MBTA Buses & Infrastructure
- Thread starter Arlington
- Start date
- Joined
- Jan 22, 2012
- Messages
- 5,078
- Reaction score
- 1,658
Bad policy focus AND corruption? Surprise surprise.
- Joined
- Dec 10, 2011
- Messages
- 5,599
- Reaction score
- 2,710
I Don't have enough evidence to make the accusation directly. All I can say is that the projects they chose look really boring and strangely distributed geographically.
The DOT TIGER process (to which this is the successor) was designed to be an apolitical, math based competition at the federal level that would encourage States to innovate and think about all projects in terms of mobility benefit per dollar spent (instead of horse trading Bridges to Nowhere)
And to ask the feds to fund the good projects that somehow had failed to get fully funded locally (often because they were a little bit weird, a little bit multimodal, or a little bit trans jurisdictional)
Recall too that under Bush and Obama secretaries Mineta and La Hood were the token cabinet secretary from the other party (Mineta D-CA and La Hood D-TX) so that transportation had had a tradition the federal level of being less partisanly administered.
There's a sprinkle of rail and Transit, but the geography and the mode choice sure looks like someone jiggered the definition of mobility.
Last edited:
- Joined
- Jan 22, 2012
- Messages
- 5,078
- Reaction score
- 1,658
If I felt like it, I could compile enough information on the McConnells federal dealings that makes a clear case for corrupt behavior. But innuendo works for now.
#bancars
Senior Member
- Joined
- Jun 1, 2019
- Messages
- 1,438
- Reaction score
- 5,653
Aside from the quasi-corruption, it's also not a good look for the feds given the demographics that these two projects would likely serve.
F-Line to Dudley
Senior Member
- Joined
- Nov 2, 2010
- Messages
- 9,259
- Reaction score
- 9,263
Well...I guess now the question is:
$15M wasn't a terribly a big ask, since the state/city were also on the hook for a matching $15M if it were committed. And the entire project has slimmed down its pricetag enormously from the 2009 28X proposal which had concrete barriers and frilly Silver Line shelters, so due diligence has already been done at serving up a functional equivalent at cut-rate price. Hopkinton interchange just had a motherlode of several times that amount dumped on it. Do they really go ahead and cancel the BHA lanes citing funding shortfall, or do Baker/Pollack go fishing around elsewhere to cobble together the missing $15M. It's not a good look either for a shovel-ready transit project in a high environmental-justice community to get turfed so they can fuckin' add a lane to the Hopkinton ramps, let alone cry poverty as the sole reason for it.
$15M wasn't a terribly a big ask, since the state/city were also on the hook for a matching $15M if it were committed. And the entire project has slimmed down its pricetag enormously from the 2009 28X proposal which had concrete barriers and frilly Silver Line shelters, so due diligence has already been done at serving up a functional equivalent at cut-rate price. Hopkinton interchange just had a motherlode of several times that amount dumped on it. Do they really go ahead and cancel the BHA lanes citing funding shortfall, or do Baker/Pollack go fishing around elsewhere to cobble together the missing $15M. It's not a good look either for a shovel-ready transit project in a high environmental-justice community to get turfed so they can fuckin' add a lane to the Hopkinton ramps, let alone cry poverty as the sole reason for it.
Pretty sure thats a feature, not a bug.
F-Line to Dudley
Senior Member
- Joined
- Nov 2, 2010
- Messages
- 9,259
- Reaction score
- 9,263
Quincy Bus Maintenance Facility MEPA ENF Scoping Session | Events | MBTA
Official website of the MBTA -- schedules, maps, and fare information for Greater Boston's public transportation system, including subway, commuter rail, bus routes, and boat lines.
www.mbta.com
Separate from today's MassDOT/FCMB meeting is a New Quincy Bus Garage scoping session where they're going to dive into the MEPA Environmental Notification Form submitted last week. Will probably entail a walk-through of renders and whatnot.
I don't even get it, even of all the highway interchanges that desperately need to be redone, the Hopkinton interchange never really was high up on my list. Very much hope Baker & Co can shake some couches to find the extra $15 Million to fund the BHA project, though.
So, um, is the T just not publishing Silver Line timetables anymore? For as long as I can remember, a specific Silver Line bus schedule has been published, listing out specific times buses were due to arrive (although if memory serves, there were indeed stretches of "every 7 minutes until" during peak hours as you still see for example on the 39 schedule today). Now though, when you go to, say, the SL5 route page, the link for the schedule just sends you to the Rapid Transit schedule (which itself has been a bit simplified).
It seems like it would be pretty lousy to show up at one of the SL3 stations in Chelsea, or at a Design Center SL2 stop and find yourself waiting 13-15 minutes.
I mean, I get it that in theory all of the branches of the Silver Line should be walk-up frequencies -- and it is true that the Washington St segment appears to be cumulatively a bus every 6 minutes all day (assuming they are scheduled evenly), the Waterfront Transitway appears to be cumulatively 4 minutes (same caveat), and the airport sees a bus roughly every 10 minutes.
But especially for SL2 and SL3, this doesn't seem great.
Also worth noting that the 111, which overlaps a bit with SL3 and which has comparable off-peak headways and much better ones during peak, has all of the times listed out on its schedule.
It seems like it would be pretty lousy to show up at one of the SL3 stations in Chelsea, or at a Design Center SL2 stop and find yourself waiting 13-15 minutes.
I mean, I get it that in theory all of the branches of the Silver Line should be walk-up frequencies -- and it is true that the Washington St segment appears to be cumulatively a bus every 6 minutes all day (assuming they are scheduled evenly), the Waterfront Transitway appears to be cumulatively 4 minutes (same caveat), and the airport sees a bus roughly every 10 minutes.
But especially for SL2 and SL3, this doesn't seem great.
Also worth noting that the 111, which overlaps a bit with SL3 and which has comparable off-peak headways and much better ones during peak, has all of the times listed out on its schedule.
JeffDowntown
Senior Member
- Joined
- May 28, 2007
- Messages
- 4,601
- Reaction score
- 3,259
Why bother? The SL has never arrived to a schedule. NEVER!
I'm struggling to remember a time when I could remember an actual timetable for SL1-2-3 though I thought they may have had one for SL4-5 a few years ago. Not sure that a time table for most of the routes is really useful...like you mentioned for SL1, SL4, or SL5 is not
I do know if you are interested in that type of information though, the MBTA website does have the scheduled trips information available if you drill down to the individual stop/station. E.g. Eastern Ave "station". https://www.mbta.com/schedules/743/..._id]=0&schedule_direction[origin]=place-estav
I found an SL1+SL2 schedule from 2018 on the Wayback Machine here: https://web.archive.org/web/2018062...oute_pdfs/2018-spring/rtSilver123-revised.pdf
The Wayback Machine is pretty hit-or-miss for PDF copies, so I can't lay my hands on a more recent one, but I'm 99% sure that I've seen full copies with all 5 routes on them.
But that is a good trick for pulling data for individual stops. And I don't deny that it may be a more effective strategy, I'm just a little surprised by this apparent move on the T's part to stop publishing the individual timetable.
The Wayback Machine is pretty hit-or-miss for PDF copies, so I can't lay my hands on a more recent one, but I'm 99% sure that I've seen full copies with all 5 routes on them.
But that is a good trick for pulling data for individual stops. And I don't deny that it may be a more effective strategy, I'm just a little surprised by this apparent move on the T's part to stop publishing the individual timetable.
I made another map: MBTA Subway + Frequent Bus.
This was prompted by looking at the Frequency Chart on the System Map, and noticing, "Boy howdy, those South Boston routes go hard during rush hour":
4 minute headways in the morning rush? That's turn up and go, right there. But neither the 7, nor the 9, nor the 11 are advertised as "key bus routes" -- which is reasonable because those off-peak headways are much less impressive. (As is the 11's PM peak headway.)
But this got me thinking: are there other routes like this? Is there an entire "turn up and go" bus network that activates during rush hour but gets hidden on the all-routes system map?
The short answer is, yes -- but the Southie network is an usually extreme case.
So I set out on project to comb through MBTA schedules and documentation and make a map of the routes which are served at "rapid transit frequencies" for some or part of the day.
Recognizing that I wanted to differentiate between all-day frequent and peak-only frequent, I created two categories of bus route, and gave them names riffing on the precious metal theme of the Silver Line:
The "Gold Line" network features buses that run at "rapid transit frequencies" throughout the day. The definition of "rapid transit frequency" is not universal across the system: in the suburbs, some lower-frequency services are included if they are relatively consistent throughout the day, and relatively consistent with the frequencies of something like the Riverside or Braintree Lines.
The "Bronze Line" network features buses that run at "rapid transit frequencies" during peak periods, but at significantly reduced frequencies off-peak. As with Gold, there is some variability in "rapid transit frequencies."
A second goal I sought to accomplish was to identify corridors where multiple routes "stack" and produce cumulatively high-frequency service. This was definitely a mixed bag -- some examples are more provisional than others, relying on uncoordinated schedules that will sometimes produce lengthy gaps. However, it still seemed worthwhile to identify them, both as opportunities for better coordination in some cases, and as areas which could benefit from bus lanes.
Every route has been annotated with its AM peak, midday, and PM peak headways. In some cases, these are more precise than others, but are still meant to illustrate the parity of certain bus routes to their much more prominent rail siblings.
A couple of caveats:
First: this is an aggregate analysis of aggregate analyses. Put another way, I'm working with summaries of summaries here. There's a lot of nuance that I am glazing over and have overlooked. In some cases, this no doubt will lead to a rosier depiction of route frequencies, and in other cases may understate the same. It also means that this map is not suitable for individual journey planning. Caveat viator.
Second: my analysis completely excludes evening/night and weekend frequencies. I did this in part to simplify things, particularly because even many of the Gold Line routes drop into 20-min-or-worse territory on weekends. (It would be an interesting map to highlight routes that have constant frequencies 7 days a week... next project!) However, it's really important to note, especially now during the pandemic, that just because a route is high frequency during the day, it doesn't mean that it doesn't have riders who would benefit from high-frequency service outside of the 9-5 workday.
A couple of invitations:
First: I highly recommend turning on and off the different layers to observe how the network changes between peak and off-peak. There really are some fascinating insights here, which I'll put into a separate post.
Second: Please give feedback! Especially if you are or were a rider of one of these routes!
This was prompted by looking at the Frequency Chart on the System Map, and noticing, "Boy howdy, those South Boston routes go hard during rush hour":
4 minute headways in the morning rush? That's turn up and go, right there. But neither the 7, nor the 9, nor the 11 are advertised as "key bus routes" -- which is reasonable because those off-peak headways are much less impressive. (As is the 11's PM peak headway.)
But this got me thinking: are there other routes like this? Is there an entire "turn up and go" bus network that activates during rush hour but gets hidden on the all-routes system map?
The short answer is, yes -- but the Southie network is an usually extreme case.
So I set out on project to comb through MBTA schedules and documentation and make a map of the routes which are served at "rapid transit frequencies" for some or part of the day.
Recognizing that I wanted to differentiate between all-day frequent and peak-only frequent, I created two categories of bus route, and gave them names riffing on the precious metal theme of the Silver Line:
The "Gold Line" network features buses that run at "rapid transit frequencies" throughout the day. The definition of "rapid transit frequency" is not universal across the system: in the suburbs, some lower-frequency services are included if they are relatively consistent throughout the day, and relatively consistent with the frequencies of something like the Riverside or Braintree Lines.
The "Bronze Line" network features buses that run at "rapid transit frequencies" during peak periods, but at significantly reduced frequencies off-peak. As with Gold, there is some variability in "rapid transit frequencies."
A second goal I sought to accomplish was to identify corridors where multiple routes "stack" and produce cumulatively high-frequency service. This was definitely a mixed bag -- some examples are more provisional than others, relying on uncoordinated schedules that will sometimes produce lengthy gaps. However, it still seemed worthwhile to identify them, both as opportunities for better coordination in some cases, and as areas which could benefit from bus lanes.
Every route has been annotated with its AM peak, midday, and PM peak headways. In some cases, these are more precise than others, but are still meant to illustrate the parity of certain bus routes to their much more prominent rail siblings.
A couple of caveats:
First: this is an aggregate analysis of aggregate analyses. Put another way, I'm working with summaries of summaries here. There's a lot of nuance that I am glazing over and have overlooked. In some cases, this no doubt will lead to a rosier depiction of route frequencies, and in other cases may understate the same. It also means that this map is not suitable for individual journey planning. Caveat viator.
Second: my analysis completely excludes evening/night and weekend frequencies. I did this in part to simplify things, particularly because even many of the Gold Line routes drop into 20-min-or-worse territory on weekends. (It would be an interesting map to highlight routes that have constant frequencies 7 days a week... next project!) However, it's really important to note, especially now during the pandemic, that just because a route is high frequency during the day, it doesn't mean that it doesn't have riders who would benefit from high-frequency service outside of the 9-5 workday.
A couple of invitations:
First: I highly recommend turning on and off the different layers to observe how the network changes between peak and off-peak. There really are some fascinating insights here, which I'll put into a separate post.
Second: Please give feedback! Especially if you are or were a rider of one of these routes!
- Does/did the day-to-day reality actually "feel" like "turn up and go"?
- Are there any routes that should be added to this? Or any that should be excluded?
- Do the various "stacked" sections actually reliably produce high frequencies?
HenryAlan
Senior Member
- Joined
- Dec 15, 2009
- Messages
- 3,963
- Reaction score
- 3,884
@Riverside, this is fantastic! I have one thought regarding the 34/34E bronze line. If you add in the 40 and 50, you might actually see gold level service extend from Rozzie Square as far as Metropolitan Ave., though I'm not 100% sure. Either way, as somebody who lives on that "route", the 40 and 50 combine with it to provide what I consider to be show up and go service throughout the day.
Some analysis:
AM vs PM
It's worth noting a distinct pattern, particularly noticeable in Bronze routes, where the PM peak frequency is lower than the AM. I'm not entirely sure why this is -- it may be that the PM peak stretches longer, so you don't have as high a demand at any given moment. But regardless, I did opt to include routes that had high-frequency peak in the AM only, but I marked them with a superscript ª.
Gold Line network
These route fall into three categories: radial, circumferential, and cumulative. The vast majority of the network are radial routes (with a few routes in the "Dorchester lattice" that are somewhat hybrid). Most of the radial corridors are familiar, with many being current or former candidates for rail service of some kind. Nearly all of them are feeder services that hook into a rail line, mostly at Harvard, Kenmore, Ruggles, and Forest Hills.
There are only a couple of circumferential routes: the 1, the 66 and debatably the 22 and 31. I'd say the most notable thing about the circumferential routes is how few of them there are -- more on that below.
Cumulative Corridors
There are about a half-dozen cumulative corridors, depending how you count them. The most well-known of these is probably the Roslindale Village-Forest Hills segment, and then the Nubian-Ruggles zigzag. Then you have three little odd segments to the north: a 1-mile stretch running into Lechmere, a Union-Sullivan segment, and then a Wellington-Everrett segment. These are all pretty clearly "accidental" corridors that arise naturally out of major bus hubs. Finally, you have a stretch from Kenmore to Nubian that could/should/may have consistently high frequency service all day; this segment is tricky to map, however, because no route actually runs all the way through from Kenmore to Nubian via Roxbury Crossing: the 8 and 19 come close, but they travel via Melnea Cass rather than Tremont + Malcolm X.
No Gold In These Here Hills
Now, one thing that is notable about the Gold Line network are the places where it doesn't exist. There's a very clear diagonal axis running roughly from Davis to Andrew: north of this axis, there are almost no Gold Line routes, except for three "accidental" cumulative corridors, and then two corridors serving Chelsea.
No single route out of Sullivan, Wellington, Malden, Lechmere, Wonderland, and only two out of Maverick, reach Gold status. I think I was most surprised to see the absence in Somerville and Everett, both communities that have made several moves in recent years to become more actively pro-transit. I was also very surprised that none of the North Shore routes made the cut, even in cumulative form -- more on that below.
I'm not saying that this is a bad thing necessarily -- all-day high-frequencies may not be warranted in those areas. But it was still surprising.
Notably High Frequencies
There are five Gold routes that I marked with an asterisk to indicate notably high frequencies -- all-day better-than-10-min, or 6-min-or-better at any one time: the 15, 23, 31, 57/57A, and 73. The 15 and 23 were unsurprising, and both hit 6-min during the AM peak. The 57/57A combo was also not particularly surprising, given that it's one of the few corridors where the T actively schedules two routes on top of each other.
The 31 was interesting, though -- from what I can see, it is the most consistently high frequency service on the entire network, advertised at 8-min headways all day. The 31 is also unusual among Gold routes in that it is not designated as a Key Bus Route.
And, for all the talk of Belmont not being friendly to transit, it's worth noting that Waverly enjoys the highest frequency Gold route, and one of the highest peak frequency routes in the entire system: the 73 operates at 6-min AM peak, 15-min-midday, and 5-min PM peak. (Yes, 15-min is a bit high for a Gold route, but my standards are lower in the suburbs.) And though it travels through Watertown and Cambridge as well, the 73's Better Bus Profile reveals consistent ridership across the route, if anything skewed slightly toward Belmont.
The Pandemic
It's also worth highlighting three routes whose Gold status has shifted during the pandemic.
The 39 previously operated at 7-min/12-min/9-min headways, akin to its bustitution sister the 57. Currently, however, its midday frequency has dropped to 15-min. Strictly speaking, that should have knocked it out of Gold using the standards I imposed for that area (which cap mid-day frequency at 13-min). However, given the historical precedent, and given that 7/15/9 isn't that far off from the 8.5/14/8.5 of the Red Line branches, I opted to keep it Gold.
The 93 previously operated at 8-min/16-min/8-min. When I started this project, prior to the pandemic, I tentatively categorized it as "barely Gold" -- those peak headways are good, and like the current 39, the off-peak isn't too far off from the Red Line's branches. However, today the 93's midday frequency is 20-min, though its peak frequency of 8-min is maintained. However, I felt that was a bridge too far, and it got shifted to Bronze.
Finally, the 111, which historically has been at better-than-10-min headways all day, now has its midday frequencies reduced to 18 minutes. This seems... odd to me. Given the historical frequencies, I've kept it as Gold.
Bronze Line network
This is the more interesting part of the map, in my opinion. If you switch on the Bronze layer after looking at the map with just the Gold layer, you'll notice several gaps getting filled in, and it starts to look something like what we'd expect a map of "major" bus routes would look like. These routes fall into a more diverse range of categories, and I'll go through them one by one.
Local "One-Seaters"
These are routes which I suspect often represent the entirety of the typical commuter's journey. These include the aforementioned Southie network (which offers a northern route to downtown via the Seaport, a middle route to Copley, and a southern route to downtown), the 93, and -- debatably -- the 65, which strikes me as mainly being useful for commuting to Longwood from Brighton and Brookline.
Radial Feeders
These are routes whose primary purpose appears to be bringing local passengers to rail hubs to continue their journey into Boston. This map reveals about half-a-dozen such hubs -- and again, it's surprising what makes the list:
Quincy Center
One of the visually tidiest parts of the map, Quincy Center has high(ish) frequency from the 225, high AM but not PM frequency on the 214 and 216 (although theoretically those two could be scheduled to provide all-day high-service on their shared segment), and then cumulatively high frequency peak service to Bicknell Square in Weymouth on the 220 + 222, which do appear to be timetabled together to form such service.
Forest Hills
Alongside the four Gold corridors, the 34/34E combination creates another borderline case. In theory, the dual 30-min headways of the two routes during midday should result in 15-min headways or better all day, which would mark this as a Gold route along the shared segment. However, the midday spacing is uneven and has some pretty long gaps, so I didn't feel it quite cleared the bar. It is notable, however, that the 34 is advertised as having 10 or 11-min headways during the peak, which on its own would merit inclusion as a Bronze Line; the addition of the 20-min 34E at peak is an interesting supplement. (Also, the printed timetable for this route is terrible at highlighting the frequencies along the shared stretch.)
Davis
A dubious contender on this list, Davis has no individual radial Bronze routes feeding into it. (I'd argue that the 89 is more of a circumferential route.) However, College Ave up to Winthrop Street shares the 94 and the 96, which theoretically could accumulate to 10-min headways in the AM peak. There's a really interesting topology of other routes that overlap this and neighboring corridors, including the 80, 89 and the 95, which is an example of a theme that's repeated elsewhere -- some areas have lots of routes, but because they need to cover so many permutations, it's much harder to treat their overlap as cumulative. And obviously, the opening of GLX will see a compete rework of this quadrant of the network.
Sullivan
An interesting array here. The aforementioned 93 boards some 20% of its inbound riders here -- the Better Bus Profile notes that these are individuals returning to their residences along Bunker Hill Street, so at this end it behaves more like a radial feeder than a local one-seater. The 89 is a bit of a hybrid of radial feeder and circumferential. The combo of the 104 + 109 to Everett is well-known at this point, thanks to Everett's bus lane efforts. But the one that surprised me was the 101, which runs a triangular route between Sullivan and Malden via Medford at 12-min peak headways. That's not a route that's on my radar much, though I can definitely see its usefulness.
(If memory serves, bus lanes on Broadway in Somerville are coming, which would be great since both the 101 and 89 run at 10-12 min headways at peak. Off-peak, they both drop to 30-min; in theory, these could coordinate to provide all-day 15-min-or-better service along Broadway, but if I recall correctly their current schedules produce too many gaps in the midday to reach Gold status.)
(continued below)
AM vs PM
It's worth noting a distinct pattern, particularly noticeable in Bronze routes, where the PM peak frequency is lower than the AM. I'm not entirely sure why this is -- it may be that the PM peak stretches longer, so you don't have as high a demand at any given moment. But regardless, I did opt to include routes that had high-frequency peak in the AM only, but I marked them with a superscript ª.
Gold Line network
These route fall into three categories: radial, circumferential, and cumulative. The vast majority of the network are radial routes (with a few routes in the "Dorchester lattice" that are somewhat hybrid). Most of the radial corridors are familiar, with many being current or former candidates for rail service of some kind. Nearly all of them are feeder services that hook into a rail line, mostly at Harvard, Kenmore, Ruggles, and Forest Hills.
There are only a couple of circumferential routes: the 1, the 66 and debatably the 22 and 31. I'd say the most notable thing about the circumferential routes is how few of them there are -- more on that below.
Cumulative Corridors
There are about a half-dozen cumulative corridors, depending how you count them. The most well-known of these is probably the Roslindale Village-Forest Hills segment, and then the Nubian-Ruggles zigzag. Then you have three little odd segments to the north: a 1-mile stretch running into Lechmere, a Union-Sullivan segment, and then a Wellington-Everrett segment. These are all pretty clearly "accidental" corridors that arise naturally out of major bus hubs. Finally, you have a stretch from Kenmore to Nubian that could/should/may have consistently high frequency service all day; this segment is tricky to map, however, because no route actually runs all the way through from Kenmore to Nubian via Roxbury Crossing: the 8 and 19 come close, but they travel via Melnea Cass rather than Tremont + Malcolm X.
No Gold In These Here Hills
Now, one thing that is notable about the Gold Line network are the places where it doesn't exist. There's a very clear diagonal axis running roughly from Davis to Andrew: north of this axis, there are almost no Gold Line routes, except for three "accidental" cumulative corridors, and then two corridors serving Chelsea.
No single route out of Sullivan, Wellington, Malden, Lechmere, Wonderland, and only two out of Maverick, reach Gold status. I think I was most surprised to see the absence in Somerville and Everett, both communities that have made several moves in recent years to become more actively pro-transit. I was also very surprised that none of the North Shore routes made the cut, even in cumulative form -- more on that below.
I'm not saying that this is a bad thing necessarily -- all-day high-frequencies may not be warranted in those areas. But it was still surprising.
Notably High Frequencies
There are five Gold routes that I marked with an asterisk to indicate notably high frequencies -- all-day better-than-10-min, or 6-min-or-better at any one time: the 15, 23, 31, 57/57A, and 73. The 15 and 23 were unsurprising, and both hit 6-min during the AM peak. The 57/57A combo was also not particularly surprising, given that it's one of the few corridors where the T actively schedules two routes on top of each other.
The 31 was interesting, though -- from what I can see, it is the most consistently high frequency service on the entire network, advertised at 8-min headways all day. The 31 is also unusual among Gold routes in that it is not designated as a Key Bus Route.
And, for all the talk of Belmont not being friendly to transit, it's worth noting that Waverly enjoys the highest frequency Gold route, and one of the highest peak frequency routes in the entire system: the 73 operates at 6-min AM peak, 15-min-midday, and 5-min PM peak. (Yes, 15-min is a bit high for a Gold route, but my standards are lower in the suburbs.) And though it travels through Watertown and Cambridge as well, the 73's Better Bus Profile reveals consistent ridership across the route, if anything skewed slightly toward Belmont.
The Pandemic
It's also worth highlighting three routes whose Gold status has shifted during the pandemic.
The 39 previously operated at 7-min/12-min/9-min headways, akin to its bustitution sister the 57. Currently, however, its midday frequency has dropped to 15-min. Strictly speaking, that should have knocked it out of Gold using the standards I imposed for that area (which cap mid-day frequency at 13-min). However, given the historical precedent, and given that 7/15/9 isn't that far off from the 8.5/14/8.5 of the Red Line branches, I opted to keep it Gold.
The 93 previously operated at 8-min/16-min/8-min. When I started this project, prior to the pandemic, I tentatively categorized it as "barely Gold" -- those peak headways are good, and like the current 39, the off-peak isn't too far off from the Red Line's branches. However, today the 93's midday frequency is 20-min, though its peak frequency of 8-min is maintained. However, I felt that was a bridge too far, and it got shifted to Bronze.
Finally, the 111, which historically has been at better-than-10-min headways all day, now has its midday frequencies reduced to 18 minutes. This seems... odd to me. Given the historical frequencies, I've kept it as Gold.
Bronze Line network
This is the more interesting part of the map, in my opinion. If you switch on the Bronze layer after looking at the map with just the Gold layer, you'll notice several gaps getting filled in, and it starts to look something like what we'd expect a map of "major" bus routes would look like. These routes fall into a more diverse range of categories, and I'll go through them one by one.
Local "One-Seaters"
These are routes which I suspect often represent the entirety of the typical commuter's journey. These include the aforementioned Southie network (which offers a northern route to downtown via the Seaport, a middle route to Copley, and a southern route to downtown), the 93, and -- debatably -- the 65, which strikes me as mainly being useful for commuting to Longwood from Brighton and Brookline.
Radial Feeders
These are routes whose primary purpose appears to be bringing local passengers to rail hubs to continue their journey into Boston. This map reveals about half-a-dozen such hubs -- and again, it's surprising what makes the list:
Quincy Center
One of the visually tidiest parts of the map, Quincy Center has high(ish) frequency from the 225, high AM but not PM frequency on the 214 and 216 (although theoretically those two could be scheduled to provide all-day high-service on their shared segment), and then cumulatively high frequency peak service to Bicknell Square in Weymouth on the 220 + 222, which do appear to be timetabled together to form such service.
Forest Hills
Alongside the four Gold corridors, the 34/34E combination creates another borderline case. In theory, the dual 30-min headways of the two routes during midday should result in 15-min headways or better all day, which would mark this as a Gold route along the shared segment. However, the midday spacing is uneven and has some pretty long gaps, so I didn't feel it quite cleared the bar. It is notable, however, that the 34 is advertised as having 10 or 11-min headways during the peak, which on its own would merit inclusion as a Bronze Line; the addition of the 20-min 34E at peak is an interesting supplement. (Also, the printed timetable for this route is terrible at highlighting the frequencies along the shared stretch.)
Davis
A dubious contender on this list, Davis has no individual radial Bronze routes feeding into it. (I'd argue that the 89 is more of a circumferential route.) However, College Ave up to Winthrop Street shares the 94 and the 96, which theoretically could accumulate to 10-min headways in the AM peak. There's a really interesting topology of other routes that overlap this and neighboring corridors, including the 80, 89 and the 95, which is an example of a theme that's repeated elsewhere -- some areas have lots of routes, but because they need to cover so many permutations, it's much harder to treat their overlap as cumulative. And obviously, the opening of GLX will see a compete rework of this quadrant of the network.
Sullivan
An interesting array here. The aforementioned 93 boards some 20% of its inbound riders here -- the Better Bus Profile notes that these are individuals returning to their residences along Bunker Hill Street, so at this end it behaves more like a radial feeder than a local one-seater. The 89 is a bit of a hybrid of radial feeder and circumferential. The combo of the 104 + 109 to Everett is well-known at this point, thanks to Everett's bus lane efforts. But the one that surprised me was the 101, which runs a triangular route between Sullivan and Malden via Medford at 12-min peak headways. That's not a route that's on my radar much, though I can definitely see its usefulness.
(If memory serves, bus lanes on Broadway in Somerville are coming, which would be great since both the 101 and 89 run at 10-12 min headways at peak. Off-peak, they both drop to 30-min; in theory, these could coordinate to provide all-day 15-min-or-better service along Broadway, but if I recall correctly their current schedules produce too many gaps in the midday to reach Gold status.)
(continued below)
Malden Center
A few notes here. First of all, though there are a lot of routes that travel from Malden to Everett, they use diverging routes to get there, meaning there's no single corridor to map directly between the two. If you're at Malden Center and you want to get to the southwestern part of Everett, and don't mind walking a bit at the end (and are flexible on how long it takes), it is true that you have your choice of the 97, 99, 104, 105, and 106, so you probably won't have to sit around waiting for long.
Second, there is a corridor along Salem St to the east that probably has stretches of decent frequency, though it's likely inconsistent. The 108 runs every 20 during peak, and the 106, 411 and 430 all run in the 30-40 minute range, so you're probably gonna have at least one bus in between the 108's, which means you probably won't have to wait more than 15 minutes, which in a suburb arguably qualifies it for Bronze. (But notice how many caveats I needed to put in there!)
Finally, it's worth highlighting that the combo 136 + 137 yield pretty regular 11-min headways in the AM peak (which are not mirrored in the 30-min PM peak) all the way up to Wakefield, with reconverging service at Reading (unmapped). This is a good example of a "hidden" corridor -- both routes are advertised at 35-min headways in the Frequency Chart, but if you look at the departures from Wakefield Square between 7am and 8:15am (targeting arrival in downtown by/before 9am), you do indeed see 9-12 min headways.
Orient Heights
Barely a "hub" on this map, but it is where the 712 + 713 radial feeder hooks in. I explained the 712 + 713's service patten in painful detail in the description on the map, so I won't repeat it here, but suffice it to say that you can turn up at Orient Heights during rush hour and be confident that you won't have to wait for long.
Wonderland
Simply put, the fact that this part of the map is so clean is in fact indicative that this area is a mess. Despite the veritable cacophony of 400-series routes serving the North Shore, this stretch from Market Square to Wonderland was the only stretch I could find that broke the 12-min barrier during the peak. (And it's worth noting that most of this stretch is along the Salem Turnpike through the marsh -- no stops.) Like in Malden, you do see the same diverging route phenomenon: if you're at Lynn Station, and you're willing to take any bus that will get you to either Wonderland or Haymarket, it is true that you have your choice of the 426, 441, 442, and 455. The 441 and 442 alone should yield 15-min headways, and the 455 is advertised at 25-min on top of that. So, you probably won't have to wait too long. But still.
(As we all know, BLX would completely transform this.)
Circumferential Service
I was surprised that the Gold network had so few circumferential routes. The Bronze network revealed a few, but it's still not too many. The most notable I'd say is the 47, which operates at 10-min in the AM and 15-min in the PM (although it is a loooooong route and, as the Better Bus Profile notes, very unreliable -- that "10-min AM peak" is likely highly aspirational). The 89 I've discussed previously, and then you have the 69 -- a classic high-frequency AM peak but not PM peak route. The 65 also behaves like a "mini-circumferential" route -- instead of traveling in to Kenmore on the B and out on the D, a 65 rider can cut across all three Green Line branches.
Express Services
These have changed significantly during the pandemic, but it's worth noting that the 501, 502, 504 and 505 all ran at 11-min-or-better in the AM peak and 15-min-or-better in the PM. Chief among these is the 501, which is advertised at 7-min headways in the AM peak. Between these and the 57/57A, it's clear that the legacy of the A Line lives on.
Cumulative Services
There are a few odds and ends leftover here, mostly stretches of high-frequency resulting from cumulative service. Aside from the aforementioned corridors out of Quincy, Forest Hills, Davis, and Malden, there is a relatively short segment running into Harvard where the 72, 74, 75 and 78 layer, running some 8 buses per hour along that stretch. However, these are all lower-frequency routes of 30-min headways or worse, and from what I can see the schedules aren't coordinated with each other, so it wouldn't shock me to see a gap of 26 minutes, followed by four buses every 60 seconds.
But, this is a point where I think it's worth pausing for a minute. One could argue that these four routes should be spaced so that one of them arrives at Harvard every 8 minutes, so that there's consistent frequencies on the shared segments. (Worth noting that the routes diverge in pairs north of here, so there would be value in coordinated 15-minute headways along those stretches.) However -- and I somehow doubt the current schedules are coordinated to this extent -- the upside to having all four routes arrive in quick succession is that they can funnel in to a single departing vehicle on the other end. (For example, if you have multiple routes feeding into a commuter rail station where the train leaves at 8:30, yeah, you should aim to have all of your routes arrive between 8:24 and 8:28 -- no need to make some riders wait a half hour for their transfer.)
Now, the Red Line -- with its theoretical 5-min-or-better peak headways -- shouldn't need to be scheduled around. But that's not the only service you might transfer to at Harvard: you might take either the 66 or the 1, both of which operate at slightly more modest headways. If the 72, 74, 75 and 78 are currently being scheduled to support timed-transfers to either the 1 or the 66, then that seems like it could outweigh the benefits of even frequencies along Concord Avenue.
Finally, the last segment is Main Street in Everett. Cumulatively, this corridor sees about 5 buses an hour, destined for either Wellington or Sullivan. (An earlier version of the map had the wrong frequency in here -- I've updated on the back end, but not sure it's come through yet.) All three routes are 30-min-or-worse, and my recollection is that the coordination is okay, not great. But still, every 12-15 minutes seemed worth including. Now, because the routes reverse-branch, technically you don't get "rapid transit frequency" to either Wellington or Sullivan, which is why this route is visualized as an orphan.
Conclusion
I've found a few overarching conclusions out of all of this. For one, it's really remarkable how much more comprehensive the network gets during peak periods. Looking at maps, it's easy to forget that the system breathes throughout the day, and this was a fascinating way to visualize that.
Second, to me an analysis like this ties directly into the ongoing discussions about where to add bus lanes and other BRT infrastructure. I'm not saying we need to paint red lanes on every road traveled by a Gold route -- but certainly quite a few of these seem like no-brainers.
Finally -- and this is more of a point for the Reasonable and Crazy Transit Pitches threads -- but if we're looking for inspiration for transit improvements, this map seems like a great place to start.
Thanks, as always, for reading.
EDIT: Three months later, I have updated this map and added some more comments.
A few notes here. First of all, though there are a lot of routes that travel from Malden to Everett, they use diverging routes to get there, meaning there's no single corridor to map directly between the two. If you're at Malden Center and you want to get to the southwestern part of Everett, and don't mind walking a bit at the end (and are flexible on how long it takes), it is true that you have your choice of the 97, 99, 104, 105, and 106, so you probably won't have to sit around waiting for long.
Second, there is a corridor along Salem St to the east that probably has stretches of decent frequency, though it's likely inconsistent. The 108 runs every 20 during peak, and the 106, 411 and 430 all run in the 30-40 minute range, so you're probably gonna have at least one bus in between the 108's, which means you probably won't have to wait more than 15 minutes, which in a suburb arguably qualifies it for Bronze. (But notice how many caveats I needed to put in there!)
Finally, it's worth highlighting that the combo 136 + 137 yield pretty regular 11-min headways in the AM peak (which are not mirrored in the 30-min PM peak) all the way up to Wakefield, with reconverging service at Reading (unmapped). This is a good example of a "hidden" corridor -- both routes are advertised at 35-min headways in the Frequency Chart, but if you look at the departures from Wakefield Square between 7am and 8:15am (targeting arrival in downtown by/before 9am), you do indeed see 9-12 min headways.
Orient Heights
Barely a "hub" on this map, but it is where the 712 + 713 radial feeder hooks in. I explained the 712 + 713's service patten in painful detail in the description on the map, so I won't repeat it here, but suffice it to say that you can turn up at Orient Heights during rush hour and be confident that you won't have to wait for long.
Wonderland
Simply put, the fact that this part of the map is so clean is in fact indicative that this area is a mess. Despite the veritable cacophony of 400-series routes serving the North Shore, this stretch from Market Square to Wonderland was the only stretch I could find that broke the 12-min barrier during the peak. (And it's worth noting that most of this stretch is along the Salem Turnpike through the marsh -- no stops.) Like in Malden, you do see the same diverging route phenomenon: if you're at Lynn Station, and you're willing to take any bus that will get you to either Wonderland or Haymarket, it is true that you have your choice of the 426, 441, 442, and 455. The 441 and 442 alone should yield 15-min headways, and the 455 is advertised at 25-min on top of that. So, you probably won't have to wait too long. But still.
(As we all know, BLX would completely transform this.)
Circumferential Service
I was surprised that the Gold network had so few circumferential routes. The Bronze network revealed a few, but it's still not too many. The most notable I'd say is the 47, which operates at 10-min in the AM and 15-min in the PM (although it is a loooooong route and, as the Better Bus Profile notes, very unreliable -- that "10-min AM peak" is likely highly aspirational). The 89 I've discussed previously, and then you have the 69 -- a classic high-frequency AM peak but not PM peak route. The 65 also behaves like a "mini-circumferential" route -- instead of traveling in to Kenmore on the B and out on the D, a 65 rider can cut across all three Green Line branches.
Express Services
These have changed significantly during the pandemic, but it's worth noting that the 501, 502, 504 and 505 all ran at 11-min-or-better in the AM peak and 15-min-or-better in the PM. Chief among these is the 501, which is advertised at 7-min headways in the AM peak. Between these and the 57/57A, it's clear that the legacy of the A Line lives on.
Cumulative Services
There are a few odds and ends leftover here, mostly stretches of high-frequency resulting from cumulative service. Aside from the aforementioned corridors out of Quincy, Forest Hills, Davis, and Malden, there is a relatively short segment running into Harvard where the 72, 74, 75 and 78 layer, running some 8 buses per hour along that stretch. However, these are all lower-frequency routes of 30-min headways or worse, and from what I can see the schedules aren't coordinated with each other, so it wouldn't shock me to see a gap of 26 minutes, followed by four buses every 60 seconds.
But, this is a point where I think it's worth pausing for a minute. One could argue that these four routes should be spaced so that one of them arrives at Harvard every 8 minutes, so that there's consistent frequencies on the shared segments. (Worth noting that the routes diverge in pairs north of here, so there would be value in coordinated 15-minute headways along those stretches.) However -- and I somehow doubt the current schedules are coordinated to this extent -- the upside to having all four routes arrive in quick succession is that they can funnel in to a single departing vehicle on the other end. (For example, if you have multiple routes feeding into a commuter rail station where the train leaves at 8:30, yeah, you should aim to have all of your routes arrive between 8:24 and 8:28 -- no need to make some riders wait a half hour for their transfer.)
Now, the Red Line -- with its theoretical 5-min-or-better peak headways -- shouldn't need to be scheduled around. But that's not the only service you might transfer to at Harvard: you might take either the 66 or the 1, both of which operate at slightly more modest headways. If the 72, 74, 75 and 78 are currently being scheduled to support timed-transfers to either the 1 or the 66, then that seems like it could outweigh the benefits of even frequencies along Concord Avenue.
Finally, the last segment is Main Street in Everett. Cumulatively, this corridor sees about 5 buses an hour, destined for either Wellington or Sullivan. (An earlier version of the map had the wrong frequency in here -- I've updated on the back end, but not sure it's come through yet.) All three routes are 30-min-or-worse, and my recollection is that the coordination is okay, not great. But still, every 12-15 minutes seemed worth including. Now, because the routes reverse-branch, technically you don't get "rapid transit frequency" to either Wellington or Sullivan, which is why this route is visualized as an orphan.
Conclusion
I've found a few overarching conclusions out of all of this. For one, it's really remarkable how much more comprehensive the network gets during peak periods. Looking at maps, it's easy to forget that the system breathes throughout the day, and this was a fascinating way to visualize that.
Second, to me an analysis like this ties directly into the ongoing discussions about where to add bus lanes and other BRT infrastructure. I'm not saying we need to paint red lanes on every road traveled by a Gold route -- but certainly quite a few of these seem like no-brainers.
Finally -- and this is more of a point for the Reasonable and Crazy Transit Pitches threads -- but if we're looking for inspiration for transit improvements, this map seems like a great place to start.
Thanks, as always, for reading.
EDIT: Three months later, I have updated this map and added some more comments.
Last edited:
Thanks! And I've gone ahead and added the 34 + 34E + 40 + 50 to Metropolitan Ave as a Gold route -- my feeling is that the ultimate arbiter here should be whether or not a corridor "feels" like "show up and go" to its riders, so am fully onboard with adding it.
And -- I mean -- if it wasn't already obvious, the entire map needs to be taken with a little bit of a grain of salt. Like I said, I'm working primarily from summaries of summaries here, and you've already seen how many caveats I've added to everything.
Now, the smart way to do this would be to take the T's live feed data and use that programmatically generate a map based on actual time-between-departures. That is unfortunately well-beyond my coding expertise at the moment, but I think it would be a worthwhile endeavor.