The trains are slow because the MTA slowed the trains, redux


In about a week or so, Andy Byford is going to reveal his big NYC Transit subway rescue plan to a public anticipating a Big Idea. Byford was brought in specifically to build this plan and execute on turning around the struggling subway system. It won’t be easy, and one of the major obstacles in Byford’s way is New York City Transit itself. The agency often can’t seem to get out of its way, and many of the current problems with fast and reliable service are self-inflicted.

One of the biggest problems, as I discussed in mid-March, are signal timers slowing down service throughout the city. These timers were a reaction to the 1995 Williamsburg Bridge crash, and in March, Aaron Gordon of the Village Voice explored how the MTA did not understand the effect the timers would have on capacity and service. A study nearly 20 years after the fact betrayed the MTA’s problems. “The 2014 study — the first time the authority had attempted to analyze the impact of any of the revamped signals, using its improved data system — found 2,851 lost total passenger hours per weekday could be attributed to thirteen modified signals alone. That was almost double the predicted impact; for comparison, the modifications of [] thirteen signals alone created 5 percent as much lost time as that experienced by riders of the entire London Underground on its average day,” Gordon reported, based on internal MTA documents.

This past week, The New York Times revised the issue with signal timers in an easy-to-understand graphic explaining how signal terms slow down service and decrease through capacity on the subways. It’s well worth your time to play with the interactive interface, and it’s worth remembering that the capacity of the system cannot exceed throughput at the slowest choke points. The Times piece delves into how the subways no longer have extra capacity because of the intentional choices the MTA has made over the past few decades. Here’s Adam Pearce on the problem:

The M.T.A. projected that the signal changes would not reduce the number of trains that could pass through a section of track each hour. But this assumed the signals would work properly and that trains would operate at the speed limit. In reality, many signals are poorly maintained and misconfigured, triggering emergency braking at speeds below the listed limit. An unpublished 2014 internal M.T.A. analysis, first reported on by The Village Voice, found that the signal changes caused a significant slowdown, more than the M.T.A. expected. Train operators face steep penalties after a number of instances of tripping a signal, like losing vacation days or being forced into early retirement…

The analysis stated that if the M.T.A. had known the signal changes would reduce the number of trains able to run on congested lines, they would not have been made. But the damage was done. After the signal changes, two fewer trains could run on the southbound 4 and 5 lines hourly, forcing the thousands of passengers those trains would have carried to squeeze into already crowded cars. Across the entire system, more than 1,800 signals have been modified since 1995.

To me, this graphic is the biggest indictment of all.

These stations in Lower Manhattan are absurdly close together and largely along straight tracks. A train operator on a downtown 4 or 5 train can see each station from the one before it, and yet, the signal timers add 15 seconds per trip from Fulton St. to Bowling Green. Over the course of a line, this adds up to a significant constraint on capacity, and delays due to “overcrowding,” an excuse the MTA has hidden behind for years.

The success of Byford’s plan will hinge on how he treats and responds to these signal timers. It’s guardedly good news that he has, as Jon Weinstein said to Pearce, “asked for an analysis of the impact of signal modifications on subway schedules.” But it’s not enough to ask; he has to respond and fix the problem (without sacrificing safety). But more on that — and the new flagging rules The Times noted — in a follow-up post.

31 Responses to “The trains are slow because the MTA slowed the trains, redux”

  1. Grandpa44 says:

    No surprise with politically controlled meta
    Another big delay is “sick passenger” – this should be limited to 5 minutes.

  2. John Landers says:

    Thank you to former train operator Robert Ray who killed several in his 14th St Union Square 4 train accident for makinh it difficult on everyone.

  3. John Landers says:

    Thank you to former train operator Robert Ray who killed several in his 14th St Union Square 4 train accident for making it difficult on everyone.

  4. TrackInspector says:

    Rather than complain about the system you should be more productive and try to understand it. I could tell you exactly why there are so many delays, but I bet you would never believe it. Keep your blog truthful instead of assuming things.

    • Robert E Lee says:

      Tell us why
      I want to know
      Not being sarcastic

      • TrackInspector says:

        Efficacy in moving the public isn’t solely based on a signal system… The track system is aging and not enough attention is being placed on our department which is what the entire structure is based on. Focusing on renovating stations because they don’t look good is fallacious but the riding public rather see pretty tile work than a great ride… Signals control the timing and traffic but do you take into consideration other variables aside from “sick passengers” which is also something that cannot be timed… Such as track fires which are in fact caused mostly due to passengers throwing garbage on the track, dropped property which is also something that causes trains to be slowed down because transit doesn’t want to have customers jump onto the track to retrieve the property… Why do you need to have your phone out as you’re walking into the train only to “be pushed” and drop it into the track. Listen I can go on and on about all the different things that exacerbate traffic but to digress and reiterate the track infrastructure needs to be worked on. Transit needs to have less bureaucracy and allow the work to be done. Oh and the last comment about cbtc is totally correct and guess what that’s for safety it is quintessential to have in place. Please feel free to ask anything else. Sorry about the grammar and such I’m writing as I drive between lights.

        • tacony says:


          Why do you need to have your phone out as you’re walking into the train only to “be pushed” and drop it into the track.

          Sorry about the grammar and such I’m writing as I drive between lights.

          • TrackInspector says:




            That’s the typical transit rider isn’t it ? Or is it a typical millennial ? Ehh who cares in the end just a sad soul.

        • AMH says:

          Good God, why are you writing while driving?

        • JarekFA says:

          Focusing on renovating stations because they don’t look good is fallacious but the riding public rather see pretty tile work than a great ride.

          The public didn’t want the ESI — the MTA board members most responsive to the riders, those which were appointed by the City — all opposed and pushed back on ESI. We want the signals to be fixed. You can say “they’re old and ancient” but when the ancient signals break down, trains can still run past them (albeit slowly); whereas when the less than 12 years old Bergen St signal breaks down, trains can’t run past it at all, causing system wide delays instead of just pushing everything back by 10-15 minutes.

          Also, just up the maintenance budget a little to get us clean tiles. It’s ridiculous that Chambers and City Hall stations look like background settings for horror films. You go to Moscow, and the subways look like museum halls.

          • TrackInspector says:

            I agree and again my argument is that it isn’t based soley on signals… There is alot of red tape between RTO Track and everyone else… RTO doesn’t let track department do their jobs. Furthermore signal is always trying to blame track for everything because they want us to do their job. Track is the red headed step child and we get scraps of the food don’t get too eat out of a plate or utensils… But they expect us to have table manners and wash the dishes oh and also pay for the meal ! Small analogy… When you buy a car doesn’t matter what kind of car you have to do maintenance not just take it to the car wash and make it look good.

            • smotri says:

              I’m sure you are right, but when I listen to the news radio stations before I go to work each day, I hear a lot about signal problems causing delays, not so much about sick passengers, etc. Faulty/broken switches also is another ‘popular’ cause of delays, it seems.

            • John-2 says:

              The desire to make the visible parts of the system more attractive at the expense of spending money on preventive maintenance pretty much dates back to the dawn of the MTA, when William Ronan’s answer to the increasing unreliability of the rails, signals and rolling stock was to re-tile the walls of the BMT’s main line, and repaint all the cars (including the by-then 30-plus-year-old R-7/R-9 trains) in the new agency’s corporate colors.

              Style over substance — paint the trains but don’t do anything to the moving parts of the trains; re-tile the walls but ignore the rails that get the trains to the stations. The graffiti epidemic of the 70s made the MTA’s skewed priorities even more absurd — they couldn’t even maintain the Potemkin Village scheme to convince passengers the system was being maintained via fresh paint and wall tile, and it took the near total-collapse of things by the early 1980s to finally force the politicians to commit real money to solving the worst of the problems. Unfortunately, that’s likely going to be the situation here; until the reliability of the system gets so bad politicians fear for their jobs, the financial support and motivation to fix those problems isn’t going to be there.

              • smotri says:

                Eventually, I will not want to stay in NYC. The mass transit system is a farce, and my tax dollars seem to go into a black hole. I – and many other New Yorkers – am not getting value for my tax dollars.

        • TimK says:

          Transit needs to have less bureaucracy and allow the work to be done.

          What bureaucracy? I think most of us are under the impression that funding is the big constraint on getting track work done.

          Sorry about the grammar and such I’m writing as I drive between lights.

          LOL! Safety first, right?

          • TrackInspector says:

            Us track guys live pretty short life spans. All those metal shavings from the trains running don’t let us think straight…

          • TrackInspector says:

            It’s silly to make one department pay another just to be able to do work… I never understood this can anyone explain this practice to me ? Our department does the most work we keep the trains on the track and we get most of the run around just to get the job done. Track is not properly managed more so today than it was 10-15-20 years ago, too much of a liberal agenda is what I blame it on.

  5. J Adlai says:

    Where these timer signals were installed to prevent collisions due to inadequate stopping distance (similar to the Williamsburg bridge incident), the effect could have been achieved by using Station Timers instead of Grade Timers. The timers would have only been active when there was an actual potential for an accident to occur.

    What might be useful to look at would be a real examination of L train running times from end to end both before and after the activation of CBTC. CBTC largely eliminates timers, as you can set an enforced speed limit on curves and trains communicate with one another, so timers due to inadequate stopping distances are unnecessary. That would give everyone a good approximation of how much time is being lost to signal timers.

    • AMH says:

      Good point about ST timers–I suspect one-shot GT timers were used because they are much simpler, which is also why they’re causing so many problems. Reactionary measures, invalid assumptions, and half-baked implementation.

  6. Stephen Bauman says:

    Timers will increase trip time. More rolling stock is required to provide the same service level (trains per hour or seconds between trains). According to information the MTA provided to the National Transportation Database, the spares ratio was 0.8% in 2016 vs. 18% in 2006. The reason is a reduction in the number of railcars available for maximum service from 6202 to 5365, while the number of rail cars required for maximum service increased from 5253 to 5324. The result has been that scheduled trips don’t leave the terminals. On a good day, only 95% of scheduled trips actually start operating, during the morning and evening rush hours.

    Where timers are placed is more important service level capacity than the timers’ presence. To delay a following train, the leader’s rear must be less than the safe emergency braking distance (300 ft). Suppose there is a 5 mph slow zone. A leader must travel 900 ft (600 ft long train + 300 ft emergency braking distance) to assure the following is not delayed by the leader in the slow zone. At 5 mph, it will take the leader 122.4 seconds to travel that distance. If the headway between leader and follower is more than 122.4 seconds (29.4 tph), the follower will not be delayed by the leader’s presence in the slow zone. Only the E and F on the Queens Blv Line have higher scheduled service levels. If the slow zone were 10 mph, the threshold becomes 61.2 seconds or 58.8 mph. The best documented service levels were on the Third Ave El, which boasted 42 tph back in 1949 (well before CBTC).

    This analysis breaks down at the entry and exit of an intermediate station. The follower will gain on its leader whenever the follower’s speed exceeds its leader’s. There are 3 distinct periods when this happens in an intermediate station: leader is braking braking, leader’s dwell time in the station and leader’s acceleration leaving the station. Each of these components has a nominal value of 30 seconds with a nominal travel speed of 30 mph and an service braking and acceleration rates of 3.0 mph/sec. This means that 90 second headways or 40 tph service levels should be possible. The theoretical service level capacity is in the 42 to 45 tph range. That’s why the #7 Flushing Line managed 36 tph before the 1970’s fiscal crisis caused service reductions.

    The terminal geometry may further reduce service level capacity. A terminal by which trains must change direction requires more time because entering and exiting trains compete for the same switch. The usual solution is to have multiple terminals or a loop terminal. The 8th Avenue terminal on the 14th St Line limited service levels to 24 tph, which is 20% greater than what’s currently operated on this line. Insufficient electrical power is the excuse du jour for not operating more trains on this line. Service levels are supposed to be increased to 22 tph, when 3 new substations are built during the 15 month closure.

  7. AMH says:

    The Times article and animations were very informative, and I’m glad this is all coming to light. 2,851 lost passenger hours per weekday is insane–that’s essentially equivalent to killing a passenger every 245 weekdays! It’s time to begin respecting passengers’ time by uncrippling the system immediately.

    • AMH says:

      Let me revise that–it’s the same as killing a small child every year, or a middle-aged adult every 6 months. This is a huge amount of our lives being taken from us (over 80 years collectively per year).

  8. Klaus K says:

    I think it is unfair to blame transit riders for the mess that the subways are in. The fact of the matter is that if the MTA has invested properly in maintaining the system, the system would be functioning better. I do not believe that there is all that much pressure from riders for pretty stations. Given a choice, riders would surely favor non-cosmetic improvements that would speed service and make it reliable over prettifying stations or adding Wi-Fi. It’s the politicians – and Cuomo especially – who like to focus on station upgrades because they offer photo-ops and ribbon-cutting opertunities.

    • TrackInspector says:

      And it makes it seem like they’re doing something… I remember one day one of those guys seeing something that “didn’t look normal” and they pulled 20+ men to go and make the corrections. This is the same situation in Washington with all these politicians talking about things they just don’t know anything about.

    • smotri says:

      Didn’t Cuomo even have the nerve to say that because of the $800 million or so ’emergency’ plan we were already seeing improvement? What improvement? Where?

  9. John says:

    Okay, let’s retime, remeausure switches and work towards efficiency. Continuous improved by repeated measurements at main and subsequent switches; hire more seasoned workers.

  10. TrackInspector says:

    Has anyone on here been on a track while s train is doing 40-45 mph passed you ? Or one uptown and one downtown at the same speed ?! Safety is there for a reason.

    • johndmuller says:

      Is it possible to modify the safety regulations so that there are different categories of jobs requiring different levels of protection? That way, some of the slowdowns might be lessened if, for example, only the track on one side of the work zone needed to be restricted; perhaps there are jobs focusing on the center of the trackbed which do not require either side to be restricted – or different levels of restrictions, still restricted, but to a greater or lesser degree.

  11. t-bo says:

    Then there are the issues before you get to the trains…such as the (already) broken escalator at 96th St Q. Let me guess: replacement parts are no longer manufactured to this spec, 18 months later.

Leave a Reply

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