MBTA Buses & Infrastructure

I'm really not sure how the Key Bus Routes came to be. Like why are the 47, 9, and the 92/93 not key bus routes? Yet the 71 somehow is a key bus route? Some local bus routes have higher ridership than key bus routes.

The 47 is the most egregious one, that bus route is the one to go directly into the LMA, and the only north south charles river connection between Allston and Copley, yet that route has 50 minute frequency intervals on Sunday. Meaning the only alternative routing is the 66 or the 1, yet those two routes are extremely far from the LMA (66 barely touches Brigham Circle).

The 1 bus also suffers from lack of direct connection with the 57 bus, so 57 bus riders have to go 1 stop on the Green Line to access the 1 bus. It also misses the Kenmore and Copley bus hubs, going via Hynes.

Local buses 34/34E, 70, 9, 86, and 31 all have more ridership than the 71 key bus route. 57 and 39 are rail replacements for the A and E branch as key bus routes, but the 92 and 93 were also GL tunnel rail replacements and OL elevated rail replacements and somehow, those aren't key bus routes. The last 2 routes are also extremely short so they should be some of the easiest to run frequently back and forth from Haymarket busway to Sullivan, but they aren't.
 
The worst leg of the 66 is the loop around Union Sq Allston. The 66 desperately needs an alternative to get from Lower Allston to Brookline without that horrendous lap.
Union Sq Allston is an important neighborhood center to directly serve and since it is unfortunately the only North-South route for about a mile each way east or west it makes sense for it to make this diversion. The other consideration is that the extra time it takes to go up to Brighton Ave is made up during rush hour by the fact that the bus doesn't have to make that left onto Harvard from Cambridge St. as that left gets backed up all the way to Lincoln. That said going northbound Union Sq is definitely a time-consuming detour since you can right on red onto Cambridge. Because of the prior point, I'd say the detour is not a huge time loss overall but I get the frustration for those on for the long haul. I think there should be a bus that runs from Central Sq to Jackson via Magazine, St. Paul, Cypress, and Perkins St to relieve some North-South transit demand.
Ive railed forever that there is a major problem with north-south transportation access and that includes a glaring need to facilitate auto traffic flow as well. The West Station / redevelopment by Harvard should absolutely have had the state step in and squash BU's resistance to flow through Babcock. There should be at least TWO roads that go from Brookline, across Comm, and connect directly into Harvard. I know people just wish the state would go to war against cars but no amount of wishing will get rid of them and the access across town between LMA-Coolidge-Allston-Harvard&Central is truly abysmal. The roads are narrow, there are hundreds of lights, and actually allowing more direct flow of all traffic would significantly mitigate the bottlenecks we have all across the area that surrounds this slice of the metro pie.
Last document I saw with a potential configuration of West Station has Malvern St extended across to Seattle St and provisions for Agganis Way to extend across in some way when they figure out how to reconfigure the throat. Doing this, however, is sure to induce more vehicular demand as you've then made driving across more easy. This is fine to better connect people across this vast gap (would be better as a transit/bike/commercial vehicle/pedestrian connection only more on that later) but it'll only add to the traffic problem down to Longwood. Making a more direct route by, what you seem to be proposing but I could be wrong, removing traffic lights and reducing sidewalk width to widen roads would further exacerbate the traffic problem. The space inefficiency of private cars is an inherent bottleneck because these destinations, Allston/Cambridge/Brookline/LMA, are of a finite size. If you encourage more people to drive via some high-speed "shortcut" to the destination all it will do is clog up the endpoint with more vehicles quicker. The only real 'solution' to the traffic along this key route is if we had a high-capacity, frequent, space-efficient, and reliable transportation alternative to driving along it. The 66 in its current state does not check off the frequent or reliable boxes because it's stuck in the same vehicular traffic. Why would you take the bus that sits in the same traffic you'd otherwise drive in? Therefore it'd need dedicated infrastructure to bypass traffic to make it the better option attracting more people to use it who can, and in turn reduce the number of drivers alleviating traffic. A bus isn't the best solution for this corridor though and there really should be a heavy rail subway line here.
I think the next worst leg of the 66 is Tremont St in Mission Hill. Unsurprisingly, this is another area where lack of any central intelligent planning has led to all traffic to be funneled onto either Ruggles or Tremont. Smith is terminated. You have Mission Hill on the other side. So there's the same problem.
Yeah old street layout from before cars and the geographic feature that is Mission Hill left only three roads including Heath to cross into the LMA from the south. At the time two direct roads only a couple thousand feet apart were probably more than enough for all the people in streetcars, walking, biking, and goods movement but now the LMA is there and it is a massive traffic draw. It'd be ridiculous to do what most of America did and bulldoze all the commercial, residential, and educational establishments surrounding this to make more room for cars that would only further the congestion problem when an actual solution would be more viable transportation alternatives like the aforementioned heavy rail line or the Ruggles Busway planned in the urban ring. What'd really rile people up is if you made Ruggles and Tremont 1-way clockwise in relation to Huntington and Columbus but with counterflow bus/truck lanes so they can travel both ways.
1701917296376.png
1701917500637.png

This would make the tens of thousands that rely on 7+ bus routes on Ruggles and the 66 on Tremont have reliable service and attract more people to switch to transit due to its viability. Would also make the Parker intersection safer and smoother by eliminating half the left-turn movements.
A better transportation infrastructure would recognize that redundancy for both cars and transit is the only way to go. In Boston there is this local control that refuses to allow streets to actually get from A to B because local councils fight and fight and for some reason are allowed to get their way. But instead of going from A to B, our roads go from A to [a few blocks and several left and right turns toward] B. This is what leads to the worst snarls across the region, and many of them could be dealt with fairly simply.
While redundancy would indeed spread out traffic across more streets, people still need to converge on the same location. The reality the rest of the world is realizing or has realized in their cities is that private automobiles are simply incompatible with density. You will never create enough space for cars to be the primary mode of transportation without heavy congestion and dedicating vast amounts of space to store them there. Creating the same redundancy dedicated to alternative higher density and capacity modes of transportation can serve a given area with a much higher bottleneck threshold.
Everything takes up space even people, that's why we built upwards, and a method that transports more people in less space that doesn't need to then be stored is the way to mitigate traffic on our roads. Better transportation infrastructure would prioritize the most efficient and suitable methods of moving people to their destinations reliably and safely.
Like: the state says, hey, there is a major, major bottleneck all across Brookline/Allston and we are literally wiping the map clean to build an entire new network. Gee, why would we not actually connect two major and dense areas that only have been separated because of bad planning?
The culmination of all I've had to say so far is Brookline and Allston's traffic woes are the result of supposed "solutions" to traffic, (fast, multilane, direct roads in the forms of I-90, Rt 9, Jamaicaway, and Rt 28) that serve to dump a large quantity of space-inefficient transportation into a small area they greatly outsize, rather than the alternatives that would've been able to handle the transportation demand. It sure was bad planning that got us here, and doing more of the same won't get us out.
 
Continuing on bus rider observations:
Another excellent roundup! Some comparisons to the 2022 ridership data:
  • Beginning at Harvard at seemingly any time of day there’s always a good chunk of people boarding
    This lines up well with how consistent the load numbers are north of BUMC
  • At Boylston a decent number get off but I don’t see a lot of people backtracking to Hynes
    This is an interesting observation. While then number of SB passengers at Boylston St is lower than NB passengers at Newbury, it's still the 2nd busiest (Behind Nubian, obviously) stop by a very good margin. It could be that I'm underestimating the number of local passengers, that you've just gotten buses full of statistically-abnormal passengers, or that they're walking to a different stop, likely Copley or Kenmore like you said.
  • About half the bus empties for BMC
    Pretty much spot on in terms of peak load numbers. During peak hours the load goes from ~8-10 people to ~20-25 after Albany and Harrison.
 
A bus isn't the best solution for this corridor though and there really should be a heavy rail subway line here.
The only issue I have is that a heavy rail line that follows the 66's route entirely will likely cost tens of billions of dollars with today's construction and planning standards. It definitely has to be deep bored almost entirely (maybe except N Harvard St), due to going through dense neighborhoods and commercial districts in Allston and Brookline.

Unless I'm mistaken, a brand new heavy rail line hasn't happened in the US for, like what, decades? Most new rapid transit projects are either light rail lines, or extensions of older heavy rail lines (like Red-Blue and NYC's Second Ave Subway). And while TBM tunnels can still be built for light rail lines (such as LA's Regional Connector*), most bored tunnels are extremely expensive in the US nowadays - when your canonical examples are the Second Ave Subway and BART's San Jose extension, that does the opposite of inspiring confidence.

Don't get me wrong, I want to be convinced otherwise that long TBM tunnels, even for new heavy rail lines, can and will happen in Boston. But at this moment, most signs point to pessimism on this.

* Actually, LA's Regional Connector seems surprisingly cheap, at $1.8 billion for 1.9 miles. However, LA is also building transit faster than any other city in the US, so it may be an exception rather than the rule.

What's a much less expensive and more realistic solution, though, is a branch from Harvard to West Station and then around the BU vicinity, before turning south to somewhere around LMA. I'm using @Riverside's own crayon map here for illustration:
1700188593430-png.44631

While this alignment obviously serves way fewer neighborhoods, where it shines is feasibility. I believe that even a grade-separated branch can be built without tunneling: @F-Line to Dudley proposed an alignment here, and he has also mentioned that Harvard is even reserving a ROW through its Lower Allston land.
 
Regarding the debate @Koopzilla24 and @FK4 are having, I think the gist of it is: Yes, we need better north-south connectivity in the Cambridge-Allston-Brighton-LMA region. But at the same time, it can also be true that the additional connectivity need not focus solely on cars -- transit, cyclists and pedestrians can and absolutely should be the focus of any new connections that we create.

The elephant in the room is that LMA sucks for both cars and transit. The very narrow streets even within LMA - most notably Longwood Ave and Francis St - means having a large number of workers commute by car is not an option, and it arguably already doesn't have capacity for additional car traffic. But if we want to encourage more LMA workers to take transit, an average walk of 10 minutes from the D and E branch stations doesn't do well at that either.

@Riverside did an analysis here to show that, despite the underwhelming transit connection to LMA, its transit share is surprisingly higher than average. 44% of LMA workers take transit, vs. 30% that drive alone; the transit share is even higher than downtown Boston.

Whether that's because of "good enough" transit or disproportionately worse road network, we don't know (though I'd guess it's the latter). But it definitely highlights the need for better connectivity to LMA, especially better transit.


Edit: Thinking about it again, a more realistic improvement to the 66 corridor will probably be shared bus/bike lanes. The roads seem wide enough for that, but not anything more (except Brighton Ave). There will still be backlash from restaurants losing their parking, though (and maybe cyclists losing their dedicated bike lanes and having to share with buses).
 
Last edited:
At Boylston a decent number get off but I don’t see a lot of people backtracking to Hynes. I’d imagine the Parcel 12 construction’s relocation of the stop has made that transfer less convenient for folks and they’d rather go to Kenmore or Copley for service downtown or out to the branches (This is in contrast to the wall of people that go from Hynes to the 1 northbound that I’ll get to later).
This might seem like nitpicking in an otherwise excellent analysis, but I really doubt that people were walking all the way from Mass Ave to Kenmore or Copley for Green Line transfers. Yes, going from the southbound 1 stop to GL Hynes requires 2 pedestrian crossings and walking over the Pike, but that's still much more efficient than walking 12 minutes to Kenmore or 12 minutes to Copley. Perhaps it may be that these crowds blend into the (vast) pedestrian traffic rather quickly, and are thus less noticeable?

Another possibility is that for many people (especially from Cambridge), their destination is in the Back Bay area, so they're not using a Green Line transfer anyway. They could be working or shopping there, and my guess is there's high demand for both, given how there always seems to be a good number of passengers waiting at northbound Hynes at all times. Whereas people taking the 1 between Hynes and points south (most likely BMC) probably aren't going to Back Bay and are instead transferring to GL, and therefore the NB alighting passengers you see immediately enter the station.

-------------------------

Also circling back to the point @TheRatmeister made earlier about the Albany St (Cambridge) and MIT stops being "underwhelming" compared to Central and Harvard, my guess is that the Albany St stop is heavily commute-focused. Most of its ridership comes from the nearby pharmaceutical companies, like Novartis, Pfizer and possibly all the way up to Area Four. A smaller group comes from apartments to the west along Albany St, most of which house MIT students (so they might be less reliant on transit and contribute less ridership). Your methodology only uses ridership for the entire week - I'm not saying that's a bad metric in itself, but I'd guess that checking only weekday (or weekday rush hours) ridership may give higher numbers.

But yeah, the triangular area bounded by Mass Ave, Albany St and Main St hosts a lot of jobs and is underserved by transit. The Red Line passes by without stopping.

The MIT stops are obviously used by MIT students and staff. I think there's decent student traffic between MIT and Harvard on weekdays, but as @Koopzilla24 said, some of them may have taken the M-2 or biked. On weekends, you'll also see some undergraduate students taking the 1 south to Back Bay for shopping, and possibly north to Harvard Square. On another note, MIT students are more likely to stay in Cambridge (even near Lechmere or along the eastern parts of GLX corridors) than across the river in Back Bay or Allston, at least according to the subleasing ads I've seen.

Anyway, it shouldn't be a surprise that they don't compare to Central's ridership, because the latter has a rapid transit connection. You can have people from Alewife and Davis taking RL and transferring to the 1 for Back Bay, for example.
 
I'm really not sure how the Key Bus Routes came to be. Like why are the 47, 9, and the 92/93 not key bus routes? Yet the 71 somehow is a key bus route? Some local bus routes have higher ridership than key bus routes.

The 47 is the most egregious one, that bus route is the one to go directly into the LMA, and the only north south charles river connection between Allston and Copley, yet that route has 50 minute frequency intervals on Sunday. Meaning the only alternative routing is the 66 or the 1, yet those two routes are extremely far from the LMA (66 barely touches Brigham Circle).

The 1 bus also suffers from lack of direct connection with the 57 bus, so 57 bus riders have to go 1 stop on the Green Line to access the 1 bus. It also misses the Kenmore and Copley bus hubs, going via Hynes.

Local buses 34/34E, 70, 9, 86, and 31 all have more ridership than the 71 key bus route. 57 and 39 are rail replacements for the A and E branch as key bus routes, but the 92 and 93 were also GL tunnel rail replacements and OL elevated rail replacements and somehow, those aren't key bus routes. The last 2 routes are also extremely short so they should be some of the easiest to run frequently back and forth from Haymarket busway to Sullivan, but they aren't.
According to Wikipedia:
he MBTA's Service Delivery Policy uses five criteria when determining if a route is part of the Key Bus Routes program.[3]

  • High ridership demand
  • Connectivity within the system
  • Geographic coverage
  • Accommodation of major new development
  • Operation as bus rapid transit
The reference page can no longer be accessed, but a 2011 archive can be found here. Here's the full addendum on Key Bus Routes:
Key Bus Route Addendum

Key Bus Routes are similar to local routes, but have policy standards for a longer Span of Service and a higher Frequency of Service. The Key Bus Route Network was designed to complement the MBTA’s light and heavy rail system and to ensure that all high-demand corridors have access to frequent transit service seven days a week.

The MBTA’s Service Delivery Policy establishes Key Bus Route Service Standards, which make the Span of Service and the peak period Frequency of Service on Key Bus Routes equivalent to light and heavy rail. This guarantee of high-frequency service provides assurance to riders that they will not have to wait long for the next bus, even if they do not know the published schedule. To encourage this kind of “walk up” use of Key Bus Routes, they will be included with the light and heavy rail system on MBTA’s “spider” maps as they are updated. Key Bus Routes will also be designated on the MBTA’s system map, schedule cards and other marketing materials.

Initially, Key Bus Routes were selected based on their demonstrated heavy demand for service on all days of the week and to provide high frequency service to areas of the region's urban core not served by light or heavy rail. Additional Key Bus Routes may be designated based on consideration of a number of characteristics. Not all of these characteristics must be present in a given route to make it eligible for Key Bus Route status. Each route will be considered in the context of the MBTA’s transit system as a whole and within available operating resources.

The characteristics that may trigger consideration of a route for Key Bus Route status include:
• high ridership demand;
• connectivity within the system;
• geographic coverage;
• accommodation of major new development; and
• operation as BRT (all BRT route segments that operate in dedicated rights-ofway will automatically be designated as Key Bus Routes).

The Key Bus Routes currently include:
• Silver Line Washington Street: Dudley Station – Downtown Crossing via Washington St.
• Silver Line Waterfront: South Station – Silver Line Way
• Route 1: Dudley – Harvard via Mass. Ave.
• Route 15: St. Peter’s Square – Ruggles via Dudley
• Route 22: Ashmont – Ruggles via Grove Hall
• Route 23: Ashmont – Ruggles via Codman Square
• Route 28: Mattapan – Ruggles
• Route 32: Wolcott Square – Forest Hills
• Route 39: Forest Hills – Back Bay
• Route 57: Watertown – Kenmore via Oak Square
• Route 66: Harvard – Dudley via Allston
• Route 71: Watertown Square – Harvard via Mount Auburn St.
• Route 73: Waverley – Harvard via Mount Auburn
• Route 77: Arlington Heights – Harvard via Mass. Ave.
• Route 111: Woodlawn – Haymarket Station
• Routes 116 & 117: Broadway @ Park Avenue – Maverick Station (the corridor in which the trunk portions of both routes operate is assigned Key Bus Route status)
The list of Key Bus Routes has not been updated since 2006, even though the document mentions possibility of additional designations.

Fortunately, the BNRD will largely replace this designation with Frequent Bus Routes that have all-day 15-min frequencies. Some of the biggest gains of 15-min corridors in BNRD include:
  • T7 (current 7 and 93 connected through downtown)
  • T8 (current 8 east of BMC and 10 between BMC-Copley)
  • T9
  • T12 (new crosstown route via Brookline Village, LMA, Nubian, BMC, Andrew and Seaport)
  • T16
  • T22 and T28 rerouted to LMA (and the latter to Kenmore), bypassing Ruggles
  • T31
  • T47 (extended north to Union Sq Somerville, but truncated to Ruggles to the south)
  • T66 rerouted more directly into LMA
  • T70 (extended to Kendall but truncated to Waltham Center)
  • T96 (absorbing current 101's Medford-Malden section, and removes Harvard but adds new 15-min corridor from Davis to Union Sq Somerville, part of current 87)
    • Looks like the T is really insistent on making Somerville Ave west of Union Sq a 15-min corridor
  • T101 (adds Sullivan-Kendall OSR and absorbs 92 in the process)
  • T104 (diverted south of Everett to give an Everett-Chelsea-Airport BL OSR)
  • T109 (extended to Harvard and replaces part of current 86)
  • T110
The only loss among current Key Bus Routes is elimination of the 117, which is completely replaced by T110 and T116.
 
Last edited:
What's a much less expensive and more realistic solution, though, is a branch from Harvard to West Station and then around the BU vicinity, before turning south to somewhere around LMA. I'm using @Riverside's own crayon map here for illustration:
1700188593430-png.44631

While this alignment obviously serves way fewer neighborhoods, where it shines is feasibility. I believe that even a grade-separated branch can be built without tunneling: @F-Line to Dudley proposed an alignment here, and he has also mentioned that Harvard is even reserving a ROW through its Lower Allston land.
Correct me if I'm wrong, but:
  1. It certainly seems like this would involve a tunnel under the river
  2. The alternative seems to be very heavy street running, which is... not great. If we want a cheap 'temporary' solution, we should just run buses into the existing bus tunnel.
  3. Even if it avoids this problem, why light rail? This is a fully grade separated route, and we'd presumably want to connect it to a LMA tunnel, which would also need to be deep bored and fully grade separated. Given that we've now spent about 90% of the cost of a new subway line, let's just build a new subway line.
 
EDIT: Ah, I see @Teban54 has already replied in Crazy Transit Pitches, my bad.

Briefly dipping in here.

It certainly seems like this would involve a tunnel under the river
F-Line outlined alternatives for both an intermediate surface-running alt and a long-term tunnel into the old Red Line Brattle Yard Leads. I’ll try to grab the links later. Yes, likely under the river, but the Charles is much narrower here than elsewhere. (And I think you could potentially do the crossing with a new bridge, though I haven’t looked closely.)
Even if it avoids this problem, why light rail? This is a fully grade separated route, and we'd presumably want to connect it to a LMA tunnel, which would also need to be deep bored and fully grade separated. Given that we've now spent about 90% of the cost of a new subway line, let's just build a new subway line.
The design above doesn’t link into an LMA tunnel, but instead uses a surface running segment along a reconfigured Park Dr, giving it a dedicated ROW along the north edge of LMA, with a couple of crossings for cars and pedestrians.

Even with an LMA tunnel, this design (briefly) interlines the Harvard <> Longwood Line with a Grand Junction Line. I suppose technically you could run one with HRT stock and the other with LRT so as to not require full grade sep on the Grand Junction, but that’s definitely a less-than-ideal solution.
 
86: The T seems to think not many people take the bus beyond Harvard, as seen by how the 86 is truncated to Harvard in the BNRD and the T109 takes over Harvard-Sullivan. However, community reaction of this change seems to suggest otherwise. Also, I'm curious whether people take it all the way from Harvard to Sullivan as a RL-OL connection, or if they use it to get to Union Sq Somerville from either rapid transit lines.
It’ll take a bit to type up more detailed analysis on my computer of these routes. But from my phone (on the 65 headed to Brighton, 2 back to 2 both standing room only leaving Longwood Ave) I can say about this I’ve only taken the 86 on a weekday beyond Harvard a couple times so I’m not the most qualified there but there’s a handful that remained that boarded along Chestnut Hill Ave and Market St in Brighton. I take the 86 the full length all the time on weekends both ways though and there was a lot more activity in that segment on those days. The primary trip generators appeared to be Harvard-Union Sq and Union Sq-Sullivan pairings. At Sullivan I see more people transferring to or from other bus routes rather than heading into the orange line or coming out of it. I’d assume its the best way to get from Malden and points north of the Mystic to Harvard or Union Squares. I take it from Brighton because it’s much faster than going Green downtown and transferring up (by 15+ min) even with Cambridge congestion. I’d imagine others who work up there do the same.
I did some analysis on the Better Bus Profile’s data on the 86 a couple of years back, looking at how many riders ride through Harvard. Let me see if I can dig it up. I remember being very surprised.
 
Last edited:
Milesintransit has a ton of detailed bus route trip reports on his site that are much better written and detailed than I hash out here but most date back to 2015 so are a bit outdated. I've only lived in the city for 4 years now so my observations are all post-pandemic shutdown for perspective.
MilesInTransit is a valuable resource and I’ll simply say that I personally find your analyses more useful.
 
I'm really not sure how the Key Bus Routes came to be. Like why are the 47, 9, and the 92/93 not key bus routes? Yet the 71 somehow is a key bus route? Some local bus routes have higher ridership than key bus routes.

The 47 is the most egregious one, that bus route is the one to go directly into the LMA, and the only north south charles river connection between Allston and Copley, yet that route has 50 minute frequency intervals on Sunday. Meaning the only alternative routing is the 66 or the 1, yet those two routes are extremely far from the LMA (66 barely touches Brigham Circle).

The 1 bus also suffers from lack of direct connection with the 57 bus, so 57 bus riders have to go 1 stop on the Green Line to access the 1 bus. It also misses the Kenmore and Copley bus hubs, going via Hynes.

Local buses 34/34E, 70, 9, 86, and 31 all have more ridership than the 71 key bus route. 57 and 39 are rail replacements for the A and E branch as key bus routes, but the 92 and 93 were also GL tunnel rail replacements and OL elevated rail replacements and somehow, those aren't key bus routes. The last 2 routes are also extremely short so they should be some of the easiest to run frequently back and forth from Haymarket busway to Sullivan, but they aren't.

Total ridership/boardings might not be the best way to compare circumfurential routes and radial routes. No one takes the 47 from Central to Broadway or the 86 from Cleveland Circle to Sullivan. People absolutely take the 71 from Watertown Square to Harvard. On the circumfurential routes the space on the bus turns over, so there can be multiple boardings for the same space on the bus; on the radial routes people tend to ride to/from the inbound terminal.

Some of the radial routes like 39 and 70 do have a handful of submarkets that they serve (e.g., Forest Hills-LMA, Waltham-Arsenal).

In the case of the 71/73, the ETBs also had fewer seats than diesel buses too.

As @Riverside has discussed here several of the corridors you mention have historically had high levels of service at peak. It's possible that their ridership is very peaky, so offering high frequencies all day may not be necessary on those routes. I'll try and remember to look at this later if I have time. On quick glance, the 47 has peak average load of 11.5 on Sundays, and the 71 has peak average load of 8.7 on Sundays.

The key bus route branding isn't that important today anyway. The weekday service levels on the 47 and 71 today is very similar (in the case of the 71, "21 minutes or less" sounds like a PITA to plan around).
 
As @Riverside has discussed here several of the corridors you mention have historically had high levels of service at peak. It's possible that their ridership is very peaky, so offering high frequencies all day may not be necessary on those routes. I'll try and remember to look at this later if I have time. On quick glance, the 47 has peak average load of 11.5 on Sundays, and the 71 has peak average load of 8.7 on Sundays.

The key bus route branding isn't that important today anyway. The weekday service levels on the 47 and 71 today is very similar (in the case of the 71, "21 minutes or less" sounds like a PITA to plan around).

Even the analysis Riverside has made in 2020 has already gotten significantly out of date since the analysis was made. The 92 and 93 used to run every 7 - 8 minutes or every 12 - 15 peak/rush hour as recently as 2019/2020, but nowadays as of fall 2023 and winter 2024 at best they run every 30 minutes during rush hour, with major service cuts hitting almost all bus routes since that time. (Would be nice to do a re-analysis of peak service again using winter 2024/Q1 2024 schedules).

The same issue I also have with MilesOnTheMBTA site. All the frequencies described in that blog are from 2013-2014, so they don't accurately represent fall 2023 or winter 2024 service levels.

The key bus routes as far as I'm aware still have the 20 minute frequency target the MBTA has, even for the 71 and 73, although in practice many key bus routes fail to hit that target as of fall 2023/winter 2024. Rapid Transit has a 15 minute service target for off peak, but the Red Line, of course, currently fails that target 7 days a week.
 
* Actually, LA's Regional Connector seems surprisingly cheap, at $1.8 billion for 1.9 miles. However, LA is also building transit faster than any other city in the US, so it may be an exception rather than the rule.
LA is also rather quietly building a significant heavy rail extension of the D-Line. I don't know whether it is better cost controlled than the New York and BART examples, or just flying under the radar, but it does seem that they have reached some level of project management expertise that significantly exceeds what we see in other US agencies.

From Wiki said:
Metro estimates that the three-section project, adding 9 miles of track and 7 new stations, will cost $8.2 billion.
So they are building an HRT subway for less than $1 billion per mile.
 
LA is also rather quietly building a significant heavy rail extension of the D-Line. I don't know whether it is better cost controlled than the New York and BART examples, or just flying under the radar, but it does seem that they have reached some level of project management expertise that significantly exceeds what we see in other US agencies.


So they are building an HRT subway for less than $1 billion per mile.
Given the insane rate (by US standards) at which LA has been expanding their system lately, I have to think there's either some economies at scale, or they managed to get a team of relatively fixed planners and/or construction crew, instead of hiring consultants who are absolutely not cost-conscious (which is the case for NYC's cost blowouts).
 
Even the analysis Riverside has made in 2020 has already gotten significantly out of date since the analysis was made. The 92 and 93 used to run every 7 - 8 minutes or every 12 - 15 peak/rush hour as recently as 2019/2020, but nowadays as of fall 2023 and winter 2024 at best they run every 30 minutes during rush hour, with major service cuts hitting almost all bus routes since that time. (Would be nice to do a re-analysis of peak service again using winter 2024/Q1 2024 schedules).

The same issue I also have with MilesOnTheMBTA site. All the frequencies described in that blog are from 2013-2014, so they don't accurately represent fall 2023 or winter 2024 service levels.

The key bus routes as far as I'm aware still have the 20 minute frequency target the MBTA has, even for the 71 and 73, although in practice many key bus routes fail to hit that target as of fall 2023/winter 2024. Rapid Transit has a 15 minute service target for off peak, but the Red Line, of course, currently fails that target 7 days a week.
Redoing my analysis is definitely on my medium-term list. However, since we are mostly using these frequencies as a measure of demand per corridor, I think the historical pre-pandemic pre-cutback frequencies are still useful.
 
Total ridership/boardings might not be the best way to compare circumfurential routes and radial routes. No one takes the 47 from Central to Broadway or the 86 from Cleveland Circle to Sullivan. People absolutely take the 71 from Watertown Square to Harvard. On the circumfurential routes the space on the bus turns over, so there can be multiple boardings for the same space on the bus; on the radial routes people tend to ride to/from the inbound terminal.
I'm not sure how much relationship this has with promoting levels of service on each route. Even if ridership on the 47 and 86 are more spread out, it's still true that making these routes more frequent benefits more riders than making the 71 more frequent.

My best guess is basically what the rest of your comment says: one may conjecture that ridership on the 47 was more peak-oriented (and likewise, ridership on the South Boston routes were definitely very peak-oriented), so it might be determined that ensuring consistent service on the 71 is better at promoting transit-friendly neighborhoods along the corridor, where people use it not just for work, but also for school, errands, etc. Whereas adding service on the 47, especially during off-peak hours, may not have such an effect.
 
According to their project page, the Summer St pilot bus/truck lane program started on Monday, December 4.
Anecdotally of course, but per my experience this week the bus lanes have been filled with general traffic. The bus I was on this morning mostly drove in the left lane, and traffic was heavy. Not sure if the program has begun in practice or there isn’t any enforcement so the bus drivers simply focus on driving down the road as quickly as possible (safely, as always).
 
I'm not sure how much relationship this has with promoting levels of service on each route. Even if ridership on the 47 and 86 are more spread out, it's still true that making these routes more frequent benefits more riders than making the 71 more frequent.

My best guess is basically what the rest of your comment says: one may conjecture that ridership on the 47 was more peak-oriented (and likewise, ridership on the South Boston routes were definitely very peak-oriented), so it might be determined that ensuring consistent service on the 71 is better at promoting transit-friendly neighborhoods along the corridor, where people use it not just for work, but also for school, errands, etc. Whereas adding service on the 47, especially during off-peak hours, may not have such an effect.

Well my hypothesis was wrong

1701966174672.png
 
This might seem like nitpicking in an otherwise excellent analysis, but I really doubt that people were walking all the way from Mass Ave to Kenmore or Copley for Green Line transfers. Yes, going from the southbound 1 stop to GL Hynes requires 2 pedestrian crossings and walking over the Pike, but that's still much more efficient than walking 12 minutes to Kenmore or 12 minutes to Copley. Perhaps it may be that these crowds blend into the (vast) pedestrian traffic rather quickly, and are thus less noticeable?
I didn’t make it clear but that wasn’t in reference to people that have been on since Cambridge transferring off the 1 to the Green Line but rather people on Marlborough, Beacon, or Comm Ave taking the 1 up to Hynes to then transfer off. Plus if you’re coming from Cambridge and want to go to a point E-W on the Green line it’s more direct to take a different bus or the Red Line. People are fickle and if their destination is east of Mass Ave and they live east of Hereford St they might just walk to Copley rather than backtrack to Mass Ave for the 1 up to Hynes unless there’s a bus very conveniently timed. I’d also imagine those living on Mass Ave itself would walk up to Hynes if they’re between Storrow and Clearway and the next bus is 10+min away. The entire Mass Ave north of Hynes in Boston is within a 1/4mi station walkshed. Then anyone south of Clearway would end up in the Symphony walkshed. The lack of reliability of the 1 and frequencies above 10min make using it as a Green Line feeder for most of Mass Ave pretty pointless. This doesn’t mean nobody does it. That’s exactly what I used it for. Since I commuted from Brighton to Roxbury down by Dudley common I would often take the Green to the 1 at Hynes then walk down Hampden.

  • This is an interesting observation. While then number of SB passengers at Boylston St is lower than NB passengers at Newbury, it's still the 2nd busiest (Behind Nubian, obviously) stop by a very good margin. It could be that I'm underestimating the number of local passengers, that you've just gotten buses full of statistically-abnormal passengers, or that they're walking to a different stop, likely Copley or Kenmore like you said.
This carries into addressing this. My commute time itself was a bit unusual as when I took this southbound for work it’d be either at ~9:50am or ~3pm. During that morning period it appeared to be entirely local traffic. Boylston St itself is a very popular destination for people because of Berklee but also Newbury St is the next block over and with the construction of Parcel 12 the relocated Boylston stop is the closest one to Newbury unless you want to walk up the hill from way back at Marlborough. The other consideration is passengers going West on the Green but my guess is unless your destination is right around Kenmore Station or just over the pike at Fenway Park you’re better off taking an alternative route outside of Hynes walkshed. Examples would be the 47 (my next focus) from anywhere around Central or the E, Orange, and back to the 47 again or other Nubian routes when farther south of Hynes.
The MIT stops are obviously used by MIT students and staff. I think there's decent student traffic between MIT and Harvard on weekdays, but as @Koopzilla24 said, some of them may have taken the M-2 or biked. On weekends, you'll also see some undergraduate students taking the 1 south to Back Bay for shopping, and possibly north to Harvard Square. On another note, MIT students are more likely to stay in Cambridge (even near Lechmere or along the eastern parts of GLX corridors) than across the river in Back Bay or Allston, at least according to the subleasing ads I've seen.
The M-2 route is one I have been the bus driver of myself during rush hour periods so while it’s not MBTA I’ve got the lowdown on exactly how that route trends. Of note though is that this is a HARVARD ONLY bus, that means no MIT students are allowed on board (ridiculous I know but they’re not a medical school and the M-2 is a LMA bus specifically for Harvard students and staff). MIT has its own cutaway bus that crosses the river that I’ve seen decently full but otherwise MIT students are mostly stuck with the 1.
 

Back
Top