Transit Planning $h!tposting (Ideas so bad, they're good)

[Riverside]

(I'd understand if one of the mods wants to change the title of this thread.)

This is a thread for transit planning ideas that are so bad, they're good amazing.

This is not a thread for actually criticizing proposals. This is also not a thread for trashing ideas that someone has proposed in good faith. This is supposed to be irreverent, and is supposed to be fun.

I'm talking about ideas that are absurd to the point of being silly. For example, my Project Blue Lace:

Or my "MMMMM Line":

As I described in the above links, these examples are what you get when you take individual segments that each seem vaguely reasonable in isolation, and then keep stringing them together without recognizing when too much is too much. If you were to pick a subset of these silly proposals (e.g. a suburban light rail line pinging between West Roxbury and Dover), they would make a vaguely reasonable proposal – maybe not a good enough proposal, but not necessarily something to reject on sight.

Shitposting is often used in cruel and callous ways, but I would argue that it also can be used as a form of Surrealism. There are benefits to setting aside concerns about quality, in favor of creative albeit "bad" ideas. Absurdist ideas can break us out of our conventions and lead us to reexamine old ideas in new light.

To be clear, a shitpost is not the same as a shit post. The ideas can be bad and irreverent, but they should be shared in the spirit of levity and creativity. A shit post is when someone abuses that creative space to have an excuse for acting like a shitty person. Let's not do that.

I'll start us off below with a silly absurd transit proposal that fails on so many feasibility measures it's downright laughable -- but which points to a few surprising points, relevant to serious transit proposals.

 

[Riverside]

You've heard of the Urban Ring, now get ready for the Urban Helix.

Imagine Type 10 supertrains zipping around Greater Boston in a triple loop -- a figure 8 drawn overenthusiastically -- running in a 40 mile circuit, bidirectionally:

1. Riverside
2. Reservoir
3. Brookline Village
4. Boston University
5. Kendall/Tech Square
6. Sullivan
7. Sweetster Circle
8. Chelsea
9. Revere Broadway
10. Northgate
11. Everett Broadway
12. Sweetster Circle
13. Chelsea
14. Airport
15. Seaport
16. South Station
17. Back Bay
18. Boston University
19. Newton Corner
20. Auburndale
21. Riverside

Now, in all seriousness, the idea of suburban "circuits" was a very real one in Greater Boston for a long time. The Highland Branch, the Needham Line, the abandoned branches to Dedham, and the B&A through Allston and Newton were all part of circuit services run by the old railroads. The "Urban Helix" combines two suburban circuits (Allston/Newton/Brookline and Chelsea/Revere/Malden/Everett) with corridors from the "Urban Circuit" also known as the Urban Ring, inspired in part by @Stlin's fun idea in Crazy Transit Pitches (itself definitely not a shitpost).

But an interesting thing happens when you split the Urban Helix into a pair of criss-crossing lines:

And, well... neither of those is actually an absolutely terrible idea. I mean, there are lots of better ideas, to be sure.

But, for example, it hadn't occurred to me until today that a radial route from the Highland Branch -- whether from Riverside as shown here or from Needham -- could feed into the Grand Junction Kendall corridor. Particularly if the other service ran into downtown (e.g. Needham <> Kendall and Riverside <> downtown), that actually wouldn't be a terrible route design. (It would have lots of other problems, of course.)

The northeast end of the map doesn't reveal as much for us, although I think it does illustrate that a Kendall line could logically extend into Revere as opposed to inevitably going over to Eastie, and potentially points to something about the distorting impact of Boston Harbor, the Mystic, and Chelsea Creek (though I couldn't say exactly what). But hey, it's a shitpost -- the fun part is finding the occasionally useful idea amongst the silliness.

 

[Teban54]

On the idea of creating the thread itself - Great idea. Just curious, where would you place proposals that may be wildly unrealistic, but also not absurd or laughable (such as the idea of turning the GL Central Subway into a fully heavy-metro route, and kicking the B and C streetcars out)?

On the specific "Urban Helix" Proposal: I hate the fact that I'm making the discussion more serious than it should be -- but it ties perfectly into a point I just raised in the Urban Ring thread: Mixing radial and circumferential routes, and in particular, in a fashion that can be called "tangential routes".

I'm talking about a conceptual route like this:

On paper, it looks perfect: The Pink Line brings radial service to the transit deserts of A and D, while adding circumferential service to B and C, which may be major hubs outside the city center. But this blogpost from Alon Levy, "Mixing Circumferential and Radial Transit", explains why they think it's a bad idea.

Alon uses three examples to illustrate their point (NYC's G train, and Shanghai's Lines 3 and 6). I happen to be familiar with all of them, and I concur: All 3 lines are generally considered flops. To be fair, there are other reasons for their failure besides the combination of radial and circumferential service (detailed in the following comment), but I think the route character still plays a big part in it.

And we have an example in Boston, too: The earlier proposal to divert some Worcester Line trains to Grand Junction and North Station. It seemingly offers circumferential service along Grand Junction, but F-Line has questioned the utility of such a service multiple times. He quoted studies that showed the proposal was "not recommended", and only helpful during a few rush-hour slots when the Red and Orange lines were overloaded. (Some of that was also because those trains would traverse Grand Junction too slowly, but not all.)

To be fair, Alon's blogpost (which was motivated by earlier iterations of NYC's Interborough Express) also discusses the lack of transfer points, which is not as relevant to @Riverside's idea above. But I think the drawbacks of such a radial-circumferential mix still apply. For one, as they said in the conclusion:

The principle in action here is that, especially when there are no compelling destinations, it’s critical to make sure the line provides connectivity between large regions. This means connecting to all or almost all radial lines in a region well-served by radial subways but poorly served by preexisting circumferential ones. [...] Mixing radial and circumferential service interferes with this principle, since the radial component has to come at the expense of completing the circle (or semicircle in a geographically-constrained city like New York). Thus, it’s harder to use the line to get to a large enough variety of points of interest to make up for the fact that it misses the city’s most important destinations. Of course, such a line is also wanting as a radial line, since it misses the center. Thus, ridership underperforms, and the line usually fails to achieve its stated purpose.

In other words: A "tangential" line tries to take on the roles of both a radial and a circumferential line, but fails to do either of them well. Using my diagram as an example:

• As a circumferential line, the Pink Line fails to complete the circle. Namely, F doesn't get any circumferential service, which hurts residents in E wanting to go to B.
• As a radial line, it's likely not the most ideal service that A and D would want. Sure, some residents in A may want to reach B, but a far greater number will want to go downtown -- both because it has the greatest concentration of destinations, and because it's the easiest way to transfer to "obtuse angle" radial lines, such as to G and H, where a circumferential line offers little time savings. Even A riders wanting to reach C may find it a hassle.
• There also other issues, such as imbalanced demand patterns (and thus frequency requirements) on the radial and circumferential sections. (For example, Shanghai's Line 6 empties out almost entirely during rush hours once it hits the main transfer point.)

You need B and C to be extremely strong secondary CBDs to overcome these issues (though it's possible: see the success of Shanghai's tangential Line 18). In terms of Boston, Kendall and LMA are definitely the strongest candidates (they're the #1 and #2 biggest job centers outside "greater downtown"), but whether they're strong enough remains a question.

Of course, this is not a hard-and-fast rule. There are a few scenarios where the Pink Line build may be the most ideal and/or the best we can hope for (feel free to add more):

1. If B and C are extremely strong, as noted above
2. If F is a very weak node that doesn't draw too much demand
3. If completing the (half-) ring from B to F is extremely expensive
4. If the network topology is shaped such that radial lines in the eastern half all have obtuse angles, making a circumferential route not worth it
   • Think of between Wonderland and JFK/UMass
5. If there's no room downtown for another radial line to be constructed easily, so A and D wouldn't get more direct radial lines regardless
6. If this route is much cheaper than all other alternatives (more or less a combination of #3 and #5)

Some of them apply to Boston in varying degrees, while others don't. The specifics of which ones apply are topics for a lot more discussions.

A few miscellaneous points:

• Regarding @Riverside's idea above, specifically the Cyan Line, I actually think of Back Bay and Seaport as part of "greater downtown", and not "outside downtown". In other words, they're at the edge of the grey circle in my diagram, unlike B and C. (This distinction may be addressed in another comment.) Regardless, the problem starts when it gets to Logan and Chelsea.
• Don't confuse "tangential" lines with another type of radial-circumferential mix, which I nickname the "L-hook lines": Radial through downtown, but circumferential further out. On the diagram, think of it as G-C-F-E-A, with the E-A hook near the end, possibly going beyond A to meet the "Blue" and "Green" Lines. Alon Levy is actually supportive of such routes when situations warrant, and they wrote a follow-up blogpost 2 years later (which I'm sure people like @Riverside will love equally as well).

 

[Teban54]

Spoiler: Explanation of other reasons for failure of NYC's and Shanghai's tangential lines (Caution: extremely detailed)

Here are my impressions for why NYC's G train and Shanghai's Lines 3 and 6 failed. The three lines actually suffer from three different failure modes.

NYC's G train: Lack of transfers. A circumferential line only works if it actually connects to radial lines (as I just argued here). But thanks to the competing nature of the three former subway systems in NYC, the G's connectivity absolutely sucks. It only connects with 4 out of 12 lines in Downtown Brooklyn; in Queens, it misses transfers to 4 lines, and has a long transfer walkway to another 3 lines. It even misses transfers to another radial line (Jamaica Line J/Z) in the middle of the route, at a pretty notable neighborhood center (Williamsburg). Such a massive waste of potential for a line connecting Downtown Brooklyn to Long Island City -- its route is otherwise much better than the currently proposed IBX for a crosstown route.

Shanghai's Line 3 (yellow on this map): Interlining with the actual ring (Line 4), and bad route choice.

• Unlike the rest of the system, Line 3 is almost entirely elevated, because it uses a former mainline rail ROW (which are rare in China). The best analogy of this ROW is "surface NSRL except it uses Grand Junction": It connects the two railroad terminals (Shanghai South and Shanghai), but veers further out of the city center.
• This nature means it was dirt cheap to build, especially at a time when few subway systems existed in China. But remember: Legacy railroads, especially for this NSRL-like purpose, typically avoid density as much as possible! So the ROW naturally misses most of the destinations downtown. But a major and fatal difference from Grand Junction is that there aren't even that many good destinations along the route. (Zhongshan Park, where it meets Line 2, is pretty much the only notable one.)
• The circumferential nature of the ROW also motivated the decision of having a ring route interline with it. IIRC, when Line 3 was being built, it was provisioned for the eventual Line 4. (Line 3 actually started with another name, the "Oriental Pearl Line"; whereas Line 4 was initially named "Pearl Line Phase 2".) But this limited frequency -- on both the rest of the ring route Line 4 (actually the busier half), and the radial portions of Line 3. Add the fact that Shanghai frequently uses short-turns, and the northernmost half of Line 3 sometimes see 10-15 min headways even during rush hours.
• The elevated character of the line also gives it inconvenient transfers to other radial lines (sometimes even out-of-system transfers), which are all underground.
• As of about 10 years ago, the city was considering serious proposals to split Lines 3 and 4, with the radial portion of Line 3 becoming a standalone line terminating along the ring. While the plans were eventually abandoned, its very existence really says something.
  • (Beijing's Line 13, another route using legacy ROW and built around the same time, was eventually also planned to be split and is already under construction.)

Shanghai's Line 6 (magenta on this map): Massive underestimation of demand and TOD, with incorrect build sequence.

• Shanghai hired foreign urban planners in the early days of planning its metro system (at a time when Line 3 was being built). They came up with a system with three layers of routes: R, M and L. This later became the holy grail of Shanghai's metro design, and most of its 18 lines today all originated from that proposal, even 20+ years later.
  • R: These lines were supposed to be commuter-rail-like, terminating way further out from the city and with large stop spacing; however, they were built as rapid transit lines, act like typical rapid transit lines within the city center (in terms of stop spacing and system integration), and have greenfield ROWs that are mostly tunneled. (Each of them was eventually broken up into two lines because they were too long. The four original R lines eventually became today's: Lines 1 and 5; Lines 2 and 17; Lines 11 and 16; Line 9 and a future Chongming Line crossing Yangtze River.)
  • M: Your typical heavy rail subway line. Almost entirely radial. Most other subway lines today were originally M lines.
  • L: Mostly similar to the "light rail" concept. They were supposed to be supplemental routes that connect the arterial M lines, and were mostly a mix of circumferential and "tangential". While they didn't literally use what the US calls LRVs, they were initially planned to use 4-car heavy rail "C" vehicles with smaller, narrower cars, whereas the M lines use 6-8 "A" car trains with much bigger cars. (China has three standard HRT car sizes, A, B, C, largest to smallest.)
• At around the same time, planning and construction for Line 5 was underway. It was motivated by Minhang District (where Line 5 almost entirely lies within), who wanted their own subway line and funded it almost entirely by themselves. (As a Boston analogy, imagine Arlington building an LRT by themselves that only connects to Alewife.) This slightly predates the R-M-L concept, so it was seen as a feeder route for Line 1; as such, it used 4-car "C" trains. (This decision presented its own capacity problems for Line 5, but nowhere near as bad as Line 6.)
• When the production line for "C" trains was introduced, the contract asked for more cars than Line 5 needed, so the city started looking into building more subway lines to use the "C" trains on. Line 6 became a natural choice, because it was originally an L line intended to connect all the parallel radial lines in the region (north to south: 10, 12, 14, 9, 2, 7, 13, 8 and 11).
  • (The other line with "C" trains was Line 8. It was originally an M line serving an extremely important corridor, but this decision was made in order to build it ASAP. That turned out to be a terrible choice that left it overcrowded, too. On the other hand, other L lines -- Lines 15, 18, and future Line 20 -- learned from Line 6's mistake: They were built only after the radial M network has been completed, and used the same rolling stock as the M lines to increase capacity.)
• But: Line 6 was accelerated in the build sequence beyond all these radial M lines! At that time, it was the only subway line serving large swarths of a rapidly developing half of Shanghai (regions east of Huangpu River). Its connection with the radial Line 2 was just two stops away from a rapidly growing new CBD (Lujiazui), and Line 2 also connects to all of the city's old CBDs... But that connection is the first transfer after 16 stops. It also doesn't help that Line 6's route south of that transfer point is not very good, due to being a circumferential line.
• Regardless, what was supposed to be a supplemental circumferential line ended up being viewed by everyone as this weird "radial-circumferential" mix. It had immediate implications: massive TOD started along the line in tandem with its construction. Residents and real estate developers don't care about the intended purpose of a line in a 30-year plan -- they want subway service now. But they didn't know about Line 6's low capacity.
• On the first day that Line 6 opened, it was packed to the brim. It remained this way in the 17 years that followed, even today (after 8 new radial lines have opened in an attempt to save it). It's the most crowded subway line in Shanghai, and one of the worst in China.

A curious reader may have started drawing some implications for Boston. On the surface, Line 3's failure would point towards building new ROWs rather than relying on legacy railroad ROWs to save cost, as well as avoiding interlining a ring route with other routes; while Line 6's failure would suggest designing routes with higher capacity to cope with projected demand. I would caution against applying these learning points verbatim to Boston due to vastly different circumstances, but these are valid points. Note that they have no bearings on the discussion here, which is about mixing radial and circumferential routes.

(Damn, I probably should have put this as a post on my website to begin with.)

 

[Longfellow]

I always thought it would be good to pick a high place (say, the top of the old state house) and have a slide with a moderate grade where you wouldn't be going too fast to a low place (say, government center). Only one direction, but the connections possible are incredible.

 

[HenryAlan]

But this blogpost from Alon Levy, "Mixing Circumferential and Radial Transit", explains why he thinks it's a bad idea.

Hey, lots of great points, and thanks for pointing out Alon's post, one I hadn't read yet. Just a quick FYI, Alon uses they/them for pronouns.

 

[Teban54]

Hey, lots of great points, and thanks for pointing out Alon's post, one I hadn't read yet. Just a quick FYI, Alon uses they/them for pronouns.

Thanks! Edited my post to correct the pronouns.

 

[Riverside]

I reiterate my previous suggestion: if Milton thinks the Mattapan Line isn’t rapid transit, they should pay commuter rail prices to board it within their borders. Capen St and Valley Road should close, and most trains should express through Milton and Central Ave stations — if it’s not rapid transit, then the 45 minute headways of the nearby Fairmount Line should be more than sufficient. Hyde Park sees, IIRC, 60-min headways during peak, and limited service off-peak — that could provide a similar model for service to the remaining stations in Milton. (The stations within Boston would of course maintain today’s frequencies, such as they are.)

And if it’s non-rapid transit rail, then probably there should be protected crossings with quad-gates at the remaining grade crossings. Combined with most trains skipping Milton and Central Ave, and closing the other Milton stops, this might actually finally allow the “High Speed Line” to live up to its name.

^^ Mostly these suggestions are tongue-in-cheek. It’s patently absurd to argue that Milton doesn’t receive rapid transit. (And, note that all of my suggestions above involve changes to the status quo, in order to bring it closer to being “not rapid transit”; the status quo itself still definitely qualifies as RT.)

Since Milton apparently wants commuter rail (not really, I know I know I know):

Mattapan ROW and Neponset Trail converted (back) to mainline rail. EMUs ping back and forth between Ashmont and Mattapan (no stops) at high frequency. Diesel spur off of Old Colony runs to a revamped Milton station (through which the EMUs express and which is therefore only served by single track commuter rail).

 

[Teban54]

Since Milton apparently wants commuter rail (not really, I know I know I know):

View attachment 48662

Mattapan ROW and Neponset Trail converted (back) to mainline rail. EMUs ping back and forth between Ashmont and Mattapan (no stops) at high frequency. Diesel spur off of Old Colony runs to a revamped Milton station (through which the EMUs express and which is therefore only served by single track commuter rail).

Don't forget that, in a typical Old Colony fashion, the diesel, single-track commuter rail runs every 60 minutes or longer, even during rush hours.

 

[Semass]

Since Milton apparently wants commuter rail (not really, I know I know I know):

View attachment 48662

Mattapan ROW and Neponset Trail converted (back) to mainline rail. EMUs ping back and forth between Ashmont and Mattapan (no stops) at high frequency. Diesel spur off of Old Colony runs to a revamped Milton station (through which the EMUs express and which is therefore only served by single track commuter rail).

Since we are shitposting, instead of EMUs, I recommend flywheel trains: https://en.m.wikipedia.org/wiki/British_Rail_Class_139#:~:text=6 External links-,Technology,uses the energy for acceleration.

 

[Delvin4519]

A hugely impractical consolidation and straightening of all street running routes. (Take 95 circuitous bus routes and consolidate and straighten them out down to 53 routes)

Many routes would now run on more straighter lines with fewer turns. Most of them now run from one end of the city to downtown, or the other side of the city.

Route duplication is also attempted to be cut down but without much success.

Is it possible to straighten out more routes (remove some unneeded turns without shifting service from an existing street to a non-existant one), or consolidate/deduplicate the routes even more, who knows?

Some of my favorite route straightening/consolidations:

1. Turkey Hill - Arlington Center- Clarendon Hill - Magoun - Sullivan - Bunker Hill - Haymarket
2. Arlington Heights - Harvard - BU Medical Center - JFK UMass - Harbor Point
3. Belmont - Harvard - Kendall
4. Downtown Crossing - Nubian - Mattapan
5. West Medford - Medford Sq. - Mystic Ave - Sullivan - Main St. - Haymarket

Spoiler: Full list of routes

In gray:

• All dedicated transitways and rail ROWs. All of them now run rapid transit service (service every 15 min or better evenings/Sundays, every 9 min or better other times) with stop spacing of approximately 16 - 18 minute walk between consecutive stops.

In blue:

• Wolcott Sq. - South Sta. via Forest Hills, Nubian & South Sta. (32/42/49)
• Oak Sq. - Kenmore via North Brighton (64/57)
• Watertown Sq. - University Park via Central (70)
• Arlington Ctr. - Lechmere via Harvard (67/78/75/69)
• Elm St. - Wellington via Malden Ctr. (99/108)
• Jeffries Point - Orient Heights via Maverick Sta. (120)

In purple:

• Mattapan - Downtown Crossing via Nubian (28/49)
• Dedham Mall - Kenmore via Forest Hills & Brookline Village (34/39/60)
• Belmont Sta. - Kendall via Harvard (74/68)
• Medford Sq. - Lechmere via Magoun Sq. & East Somerville (101/80)
• Elm St. - Wellington via Fellsway (100)
• Linden Sq. - Sullivan via Malden Ctr. & Main St. (108/106/105)
• Woodlawn - Maverick via Bellingham Sq. (111/116)

In pink:

• Ashmont - Prospoect Hill via Nubian, Ruggles, BU West, Kendall & Union Sq. (26/23/CT2/85)
• Millennium Park - Mattapan via Roslindale (36/30)
• Brighton Ctr. - Russel Field via Central Sq. (57/64/83)
• Turkey Hill - Haymarket via Clarendon Hill, Powderhouse Sq, Magoun Sq., Sullivan Sta. & Bunker Hill St. (87/89/93)
• Oak Grove - Airport Sta. via Malden Ctr. (131/104/112)
• Linden Sq. - Beachmont via Revere Ctr. (119)

In magenta:

• Airport Sta. - South Sta. via Ted Williams Tunnel (SL3)
• Arlington Heights - Harbor Point via Mass Ave. & JFK UMass (77/1/8)
• Neponset - Watertown Yd. via Neponset Ave., Fields Corner, Nubian, Brookline Village, Brighton Ctr. & Oak Sq. (202/210/15/66/65/57)
• Wolcott Sq. - Putterham Cir., via Cleary Sq. & Roslindale Sq. (24/50/51)
• West Medford - Haymarket via Medford Sq., Mystic Ave., Sullivan Sq. & Main St. (134/95/92)
• Malden Ctr. - Wellington via Hancock St. (97)

In red:

• Forest Hills - Downtown via Andrew, South Boston, & Summer St. (16/10/7)
• Wren St. - Broadway Sta. via Soldiers Monument, Jackson Sq. & Nubian (38/41/47)
• Cleveland Cir. - Kenmore via Coolidge Corner (C)
• Clarendon Hill - Lechmere via Davis & Union Sq. (87)
• Arlington Ctr. - Oak Island via Medford Sq., Wellington, Woodlawn & Wonderland (95/134/110/411)

In orange:

• Design Ctr. - North Sta. via Seaport Blvd. (4)
• City Point - Assembly Row via Andrew, Ruggles, BU West, Central Sq., Union Sq. & Sullivan (10/CT3/47/91/90)
• Wolcott Sq. - Roslindale Sq. via Georgetowne (40)
• Jackson Sq. - Wonderland via Brookline Village, Harvard, Porter, Davis, Medford/Tufts, Medford Sq., Malden Ctr., Linden Sq. & Revere St. (14/66/96/101/108/411/116)
• Wellington - Airport via Gateway Ctr., Quigley Hospital, Market Basket, Admirals Hll, Bellingham Sq. & Lexington St. (97/112/114/121)

In yellow:

• Logan Airport - South Sta. via Ted Williams Tunnel (SL1)
• Wolcott Sq. - Andrew via Georgetowne, Cleary Sq., Mattapan Sq., Ashmont & Fields Corner (33/24/18)
• Roslindale Sq. - Back Bay via American Legion Hwy, Humboldt Ave., Nubian & Boston Medical Ctr (14/44/10)
• Watertown Sq. - Harvard via Mt. Auburn St. (71)
• West Medford - Lechmere via Boston Ave., Medford/Tufts, Powderhouse Sq., Davis & Highland Ave. (95/94/88)
• Wonderland - Haymarket via Revere St., Revere Ctr., Broadway & Bellingham Sq. (116/111)

In lime:

• Neponset - Forest Hills via Gallivan Blvd & Morton St. (201/215/21)
• Ashmont - Boston Common via Jackson Sq. & Ruggles (26/22/43)
• West Fenway - Copley via Boylston St. (55)
• Reservior - Northgate Shopping Center via Brighton Ctr., Harvard, Union Sq., Sullivan Sta., Broadway & Linden Sq. (86/109/119)

In green:

• City Point - North Station via East 8th. St. & Atlantic Ave. (11/4)
• City Point - Brookline Hills via Broadway, Back Bay & Huntington Ave. (9/39/60)
• Neponset - Ruggles via Adams St., Fields Corner, Blue Hill Ave., Nubian & Roxbury Crossing (202/201/19/45)
• Dedham Mall - Baker & Vermont St. via West Roxbury Sta. (35/37)
• Boston College - Kenmore (B)
• Waverley - Harvard via Mt. Auburn St. (73)
• Arlmont Village - Alewife (78/84/67)
• Lebonan Loop - Oak Grove via Newland St., Commercial St., & Malden Ctr. (106/105/97/132)

 

[Delvin4519]

Hugely impractical/unrealistic and would never happen in a million years ever (this is just for fun).

NSRL, but for HRT lines within the inner core of Boston only. Full segregation of all inner core railway ROWs from "streetcar" tunnels.

(BL and RL tunnels are exceptions due to the Riverside and Braintree Branches already existing, and the need for Lynn/Waltham extensions, also Boston Harbor makes the BL the odd one here anyways).

Extensions included for good measure:

• GL to Nubian/Blue Hill Ave.
• Huntington Ave. subway extension to Brigham & E branch extension to Hyde Sq. (removal of Prudential and NEU stops ... use the 39 bus for local service)
• BLX to Lynn, Kenmore, Riverside (and Needham Heights)
• RLX to Waltham Center (no other line was available for extension to Waltham).

Yes, I know some station names may have better alternatives, plus there's better ways to map it out than using coarsely gridded cells/boxes, (I needed an easy way to shift the map around if I messed up). Don't nag me about it, this is the "bad/terrible transit ideas" thread after all.

 

[Riverside]

^ I think it’s really interesting what happens to the Orange Line corridor in this concept. The branching on the southern end is modestly parsimonious in terms of decreasing frequencies against decreasing density; the northern end a bit less so, giving that Malden Center is currently a major bus hub. But, I wonder how a rail service through eastern Malden to Linden Square would impact (reduce?) transfer pressure at Malden Center.

(Is that a Linden Square <> Davis Square service I spy? )

I've also occasionally thought about rerouting the IRL Orange Line out of the Southwest Corridor to free up extra tracks + flexibility for the mainline services. In real life, I think the negative impact of service to Ruggles is insufficiently offset by the modest increase in flexibility, but this map does a good job of illustrating what a "best case scenario" for that might be (including a Congress St NSRL that provides good transfers to all other lines).

Maybe this is recency bias, but I really am struck by the way that the changes shown on your map align with the thesis I'm trying to get at in my last blog post. Your map dismantles large swaths of the Orange and Red Lines, but most notably keeps intact the Red Line to Harvard, plus the downtown subways, plus the East Boston Tunnel. Even with full electrification and high frequencies on mainline tracks, those pieces remain fully necessary.

I dunno, maybe I’m just rambling about something obvious. But it feels somehow significant that the Orange Line traverses highly urbanized areas almost entirely along a converted mainline route, while the Red Line only utilizes mainline ROWs beyond the Inner Belt (ie JFK/UMass).

(What about Green and Blue? Blue’s “Inner Belt” boundary is debatable, but is probably at Maverick or Airport, so, like Red, the Blue Line really only switches to mainline ROWs outside of the Inner Belt. I put the Green’s southern Inner Belt boundary at Kenmore, so it also passes the Red Line’s test of avoiding mainline ROWs within the Inner Belt. The northern Green Line has an unclear boundary — historically it obviously would be Lechmere, in which case it would pass this test. Now it’s potentially been pushed to Union Sq and East Somerville, which would mean… well, that’s debatable actually. Especially in East Somerville’s case, GLX doesn’t really use the mainline ROW until actually arriving at the station itself. Anyway, all this to say that the Orange Line is unusual if not unique for using mainline ROWs and corridors within the Inner Belt, almost to downtown.)

 

[Delvin4519]

(Is that a Linden Square <> Davis Square service I spy? )

Yes, that is the case. The Saugus Branch RR seemed to have a historical connection to Edgeworth and Wellington Stations on the B & M Haverhill mainline route. It is quite a bit overwritten like the old mainline Harvard Branch, but one can still make out traces of the old connection.

There's also a rail route that bypasses downtown completely and stays within the inner belt, looping from one radial route to another. As such, the north side weirdly enough has a extra pair of (abandoned) railway ROWs for a 2nd crosstown route on the north side (which the south side has essentially none).

Maybe this is recency bias, but I really am struck by the way that the changes shown on your map align with the thesis I'm trying to get at in my last blog post. Your map dismantles large swaths of the Orange and Red Lines, but most notably keeps intact the Red Line to Harvard, plus the downtown subways, plus the East Boston Tunnel. Even with full electrification and high frequencies on mainline tracks, those pieces remain fully necessary.

Yes, that was the whole point of my map The idea of how modern day transit proposals are stuck to historical mainline railway ROWs, and tunnels built downtown primarily for the purpose of unclogging surface streets of streetcars, not connecting mainline railway routes that terminated on the edge of downtown.

I took this idea from the blog post, and pushed it to the most radical extreme example I can think of, for this map I made. The whole map is redrawn so that NSRL link and all the northside and sothside mainlines are down in mostly direct routes. It is no longer centered on the 4 subway tunnels like how modern MBTA system maps are (which would distort the NSRL link).

(What about Green and Blue? Blue’s “Inner Belt” boundary is debatable, but is probably at Maverick or Airport, so, like Red, the Blue Line really only switches to mainline ROWs outside of the Inner Belt. I put the Green’s southern Inner Belt boundary at Kenmore, so it also passes the Red Line’s test of avoiding mainline ROWs within the Inner Belt. The northern Green Line has an unclear boundary — historically it obviously would be Lechmere, in which case it would pass this test. Now it’s potentially been pushed to Union Sq and East Somerville, which would mean… well, that’s debatable actually. Especially in East Somerville’s case, GLX doesn’t really use the mainline ROW until actually arriving at the station itself. Anyway, all this to say that the Orange Line is unusual if not unique for using mainline ROWs and corridors within the Inner Belt, almost to downtown.)

The Lechmere Viaduct seemed to have been isolated from the mainline rail network prior to GLX. It was initially linked with only the current bus routes along Cambridge St. or McGrath Hwy. In order for GLX to be extended from the relocated Lechmere station, the 21 Water Street facility initially blocked the route from Lechmere to Union Sq. and East Somerville, and as such the facility had to be demolished to make room for a path for the Lechmere Viaduct to reach/access the mainline rail network to access East Somerville/USQ.

 

[The EGE]

Yes, that is the case. The Saugus Branch RR seemed to have a historical connection to Edgeworth and Wellington Stations on the B & M Haverhill mainline route. It is quite a bit overwritten like the old mainline Harvard Branch, but one can still make out traces of the old connection.

View attachment 49556 View attachment 49551

This was the original alignment of the Saugus Branch - it was a B&M branch intended to steal Lynn traffic from the Eastern. It opened in February 1853; the Eastern bought the line in 1855 to end the competition. That October, they extended it south to Everett Junction and removed the connection to the B&M. This was one of the first handful of rail segments to be abandoned in New England. The western portion was reused for an industrial spur soon afterwards (which lasted until the last few decades), which is why that portion of the alignment is still visible.

 

[Riverside]

There's also a rail route that bypasses downtown completely and stays within the inner belt, looping from one radial route to another. As such, the north side weirdly enough has a extra pair of (abandoned) railway ROWs for a 2nd crosstown route on the north side (which the south side has essentially none).

Yeah, this is one of my favorite routes to crayon, actually. The potential for a "Brickbottom Bypass" opens up interesting possibilities such as a Porter <> Sullivan service.

Yes, that was the whole point of my map The idea of how modern day transit proposals are stuck to historical mainline railway ROWs, and tunnels built downtown primarily for the purpose of unclogging surface streets of streetcars, not connecting mainline railway routes that terminated on the edge of downtown.

I took this idea from the blog post, and pushed it to the most radical extreme example I can think of, for this map I made. The whole map is redrawn so that NSRL link and all the northside and sothside mainlines are down in mostly direct routes. It is no longer centered on the 4 subway tunnels like how modern MBTA system maps are (which would distort the NSRL link).

Oh awesome! I didn't want to arrogantly presume

I'm particularly appreciative because I actually started to draw a fantasy map of my own for this, with a similar concept -- electrified mainline service running Tokyo-style through the core -- but in my map the north-south connection was going to be through the Washington St Tunnel (or something similar). (More details in the spoiler box.) But I couldn't really figure out a way to make it interesting/informative/provide insight, so I set it aside.

Spoiler: a map in words

This map melded a few ideas together -- it was loosely intended to be alternate-history and therefore have an internally consistent lore, but I couldn't quite get the pieces to gel:

Cambridge <> East Boston Line: under the theory that the Cambridge and East Boston corridors were vital "legs" of the network that would basically have to be built as rapid transit, this map had an east-west subway line from Harvard to Maverick/Airport (couldn't decide)

Washington St Subway as NSRL: the vision here was like yours: funnel all services on mainline ROWs through a core downtown tunnel. (Washington seemed narrow for doing this but, like I said, I didn't get that far)

The El... but light rail: in addition to the sole subway line and the Tokyo-style mainline routes, this map was gonna have lots of subway/el-streetcar lines. All three els got built, but were instead used by streetcars as express routes to downtown -- the way that the Boylston St Subway provided an express route for A, B, and C trolleys. This was inspired by my "Curious Case" map. The Atlantic Ave El created a full loop through downtown, and the Park St Loop and Brattle/Adams Sq Loops served as terminals for "wraparound" services -- e.g. Grove Hall <> Dudley <> Rowes Wharf <> Commercial St <> Scollay/Adams.

The "artistic intent" of the light rail system was for it to kinda be a disaster, taking cues from today's Green Line and SEPTA's trolley lines. And because the downtown Loop is loadbearing for north-south transit service, they have no choice but to keep it in place.

Rough ideas of routes:

• Kenmore Portal
  • Watertown
  • Commonwealth
  • Beacon
• Pleasant St Portal (west)
  • Village Sq (Brookline) via Tremont, Massachusetts, & Huntington
  • Hyde Square via Tremont, Columbus, & Centre St
  • Egleston via Tremont
• Nubian Incline (via east Pleasant St Portal and Washington El)
  • Harambee Park via Grove Hall & Blue Hill Ave
  • Codman Square via Grove Hall & Washington
  • Fields Corner via Geneva
• Sullivan Incline (via Charlestown El)
  • Clarendon Hill via Broadway
  • Malden Center via Main St
  • Glendale via Everett
• Lechmere Incline
  • North Cambridge via Union Sq & Somerville Ave
  • Medford Sq via Medford St

Maverick and Harvard both had feeder streetcar routes running into transfer stations, and like IRL they were eventually bustituted. The BRB&L is a Mattapan-style "high speed line" for Reasons I Couldn't Figure Out But A Harvard <> Lynn Line Would've Disrupted The "Aesthetic" Of The Map So I Didn't Want It.

The Lechmere Viaduct seemed to have been isolated from the mainline rail network prior to GLX. It was initially linked with only the current bus routes along Cambridge St. or McGrath Hwy. In order for GLX to be extended from the relocated Lechmere station, the 21 Water Street facility initially blocked the route from Lechmere to Union Sq. and East Somerville, and as such the facility had to be demolished to make room for a path for the Lechmere Viaduct to reach/access the mainline rail network to access East Somerville/USQ.

Yeah, to be clear, if the test is, "Does a rapid transit line avoid using a mainline ROW to go from the Inner Belt to downtown, a la Kenmore, Sullivan, or Harvard?", then the northern Green Line definitely would have passed before the opening of GLX.

With GLX open, I think it's less clear where the "Inner Belt" is; if we keep it at Lechmere, then Green continues to pass the test. But if we think the Inner Belt is actually the Harvard <> Sullivan axis along Washington St, then both the Medford and Union branches would be debatable at best. (Particularly Union, which relies heavily on the Fitchburg RR ROW.)

 

[Delvin4519]

Yeah, this is one of my favorite routes to crayon, actually. The potential for a "Brickbottom Bypass" opens up interesting possibilities such as a Porter <> Sullivan service.

The Inner belt is quite unique in having 3 different loops that all allow for route combinations that bypass Downtown Boston, as well as the radial routes going off into 3 primary forks and additional secondary forks all in quick succession within the same area as these loop tracks. (the destinations listed here are the last stops before their respective junctions).

• Grand Junction or Prospect Hill -> Magoun Sq.
• Grand Junction or Prospect Hill -> Assembly
• Magoun Sq. -> Assembly

The south side has much less flexibility, where there's only a single loop track that bypasses South Station from WB routes to SB routes, and there is only a 2 way fork in the general area of the loop track.
(The initial image did not show the 3rd loop track utilized, I've updated the original version to use all 3 loop tracks)

I'm particularly appreciative because I actually started to draw a fantasy map of my own for this, with a similar concept -- electrified mainline service running Tokyo-style through the core -- but in my map the north-south connection was going to be through the Washington St Tunnel (or something similar). (More details in the spoiler box.) But I couldn't really figure out a way to make it interesting/informative/provide insight, so I set it aside.

I've thought and am still thinking about making a second version of this map that has a single W - E route connecting Harvard, Cambridge, and East Boston, where the Blue Line feeds directly into the Red Line at Charles MGH. Interesting that we've thought about similar concepts and ideas, relating to the usage of the mainline railway ROWs, and separation from street running streetcar/buses routes/tunnels.

Washington St Subway as NSRL: the vision here was like yours: funnel all services on mainline ROWs through a core downtown tunnel. (Washington seemed narrow for doing this but, like I said, I didn't get that far)

I'm seeing that after the time the Tremont St. tunnel was built in 1897, that all the southside mainline ROWs terminate at South Station, and that original reverse forking/branching of the Back Bay terminal was long gone. I'm seeing the South Station terminal opened in 1899 with the Boston & Providience line rerouted to serve it. The Els began appearing in 1901 and the Washington St. tunnel not until 1908.

In such a case, the only way to connect the mainline railways in the interim after the opening of South Station in 1899, but before the Washington St. tunnel in 1908, would be the NSRL link as we know it today, connecting South and North Stations.

I will either edit this comment, or post a follow up comment, with a modified map that is closer to your ideas. I might not include all the proposed changes/ideas, but Red-Blue being a one seat ride is one of those changes. As I was thinking about it, connecting Harvard and Maverick directly would allow me to sketch a much "cleaner map" in a way. (I initially drew the first version of the map being a modified version of present day routes and proposals, mostly as a thought exercise and not as an alternate history). The alternate history idea seems quite attractive to me.

Yeah, to be clear, if the test is, "Does a rapid transit line avoid using a mainline ROW to go from the Inner Belt to downtown, a la Kenmore, Sullivan, or Harvard?", then the northern Green Line definitely would have passed before the opening of GLX.

With GLX open, I think it's less clear where the "Inner Belt" is; if we keep it at Lechmere, then Green continues to pass the test. But if we think the Inner Belt is actually the Harvard <> Sullivan axis along Washington St, then both the Medford and Union branches would be debatable at best. (Particularly Union, which relies heavily on the Fitchburg RR ROW.)

I like to use the idea of "the fastest crosstown loop that follows only existing bus routes" to find the inner belt loop (and the outer belt loop, where I don't exit original BERy territory).

Such maps give me these results (black lines are inner and outer belt, red are major bus termini or intersections). Kendall MIT being a growing hub these days makes it bit harder now to determine if Central should still be considered the bounds of the inner loop, or if it should be shifted to Kendall. BNRD rerouting to Charlestown will only make this more messy.

I've also thought about the outer loop of the inner core quite a bit. Revere kinds of makes it tricky with Revere St./Revere Center. I'm tempted to route the outer loop to Arlington, but since the 74, 75, and 73 buses in Belmont don't meet until Harvard, it essentially causes Harvard to be considered "the outer loop". The same issue also exists for Brookline Village and urban density dropping off SW of Brookline Village.

 

[Delvin4519]

I'm particularly appreciative because I actually started to draw a fantasy map of my own for this, with a similar concept -- electrified mainline service running Tokyo-style through the core -- but in my map the north-south connection was going to be through the Washington St Tunnel (or something similar). (More details in the spoiler box.) But I couldn't really figure out a way to make it interesting/informative/provide insight, so I set it aside.

Spoiler: a map in words

This map melded a few ideas together -- it was loosely intended to be alternate-history and therefore have an internally consistent lore, but I couldn't quite get the pieces to gel:

Cambridge <> East Boston Line: under the theory that the Cambridge and East Boston corridors were vital "legs" of the network that would basically have to be built as rapid transit, this map had an east-west subway line from Harvard to Maverick/Airport (couldn't decide)

Washington St Subway as NSRL: the vision here was like yours: funnel all services on mainline ROWs through a core downtown tunnel. (Washington seemed narrow for doing this but, like I said, I didn't get that far)

The El... but light rail: in addition to the sole subway line and the Tokyo-style mainline routes, this map was gonna have lots of subway/el-streetcar lines. All three els got built, but were instead used by streetcars as express routes to downtown -- the way that the Boylston St Subway provided an express route for A, B, and C trolleys. This was inspired by my "Curious Case" map. The Atlantic Ave El created a full loop through downtown, and the Park St Loop and Brattle/Adams Sq Loops served as terminals for "wraparound" services -- e.g. Grove Hall <> Dudley <> Rowes Wharf <> Commercial St <> Scollay/Adams.

The "artistic intent" of the light rail system was for it to kinda be a disaster, taking cues from today's Green Line and SEPTA's trolley lines. And because the downtown Loop is loadbearing for north-south transit service, they have no choice but to keep it in place.

Rough ideas of routes:

• Kenmore Portal
  • Watertown
  • Commonwealth
  • Beacon
• Pleasant St Portal (west)
  • Village Sq (Brookline) via Tremont, Massachusetts, & Huntington
  • Hyde Square via Tremont, Columbus, & Centre St
  • Egleston via Tremont
• Nubian Incline (via east Pleasant St Portal and Washington El)
  • Harambee Park via Grove Hall & Blue Hill Ave
  • Codman Square via Grove Hall & Washington
  • Fields Corner via Geneva
• Sullivan Incline (via Charlestown El)
  • Clarendon Hill via Broadway
  • Malden Center via Main St
  • Glendale via Everett
• Lechmere Incline
  • North Cambridge via Union Sq & Somerville Ave
  • Medford Sq via Medford St

Maverick and Harvard both had feeder streetcar routes running into transfer stations, and like IRL they were eventually bustituted. The BRB&L is a Mattapan-style "high speed line" for Reasons I Couldn't Figure Out But A Harvard <> Lynn Line Would've Disrupted The "Aesthetic" Of The Map So I Didn't Want It.

Following up on this. I've took this description and incoperated most, but not all of this information, into a modified version of my original map, that incoperates some, but not all, historical aspects.

As mentioned in my original post, South Station became the hub for southern and western mainline railway routes in 1899, but the Washington St. tunnel was not operating until 1908. As such, I've left the mainline rail network as is. (including SL tunnel, Track 61, and Harvard Branch. Also note that the 77/79 streetcar is omitted on this map.)

My modified map still emphasizes the mainline rail network and having mainline rail routes appear as straight/direct paths, while distorting the appearance of the remaining streetcar routes.

In my view, I find this map to be slightly cleaner in having only the mainline routes show within the inner core of Boston (i.e. hiding the Braintree/Riverside branches from view). Showing the streetcars feeding into the Green Line's tunnels and Els does start to make the map look messy.

If the Green Line tunnel is essentially serving the role of the Washington St. tunnel (with the mainline NSRL tunnel link serving South Station instead of Washington St.), that freed up the light shading of orange for use. I've repurposed the light shading of orange for streetcar routes that don't feed into the Green Line tunnel.
With lines like the Braintree/Riverside lines outside the inner core hidden from view, I've opted for a darker shade of Blue (but not as dark as the mainline NSRL routes) for the Blue Line tunnel extending to Harvard (subsituting what is today the Red Line tunnel). I think the stark contrast between the light shade of Orange and the darker blue for the Harvard and Maverick terminals emphasizes that these streetcars do not feed into the tunnel.

I am pretty satisified with my modified version of the original map, so I'm fine with other versions of this map being created and adapted with other changes as seen fit.

Version with streetcars shown:

EDITED: Added a Google MyMaps version:

Clean version (no streetcars/els) - easier to readapt

 

[Plen-T-Pak]

Put these everywhere

 

[Delvin4519]

Following up on this. I've took this description and incoperated most, but not all of this information, into a modified version of my original map, that incoperates some, but not all, historical aspects.

As mentioned in my original post, South Station became the hub for southern and western mainline railway routes in 1899, but the Washington St. tunnel was not operating until 1908. As such, I've left the mainline rail network as is. (including SL tunnel, Track 61, and Harvard Branch. Also note that the 77/79 streetcar is omitted on this map.)

My modified map still emphasizes the mainline rail network and having mainline rail routes appear as straight/direct paths, while distorting the appearance of the remaining streetcar routes.

In my view, I find this map to be slightly cleaner in having only the mainline routes show within the inner core of Boston (i.e. hiding the Braintree/Riverside branches from view). Showing the streetcars feeding into the Green Line's tunnels and Els does start to make the map look messy.

If the Green Line tunnel is essentially serving the role of the Washington St. tunnel (with the mainline NSRL tunnel link serving South Station instead of Washington St.), that freed up the light shading of orange for use. I've repurposed the light shading of orange for streetcar routes that don't feed into the Green Line tunnel.
With lines like the Braintree/Riverside lines outside the inner core hidden from view, I've opted for a darker shade of Blue (but not as dark as the mainline NSRL routes) for the Blue Line tunnel extending to Harvard (subsituting what is today the Red Line tunnel). I think the stark contrast between the light shade of Orange and the darker blue for the Harvard and Maverick terminals emphasizes that these streetcars do not feed into the tunnel.

I am pretty satisified with my modified version of the original map, so I'm fine with other versions of this map being created and adapted with other changes as seen fit.

Version with streetcars shown:
View attachment 49663
EDITED: Added a Google MyMaps version:
View attachment 49753

Clean version (no streetcars/els) - easier to readapt

I got carried away and converted the aformentioned map to the MetroDreamin interface as well as Google MyMaps (edited the original post above to add the Google MyMaps version). (Yeah, I spent a few days of my time converting the map)

Link to converted map: https://metrodreamin.com/view/cXhmSU9IMGtIcFE4QWcyeHZyV3ZEQTNYU1gyMnww (Map is best viewed in light mode (disable dark mode) and open fullscreen. I also recommend hiding buses/trams/streetcars to just see the lines)

Apparently, Metrodreamin simulates a single roundtrip for every single dream subway/streetcar/ferry/bus line drawn on it's interface. So I converted the map to use the format mostly to see the satisfaction of watching the subways on all the mainline ROWs go super fast/high-speed on the B & A and B & L right-of-ways, while all of the buses creep along street-running in Allston-Brighton and Somerville.

Also I deleted a few lines and rearranged others. Since it's a metro map maker, I essentially just drawn the lines as I saw fit with no regard to realism. (It should be easier to clone and reconfigure the map on this platform than Google MyMaps as well)