Reasonable Transit Pitches

What about closing the Commuter Rail Connection at Malden and Oak Grove on the Orange line and building a brand-new Shoulder Regional Rail station at Sullivan Square? It's the first stop that I can see where Newburyport/Ipswich trains and Haverhill trains meet before continuing to North Station. With all the new development planned around Sullivan Square, it seems like this could be the next Ruggles Station-shoulder station that has access to Regional Rail and local subway connections. People coming from Newburyport/Ipswich or Haverhill could transfer one stop to the growing job center of Assembly Square as well. Currently, there is no way to accomplish this from Newburyport/Ipswich unless you travel all the way to North Station and transfer to an Orange line train to Assembly.

Plus, that 1970's Station for Sullivan Square could use a makeover!

This seems like a "reasonable" transit infrastructure project on our way to a more Regional Rail system. Interchange stations of JFK, Quincy Center, Braintree, Ruggles, Back Bay, Forest Hills, and Porter already exist. (I actually don't know how much the Oak Grove and Malden interchanges are used by riders.) However, I think more connections between the mainline subway and regional rail lines is a good thing. (West Medford and GLX maybe at some point?)

What do people on this forum think?

I'm not sure that closing Malden Center is a good idea. It's the last major inner core town/suburb in the north Boston, and a bus hub and that

Oak Grove is mostly empty and surrounded by 2 forested areas, that one could be closed in favor of Sullivan Sq. The latter has multitude of bus connections to Everett, Somerville, and the junction to the Commuter Rail to the North Shore.

West Medford is a bit more tricky. It's the last major square/downtown/center in the inner core of Boston, with two bus connections. Realistically, the rails need to be buried for both the GLX and Commuter Rail to meet there, but that's no easy task. There's only enough space for a grade level crossing for two tracks there. Route 16 is easier than West Medford, since one can just eat off the parking lots, but there's a narrow bridge there, and there's essentially nothing but the Mystic River and a traffic sewer there.
 
Last edited:
Is an “office car” a reasonable idea for the commuter rail? I’m thinking something with seats with desks, with partitions between them to offer a moderate amount of privacy, top tier wifi, and all the most common outlets (including usb and usb-c). It could even require a premium ticket.

Basically, market it to people who want to get a legit amount of work done on the train in/out.

The MBTA Commuter Rail from Boston to Worcester, Lowell, New Bedford, and Fall River doesn't really extend out that far. These are distances of about 30 - 50 miles, which only take 40 - 55 minutes in the Netherlands (Amsterdam -> Rotterdam/Den Haag (The Hague)/Zwolle/'s-Hertogenbosch). These commutes are about the same length of time as the MBTA's Red Line from Quincy to Alewife in 2018, when the slow zones are removed. They aren't really long enough to get that much done, especially if one needs to set up and wind down in time to get off at their stop. It's why Nederlandse Spoorwegen doesn't really have that much amenities on their electrifed trains, such as food etc., since most commutes simply just aren't that long enough.

For a 90 minute commute, you're looking for distances of 80 miles or more. This is a Boston - Springfield commute, or Pittsfield to Worcester commute (equivalent to Den Haag -> Zwolle). This is out of bounds for the MBTA, and it's not possible North - South Shores in Massachusetts with the existing North and South Stations in Boston.

Most riders don't have long enough commutes on CR to do "legit amounts of work" in a biz car. That's a perk for single-seat commutes of like 90 minutes or more, which is pretty firmly outside of what the Purple Line offers. Less than that you're probably only doing small tasks that don't require more than a phone to do. Maaaaybe Metro-North could offer this for some of its super-commuters at the endpoints of the system (though I kind of doubt there's enough of them to make it fly), but getting serious work done on the train is pretty firmly in the court of Amtrak business class...not Commuter Rail.

Maybe a one seat train ride ride from Pittsfield to Hyannis might be long enough to have such amenties entirely within MA state bounds, but it's way out of reach. :)
 
If you watch this video on "Shoulder Stations" for good regional rail, it got me thinking:
(3) The SECRET to Good Regional Rail - YouTube

What about closing the Commuter Rail Connection at Malden and Oak Grove on the Orange line and building a brand-new Shoulder Regional Rail station at Sullivan Square? It's the first stop that I can see where Newburyport/Ipswich trains and Haverhill trains meet before continuing to North Station. With all the new development planned around Sullivan Square, it seems like this could be the next Ruggles Station-shoulder station that has access to Regional Rail and local subway connections. People coming from Newburyport/Ipswich or Haverhill could transfer one stop to the growing job center of Assembly Square as well. Currently, there is no way to accomplish this from Newburyport/Ipswich unless you travel all the way to North Station and transfer to an Orange line train to Assembly.

Plus, that 1970's Station for Sullivan Square could use a makeover!

This seems like a "reasonable" transit infrastructure project on our way to a more Regional Rail system. Interchange stations of JFK, Quincy Center, Braintree, Ruggles, Back Bay, Forest Hills, and Porter already exist. (I actually don't know how much the Oak Grove and Malden interchanges are used by riders.) However, I think more connections between the mainline subway and regional rail lines is a good thing. (West Medford and GLX maybe at some point?)

What do people on this forum think?

It should absolutely be on the T's medium-term plan given all the development attention the area's getting -- TBH at the rate the I-90 realignment is going, it's more likely going to be the next big growth center, not the West Station area -- and eyeballing it from the Washington Street overpass, I *think* there's space for a joint MBCR-Orange Line island platform to the west of the tracks, but it looks an awful lot like you'd need to knock down the retaining wall that supports the upper busway to make room for an outbound platform.

(the logic behind not converting the OL test track/platform to MBCR is that you probably want to save what's currently the SB OL track for Grand Junction/Urban Ring service)
 
(the logic behind not converting the OL test track/platform to MBCR is that you probably want to save what's currently the SB OL track for Grand Junction/Urban Ring service)

Heavy Rail on the GJ is def not happening. You'd have to cut off the roads completely... and there's no way Cambridge approves of that. It's pretty much Green or CR (or nothing)
 
That all being said -- overall I think this could actually really work:
  • The Newtons, Framingham, and Worcester get clockfacing half-hourly peak direction headways.
  • Most riders see comparable travel times to today.
  • No extra equipment or capital investments are required.
  • And no station sees a reduction in service -- the suburban stations between Worcester and Framingham and 128 continue to see hourly peak headways.
As ever, further details available on my blog.

See the flaw in this is that the posted times for the WL are not realistic in practice. I assume the issue is mainly the stoppage time and pax boarding times... IOW the posted 13 minute difference between stopping and not stopping at West Natick-Boston Landing is probably double that at a minimum.

Bringing back 2019 service levels would be the easier thing to do, if ridership improved to the point where you could justify bringing it back. And then perhaps talk about service from Riverside if there was actually room for it.
 
See the flaw in this is that the posted times for the WL are not realistic in practice. I assume the issue is mainly the stoppage time and pax boarding times... IOW the posted 13 minute difference between stopping and not stopping at West Natick-Boston Landing is probably double that at a minimum.
I mean, we have reliability data on OTP for the Worcester Line. Even filtering down to peak trains only, the OTP comes out to ~83%, meaning at most 1 in 5 trains fails the OTP test (arrive no more than 5 minutes late), which puts a cap on how much of a difference there can actually be between the posted times and the actual times. (Unless you want to argue that the T's data is itself suspect, which would be a strong claim I'd be skeptical of without evidence.) If you can show me that trains actually take significantly longer than the posted schedule, then I'm happy to reconsider the proposal.

Likewise, feel free to make a copy of the spreadsheet and experiment with different numbers. For example, I just ran a test with more modest time-savings of 80% instead of 67%, and the schedule still works:

1683662637462.png


And in some ways it's not surprising that changing the stop-savings doesn't alter the results too much. In all of these scenarios, swapping a zonal express model for a skip-stop model still means that, either way, the trains will need to travel 44 miles and make roughly ten stops, taking a bit more than 80 minutes to do so. All I'm doing is proposing reallocating some of that stopping time to the Newton stations.

It's true that the 2019 schedule was better, but it still saw 35-45 minute headways in Newton, and my schedule provides twice as much service to Worcester, one of the highest-ridership stations on the commuter rail, but currently receiving (I believe) the least frequent service among its high-ridership peers.

I would also have to look, but I don't know if the 2019 schedule can be run using today's number of trainsets -- my proposal appears to be rolling stock-neutral in that respect.
 
Last edited:
The Urban Ring Phase II DEIR specced a Commuter Rail platform at the Sullivan superstation, owing to the supersizing of the hub. The Rail Vision, however, hasn't listed one there for Regional/Urban Rail.

The problem is ultimately cost. To fit platforms in there amongst the I-93 girders you have to excavate out a large retaining wall on the busway side, move an Orange Line substation, and re-brace the I-93 supports while undergoing construction. So it'll be an expensive job, and such an expensive job requires a really robust ridership forecast to justify its existence. The Rail Vision didn't think that :15 Urban Rail would justify it with the current OL/bus audience amidst the complications of additional congestion on the critical shared portion of the Eastern and Western Routes (though I suppose if the terminal district slow zones were well-optimized that wouldn't be such an issue). You might actually need the heft of the Urban Ring's demand to put this project over the top.
 
I would also have to look, but I don't know if the 2019 schedule can be run using today's number of trainsets -- my proposal appears to be rolling stock-neutral in that respect.

I did a quick sketch: I can see a way to run the 2019 schedule with 9 trainsets, which may be equal or slightly more than today.

1683664686773.png


The mappings are by no means exact (for example, I'm pretty sure the outbound 7:20am and 7:30am sets should be swapped), and it's quite possible the trainsets were intermingled with other Lines. But, from what I see, you'd need at least 9 sets to run the 2019 service, whereas I think 2023 service can be done with 8 sets, and my proposal could maybe use as few as 7.
 
I mean, we have reliability data on OTP for the Worcester Line. Even filtering down to peak trains only, the OTP comes out to ~83%, meaning at most 1 in 5 trains fails the OTP test (arrive no more than 5 minutes late), which puts a cap on how much of a difference there can actually be between the posted times and the actual times. (Unless you want to argue that the T's data is itself suspect, which would be a strong claim I'd be skeptical of without evidence.) If you can show me that trains actually take significantly longer than the posted schedule, then I'm happy to reconsider the proposal.

My own experience taking it. Let me put it this way, I targeted my arrival to be when the train was supposed to leave. Most of the time I even had time to chill for a bit.

The situation might be better currently because of the depressed ridership.
 
My own experience taking it. Let me put it this way, I targeted my arrival to be when the train was supposed to leave. Most of the time I even had time to chill for a bit.

The situation might be better currently because of the depressed ridership.
I went and looked -- the data I used to calculate the OTP includes all peak trains from 2016 to present. I filtered down to 2019 only and re-calculated the OTP: 82%, basically unchanged. 2018 was actually better at 85%.

I could be convinced that the schedules are/were not realistic, but claiming that West Natick to Boston Landing non-stop must take at least 25 minutes just seems hard to accept on anecdotal evidence alone.
 
You might actually need the heft of the Urban Ring's demand to put this project over the top.
If the urban ring extends to at least Chelsea via the Eastern and the Reading line is converted to OL, I would wonder if it’s even worth it to go through all the trouble of building a CR stop at Sullivan. One could transfer to UR at Chelsea and OL north at NS. It would be less convenient, sure, but a hell of a lot cheaper. Maybe once the 93 Expy viaduct is rebuilt, they could provision in a wider ROW to allow for a station, but you’d still need to excavate the busway retaining wall. I think the OL “test track” should be converted to it’s original intended use as an express track once the OL is extended beyond Oak Grove, so I would be hesitant about stealing that part of the ROW for CR.
 
I would argue it is, only if you want to future-proof things.

1) Sullivan Square's parking lots are going to be redeveloped sooner rather than later, and once it's done that will make it harder (logistically, bureaucratically, legally when it comes to getting the ground-lessors on board) to doing the excavation for a platform.
2) This is render of how Boston wants to rezone the Sullivan Square area. FARs of 4 and 5, max heights of 150-350, on the scale of the kinds of things that will be built around a future West Station. That's not even taking any of Somerville's rezonings into account, either. It's hard to overestimate the impact this will have on where big developers will want to build investment-grade properties.

1683722623473.png


Whether or not we get NSRL or Urban Ring, we'll need a Sullivan Square MBCR stop to support this density, prevent future crowding on the OL by providing a form of "express" service, and let people living in the suburbs or Gateway Cities to take regional rail to work.
 
See the flaw in this is that the posted times for the WL are not realistic in practice. I assume the issue is mainly the stoppage time and pax boarding times... IOW the posted 13 minute difference between stopping and not stopping at West Natick-Boston Landing is probably double that at a minimum.
A lot of this is due to the platforms being low and on one side with stations too close to each other. Trains need to slow to change tracks to access the stations and/or wait for another train to clear the tracks if there’s conflict. Future consolidation into a full high level both 2 track stop at Newton Corner and eliminating Auburndale and Newtonville would speed up travel through this section of the line. Getting rid of Newtonville isn’t ideal as it’s the busiest of the Newton stops by far and has key destinations at it but its proximity to Newton Corner and the increased and more frequent bus connections there outweigh the negatives of removing Newtonville. There’s also the potential that if a West Station is built with GJ light rail to Cambridge and Longwood, some trains could terminate at the station.
Though this all becomes less relevant if the T can do as @Delvin4519 mentioned and get travel times Worcester-Boston down to under 60min with electrification, EMUs, and higher track speeds.
 
A lot of this is due to the platforms being low and on one side with stations too close to each other. Trains need to slow to change tracks to access the stations and/or wait for another train to clear the tracks if there’s conflict.

It doesn't do this during peak. It just simply doesn't stop at the Newton stops in the reverse direction.

I could be convinced that the schedules are/were not realistic, but claiming that West Natick to Boston Landing non-stop must take at least 25 minutes just seems hard to accept on anecdotal evidence alone.

To be clear, I was saying that the time difference between the Semi Express and the Local is probably closer to 25 minutes.
 
It doesn't do this during peak. It just simply doesn't stop at the Newton stops in the reverse direction.
On this topic, I’ll also mention that my proposal requires no overtakes, and because no trains turn at Framingham, the path is never blocked by a turning train.
To be clear, I was saying that the time difference between the Semi Express and the Local is probably closer to 25 minutes.
Sorry, I’m still not clear — which do you mean by “Semi Express”?
 
Whether or not we get NSRL or Urban Ring, we'll need a Sullivan Square MBCR stop to support this density, prevent future crowding on the OL by providing a form of "express" service, and let people living in the suburbs or Gateway Cities to take regional rail to work.

This is pretty much the exact same argument that was made in the RM Transit video I posted with the Sullivan idea. Is MBCR short for (Mass Bay Commuter Railroad?) Didn't that go away with Keolis winning the contract?

Video link for reference:
 
It doesn't do this during peak. It just simply doesn't stop at the Newton stops in the reverse direction.
Currently the WL is boarding all trains on the outbound tracks from West Natick and then inbound at BB and Landsdowne. This means inbound trains need to switch to the inbound track after Wellesley Farms to serve Newton platforms and then possibly wait for outbound trains to clear the Landsdowne and BB inbound platforms before proceeding. Keolis also prioritizes peak direction and will often make the off peak train wait for the peak to clear even if it arrives first giving advantage to that peak time. My experiences with the Worcester Express v Local is typically about a 15min difference granted that’s the outbound peak #525 v #595. Train #525 typically remains on the Newton platform tracks express to then switch before Natick Center because the inbound train it passes skips the Newton stops but serves Wellesley. There’s some very screwy patterns going on right now that makes things complicated.
 
Doesn't stop between West Natick and Boston Landing.
Okay, so you are asserting that the Framingham Locals are routinely more than 10 minutes behind schedule by the time they arrive at Boston Landing, is that right? And that the all-stops travel time between West Natick and Boston Landing is 45+ minutes rather than the scheduled 35 minutes, yes?

Stepping back: in general, yes, it seems plausible that the depressed ridership may be partially mitigating some of the delays you are describing. That being said, I'd say that the current conditions today seem likely to prevail for the foreseeable future, so in some ways I'm not too worried about what performance was like pre-pandemic.
Currently the WL is boarding all trains on the outbound tracks from West Natick and then inbound at BB and Landsdowne. This means inbound trains need to switch to the inbound track after Wellesley Farms to serve Newton platforms and then possibly wait for outbound trains to clear the Landsdowne and BB inbound platforms before proceeding. Keolis also prioritizes peak direction and will often make the off peak train wait for the peak to clear even if it arrives first giving advantage to that peak time. My experiences with the Worcester Express v Local is typically about a 15min difference granted that’s the outbound peak #525 v #595. Train #525 typically remains on the Newton platform tracks express to then switch before Natick Center because the inbound train it passes skips the Newton stops but serves Wellesley. There’s some very screwy patterns going on right now that makes things complicated.
Yeah, I haven't been able to keep up with all the latest screwiness on the Worcester Line, it really is nuts. What I had been envisioning was that trains beyond Framingham would use the normal inbound/outbound track assignments, and then between Framingham and Boston, all trains in the peak direction would operate on Track 2 (the "inbound track") and all reverse peak trains would operate on Track 1 (the "outbound track"); this would mean that no crossovers would be required in the morning peak, and that only inbound reverse peak afternoon trains may need to hold, west of Framingham, to allow outbound peak trains to switch tracks. That's not ideal, but seemed manageable enough, especially since there won't be any Framingham short-turns blocking the tracks.
 
I would argue it is, only if you want to future-proof things.

1) Sullivan Square's parking lots are going to be redeveloped sooner rather than later, and once it's done that will make it harder (logistically, bureaucratically, legally when it comes to getting the ground-lessors on board) to doing the excavation for a platform.
2) This is render of how Boston wants to rezone the Sullivan Square area. FARs of 4 and 5, max heights of 150-350, on the scale of the kinds of things that will be built around a future West Station. That's not even taking any of Somerville's rezonings into account, either. It's hard to overestimate the impact this will have on where big developers will want to build investment-grade properties.

View attachment 37702

Whether or not we get NSRL or Urban Ring, we'll need a Sullivan Square MBCR stop to support this density, prevent future crowding on the OL by providing a form of "express" service, and let people living in the suburbs or Gateway Cities to take regional rail to work.
Preventing future density-induced crowding on the Orange Line is being taken care of by Orange Line Transformation's target of 3-4 minute peak headways on Orange, a dramatic capacity increase over anything we've ever seen. It's not like there's no movement afoot to tackle that. The frequencies to handle explosive growth at Sullivan and Assembly are in the works. And that's in large part the reason why the Rail Vision didn't see a need for a Sullivan stop even in a universe where 2 mainlines would be pulsing 15-minute frequencies through there. It's still projects too easy to grab a once-every-3 Orange train out of NS (or, for Haverhill/Reading riders...a once-every-3 OL train @ Malden). Yes, maybe it's a little easier to have a shoulder stop directly there...but it's definitely not hard. And since this is going to be a very expensive build, you need to clear a bigger hurdle than "a little bit easier".

The game-changer is the Urban Ring super-sizing the linked-trip audience hubbing at Sullivan. If you've got a rapid-transit caliber one-seat trip linking Sullivan with Kendall, the Universities, Kenmore, and LMA...you've got ease-of-trips that either can't be approximated at all by a NS transfer or approximated well (thru-the-gut Green Line) projected over utterly massive demand numbers. That one the Phase II DEIR definitely did see a driver large enough to push all the construction expense.

In the end you can get a convergence of things calling for the station: density + Regional Rail frequencies + Urban Ring (even Phase I's crosstown bus system) conspiring to make it a priority. But outside of the real-deal rapid transited UR later phases there's no singular change here decreeing "this must be done or else." Not even the most optimistic densification of the Square rises to that level, because Orange is eventually going to rise to that level.
 

Back
Top