I often pity the poor F train. It reaches from Coney Island to Jamaica, Queens, with an extended local stint in Manhattan, and touches the lives of so many people. Yet, these straphangers by and large can’t stand the F. It’s slow; it’s crowded; it doesn’t offer Brooklyn express service; and the waits can seem interminable.
Last year, under pressure from State Senator Daniel Squadron, the MTA conducted an internal review of F train operations, and the results were decidedly not pretty. With old rolling stock, poorly maintained infrastructure and decaying assets, the F line was literally falling apart. A year later, a follow-up reveals that Transit has greatly improved the F, but the line still lags far behind the average subway performance.
Michael Grynbaum of The Times secured a copy of the follow-up report, and I’ve posted it in full after the jump below. It contains few surprises and perhaps a few twisted numbers as well. The F is doing better, Transit insists, but it’s still not there. “It feels like you get the diagnosis again, and it’s not specific about the solution,” Squadron said to The Times.
In general, the various metrics that tracked the F train’s performance showed marked improvement. The weekday on-time rate is now at 78.1 percent, up from 62.3 percent in 2009 but off the systemwide mark by over eight percent. The weekend and on-time assessments show similar improvements, but the MTA cautions that it has changed the way it calculates on-time performance to “improve transparency and to more closely align measures to customer experience and management priorities.” The trains aren’t necessarily more on time in an absolute sense, but they are on time if we consider only the MTA’s tracking methods.
Most notable is the vast improvement in the mean distance between car failures. Over the last year, the rolling stock options along the F declined from five — with some of the oldest cars running along this route — to two, and thus the MDBF shot up from 148,257 to 703,159 in July. “Improvements in F performance statistics over the past year can be attributed to the replacement of older R46 cars on the F train with newer R160 cars, as well as the completion of trackwork in Queens that had been underway for several months in 2009,” says the report. “The replacement of V with M service in Manhattan and Queens may also have positively affected performance of the F train.”
Ah, yes, the semi-controversial, semi-obvious M train rerouting. The report couldn’t assess the F train without contemplating the V, and it does so in depth. Transit believes the elimination of the V train and the replacement with the M between 47th-50th Sts. and Broadway/Lafayette St. can be a boon for the F. No longer with V trains interfere with F train operations at 2nd Ave. as the M bypasses that stop and trains no longer terminate there. Additionally, the reduction in the number of passengers transferring between the J/M/Z and the F at Delancey St./Essex St. has speeded up trains into and out of that station. As a result of the M train’s continuing along the 6th Ave. line, the F trains are also relatively emptier out of Delancey/Essex as well.
At some point in the near future, we’ll have enough data to assess fully the successes of the new M train, but so far, the results have been positive. Anecdotally, the northbound M trains leaving Broadway/Lafayette and West 4th St. have been more crowded than the V trains were, and with so many passengers from Middle Village and Brooklyn bound for Midtown, the one-seat ride has proven to be a welcome addition to the subway map. It’s a change that should have come sooner. But I digress.
F train load analysis results for the AM rush.
Finally, Transit is still working to assess load demands along the F at the AM and PM rushes. The tables above show that train car crowding levels are within acceptable parameters, but again, it’s worth noting that the MTA has changed its load guidelines since the initial F assessment. In 2009, a train was considered 100 percent full if every seat was taken. Today, a train is considered 100 percent full if every seat is take and around 15-18 passengers are standing. Thus, at times when the F cars are at 80 percent capacity, there are no seats to be had.
Ultimately, the F is still suffering from years of neglect and few adjustments to the operation schedule as the train’s popularity has peaked. Transit didn’t need to add more layers of management as they threatened to do last year, and key capital improvements and signal modernization programs are still awaiting funding. Concerns over the timetable and frequency of off-peak service haven’t been addressed, and the F is still very much a work in progress.
Postscript: For those of us still holding out hope for an F Express train in Brooklyn after the interminable Culver Viaduct project wraps in 2014, Transit has promised at least a feasibility study. “The review of express service in Brooklyn will be undertaken closer to the completion of the Culver Viaduct project,” the document says. There’s hope for it yet.
After the jump, read the F train review follow-up in full.
26 comments
It will be interesting to see how the MDBF on the R-46 cars looks a year from now, after they’ve been serviced by Pitkin Yard on the A for a while. If the problem was those trains were just getting old and less reliable that’s one thing; if it’s because Jamaica Yard wasn’t doing a good job maintaining them, then the F’s MDBF will start trending down again, as the R-160s begin to age.
Meanwhile, it’s hard to see how you ever get F express service back between Jay and Church now that the M service has been established, since it would create the same complaints from people in Carroll Gardens and Park Slope had 40 years ago. The original thought was to extend the V rush hours at least to Church and either have it or the F serve as the Brooklyn express, but the revised M service makes that impossible.
A split express/local F from Church (or Kings Highway) to Jay, as was done from 1968-76, would mean less one-seat ride rush hour service from Manhattan for local station customers, while the AM rush would irk passengers at Bergen trying to decide whether the F express downstairs or the F local upstairs is going to arrive first (though the new electronic arrival boards could solve part of that problem). And it would be even worse if you had full F express service while the local stations were served by only the G, with the track layout at Bergen making it impossible to do it the other way around.
John raises the same points about the Culver express that I was going to make.
The new M service is going to have staying power. It fulfills the original promise of the Chrystie Street project, providing a one-seat ride to midtown for the Jamaica/Myrtle customers, which has been lacking since the K was discontinued decades ago.
But without a V train operating in Brooklyn, I can’t see how you would institute a Culver express without a seriously disadvantage to customers at the local stops, as John has outlined.
Why would anyone want it the other way around?
The downstream Brooklyn F service is terrible. This has lead to reduced property values and density along the F line south of kings highway. The slow F service adds riders to the overcrowded Brighton line and the N line that has seen big ridership gains due to the large Chinese community along it’s Brooklyn Route
IF the Mayor wants to take more people out of cars the mta should consider running the F and D line as one large loop through stillwell ave Coney Island. F train enter stillwell and exit up the west end line and D train enter stillwell and ride up the F line. This will shave off time big time for inter Brooklyn travel and reduce wait times as train enter and leave stillwell terminal .
It would take me out of my car as it would reduce my commute to work by 10 (mass transit that is ) . Many MANY people transer from the D to the F at stillwell in the morning having to transverse two flights of steps and multiple delays
It’s a pretty big detour to through-route lines like that. The line would look like a gigantic loop. It doesn’t make it a bad idea, and you’re right that the MTA should look into it, but it may not be the greatest.
In the medium term, the best solution for that area is to use the Bay Ridge freight line’s ROW to build rapid transit cheaply; it would intersect both the D and the F, connecting not just those lines but also Brooklyn College, Broadway Junction, Middle Village, Elmhurst, Astoria, and Yankee Stadium.
I’d classify that as long term, not medium term. A new ROW would need to be found for the existing freight service; tracks and signals would need to be brought up to passenger standards; stations (ADA!) would need to be built; rolling stock would need to be purchased (what type of car would the line use? where would cars be stored and maintained?); etc.
It’s not a bad idea, but there are a lot of details that need to be worked out.
Freight could keep using the same ROW – it’s wide enough nearly everywhere, and could be widened at the few other sections for 1-2 orders of magnitude less money than building a subway. The only real problem section is Hell Gate Bridge, where CSX would need to be moved to the underused Amtrak tracks. (Not a problem ever, with decent time separation.)
The rolling stock costs some money, but the MTA buys trains regularly; it could add a little bit for the order, and it wouldn’t be too much. The cost of this would be dominated by rolling stock and short tunneling near Yankee Stadium, which could be left for a future phase in a crunch.
Everything else is pocket change. The Germans build accessible commuter rail stations, with medium-height (about 2.5′) level boarding, for $200,000 each. It would be more expensive in New York because of the need for two platforms and fare barriers, but not by enough to make a dent in the budget. The actual laying of tracks is dirt cheap, too – when the ROW is already there, the Germans do it for $1-2 million a mile. It would be more expensive in New York, because it’s urban and in a cut, but again not by enough to be worth much. Electrification built to high-speed rail standards is around $3 million per mile, and state of the art signals are about as costly.
I’m not as convinced as you are of the costs.
Nor am I convinced that there’s much of a market.
Do you know why Paris ultimately decided to build its T3 tram at street level rather than in the Petite Ceinture cut?
The market for such a service is tiny. The tracks aren’t set up for it at Stillwell, and even if they were, it would be an operational nightmare. Both the F and D are long lines, and they have recovery time at each end so that a southbound delay doesn’t automatically translate into a northbound delay. If you were to through-route them, presumably you’d cut out the recovery time (otherwise the train would sit in the station for 10+ minutes – not very attractive to the through riders you’re trying to attract!). But that means that West End riders trying to get into Manhattan (who probably outnumber your down-and-back-up riders by several orders of magnitude) would be held hostage to delays on the Queens Boulvard line!
The south end of the Culver line is lightly used because for the most part it passes through relatively low-density areas and because the buses from the east all reach the Brighton line first. South of Kings Highway it’s also very close to the Sea Beach line. (I would almost argue that the line south of Church is dispensable, but it would be good to preserve one track to maintain yard access, and once one track is being preserved, the other two might as well stay too.)
I’d also like to see what has happened to the R with the loss of the M with respect to increased crowding and waiting times along part of 4th Avenue. Of course the demand that shifted to the Q to Canal with a transfer southbound to the R, 6 or J won’t be able to be determined without passenger surveys which will never happen.
If every single person who used to ride the M is now riding the R, trains are still not overcrowded.
N and Q riders have always ridden over the bridge and then transferred to southbound trains. It’s a quicker trip to the City Hall area that way than by slogging up on the R.
Passenger surveys are expensive and have small sample sizes. For most purposes, there are better ways to gather data. (In this case, I’m not even sure what you’re looking for.)
I live in Forest Hills and work in Midtown; naturally I heavily depend on the (F) as it provides a one seat (or standing) ride from 75 Avenue to 57 Street. Despite the general lack of seats in the morning and the impossibility of a seat in the evening, when (F) service is operating normally, it is reasonably quick. The one thing I would have to complain about is frequency of service in the evenings especially. By the time the (F) arrives at 57 Street, it is usually extremely packed, sometimes it is impossible to board even in the front of the train.
The rerouting along 63rd Street is definitely a plus, especially going into Manhattan as things run pretty smoothly especially after Roosevelt Ave/Jackson Heights.
The Brooklyn Express issue will definitely prove to be controversial; in order to bring an Express train to the Crosstown Line south of Bergen, service would have to be provided from 6th Avenue, making local passengers to Manhattan along that route rather unhappy.
If there was some reason to justify additional Eighth Ave. local service in Manhattan, but not additional Fulton local service in Brooklyn, I suppose you could steer selected C trains (or re-introduce the K-8th Ave. Local) that could be shifted over to the F line at Jay Street and then send those trains as express to Church (you could also do the switchover south of West Fourth Street, but that would jam three lines onto the Sixth Ave. local tracks between West Fourth and B’way-Lafayette). Since the Eighth Ave. local track isn’t at capacity, that’s the only place where you could add additional service in Brooklyn and get it into Manhattan, via the Cranberry Tunnel.
The A/C already do their track switch at Canal in Manhattan in conjunction with the E, so this would just be the same thing in Brooklyn, only with the F replacing the E. It would allow for use of the express track without taking away F service from Park Slope/Carroll Gardens local station passengers, as well as giving them an express ride downtown, though odds are most would prefer direct express service to Sixth Avenue instead of Eighth.
I wouldn’t want to try to cram any more trains into the Cranberry Tunnel. It’s pretty crowded with the A and C as they exist today.
Agree it would be tough, but you’ve got more TPH on the express track between 59th and 145th with a A and D trains sharing trackage than you do in Cranberry with the A/C combo. So there is a little leeway there to jam in a few more trains that could be redirected to Culver express service (the Rutgers tunnel has loads of spare capacity, of course; it’s just the bottleneck it would create between B’way-Lafayette and West Fourth that makes that option unworkable).
True, the C isn’t as frequent as the D. But there still isn’t much room to spare, and the additional merges would make it difficult to operate in such a tight corridor.
If the F express were introduced, I imagine it would be a very infrequent service operating for just a few hours each day. It would probably be something like every fourth train would skip the stops between Church and Jay, stopping at 7th Ave. This would only happen between 7-9 am and 4-6 pm. It would be similar to the A service to 116th St, where only 5 A trains take this route in the AM rush and 5 in the PM rush (more or less).
The only way to have service match a much higher demand would be to bring back the V and return the M to it’s previous route. The tracks just aren’t configured to allow for maximum capacity if the Chrystie St cut is used this way, which is unfortunate.
I really don’t understand why the M allows the F to run so much more smoothly than the V. Except I do know: the MTA is dumb. I remember taking the F to 14th St everyday from Carroll Gardens. The way the MTA should have operated it is that the V leaves right after every F (or maybe every 2 out of 3 Fs). Instead, the V always left per its schedule, meaning the F would pull up and you’d see the V leaving and you’d have to sit there on a crowded train for a couple minutes until the empty V got far enough ahead. Why the F didn’t have automatic priority boggles my mind. When the M and E pull up beside each other at Queens Plaza headed into Manhattan, the E always leaves first. That makes sense.
I disagree strongly. The V should have had priority, so that people waiting at stations shared by both lines would first be offered an empty V instead of a crowded F.
At Queens Plaza, the M should go first, so people waiting at 23rd-Ely don’t have to squeeze onto a crowded E. M riders should also be encouraged to stay on their train instead of piling onto the E. And, operationally, holding the M is more likely to delay the train behind it (which is only 3 minutes away, by the schedule) than holding the E (the next E is 4 minutes away).
I’m not 100% sure this is possible, but if I were in charge this is how I would implement the F express:
Switch the B to the local track (I know this is possible) and the F to the express (not so sure if this is) on the Sixth Avenue line. It doesn’t matter much since the line has only two local stops anyway, and it would make both trains more consistent (F is now a full express, B is completely local in Manhattan). Since the B and M both don’t run very frequently, add another local service (call it the V) that runs local along the Culver line, and make the F run express.
During the weekend, make the F run local to replace the B and M.
Brighton riders would never go for that.
Why so? This would add a grand total of two stations to the B’s route. Nothing major at all..
It’s physically possible, but it would totally gum things up as trains wait for their chance to cross behind other trains. If you start crossing trains left and right on a busy Manhattan trunk, you’d have to reduce service, but you’re trying to increase service.
A look at some track maps indicates that they’d have to construct one or two junctions to make it work–but that has to be a fraction of the price of the culver renovations..
On-time performance is irrelevant to riders on a frequent subway system. Virtually nobody consults a schedule and aims to catch a particular interval, with the goal of arriving at the end of the line within five minutes of the scheduled time – and that’s what OTP measures. (The easiest way to improve OTP is to pad the schedule. That does nothing to actually improve service, but it increases operating costs and it slows down service.) The document mentions a new “Service Key Performance Indicator” – that looks far more interesting, since it incorporates wait assessment, which is something riders actually care about.
More importantly, the loading guidelines that were changed were the off-peak loading guidelines! Rush hour guidelines haven’t changed. They called for 145 passengers on a 60 foot car or 175 passengers on a 75 foot car – and they still do. They never called for a seated load!
Finally, since F trains in Brooklyn carry well below capacity loads, there is simply no chance that NYCT will increase service on the line for the purpose of running an express. That means that the only way an express will run is if some existing F’s are sent up the express track. That seems like a very bad idea to me, since the stations south of Church are lightly used while there are some very busy local stations closer to Manhattan that would lose service. I don’t understand where the obsession with express service here comes from, but it really doesn’t make sense.
I don’t understand where the obsession with express service here comes from, but it really doesn’t make sense.
It comes from the idea of making commutes faster and more convenient while alleviating overcrowding on local trains. It’s not a good idea if it involves cutting other services, but it stems from the same theories that saw the 1 and 9 run skip-stop service in Northern Manhattan and the Bronx before demand grew too high.
F trains are still quite empty when they reach Church (I used to live on the line and ride them every day). Expresses would be very lightly loaded. I dare you to go to Carroll Street and inform the people waiting there that, to improve their service, half of their F trains won’t be stopping at their station anymore. They’ll soon notice that, not only are they waiting twice as long for their trains, but those trains are substantially more crowded than they had been.
Based on last year’s report, the F in Brooklyn is carrying 75% of a capacity load in the morning and 62% in the afternoon. NYCT is not going to add service to a line that’s carrying 75% of a capacity load. (And that’s based on the actual rush hour loading guideline, which hasn’t changed.) Sorry. It’s just not going to happen. NYCT just cut service on the West End; why would it now add service to the Culver?
As your graphic shows, the first car is crowded. Ride in the middle or back of the train and it isn’t hard to find space. No, you won’t get a seat at Bergen, but you’re not supposed to get a seat when you board a crowded rush hour train.