MBTA Commuter Rail (Operations, Keolis, & Short Term)

action.transitmatters.info

What kind of Fairmont # where "they" expecting vs what they got?
 
Some scattered thoughts:

It's interesting to see how much intra-line traffic there is on the Fairmount Line. Most ridership at Newmarket and Uphams Corner, and 40% at Four Corners/Geneva, is from the south. With good service you'd probably see a lot of short trips on the line.

Boston Landing is doing better than I expected - already ahead of the Newton stops and two Wellesley stops - which indicates just how badly Allston-Brighton needs frequent and speedy transit. 82% of boardings are inbound; it's primarily serving commuters from the local neighborhood rather than the development.

Yawkey is doing incredibly well - exceeding projections (937 daily boardings+alightings) already. It's now at 2,491 (1200+ round trips) daily - roughly triple what it was in 2012. Approximately half of the new trips on the Worcester Line can be directly attributed to Yawkey and Boston Landing (though Yawkey likely poached a few from Back Bay), and those two stations really show how RER-type service (particularly that the NSRL could provide) would be transformative.

The big winners on the Fitchburg Line are Littleton/495 and South Acton - unsurprisingly, the two stations with proven demand that got major upgrades (and, for Littleton, a frequency boost) - which drove almost all the line's increase. Impressively, this is at two stations that already had full parking lots every day - these increases are coming entirely from carpools, dropoffs, and walk/bike. The speed increases didn't help other stations much, and Wachusett is at just 130 riders a day (but growing - in February 2017 it was 45-75 daily). Brandeis lost a lot of reverse commute ridership - perhaps this count was in the summer.

Side note: would be very interesting to see for how many passengers the full-length accessible platforms are a factor in deciding to ride. I would guess it is not insignificant.

Despite the double-tracking, the Haverhill Line had basically no growth.

Plymouth is down to just 20 riders per day. The MBTA should really cut their losses there and make it weekend-only or something.

Chelsea inbound ridership has tripled, with only modest change to the ~120 daily riders from further out who alight there. That indicates that SL3 is not being used as a Seaport connection, but may be helping feed CR. Salem had a decent ridership increase; Beverly didn't.

Wickford Junction and TF Green are still struggling - around 250 riders per day each.

Nothing of too much note on the Franklin, Needham, Lowell, and Middleborough/Lakeville lines - modest growth, with relatively minor year-to-year variation at most stations.

The unsurprising overall message: frequency matters, a lot. As does having stations sited to meet demand. Building new parking structures doesn't guarantee riders will come (North Leominster, Beverly, Wachusett, TF Green, Wickford) but improving in-demand stations and increasing service seems to (South Acton, Littleton, Chelsea, Salem). The investments that really make sense for the system are to increase frequency everywhere (and speed travel when you do so), and to focus station improvements on untapped demand (and accessibility!).
 
It's interesting to see how much intra-line traffic there is on the Fairmount Line. Most ridership at Newmarket and Uphams Corner, and 40% at Four Corners/Geneva, is from the south. With good service you'd probably see a lot of short trips on the line.

You're also seeing a big shift from Hyde Park to Fairmount down the street because the characteristics of Fairmount Line service better fit the neighborhood than Providence/Stoughton. The great Bus Reconfig-A-Thon should take note of this and try to recalibrate Cleary Sq. routes 32, 33, & 50 to loop at Fairmount Station as best they can, with City of Boston smoothing things over on use of couple of side streets for the pingback to HP Ave. Would be very easy for the Cleary-terminating 50, OK/non-invasive for the 33, and somewhat ham-fisted for the 32.

Better frequencies are only going to accelerate this trend towards Fairmount and away from HP. The NEC's unable to add more trips to HP, if South Coast-via-Stoughton is ever built it would lose half its frequencies through skips, and Amtrak's future quad-tracking of Forest Hills-Readville would squish the HP platforms off to one side where it's convenient only for Franklin trains (which hardly ever stop there). It's probably time to think about outright closing that stop instead of rebuilding it.

Boston Landing is doing better than I expected - already ahead of the Newton stops and two Wellesley stops - which indicates just how badly Allston-Brighton needs frequent and speedy transit. 82% of boardings are inbound; it's primarily serving commuters from the local neighborhood rather than the development.

Yawkey is doing incredibly well - exceeding projections (937 daily boardings+alightings) already. It's now at 2,491 (1200+ round trips) daily - roughly triple what it was in 2012. Approximately half of the new trips on the Worcester Line can be directly attributed to Yawkey and Boston Landing (though Yawkey likely poached a few from Back Bay), and those two stations really show how RER-type service (particularly that the NSRL could provide) would be transformative.
The hyper-local trip origins are big evidence in support of Urban Rail. Since Riverside can safely chain off a max-out upgrade of Sharon substation for Providence + Fairmount electrification the drumbeat needs to get loud to include both intra-128 Urban Rail lines in the initial buildout. These trip characteristics are also big evidence in support of fixing the crippled Newton trio sooner rather than later. As well as building the Newton Corner infill, and building the west-region bus garage at Watertown to substantially boost bus frequencies out there. If paired with enhanced Fitchburg 128 short-turn service you'd also need to do a second-wave bus revamp out west. Things like trimming the increasingly traffic-clogged Pike express routes, plugging in TT wire into the still-active A Line feed under Galen St. to extend the 71 to loop at NC, redrawing some of the duplicating Waltham-Newton routes for better frequency & distribution, etc.

The big winners on the Fitchburg Line are Littleton/495 and South Acton - unsurprisingly, the two stations with proven demand that got major upgrades (and, for Littleton, a frequency boost) - which drove almost all the line's increase. Impressively, this is at two stations that already had full parking lots every day - these increases are coming entirely from carpools, dropoffs, and walk/bike. The speed increases didn't help other stations much, and Wachusett is at just 130 riders a day (but growing - in February 2017 it was 45-75 daily). Brandeis lost a lot of reverse commute ridership - perhaps this count was in the summer.
Better reverse-commutes was one of the supposed upsides of the Fitchburg upgrades. Littleton is an especially ripe target because of the MART Littleton-Westford shuttle bus hitting all the major tech employers along 495. Frequencies on that bus are quite good during peak 9-5 commute hours, though it could badly use a service hours expansion for shift employees. Wachusett figures to be a grower mainly for the MART routes from Gardner and excellent overall bus coverage in Gardner enabling those pipes. South Acton has an in-town shuttle pinging between West Acton, North Acton, and South Acton, as well as a SA-Maynard shuttle. All are skeletal operations that need better frequencies (Acton isn't a member of any RTA district, so their shuttles are solo efforts), but more/better connecting bus service is probably going to be their salvation from the perpetually overfull parking lot. Not to mention positive impetus to bring back the old West Acton stop.

It says these were "Spring/Fall" counts, but without knowing dates it's impossible to tell whether students were on-campus for the Brandeis counts. I'd say highly likely the Spring counts were either sampled during Spring Break or in May post-graduation, because the degree of ridership crash there would show itself at Waltham Ctr. too (it didn't) as well as a +/- change in the Waltham bus ridership (haven't been any wild swings there). Sounds like a statistical quirk of timing.

Also...holy moly, Lincoln! Where did those riders come from???

Despite the double-tracking, the Haverhill Line had basically no growth.
Not surprising, because they're still shackled to the same schedule ceiling as long as they're stuck at inadequate Bradford Layover. Cycling trains is difficult to begin with on the outer half of this line because of the jam-packed schedule of freight trains and Downeasters keeping the equipment rotations more irregular than most lines. The double-tracking was supposed to dovetail with the now-canceled Plaistow extension with more spacious layover digs. Although 1 of the 3 yard options under consideration for Plaistow was just on the MA side of the border and could be pursued independently, they haven't done so. The only thing the DT has done has fished Haverhill out of the bottom division on delays caused by other railroads; it has yet to improve service. And the conjoined line with Reading still makes travel times to Haverhill borderline excruciating, especially with all the dwell time problems at the non-ADA platforms south of Ballardvale and the equipment shortages.

The Reading Line portion is up modestly, and that should be a key target of RER for the neighborhood characteristics. What they need to do immediately is cue up an ADA bundle to redo the platforms Wyoming Hill-Wakefield in a big wad, since they all sit at grade crossings and should be able to be mowed down prefab-style. Probably will involve flipping the short Wakefield platform from the north to the south side of the Albion Ave. grade crossing so it can be a full-length, but other than that it's a cookie-cutter job. Reading's difficult because of the attached depot building; maybe they can build a single full-high on the unused 2nd-track berth, flip the track to the other side, and (very) temporarily disuse the depot side until they figure out an architecturally non-invasive way to redo that platform. Use of auto doors to Reading would do wonders for the dwells.

Then they need to start paying the going rate to up the capacity for future Urban Rail short-turn service. Reading Jct. in Somerville needs to be re-graded into a 2 x 2 track split with the Eastern Route. The Wellington-Malden passing siding needs to be upgraded for full service. The Reading pocket track needs to be upgraded for layover yard. Crossing protection needs to be upgraded, especially at places like Greenwood that still retain a staffed crossing tender.

Then for the outer half of the line they NEED that Bradford-replacement layover yard at the state line. And they'll need to start running a lot more Haverhill trains via the Lowell Line + Wildcat Branch to both increase frequencies on each of the currently conjoined halves, ease the dwell/overcrowding problem, lower the painfully long travel times to Haverhill, and have a less brittle schedule for surviving freight/Downeaster delays. They'll need to reconfigure the Wilmington platforms on the Lowell Line so both platforms are south of the MA 62 overpass accessible to Wildcat trains, instead of staggered and partially inaccessible like today (bonus: south of the overpass there'd be room for full-highs + a center passing track for freights/Downeaster). They *may* need to double-track the Wildcat. If the service shift substantially decreases frequencies at N. Wilmington, they *may* need to expedite sooner rather than later a rebuild of Salem St. station on the Wildcat as a replacement. And they should, if Bradford Layover is vacated, rebuild Bradford Station as full-highs with passing tracks using the yard real estate.

Plymouth is down to just 20 riders per day. The MBTA should really cut their losses there and make it weekend-only or something.
Which sucks because the TOD at Cordage Park isn't totally awful anymore. Still light years from where it needs to be, but there's at least a cornerstone of density-supporting tenants in place.

The decision to fork the terminals and cut back from Downtown Plymouth is now in its 22nd year of biting them square in the ass. Cordage is an afterthought, and an operational disaster...but the connecting bus transit really isn't bad. Kingston is getting OK utilization as a parking sink, but it's in the middle of a sand pit access-separated from the McMansion subdivision and flanked to the south by transient big-box retailers. With Kingston Collection in particular white-knuckling it through retail market volatility and distressingly high turnover rate for mall tenants. And while it also has OK bus access, the wall of highways surrounding it makes the access severely constrained. It's tough to see what kind of future that stop has staking itself to so much sprawl and bunkered-in surroundings. The odds of the mall being a ghost town in 12 years are astronomical, and the boost from RER service levels is largely going to leave the sand pit station stuck in prehistoric times.

Unfortunately I don't think there's any way to fix this other than doing what they should've done from Day 1: extend the last 1.2 miles to Downtown Plymouth (the townies seem to be realizing the folly of their initial opposition as they try to stimulate the area), close the Kingston sand pit, replace with a reanimated Kingston Depot on 3A in the walkable part of that downtown, shuttle buses FROM the walk-up density TO the malls not the other way around. They can keep the Kingston Branch for the layover yard. A bunch of messy loose ends to tie up to make it work, but the current setup really doesn't work for RER and it's really not a large sum at all to lay a mile-plus of track and build 2 outlying stations.

Chelsea inbound ridership has tripled, with only modest change to the ~120 daily riders from further out who alight there. That indicates that SL3 is not being used as a Seaport connection, but may be helping feed CR. Salem had a decent ridership increase; Beverly didn't.
Beverly basically had no improvements whatsoever except for the underutilized parking garage down the street, while Salem got a full revamp not just on parking but platforms, car/ped egresses, busway, amenities. This may be an important tell on where the cutoff for Urban Rail is on the North Shore. There's indecision on whether it should be Salem or Beverly, with the T penciling in Beverly by default. The single-track platform coming out the tunnel + bridge openings may make Beverly a bit of a reach, while using the Peabody Branch turnout + a new flanking platform @ Salem may simply end up being the operationally easiest way to swing it. If the ridership data pointed a little more strongly to Salem too, that probably cements it. Also...Salem is a mini- bus hub in its own right so all else we're seeing in these data about the enhancing value of connecting transit puts more weight on Salem for Urban Rail.

Wickford Junction and TF Green are still struggling - around 250 riders per day each.
Eh, RIDOT's still playing a very long game here with its intrastate buildout. Those are still placeholders until enough infill station builds get cued up that true Pawtucket-Westerly intrastate service can be run. It's still way too early to judge.

Nothing of too much note on the Franklin, Needham, Lowell, and Middleborough/Lakeville lines - modest growth, with relatively minor year-to-year variation at most stations.
I would say Middleborough Station's robust growth spits right in the face of the state's attempt to close it to shoehorn South Coast FAIL on that useless M'boro alignment. Also...the bus hubs of Quincy & Brockton are really stepping up, continuing the theme of connecting transit driving ridership increases. Quincy you'd need to dig deeper to see how many people are just sidestepping the zone increase to SS by transferring to/from Red there, but if BAT Centre in Brockton is slugging almost equal increase there's definitely something cooking behind the bus angle. I always thought BAT needed to expand its spider map and span the whole area between the Stoughton Line to the west (with increased frequencies on BAT 14 and touches to the South Coast Rail stops in Easton if they're ever added)...and the Plymouth Line to the east with full service hitting Abington and Whitman. BAT's intra-city frequencies are so good and their route map so logical they really are the best-kept secret of the state's RTA's, and an asset to exploit for commuter rail.
 
You're also seeing a big shift from Hyde Park to Fairmount down the street because the characteristics of Fairmount Line service better fit the neighborhood than Providence/Stoughton.

Fairmount is also way cheaper than Hyde Park (Zone 1A vs Zone 1). $84.50/month versus $200.25/month, or $2.25/trip versus $6.25/trip. A ridiculous discrepancy for two stations separated by an 8 minute walk.
 
Ruggles also got quite a big boost in hyper-local CR (ie, riding to South Station and back) in those numbers and honestly I think they under counted because the past few months I've seen a big uptick in people boarding there heading inbound, looks like mostly NU students but the uptick has been noticeable. I find it an interesting trend given the direct connection to the OL but it seems like the direct connection to RL and SL combined with less crowding and less broken trains than the OL...
 
I found errors in their data, and I'm unsure how to report that information.

For example:

Error in Fitchburg Line data

Here, they "counted" 0 passengers alighting or boarding at Waltham on Trains 494 and 424. Yet, they counted dozens of people using the more lightly used Brandeis/Roberts stop at a flag stop on these sames trains.

I've personally boarded Trains 494 and 424 at Waltham many times, and never has there been fewer than 10 people alighting or fewer than 10 people boarding. There is generally a line of people on the ramp and next to the ramp waiting for people to finish exiting before boarding.

The fact that the count is 0 (for boarding and alighting for both runs), rather than another low number, makes me believe there was an error in data processing or collection.

Who would I direct this to so this it is rectified? Ideally, for mass transit to work for everyone, decisions need to be made based on accurate data. GIGO.
 
Apologies for the double post:

Upon closer examination of the editable data in Excel, Trains 494 and 424 are the only inbound Fitchburg Line trains that are mysteriously missing from the count date table. So, these two trains clearly have false data with no record of when the data was collected. That certainly raises some interesting questions.

My best guess is that they used all Zone 2 purchases during those runs and arbitrarily assigned them to Brandeis/Roberts. That would explain the inflated Brandeis/Roberts counts, the completely empty Waltham counts, and the complete lack of record for when and how the counts were obtained.

Anybody know how I can follow up on this oversight?
 
I found errors in their data, and I'm unsure how to report that information.

For example:

Error in Fitchburg Line data

Here, they "counted" 0 passengers alighting or boarding at Waltham on Trains 494 and 424. Yet, they counted dozens of people using the more lightly used Brandeis/Roberts stop at a flag stop on these sames trains.

I've personally boarded Trains 494 and 424 at Waltham many times, and never has there been fewer than 10 people alighting or fewer than 10 people boarding. There is generally a line of people on the ramp and next to the ramp waiting for people to finish exiting before boarding. (Or maybe that was when the person recording had to run to the john to drop off a loaf?)

The fact that the count is 0 (for boarding and alighting for both runs), rather than another low number, makes me believe there was an error in data processing or collection.

Who would I direct this to so this it is rectified? Ideally, for mass transit to work for everyone, decisions need to be made based on accurate data. GIGO.

Definitely looks like an error. Given that these counts are done manually, often with a clipboard and someone making hash marks by hand on paper, it makes me think there was a transference error somewhere between the person in the field and the person doing data entry.
 
Last edited:
Apologies for the double post:

Upon closer examination of the editable data in Excel, Trains 494 and 424 are the only inbound Fitchburg Line trains that are mysteriously missing from the count date table. So, these two trains clearly have false data with no record of when the data was collected. That certainly raises some interesting questions.

My best guess is that they used all Zone 2 purchases during those runs and arbitrarily assigned them to Brandeis/Roberts. That would explain the inflated Brandeis/Roberts counts, the completely empty Waltham counts, and the complete lack of record for when and how the counts were obtained.

Anybody know how I can follow up on this oversight?

The Brandeis numbers are definitely not inflated. I personally know at least 15 people that regularly take the 494 and 424 from Brandeis, 40 definitely doesn't seem unusual. Though given the aforementioned observation that these counts seem to be from summer/Passover (this data shows a 50% drop in inbound boardings at Brandeis from 4 to 7PM) those numbers might be a little high.
 
The Brandeis numbers are definitely not inflated. I personally know at least 15 people that regularly take the 494 and 424 from Brandeis, 40 definitely doesn't seem unusual. Though given the aforementioned observation that these counts seem to be from summer/Passover (this data shows a 50% drop in inbound boardings at Brandeis from 4 to 7PM) those numbers might be a little high.

That's fair. I only want to attest to the mistaken Waltham numbers on those runs. The Zone 2 speculation was only a tiny bit of speculation. The main point is the incorrect Waltham data for those two trains.
 
That's fair. I only want to attest to the mistaken Waltham numbers on those runs. The Zone 2 speculation was only a tiny bit of speculation. The main point is the incorrect Waltham data for those two trains.

I agree on that count, and can also personally attest to "0" being suspiciously low.
 
Mishawum barely gets trains, but it seems to get decent ridership on the trains that do stop there.
 
Fairmount is also way cheaper than Hyde Park (Zone 1A vs Zone 1). $84.50/month versus $200.25/month, or $2.25/trip versus $6.25/trip. A ridiculous discrepancy for two stations separated by an 8 minute walk.

Yep. But Hyde Park is way faster to SS and also hits Back Bay as well.
 
How many trains can layover at the Wachusett Yard in Westminster? I see that there are six tracks that are 1,000 feet long. Does this mean only six trains can layover there, or is there room for more?
 
Mishawum barely gets trains, but it seems to get decent ridership on the trains that do stop there.
Yes! It makes a pretty good TOD stop as it is. Tightly nestled into what, in the burbs, passes for a kind of walkable density.

Really should get its own little TOD neighborhood (given that Anderson is just Acres of Asphalt and always will be)
 
How many trains can layover at the Wachusett Yard in Westminster? I see that there are six tracks that are 1,000 feet long. Does this mean only six trains can layover there, or is there room for more?

6, with expansion space set aside for a 7th if it's ever needed. The ruling limit for layover capacity is how many trainsets have to be staged for overnight storage to feed the next weekday morning's inbound schedule. During the service day the layovers are mostly empty, so their capacity needs are pretty narrowly-defined around the logistics of daily service starts.

6-7 trainsets is all Wachusett will likely ever need. Shared freight traffic Ayer-west puts a finite limit on how close-packed A.M. rush trains can be, so any further step-ups for A.M. rush surges would likely necessitate a pocket idling track near Littleton to stage short-turn inbounds instead. And as you move closer to RER service characteristics the trains during the day are staying in motion by balancing inbound/outbound service levels...meaning that even with substantial outright traffic increases few trains are going into the yard except for very short stopovers to change crews.

Therefore, outer layovers are always going to have their needs set by one thing: how much equipment does it take to briskly wake up the system from an overnight shutdown.
 

Back
Top