The Official MBTA System Map

ra84970

Active Member
Joined
Jun 23, 2020
Messages
312
Reaction score
264
I could not find a thread about the Official MBTA System Map (TM). (Apologies if there is one, but the search function revealed the map contest from a few years ago and the (unofficial) scaled track map.)

The official cartographer and graphic designer who makes the official map will be doing a webinar in a couple of days. I heard about this event last week and finally am getting to post this. https://us02web.zoom.us/meeting/register/tZIuce2uqDwoEtcJYwcfuNT0yLJEio7kbCFM
 
I don't think that's the guy who made the current map. I forget his name but I'm pretty sure he's Russian. He actually emailed me an early copy during the contest the MBTA held and I gave him some notes.
 
I don't think that's the guy who made the current map. I forget his name but I'm pretty sure he's Russian. He actually emailed me an early copy during the contest the MBTA held and I gave him some notes.
From what I take, this is the person who took that Russian entry and made it work within the MBTA's graphic formats.
 
The subway map desperately needs an update to fix the distortions and the bus routes. Obviously you have distortions in every subway map, but the MBTA has some that are particularly egregious:
  • The 77 is on the wrong side of both Davis and Alewife
  • The 1 is horribly wrong between Central and the river
  • The fan shape of the Green Line branches is utterly wrong. The C is almost perfectly straight from Cleveland Circle to Boylston, and the map should reflect that - it's a perfect built-in axis. The D needs to be longer than the B and C, and there needs to be an indication that Reservoir and Cleveland Circle are adjacent.
  • The Worcester Line crosses the B in the wrong place (forgivable) and appears to have a connection at Harvard Avenue (not forgivable)
  • The Green Line is shown crossing the Orange Line, which is unnecessary and wholly wrong
  • The 66 doesn't cross Huntington Avenue - it runs along it for several stops
  • The 32 is on the wrong side of the commuter rail tracks - a big deal when there's only a way to cross the tracks about every half mile.
  • The whole Back Bay and South End area is a mess. The Silver Line loops take up way too much space, the 39 is a disaster, and Copley and Back Bay are too far apart.
Until blatant issues like these are fixed, I'm not sure how much adding Columbus Avenue would help.
 
The subway map desperately needs an update to fix the distortions and the bus routes. Obviously you have distortions in every subway map, but the MBTA has some that are particularly egregious:
  • The 77 is on the wrong side of both Davis and Alewife
  • The 1 is horribly wrong between Central and the river
  • The fan shape of the Green Line branches is utterly wrong. The C is almost perfectly straight from Cleveland Circle to Boylston, and the map should reflect that - it's a perfect built-in axis. The D needs to be longer than the B and C, and there needs to be an indication that Reservoir and Cleveland Circle are adjacent.
  • The Worcester Line crosses the B in the wrong place (forgivable) and appears to have a connection at Harvard Avenue (not forgivable)
  • The Green Line is shown crossing the Orange Line, which is unnecessary and wholly wrong
  • The 66 doesn't cross Huntington Avenue - it runs along it for several stops
  • The 32 is on the wrong side of the commuter rail tracks - a big deal when there's only a way to cross the tracks about every half mile.
  • The whole Back Bay and South End area is a mess. The Silver Line loops take up way too much space, the 39 is a disaster, and Copley and Back Bay are too far apart.
Until blatant issues like these are fixed, I'm not sure how much adding Columbus Avenue would help.

I think there are a lot of necessary compromises regarding geographic accuracy. The whole design theme relies on orthogonal linework which helps legibility. But unfortunately the Boston metro is more like a pile of spaghetti so the disconnect is unavoidable.

That video from a couple weeks ago sheds some good light on the design process and the kind of constraints of layering so much information. The Green Line was particularly difficult because of the number of stops. B C and E are just crazy dense with stops, especially compared to the longer D branch. The lines are at a 45 degree angle and spaced evenly so they can fit all the text in horizontally (past iterations with angled text were hard to read). There's a real art to spacing everything out so it's legible and aesthetically easy on the eyes - while also being grounded in some reality. There's been a few attempts at a more "accurate" scaling of the GL, but it ends up stretching the aspect ratio wider and puts too much visual emphasis on a relatively minor segment of the T (in the grand scheme of things)
Boston_Current_NoBus_2017_1200px.jpg
.

The Green-Orange crossing is one I definitely agree with though and it's planned to be fixed when the new map fully ships:

NewMBTAMap.jpg
 
@The EGE if you haven't watched the presentation linked up thread, I highly recommend, as I think you'll enjoy it.

@bigeman312 The Columbus Ave bus lanes pose an interesting question in terms of the map. You're not wrong -- they'll be higher frequency and more rapid-transit like than some services that do appear on the map (*cough* SL4/SL5). On the other hand, there's no particular intention of "placemaking" behind their construction (i.e. it wasn't built to be "a line"), and moreover the current design language of the map isn't well-suited to showing "a high frequency BRT spine that is fed by multiple decidedly non-BRT branches". (Also, what would we show it as? Another disconnected branch of the Silver Line? A "rapid-transit" line [in thickness and weight] of a brand new color? A new unique variant on the current key bus route lines?)

Overall, I think this (plus the issues The EGE lays out) highlights that the MBTA has outgrown the diagrammatic map and needs to shift to something more geographic. A renewed emphasis on multimodal journeys with walking connections, fare integration with regional rail (fingers crossed), the growing expansion of bus lanes -- all require some level of geographic fidelity in the map that we currently lack. Our current map is more like the WMATA's, but I think we need to look more toward NYC's.

Taking an even bigger step back: I think we're arriving at the point where the T needs to move on from the idea of the subway diagram -- in particular the shapes of the subway lines -- as a piece of branding. The Cambridge Seven Associates' spider map was iconic, but I think even our current map still tries to capture some of the elegant simplicity of that design, and ends up hamstringing itself. Better to de-emphasize the map as a piece of marketing, and re-emphasize the practical aspects.

While I'm on this soapbox, here are some ideas for the next generation of T map:
  • Emphasize frequency, de-emphasize mode (though be sensitive to travel times -- need to show that the 23 comes as often as the Orange Line, but that is also will be slower)
  • Visualize frequency according to "actions required":
    • "Show up and go" (every 10 min or better)
    • "Maybe check a schedule, if you hate waiting" (e.g. every 15-20 min)
    • "Definitely check a schedule" (e.g. waiting more than, I dunno, 25 minutes... this threshold would probably not be hard-and-fast)
  • Do not label minor LRT, BRT and bus stops
    • This one is controversial, as I believe there is user data that suggests that riders like having all of the B/C/E stops labeled
    • But on the other hand it seems absurd that Hawes St (to say nothing of Back of the Hill) is shown with the same visual language as Copley
  • Consider labeling squares on key bus routes (somewhat done today, but it appears unintentional -- Boston and Cambridge are cities of squares, and it'd be great for our map to reflect that)
  • Consider "zooming out" to show high-frequency regional rail service to Lynn and Brockton (beyond the edge of the current map)
  • From North Station to Ruggles and MFA, the Orange and Green Lines are never more than a 10 min walk apart (often less than 5). Bringing the lines visually closer together would better depict walking connections and would make it much easier to solve the thorny issues of the 39 and the Silver Line loops
    • Hot take: don't bother with the SL5 loop -- show it as a line running between the Green and Orange Lines with barbell connections to both Boylston and Chinatown; install clear street-level signage at both to indicate that southbound buses board on Tremont St, and call it a day
  • Super hot take: color key bus routes based on the hub they feed in to -- e.g. Red for Harvard, Central, Quincy, Orange for Forest Hills, Ruggles, Sullivan, Malden, Blue for Maverick and Wonderland, Green for Lechmere, Haymarket (arbitrarily -- need to balance out the overload of Orange with this concept), Copley, Kenmore
    • Begin to present these routes as, for example, part of the "Orange Line ecosystem" -- not quite a branch of the Orange Line, but definitely related
    • We have some precedent for this concept: the Mattapan High Speed Line. Is it part of the Red Line? Yes. Is it a separate service from the rest of the Red Line? Also yes. Is it essentially a feeder service into Ashmont? Also yes.
 
I totally agree with most of your points. I think you may run into accessibility issues concerning the last one, though, unless you add labels to an extent that will end up being a net-negative for readibility.
 
Coloring the key bus routes by hub would be a really cool thing for someone to draw. I'd do it if I had the skill.
 
I understand that the Green Line branches are difficult, but the present style is terrible. The detailed shorelines imply some level of geographical accuracy, and turning arrow-straight rail lines into 90-degree bends is utterly contrary to that. Hopefully stop consolidations, and eventual removal of the access icon when all stops are accessible, will free up some space to do the branches correctly.

Basically everything else I mentioned above is a pure error. Things like the 77 being mirrored from its actual path, and the 1 being shown going through Cambridgeport, should never have made it past the most basic review.

As part of the late-1960s rebranding, bus rollsigns were actually colored with the destination station's line. I don't think it latest long (and might not have even gotten beyond a test). The issue with coloring is that many routes feed into multiple hubs. Are the 1 and 66 red or silver?
 
The MBTA has previously shown the ability to display the Green Line Branches in a far more palatably accurate manner:

OLD_Map.jpg
 
Some (i.e. people like me) would suggest that that is a benefit of the design, not an issue.
Oh definitely. I prefer this over geographic realism. Including the bodies of water provides a point of reference, but from there everything else should (and is) about disseminating information.
 
I totally agree with most of your points. I think you may run into accessibility issues concerning the last one, though, unless you add labels to an extent that will end up being a net-negative for readibility.

Yes, there would need to be some clever design to make it work. I'm not a designer, so I won't pretend to know the best way to accomplish it, but I think it should be considered as a goal -- how do you visually unify the feeder routes with their "parent" subway line?
The MBTA has previously shown the ability to display the Green Line Branches in a far more palatably accurate manner:
That map, in my opinion, also does a significantly cleaner job with the Dorchester bus network than the current map does.
I understand that the Green Line branches are difficult, but the present style is terrible. The detailed shorelines imply some level of geographical accuracy, and turning arrow-straight rail lines into 90-degree bends is utterly contrary to that. Hopefully stop consolidations, and eventual removal of the access icon when all stops are accessible, will free up some space to do the branches correctly.

That is a good point about the mismatch between the geographic fidelity of the shorelines vs the purely diagrammatic nature of the rest of the map. That's a valid concern.
As part of the late-1960s rebranding, bus rollsigns were actually colored with the destination station's line. I don't think it latest long (and might not have even gotten beyond a test). The issue with coloring is that many routes feed into multiple hubs. Are the 1 and 66 red or silver?

I'd never heard about the rollsigns, that's fascinating! As for the 1 and 66, I actually gave that some thought... basically they'd both be red, in my opinion. A few overlapping reasons -- most admittedly arbitrary, but I think with some consistent logic:
  • I wouldn't have any of the feeder routes be Silver. Several reasons for that, but the big one is that a huge chunk of the Dorchester network would/"should" be Silver insofar as it feeds into Nubian, but the reality is that nearly all of those routes continue on to Ruggles, precisely so that they would connect to Orange. So those are really part of "an Orange Line network" rather than "a Silver Line network".
  • You mention the 1 and 66, but there are several other routes that run between hubs -- e.g. the 23 and 28, or the 39. In those cases, I'd arbitrarily choose the color of the "inboundmost" hub, which ends up being a reasonable tiebreaker for most of the radial routes
  • That still leaves the true crosstown routes in limbo -- the 1 and 66, the 86 and 91, the 8, and the 47. One option would be to color these routes differently altogether. However, in general I think the problem can be solved by applying two principles: 1) Try not to overload the map with too many routes of the same color, and 2) Limit the number of "color-originating hubs" as much as possible
    • 1 and 66: Choice between Red and Silver (or maybe Orange) yields Red from Harvard -- a massive center of gravity on the network
    • 86 and 91: Choice between Red and Orange, yields a complete coin toss; Sullivan, Harvard, and Central are all going to be major hubs with many routes originating from each. But, the map will have a lot of orange (Forest Hills, Jackson Square, Ruggles, Sullivan, Malden) so I'd tip the scale toward Red for variety
    • 8: Choice between Green and Red yields Green from Kenmore -- JFK/UMass has virtually no originating routes of its own, and there's a core of routes feeding into Kenmore that run along Brookline Ave, of which the 8 would be a part
    • 47: Actually is Red from both ends, so an easy solve. The 47 is ridiculous though, so if it ever were split into Central-Ruggles and Ruggles-Broadway routes, I'd say the Central route gets Red (from the Central hub) and Broadway gets Orange (from Ruggles)
I think in my head I've come up with the following "color-originating hubs":
  • Red
    • Alewife
    • Harvard
    • Central
    • Kendall
    • South Station
    • Andrew
    • Fields Corner (for routes from south)
    • Ashmont (for routes from south)
    • Mattapan (for routes from south)
    • Quincy Center
  • Orange
    • Malden Center
    • Wellington
    • Sullivan
    • Ruggles
    • Jackson Square
    • Forest Hills
  • Blue
    • Maverick
    • Wonderland
  • Green
    • Lechmere
    • Haymarket (arbitrary, but recognizes the streetcar heritage of the bus routes there, in particular the "foreign cars")
    • Copley (yes, the 39 should be Green, and likely also the 9 and 10)
    • Kenmore
And if anyone is curious what the Key Bus Routes look like in a system like this...

Screen Shot 2021-05-31 at 11.56.57 PM.png


(Boy it sure looks like I spend a long time thinking about these things... the scary thing is that it was the write-up that took a while! The thinking part was relatively straightforward and speedy... And the map, I'd put together previously)
 
Yes, there would need to be some clever design to make it work. I'm not a designer, so I won't pretend to know the best way to accomplish it, but I think it should be considered as a goal -- how do you visually unify the feeder routes with their "parent" subway line?

That map, in my opinion, also does a significantly cleaner job with the Dorchester bus network than the current map does.


That is a good point about the mismatch between the geographic fidelity of the shorelines vs the purely diagrammatic nature of the rest of the map. That's a valid concern.


I'd never heard about the rollsigns, that's fascinating! As for the 1 and 66, I actually gave that some thought... basically they'd both be red, in my opinion. A few overlapping reasons -- most admittedly arbitrary, but I think with some consistent logic:
  • I wouldn't have any of the feeder routes be Silver. Several reasons for that, but the big one is that a huge chunk of the Dorchester network would/"should" be Silver insofar as it feeds into Nubian, but the reality is that nearly all of those routes continue on to Ruggles, precisely so that they would connect to Orange. So those are really part of "an Orange Line network" rather than "a Silver Line network".
  • You mention the 1 and 66, but there are several other routes that run between hubs -- e.g. the 23 and 28, or the 39. In those cases, I'd arbitrarily choose the color of the "inboundmost" hub, which ends up being a reasonable tiebreaker for most of the radial routes
  • That still leaves the true crosstown routes in limbo -- the 1 and 66, the 86 and 91, the 8, and the 47. One option would be to color these routes differently altogether. However, in general I think the problem can be solved by applying two principles: 1) Try not to overload the map with too many routes of the same color, and 2) Limit the number of "color-originating hubs" as much as possible
    • 1 and 66: Choice between Red and Silver (or maybe Orange) yields Red from Harvard -- a massive center of gravity on the network
    • 86 and 91: Choice between Red and Orange, yields a complete coin toss; Sullivan, Harvard, and Central are all going to be major hubs with many routes originating from each. But, the map will have a lot of orange (Forest Hills, Jackson Square, Ruggles, Sullivan, Malden) so I'd tip the scale toward Red for variety
    • 8: Choice between Green and Red yields Green from Kenmore -- JFK/UMass has virtually no originating routes of its own, and there's a core of routes feeding into Kenmore that run along Brookline Ave, of which the 8 would be a part
    • 47: Actually is Red from both ends, so an easy solve. The 47 is ridiculous though, so if it ever were split into Central-Ruggles and Ruggles-Broadway routes, I'd say the Central route gets Red (from the Central hub) and Broadway gets Orange (from Ruggles)
I think in my head I've come up with the following "color-originating hubs":
  • Red
    • Alewife
    • Harvard
    • Central
    • Kendall
    • South Station
    • Andrew
    • Fields Corner (for routes from south)
    • Ashmont (for routes from south)
    • Mattapan (for routes from south)
    • Quincy Center
  • Orange
    • Malden Center
    • Wellington
    • Sullivan
    • Ruggles
    • Jackson Square
    • Forest Hills
  • Blue
    • Maverick
    • Wonderland
  • Green
    • Lechmere
    • Haymarket (arbitrary, but recognizes the streetcar heritage of the bus routes there, in particular the "foreign cars")
    • Copley (yes, the 39 should be Green, and likely also the 9 and 10)
    • Kenmore
And if anyone is curious what the Key Bus Routes look like in a system like this...

View attachment 13478

(Boy it sure looks like I spend a long time thinking about these things... the scary thing is that it was the write-up that took a while! The thinking part was relatively straightforward and speedy... And the map, I'd put together previously)

This map makes it really clear to me that the area just north of Boston needs some higher-quality transit!
 
This map makes it really clear to me that the area just north of Boston needs some higher-quality transit!

Well, yes, but also that map is misleading in some ways. It only shows the Key Bus Routes, and which bus routes receive the designation "key" is a bit of a mystery to me -- certainly I wouldn't disagree with any of the existing designations, but I think some of the omissions are... surprising.

You might be interested in some ridiculous posts I put together over the last year, as background/commentary on this map of frequent bus routes (pre-covid). My original posts were in September and then I did a follow-up in December, including a deep dive on Dorchester's network.

When we include routes that are high-ish frequency at peak hours only, the network looks something like this (gold = frequent all day, bronze = frequent only during rush hour, sometimes only during morning rush):

1622521534258.png


North of Boston still gets hosed (seriously, the frequencies for North Shore buses in particular are amazingly low), but it's not quite as dire as the map in my post above implies.
 
I still think that even with the inclusion of a "bronze" (sub-Key Bus Route) category shows the corridors that the BNR better be investing more resources in - they should basically be Key Bus Routes. The Malden-Medford-Everett and the North Shore are clearly areas that need more service if they already have "bronze" level service with peak-hour frequent buses. Not that I want every bus route shown, but, I do want more key bus routes on the map :)
 

Back
Top