in major signal work and the need to modify structure (removing columns often means you need to use deeper beams, and there’s probably no additional clearance in the tunnel) and it becomes clear that this would be a very difficult project. I can think of a few ways to make it happen, and I agree with you that it should happen, but it won’t be easy, especially when you need to find funding.

I thought about ‘straight railing’ the QBL, and came to some of the same conclusions you reached. In addition to what you mentioned, making the 8th ave line the express forces all of the local riders between Queens Plaza and Roosevelt to use 63rd, which is pretty… blech. You could fix that by extending the crosstown back to 71st, but then the number of 6th ave locals that could be operated goes down. So 6th ave needs to be the express service, but then that means that 53rd goes from 25ish TPH to 20 TPH. Plus, that means that WTC is seeing a level of service it still can’t handle, so some of those trains would need to be sent to Brooklyn.

The system was designed to be interlined. Detangling some lines would improve service, but others would be harmed.

Brooklynite December 15, 2015 - 7:49 pm

They also said that air conditioning subway cars was impossible. And that reopening Old South Ferry after Sandy was extremely difficult. In both cases it was done; take their word with a grain of salt.

That said, you’re right. The beam span would effectively double so there might need to be a few inches of clearance added. That’s not the end of the world though – the tracks are below the middle roadway of Eastern Parkway. The roadway can be opened up, or even raised a few inches, if necessary. Heck, it can be made a speed bump to get Vision Zero funding!

Funding shouldn’t be THAT much of a problem. The capital plan is $30 billion, and it definitely has enough bloat to cover a project as (relatively) minor as this one. That said, it WOULD need to be procured.

If the 8th Avenue line goes express, there can still be a 60th Street service on the local track. If there’s too much F service some of it can be sent up to 96/2 and give it something better than 8tph during the rush hour. 6th Avenue being the express doesn’t work for two reasons: express riders lose Lex access (assuming no tunnel from 63 to 59 is built) and the M is 8 cars (which shouldn’t be running on the express).

Regarding 8th Av, if you make the QBL express 8th Av express then you can send all of them to Brooklyn and not worry about WTC terminal. This would effectively eliminate the A/C and bring back the K, but oh well!

mister December 15, 2015 - 9:37 pm

Have you seen what an interlocking replacement costs MTA? I’m not concerned with the clearance between the top of the tunnel and the road, I’m sure that there’s room to add deeper beams/girders as needed. It’s the cost of digging up the road and installing this new structure that makes me think that the budget will get blown. Yes, it would be a small percentage of a $30B budget, but what are we going to cut out to make this happen?

For the record, I’m onboard with your suggestion. I just think it’s going to cost a lot of money that no one is willing to put up right now.

So to improve QBL, you’re thinking:

E – 179th/Sutphin to Brooklyn/Queens terminals
F – 71st Continental to Coney Island. Supplemental service to 96/2nd.
R – Unchanged.

The best thing about this is that it makes it easy to rationalize straight railing the A/D on CPW as well.

The biggest problem I see with this is that running full length, 30 tph through Cranberry to Fulton is probably not feasible with the signal changes, most likely overkill and definitely a budget breaker.

Brooklynite December 15, 2015 - 10:25 pm

It’s on the order of 100-250 million, depending on the work and complexity of the interlocking. New interlockings have the advantage of not having to remove old stuff, but the problem of all the systems being entirely new. I doubt there are too many utilities between the road surface and the tunnel, which should help. Aside from that how much can it cost to dig up a road? A few guys with backhoes and tools should be able to do it. If we assume typical MTA costs for everything, does a billion dollars sound reasonable for the entire project? As for sources of funding, if it’s not pared out of the next capital plan money could probably be found by placing harsher penalties on contractors for delays. Either they finish early, saving money, or they don’t and pay up.

Regarding QBL yes, that was essentially the conclusion I came to.
E from Jamaica to Euclid/Lefferts/Rockaways would be quite a route though, and very delay prone. I disagree about the overkill part – during the rush hours the A/C can be quite crowded even though they run 26tph (per the A/C line review). The area along the C line is starting to gentrify as well, so despite what the report says service will need to be added pretty soon. Besides, an extra 4tph is not as budget-breaking as it sounds, especially if the service will increase regardless of interlining or lack thereof.

Regarding train length, ten-car trains do not cost more to operate than eight-car (other than acquisition and maintenance costs). It’s worth noting, though, that the interlocking at Hoyt supposedly can’t handle more than 26tph, which is pretty sad to be honest. Perhaps an interlocking upgrade (=\= CBTC) is in order there.

The F and M would, in some combination, serve Continental and 96/2. QBL local would see a service boost, unless it’s deemed reasonable to give SAS over 20tph. For three stops that does seem to be too much.

The R would stay as is.

Regarding CPW, yes, that would be straight railed as well. It’s actually the perfect place to do that because of the cross-platform transfers on both ends of the route. With the A and C eliminated, some combination of the B, D, and K would need to provide service to Inwood and the Concourse line. The B and D would most likely run express to 145th, with the B going Concourse express and the D making all stops from there to Inwood. The K would be left as the full-time Concourse local to 205th. The B/K interlining is less than ideal but there isn’t much of a better option – sending the K to 207 would leave everything north of 145th with two locals (1 and K) while giving the Jerome/GC corridor two expresses (4 and D). This evens that out somewhat.

mister December 15, 2015 - 11:47 pm

I’m thinking the $250 Million is the low side estimate for an MTA project, but without knowing what’s in the street, it’s hard to call it.

CPW could simply be served by two services:

A – 207 to WTC via local
D – 205 to Coney Island via express

When you straight rail the interlockings, you’re not worried about a particular branch not having express service; you’re running direct trains and people can transfer if they want/need to.

The problem with this strategy in NYCT is that it results in additional frequency in places where you just don’t need it. For example, with sending a 30tph E train to Brooklyn: this would result not just in an extra 4 tph going northbound in the AM peak, but also an extra 10 tph coming into Brooklyn during that same time period. Looking at the D line that I proposed above, You are probably operating 20+tph into Brooklyn onto a branch line that only needs half that much service.

That could be rectified by having a more heavily used branch on the northern end matched up with two less used branches on the southern end (i.e. – the D serves both West End and Sea Beach), but if you can run a relatively efficient operation, then you’re offering a better service to those customers if you reverse branch, which brings us back to where we started…

Brooklynite December 16, 2015 - 9:32 am

$250 million is the most expensive interlocking rehab in that link you posted, I’m assuming the entire job would be around a billion. Not pocket change, but not some inordinate sum either.

And if you’re straight railing the interlockings, you may as well give people the service they want. If >50% of the people will be transferring you’re doing it wrong.

And regarding the D having 20tph, don’t forget that half of that will have to run via Brighton Express (unless, of course, you deinterline Gold St Junction too…). And the southbound QBL expresses could layup in Pitkin Yard, instead of turning around at WTC or going, virtually empty, to Coney Island, as they currently do. There is not that much more half-empty running.

mister December 16, 2015 - 9:23 pm

By removing the one-seat options, you’re practically guaranteeing that more people will transfer. I.E – on WPR, de-interlining would result in a lot more transfers than presently exist.

It’s a lot more empty running because you would now have:

30 empty tph through Cranberry,
20 empty tph coming down from the QB Local onto 6th ave and down to Culver.
Both lines would now have more empty reverse peak trains than they presently do.

I don’t see the point of the de-interlining exercise if we’re only going to do one line.

Brooklynite December 17, 2015 - 9:46 am

Obviously de-interlining will introduce some transfers, but what I’m saying is that if more than 50% of passengers are transferring you should probably flip the services, as is feasible on CPW because of “Homeball Alley” junction south of 145th. On WPR yes, there would be a significant amount of transferring at 149th/GC if the 5 were eliminated, and I’m not sure if the station can handle it.

And regarding the empty counterpeak trains, yes there would be some extra half-empty running, but it wouldn’t be as bad as you’re saying. Currently (during the AM rush for simplicity) the QBL expresses either turn at WTC, becoming empty counterpeak trains to Queens, or run counterpeak all the way to Coney Island. Running them to Euclid or Lefferts/Rockaways instead is not that much of a change. Yes Cranberry would see a service boost, but it’s more of a reroute than an addition of empty trains IMO.

Larry Littlefield December 15, 2015 - 2:23 pm

Great map.

Safety improvements added since then have reduced capacity no doubt.

But one thing for sure. We miss those 250 or so cars per hour (which would be 25 subway trains per hour) on the Third Avenue El in Manhattan.

adirondacker12800 December 15, 2015 - 3:56 pm

El trains weren’t ten cars long.

Stephen Bauman December 15, 2015 - 5:45 pm

For the record the BOT reported that in August 1949 the Third Ave El operated 264 cars in peak service. That was 42 tph. There were 21 locals with 117 cars terminating at South Ferry and 21 expresses with 147 cars terminating at City Hall.

Larry Littlefield December 15, 2015 - 6:09 pm

The map includes the number of cars as well as trains, and has 251 cars.

Mike December 16, 2015 - 3:51 pm

Probably the easiest place to de-interline would be Gold St Junction, where the B, D, N and Q trains converge/diverge, so I think that would be the best place to start. One option could be to swap the D and Q lines, so the Q goes via the West End Line and the D returns to the Brighton Line Local. This would certainly require a lot more people to transfer at DeKalb and Atlantic, although DeKalb would once again have a 6th Ave service stopping there 24/7 (it currently doesn’t on weekends when the B doesn’t run).

mister December 16, 2015 - 9:46 pm

This would also probably increase Gold’s capacity. However, the problem is that West End/Sea Beach Riders who wanted sixth ave would either need to make a really inconvenient transfer at Pacific-Atlantic, or transfer to the R first. Brighton Riders who want Midtown Broadway would either need to make the same inconvenient transfer, or ride the R through Lower Manhattan.

Fbfree December 17, 2015 - 9:45 am

Other than for customers taking the train to/from Canal St or Grand, how much of a problem is this extra 1 block walk? Through customers can connect at Herald Sq.

Mike December 17, 2015 - 1:04 pm

Or another option for decongesting Gold St Junction could be to switch the B and N lines, so that the 4th Ave line can still have direct access to both Broadway and 6th Ave. That would also involve less switching at Gold St, so it would be more “straight-railed.” Fewer switches would need to be used if the B and D both run on 4th Ave and if the N and Q both run on Brighton. Brighton Line riders keep their direct link to the Broadway Line with the N and Q trains and West End riders still have a direct line to 6th Ave on the D, although Sea Beach Line riders who want Broadway would have to transfer from the B to the N, Q or R at Pacific-Atlantic.

With both the B and D skipping DeKalb, that would result in DeKalb Ave being served by Broadway lines only. Also, if the B and N trade routes in Brooklyn, they’d most likely also have to trade service hours. That would also affect the current service patterns on the Q (and possibly the R) as well to compensate for the loss of the N on weekends and late nights. The new patterns can go something like this:

B – Bedford Pk Blvd to Stillwell Ave via 4th Ave express and Sea Beach Line during rush hours. Middays/Evenings/Weekends terminates at 145th St. Late nights terminates at 36th St.

D – Unchanged

N – Ditmars/Astoria to Brighton Beach via Broadway local and Brighton express, weekdays. Because the N would be the new Brighton express, it wouldn’t be needed in Brooklyn outside of weekdays. So weekends and late nights, the N could run from Ditmars/Astoria to Canal St, terminating on the upper (R) level.

Q – Unchanged, except it would now run via Lower Broadway and the Montague Tunnel late nights, replacing the N.

R – Unchanged, except the late night shuttle would be extended to Pacific to connect with the Q, 2 and 4 trains since it would no longer have a cross-platform transfer to the N.

If this is unworkable or unacceptable, then I guess we’re just going to continue to have to put up with the numerous switching delays we currently face on the B, D, N and Q lines at Gold St Junction.

aestrivex January 4, 2016 - 3:40 pm

Where would it turn at Canal? I thought the inner tracks at Canal leading up to the lower level at city hall are permanently decommissioned.

The nearest place that I can think of to turn the N service you are describing is whitehall, which would make it exactly the same as the former W.

Also, since the entirety of the route is superseded by the Q, most likely the MTA would just not run the N off peak under this configuration, just as they didn’t run the W and don’t run the B.

Brooklynite December 17, 2015 - 8:02 pm

I’ve thought about this and there’s a relatively simple solution to the problem of the transfer at Atlantic.

Near Dekalb Av, the tracks that the R run on first rise from a lower level, then platform, then descend back to the lower level. If these tracks were kept on the lower level throughout, Dekalb station would have a new pair of platforms added under the current ones. The lower level platforms would serve the R, while the upper level platform would be extended over the former R tracks (now the space above the R tracks) and give the bypass a platform. Voila, cross-platform transfer between 4th Avenue and Brighton services.

With that construction completed, Gold St can be de-interlined without riders complaining about the ridiculous transfer between trunk lines at Atlantic (as would otherwise kill the idea). The N and Q, both full-time lines, would serve Sea Beach and West End respectively, while the D would go back to Brighton Local and the B would remain the part-time Brighton Express.

P.S. Calling the R “access to Broadway” is stretching it. If I’m at the 4th Av platform at Atlantic and my choices to get to Union Square are taking the R or walking all the way to the Brighton platform and catching the N/Q, the N/Q would win. The R’s headways and tunnel route are slow.

Michael December 19, 2015 - 10:48 am

From a previous message:

“If I’m at the 4th Av platform at Atlantic and my choices to get to Union Square are taking the R or walking all the way to the Brighton platform and catching the N/Q, the N/Q would win. The R’s headways and tunnel route are slow.”

If I’m at the Atlantic Ave/Pacific Street stop on 4th Avenue, and I need to go to Union Square but I don’t want to take the R-train all of the way there, but would rather take the N/Q that’s been re-routed to the Brighton line. I’d simply take that R-train ONE STOP to DeKalb Avenue for an “across the same platform” transfer to the N/Q trains RATHER than trying to walk through the crowds, plus the up and downstairs to get to the N/Q trains.

Every train route that travels along the Brighton Line MUST stop at DeKalb Avenue, it has been that way since the mid-1960’s!

Why make things complicated?

Just saying.

PS – It is truly doubtful that pairing the N & Q trains along the Brighton line will ever happen. Many of the “proposed inter-lining schemes” really do not make it better for the riders. The subways can be difficult enough – why make things even more complicated?


Brooklynite December 19, 2015 - 11:30 am

Waiting for that R train to take it one stop would take quite a while, and potentially be even slower than walking through the passageway and catching the Brighton trains. Although if the system was theoretically fully deinterlined, the R would be the sole Astoria service and would presumably run more frequently.

The point I was trying to make was that transfers between 6th Avenue and Broadway trains would be easier with the new platforms at Dekalb. My thoughts about your question regarding the necessity of deinterlining are on your other post, below.

Alon Levy December 16, 2015 - 11:52 pm

Over at my place, one of the commenters notes that in Tokyo, the heavily interlined Hanzomon Line sees many delays, while the non-interlining, non-branching Ginza Line sees none.

Michael December 18, 2015 - 12:07 pm

After reading all of the various messages and responses – I am beginning to think that the current system of subway routes, inter-linings, and route services is the best possible that can be obtained given the circumstances.

I started out wanting to respond that there was a difference between “theory” and “practice” – and that what the folks at the MTA have is the day to day “practice” where beyond all “theories of how things should work” they are there doing to the day to day work.

Considering many of the various ideas of “de-interlining” the system to “make it better” and just how few of those proposed ideas actually consider the effects and dis-advantages to riders that would actually have to ride such grand schemes – I say leave it alone.

All in all, this message stream, to me anyway, sounds like a “leave well enough alone” situation.


Brooklynite December 19, 2015 - 11:53 am

Obviously what is best in theory may not work best in practice. People are accustomed to their routines and disrupting those has a cost.

At the same time, while interlining everything as is done today may work in theory, it doesn’t do so well in practice. Even today, with the only place on the system serving 30tph (aside from that brief segment of Rogers Junction) being the Queens Boulevard express tracks, there are plenty of delays due to “train traffic.” It’s possible to schedule trains to arrive at merge points no less than 90 seconds apart, but sticking to that kind of a schedule is something we, unlike the Germans or Japanese, have not managed. Even assuming perfect OTP, the numerical acrobatics that schedulers need to perform to get everything to line up are somewhat astounding, as referenced several times in the A/C report. A typical C train has to fit in service gaps in the A, E, and B services on each trip, and those services in turn affect others. The reduced capacity that such complex interaction between services entails is most likely no longer feasible as ridership continues to skyrocket.

Reliability, both of service and of trip time, is another concern. If there’s a delay on one line it will affect service along others, which will affect the trains still running along those and delay those too. Making the different services more self-contained will perhaps worsen the initial delay (if the dispatchers decide not to run a shuttle on the unaffected segment) but will improve recovery because there will be fewer service patterns to juggle. Trip time will get longer in some cases, that is true. However, it will be predictable. There will no longer be the uncertainty of whether the first or third train to come will be the one you need, because they will all be going to the same place. You may need to transfer, but the trains will come more often and you will always be getting on the first train as opposed to waiting for another one.

The status quo is, for the short term, holding up. However, as ridership continues to skyrocket and capacity needs to be increased without waiting for projects like SAS, new solutions need to be considered.

Mike December 21, 2015 - 4:35 pm

What disadvantages would those be? More frequent trains throughout the day? Not having to stop in the tunnel at Gold St junction to wait for another train to merge onto the same tracks as yours? The current setup may look “well enough” on paper, but what’s the point of a one-seat ride if it requires you to wait in the tunnel to let another train either pass through or merge onto the same tracks as your train? And on trains that run less frequently due to limitations placed on them by all the merging.

adirondacker12800 December 21, 2015 - 6:05 pm

Waiting 45 seconds in the tunnel is shorter than waiting 120 on the platform to change trains?

Alon Levy December 21, 2015 - 6:28 pm

The 2 extra minutes of transfer time cancel out with 2 extra minutes of waiting at the beginning of your trip for a less frequent one-seat ride. (Think why.)

adirondacker12800 December 21, 2015 - 8:51 pm

How much is it going to cost to rebuild the platforms at the stations where all this merry cross platform transferring goes on, to accommodate the crowds?

Nathanael December 22, 2015 - 12:12 am

DeKalb Avenue is plenty big enough.

Queensboro Plaza is too, and needs reconstruction anyway.

Michael December 26, 2015 - 7:06 pm

As just a simple online transit fan there’s one thing that I’ve learned about other transit fans.

Rule #1 – Transit Fans Pitch & Moan About Everything!

To the statements that some folks wants to “rationalize” and “de-interlink” and ….

Go back, and re-read Rule #1!

Did transit fans “pitch and moan” loudly for months on end about the removal of the F-train off of 53rd Street by the MTA in favor of the V-train? YES, TRANSIT FANS PITCHED AND MOANED BIG TIME!


The rationality, the rightfulness of the change, the de-interlinking, etc. – plenty of transit fans were hearing none of it! Transit fans pitched, moaned and groaned about the V-train! Transit fans pitched, moanded, groaned and gave birth to litters of kittens with the complete removal of the G-train off of the Queens Blvd line!

Did transit fans “pitch and moan” loudly for months on end about the switching of the B and D trains between the West End line and the Brighton line in Brooklyn? YES! AGAIN, TRANSIT FANS PITCHED AND MOANED BIG TIME FOR MONTHS ON END! There were some who were stoking the fires under their boiling cauldrons they were that mad!

Did transit fans “pitch and moan” about the switching of the northern terminals of the B and C trains, ending forever the symmetry of the A & B routes to Washington Heights, and the C & D routes along the Concourse line in the Bronx? There were plenty of appeals to history, etc. YEP! Are there other examples that I could cite, if given enough time? YEP!

The bottom line is simple – every new-fangled idea of different route patterns can be defended on the grounds that they are “rational”, “they are the right thing to do”, and that one wants to provide “better experience for the riders” and that by “de-interlinking the various routes we will achieve ….”

I again refer to Rule #1: Transit Fans Pitch & Moan About Everything!

Rule #2: The number of proposals or schemes to “improve” or “re-arrange” the Central Park West segment of the IND lines in Manhattan will be two to three times greater than the number of proposals or schemes to “improve” the IRT number lines. While the west-side IRT number line stations handle twice the ridership at all hours compared to the nearby IND Central Park West segment of Manhattan – every year there will be several proposals to “rationalize” Central Park West train routes. Is this due to the greater possibility of schemes or the needs of the real every-day riders?

Rules #3, #4 and #5 will be revealed at another time. Happy Holidays!


Leave a Comment