Last January, I asked the MTA:
… are there any plans for changing Roosevelt Island subway service with the M Local either replacing the F train or adding to the F train service through the 63rd street tunnel?
The Roosevelt Island F platform is jam packed during the morning rush hours coming from Queens. Residents often cannot get on the train and have to let 2-3 trains go by before being able to get on.
Some have suggested the M Local would be less crowded coming into Roosevelt Island and might be a better alternative.
Is that a possibility?
A MTA spokesperson replied:
We are always evaluating ridership and potential ways to improve service. I have shared with our operations planning team for awareness.
Last week, I started hearing rumors that the MTA was planning re-routing the M Train to include Roosevelt Island. I again asked the MTA:
I am following up on my January 24, 2025 email below regarding M Train Service coming to Roosevelt Island,
Are there any plans for changing Roosevelt Island subway service with the M Local either replacing the F train or adding to the F train service through the 63rd street tunnel?
The Roosevelt Island F platform is jam packed during the morning rush hours coming from Queens. Residents often cannot get on the train and have to let 2-3 trains go by before being able to get on.
Some have suggested the M Local would be less crowded coming into Roosevelt Island and might be a better alternative.
Has the MTA agreed to bring M train service soon for Roosevelt Island.
When will it start?
Any comment from the MTA?
I’ve confirmed the MTA will re-route the M train to Roosevelt Island as shown in the map below.
The F and M Train routes will be swapped weekdays from approximately 6 AM to 9:30 PM beginning in December 2025.
Currently, the F Train arrives at the Roosevelt Island subway station jam packed during weekday rush hours and residents often have to wait for 2,3 or 4 trains before being able to board. The hope is that the M Local from Queens will not be as crowded by the time it gets to Roosevelt Island as the F is now so residents will be able to board the incoming train or maybe even get a seat.
Here’s more on the MTA’s plan to swap the F and M train routes.
About the current Roosevelt Island F Train station morning rush hour, residents say:
- Very crowded and I’ve had to wait for 20 minutes for the next train! Sometimes they come every 8 minutes but you can’t get on because it’s full. This happens usually from 8:30 am to 9 am.
- The 7:30 is pretty bad too
- Yes, and I have had to wait as long as 40 minutes to board a train because it was too crowded, and some days there is a 12-20 minute wait between trains
Here’s an interesting video by the Joint Transit Association advocating for the M and F Train swap.
Stay tuned for more about the Roosevelt Island M and F Train Swap
I am not buying this.
1) None of the docs are in the style the MTA usually posts.
2) Would cause too much confusion having differing service patterns.
3) 🅜 trains are two cars (120 feet) shorter in length than 🅕 trains, so it would be a service cut.
It’s really not that confusing at all. It’s actually the most sensible option. For one, only a few stops are being changed, while the rest of the route stays the same on both lines. The shorter trains can be offset by more frequent service, which can now be run now that the M has one less merge to make. The F train’s infrequency results in overcrowded trains on 63rd Street, but that comes from other parts of the line. The M would bring in less passengers through that area, allowing more people to get on. Also, most transfers are kept, albeit at different stations.
The F and M swap is good because it removes a nasty merge at Queens Plaza, and will help way more people than it will hurt.
You may say, “None of the docs are in the style the MTA usually posts”, but that does not matter, as anyone can tell that removing a merge between two trains will improve service.
You may say it “Would cause too much confusion having differing service patterns”, but confusion can happen at any time that the MTA has to reroute trains, and reroute information can be better displayed at stations so people understand.
You may say, “🅜 trains are two cars (120 feet) shorter in length than 🅕 trains, so it would be a service cut”, but that is not true. This is because people think the (F) being express is faster, so a lot of people crowd on to the (F) Because Qns Blvd riders crowded on to the (F), there is not a lot of room on trains for 63rd St riders. But by sending the less crowded (M) train there, 63rd St riders have more room on the train.
The F train should be kept as is and the M line should be added. As simple as that. This would provide much needed additional service and avoid the confusion MTA is creating with these weekly swaps. I don’t know if it is technically feasible, but if it is, I don’t see why the MTA wouldn’t go that way instead.
And this will cut service to the 53rd St Tunnel, which currently sees twice the passenger volume that the 63rd St Tunnel sees. Although the M doesn’t do a great job taking in riders as most passengers would crowd onto the E instead, it is still an option for Court Sq and Queens Plaza riders and helps decongest crowding somewhat for the E. Removing the M and making only the E provide is not a feasible idea.
Not to mention the F/M Swap is the superior way to run service. It removes one the worst choke points in the system, decreases crowding on both the 53rd and 63rd St corridors (F being an express just like the E balances the train and platform loads at 53rd St stations, M being way less crowded after 36th St allows more room for Queensbridge and RI commuters), increases service, and doubles one seat ride connections for QBL riders.
Funny how these reports make absolutely no mention of how this service change is just going to move all these merging delays to midtown Manhattan, while you have both F and M trains stacking up from Rockefeller Center down 6th Ave while the more frequent Fs have to wait for a slot to merge with the Es!
Because what you mentioned is a false equivalency. Currently, the merge near Rockefeller Center involves three trains: the E, F, M. It is already overworked, sorting the E, F, and M trains on their respective tracks. The question is whether you want to overwork a second junction near Queens Plaza, which is currently one of the worst choke points in the system that eats into capacity.
As a long-time resident of Roosevelt Island, I strongly oppose the proposed F and M train swap. This plan is shortsighted and does not address the core issue: the severe overcrowding during morning rush hours. Instead of rerouting lines, we should be working with the MTA to increase the frequency and capacity of F trains.
Swapping in the M train — which may appear less crowded now — offers no guarantee of better service. Who can assure us that the M will remain less crowded during peak times or maintain reliable headways? Moreover, the M is a local train and far more prone to delays in Queens compared to the express F train, making it a less dependable option.
This change would also severely impact Roosevelt Island residents traveling to the airports during weekdays. Currently, we have convenient, environmentally friendly, and cost-effective options: two stops to transfer to the Q70 bus for LGA, or express service to connect with the JFK AirTrain. Replacing the F with the M would significantly worsen access to both airports, pushing more residents to rely on Uber or private cars — adding to traffic congestion on the island.
In short, this proposal feels out of touch with the actual needs of Roosevelt Island commuters. We need a thoughtful solution, not a stopgap. I urge the MTA to reconsider and engage directly with the community to find better, sustainable alternatives.
“This plan is shortsighted and does not address the core issue: the severe overcrowding during morning rush hours.”
It literally does. The M train is no where near as crowded as the F train is past 36th St.
“Instead of rerouting lines, we should be working with the MTA to increase the frequency and capacity of F trains.”
Yeah and there is absolutely no way you can increase F frequency with the current state of how trains are routed. There are way too many merging conflicts, which delay trains. Those delays eat up capacity which is why the MTA ran fewer trains today than they did prior to 2001. By removing those merging conflicts, that means less opportunities for delays, which means more trains can be run on QBL. And the F/M Swap does exactly that, targeting one of the worst choke points in the entire system.
“Who can assure us that the M will remain less crowded during peak times or maintain reliable headways?”
NYC Commuters. It has long been documented that NYC Commuters would switch over to express trains the moment they have the opportunity. That is why the F is more used than the M, and will continue to be that way. Not to mention less merging conflicts is the literal recipe to more reliable service, which is what the swap achieves.
“Moreover, the M is a local train and far more prone to delays in Queens compared to the express F train, making it a less dependable option.”
There is no evidence to suggest that, you are just making that up.
“This change would also severely impact Roosevelt Island residents traveling to the airports during weekdays.”
Except they aren’t the majority. The majority on Roosevelt Island, according to Census Maps, are going to Manhattan, where commutes are going to improve, thanks to less merging conflicts and less crowding that the M sees.
Spoken like someone who doesn’t ride the Subway at all. Let alone go into Queens.
Spoken like people do not actually ride the F line into Queens. I have to offer counter points to this “idea” of less crowded trains.
#1 – The M train is shorter than the F, meaning any hope of lessening crowded trains is not happening due to lack of actual space on the train. I didn’t see plans to extend the train to match the F’s capacity.
#2 – Riders from the R, M, and E get on the F to go to it’s stops in Manhattan from 63rd which leads to crowded trains. 63rd is a major stop and that isn’t changing anytime soon.
#3 – It’ll add to delays for passengers who need to transfer. For those who have to travel from Kew Gardens or beyond, we have to time the trains (which is impossible to plan), otherwise we have to wait 5-10 minutes for a train (which will be crowded) from Roosevelt Ave going in either direction. This is from experience of transferring at Queens Plaza last year when the F was removed from 63rd for maintenance. While the R is not the M train, again it’ll lead to crowded trains no matter what happens.
#4 – There is no local service to Kew Gardens from the M or the R…so again we have to transfer and wait for the train to arrive. See point #3.
#5 – There aren’t plans from the article in question about increasing the amount of trains onto the route. If the M had this route all to itself it might not be as bad, but because it’s shared with the R increasing trains will be difficult to figure out.
So please explain how this will “lessen” the crowded trains? I genuinely do not see how this will be a positive step. If anything, this feels like a regression.
So right off the bat, you attack me based on my character instead of on the substance? Wow, really effective strategy you used right there.
So let me make this clear, I am a Queens resident of 10 years. I take F train on a regular basis. So if you want to attack me based on my character, please try again.
Now onto your points.
1. That is one way to look at it. The other way to look at the situation is passenger loads: the M is way less crowded than the F train west of 36th St. This means instead of a packed F train showing up 21st Queensbridge or Roosevelt Island, it would be a half full M train showing up. That is because how commuters perceive express trains to be faster, which means the moment that local riders further in Queens can get on the E or F trains, they do so.
And this isn’t my experience riding both the F and M trains for 10 years, it is proven through studies. The F load through 63rd is about 1.2, while the M load through 53rd is 0.4. This means if we swapped the M train to go via 63rd, 63rd St Tunnel commuters will get that 0.4 load train instead the 1.2 load train.
2. That is not true. According to Census Maps, areas around the 63rd St corridor don’t even register as a top 50 destination for the entirety of the Queens Blvd Line.
3. That is incorrect. By removing one of the worst merging conflicts in the system, it removes one of the biggest delay generators in the entire system. This helps dispatchers schedule and move trains around the system faster.
According to delay charts made by MTA employees, the top delay generator for the E and F trains is literally the junction at Queens Plaza and 36th St. The F/M Swap targets that one merge, and removes it.
4. If you want the actual numbers on who is traveling to Queens as opposed to Manhattan, 67 percent of Roosevelt Island commuters travel into Manhattan. Only 7 percent travel into Queens. So if you want to make up hypotheticals, then make sure you make them up with what the majority sees.
5. You sort of get the concept of deinterlining with the last sentence you said: “If the M had this route all to itself it might not be as bad, but because it’s shared with the R increasing trains will be difficult to figure out.” Deinterlining is when you remove merging conflicts so that trains are more reliable. When trains are more reliable, dispatchers can add more trains. Prior to 2001, the E/F ran a combined 30 tph. Today, that number dipped to 27 tph. While CBTC helped a little bit, that only got it back up to 29 tph. With the F/M Swap being a deinterline, that means the E/F will be increased back to 30 tph, and CBTC can further boost it to 32 or 33 tph.
Now, as for the M, read the first point on passenger loads versus the F and M trains. Not to mention the MTA is currently making the Forest Hills terminal more efficient, and the R will likely be increased to about 10-12 tph. That means the M can use the other 10-12 tph, which translates into a train every 5-6 minutes, as opposed to 8 minutes today.
Stop falling for these lies just because they are your fantasy service patterns! This will not improve service at all for a wide variety of reasons!
List them then
I get on at Queens Plaza and Court Sq often to go deeper into Queens, so this change seems really beneficial since both expresses will stop at those stations rather than just the E. Plus the E won’t be held up anymore by an F merging at 36 St! Excellent
Can’t wait for this change to into effect
How about traveling from RI to Queens? Four additional stops before Roosevelt Avenue. And coming back from Manhattan to RI – fewer and shorter trains will exacerbate the crowding, on the M, especially during rush hour. Did the MTA really think this out?L
Luckily, only 7 percent of Roosevelt Island commuters travel into Queens, per Census data. 67 percent of Roosevelt Island head into Manhattan. Not to mention, M loads west of 36th St are way lower compared the F train, which contributes to lower overall crowding on the 63rd St corridor.
Agree it is a dumb plan. This is actually a disaster for those of us who travel towards Queens during rush hours and for access to the airports! It will add 10 min to our commute in each direction and force those who go beyond Roosevelt Ave to change trains to transfer to an express to go to Forest Hills and beyond. If you add up all the people who are now going to add 20-25 min to their commute to Queens per day, it likely is as many as those who have to wait for two trains in the other direction and they usually wait only 5 min and certainly the overall minutes lost is greater. Moreover, M train service will be much less frequent than F trains during non-rush hours, which means returning between 8 and 9:30 pm will also mean a longer wait. Currently the F train is scheduled to run about every 5-6 min in the morning rush (today, 8 trains in 42 minutes starting at 8:20 am, which is typical) so the schedule they sent you is incorrect and it is likely the M trains will run less frequently than every 6 minutes in both directions.
Has anyone commented on the 8 car m service against the 10 car f…53 st overcrowding…why not stagger m trains on 63 or bring back the v for rush hour service to 2ndvave that’s a option to alive 63 st passenger complaint during rush hours
That is way too complicated and will result in worse operation near Queens Plaza, which is what the swap solves. Already, the delays here are so bad that it eats into capacity. Adding another merge point near 36th St for some M trains to enter 63rd will make things even worse.
Also, the MTA is looking into short turning some M trains at 2nd Ave.
“This is actually a disaster for those of us who travel towards Queens during rush hours and for access to the airports!”
And that is a minority of people. However way you route trains, you are going to run into different scenarios that it will hurt people. No one talks about the current status quo, where trains get held up at Queens Plaza so frequently that it cuts E and F service.
” If you add up all the people who are now going to add 20-25 min to their commute to Queens per day, it likely is as many as those who have to wait for two trains in the other direction and they usually wait only 5 min and certainly the overall minutes lost is greater.”
Not true. According to Census Maps, 67 percent of Roosevelt Island commuters are heading into Manhattan. Only 7 percent enter further into Queens. That means for the vast majority of Roosevelt Island commuters, they will see little change other than the fact that their trains will be less crowded thanks to the M being way less crowded than the F.
“Currently the F train is scheduled to run about every 5-6 min in the morning rush (today, 8 trains in 42 minutes starting at 8:20 am, which is typical) so the schedule they sent you is incorrect and it is likely the M trains will run less frequently than every 6 minutes in both directions.”
That is just vibe based reasoning you used. First of all, with less merging conflicts, this means less delays, which translates into more trains that can be run. That is where the extra M trains are going from. Not to mention the MTA is also looking into short turning some M trains at 2nd Ave, which would also add some more trains to 63rd St.
In short, M frequency will likely match or be slightly lower than F frequency today. And the tradeoff is way less crowded trains and more reliable service thanks to eliminating one of the worst choke points in the entire system. Not to mention with CBTC installation, which will speeds trains by 7 percent, it means that you are likely to make back your extra minute of wait time at Roosevelt Island. That is if the F train isn’t bunched and makes you wait 15 minutes.
The bottom line is to get rid of the chokehold at Queens Plaza to improve service along QBL and 63rd Street. I’m for this plan.
Again. 7% of commuters, those going to Queens, will see an average 20-30 min increase in their travel time overall daily, whereas at the very most 5% overall going to Manhattan during the entire rush hour, (since the majority do not travel between 8:30 and 9) wait an average of 3 min for another train. THIS IS A POOR SOLUTION! More F trains is the answer — running them every 4 min as the table says (they run now every 5-6 minutes) will resolve the problem.
You can’t even get the numbers right. According to Census data, it is 67 percent of Roosevelt Island commuters are going to Manhattan. About 7 percent go into Queens, which means people going to Manhattan outnumber people going to Queens more than 9 to 1.
Not to mention, what you said is hyperbolic pandering. According to timetables, you save 4 minutes on the express, not 20-30 minutes. Factor in wait times if you want to transfer, and that adds 2-4 minutes. We are talking about a 6-8 minute increase at the worst case scenario, which is far from 20-30 minutes. And because now the Queens Plaza Junction is streamlined, it means no more heavy delays there, which means the 6-8 minute increase may actually be less.
And your solution of running more F trains is completely unachievable without the F/M Swap. Currently, the Queens Plaza merge is the biggest delay point that the F train sees. Only Roosevelt comes close to it, and that is a separate problem. These delays are so bad that they eat into capacity, because capacity is measured by how many trains cycle through a given a point. Having trains stop for two or three minutes because of another train merging ahead is the antithesis of that, which is why less E/F trains run to Queens now than 2001. The F/M Swap simply targets that merge and removes it, which allows more trains to be run.
I think this swap is good for the Queens Boulevard line because it removes the tangling and the F trains at both Court Square AND Queens Plaza would be good because stations with a large ridership would favor the express trains heading to Jamaica. These two stations are both big transfer stations, but if the Roosevelt Island M train passengers want express services, they would have to transfer at Jackson Heights Roosevelt Avenue for an express train. If passengers from RI want service to Queens Plaza for an express train going towards Manhattan to 53rd street (the E/F trains), they need to transfer to the R train at the next available stop and can stay on the Manhattan-bound R train if they want service to 59th street or Broadway. If the RI passengers want Sixth Avenue service, they can usually take the Manhattan-bound M train as usual is now with the F train. I would say that overall, this is an 80/100 idea that is likely to go into effect.
Funny how this plan completely ignores the reason it came about. We were supposed to get a Queens Super Express. We didn’t. So the best the MTA could do is split the load of the E, F between the 53rd and 63rd street tunnels. It was and still is the best solution absent the Super Express track. This plan as proposed would be a disaster for both 53rd and 63rd Street.
This swap is an excellent idea for Queens Boulevard Line. The (F) should run via 53rd Street Tunnel exclusively with the (E) 24/7. The (M) should run via 63rd Street Station Tunnel exclusively, plus have the (M) end at 21st Street Queens Bridge in Queens on Late nights weekends and holidays, if the MTA NYCT could that, till Queens Link gets built in The Rockaway Beach Line for the (M) to The Rockaway’s with the (A), plus replace the The Rockaway Park Shuttle.
Actually, you are wrong. When the MTA did the EIS back in 1989-92 on what to do with the stub line of 63rd St, they found that the F/M Swap will further decrease crowding by 22 percent, save twice as many transfers than the status quo, all while avoiding creating a second nasty merge at Queens Plaza.
The reason why they didn’t is because the MTA wanted the track connections to the 63rd St Tunnel to be accessible from both the express and local tracks, and wanted a stronger case to the feds and the state government. Using the F train, which has a much higher passenger load than the V (or the M), did the trick. Unfortunately, the MTA soon confirmed the results from their own study: making the F run via 63rd was a bad idea and destroyed reliability on the whole of QBL.
Sorry but this is a bad idea for starters, there are not enough cars to staff additional “M” trains to make up the difference in the lost number of “F” trains. It is more than the two additional cars per train, it is entire trains themselves. The “M” train yards are at the end of the line in Middle Village and those yards can only hold four sets of trains.
You can’t just use cars from other lines due to those trains that ride into Brooklyn.
Once upon a time the “F” train ran 18 trains during rush hour but that was reduced to 15. Those three trains were difference makers.
Sorry but the MTA is selling you folks a bill of goods as they are looking for more ways to cut back on service not add to it.
This is going to cause more people to go back into cars
“…they are looking for more ways to cut back on service not add to it.”
You have no evidence at all, but I have evidence to support the opposite of what youre saying. My evidence is that the F and M swap will get rid of a nasty merge at Queens Plaza, where uptown M trains merge with R trains at Queens Plaza to run Qns Blvd local, and southbound M trains merge with E trains to go via 53rd street to Manhattan. If two of the trains that must merge show up at the merge point at the same time, one train must be delayed, which isnt good. But by swapping the F and M trains, the F and E will stay on the same track at Queens Plaza, with no merging conflict. To sum up, the F and M swap gets rid of a nasty merge at Queens Plaza, so more trains can run on Qns Blvd.
Swapping the F and M trains is good because it gets rid of a merge at Qns Plaza where northbound M trains must merge with the R to go local on Qns Blvd. By swapping the F and M train tunnels between Manhattan and Queens, the F stays on the express track with the E at Qns Plaza, with no merging. I will now respond to the things u said, and why theyre wrong.
“…there are not enough cars to staff additional “M” trains to make up the difference in the lost number of “F” trains.”
There is no need to make up for any difference because F trains are jam packed when sent thru the 63rd St tube. The reason its jam packed is cuz people think the F being express makes it faster, so they crowd on to the F express train. Because The F is packed from all those people, theres less space for 63rd St residents to board the train. But by sending the M train thru 63rd St, the train is not very full, so 63rd St residents have more room on the train.
“The “M” train yards are at the end of the line in Middle Village and those yards can only hold four sets of trains. You can’t just use cars from other lines due to those trains that ride into Brooklyn.”
I think youre talking about the fact that the M wont have as many trains as the F, leading to 63rd St being more crowded, but I just talked about that.
“Once upon a time the “F” train ran 18 trains during rush hour but that was reduced to 15. Those three trains were difference makers.”
Congratulations, you just gave a reason that goes entirely against your opinion. The reason the F train used to run more trains than it does now is because it used to run along 53rd St with the E and had no merging conflicts along Queens Plaza. Now, there is another merging conflict at Queens Plaza, which could cause a chain reaction that leads to the F train getting delayed. That chain reaction is that a southbound E train could get held at Qns Plaza because of an M train merging, and then more E trains get delayed, and the F train is affected as it shares tracks with the E.