Harvard Square is my home station and I use the Red Line regularly, but I would welcome a full closure north of Kendall for a while if it actually fixed the slow zones. At this point, the crawl from Central to Harvard has gotten so slow that I have genuinely chosen to walk because it is faster for any destination that is more than one block from the station exit. This is over an inter-station stretch that is almost a mile long!
The nice thing about the Ashmont shutdown (assuming the slow zones don't come back) is that it's now much more politically feasible to fix other parts of the system with short-duration full closures. The public will see the effectiveness of shutdowns and make statements exactly like yours, unlike the period after the Orange Line shutdown last year.
I personally know someone who commuted from Porter to MIT (reasonably close to Kendall station) last year, and she almost always biked. It's a shame, really.
If the slow zones doesn't come back. I think the big difference between the Red Line shut down versus the Orange Line is planning. The Orange Line did a month long shutdown, but it seems it was only decided roughly a month ahead. The Northern portion was planned for months before it was cancelled. Yet the expansion to repair the entire line was had only a month to plan out.
Also along with time to plan, was the act of planning at all. The Orange Line found a bunch more issues as the repairs were executed as supported by previous MBTA statements and also personally derived by MBTA news reports alot of issues were not documented (or worse). Following the chatter of people like that redditor tracking the slow zones, it seems the MBTA spent a lot more time inspecting. This "caused" the Red Line August slow zone surge, but it also mean actually knowing all the issues and thus have the labor, preparation, and materials all laid to install without finding any new issues to also scramble to fix.
Now I hope I don't end up eating those words. Please this time finally see some breaks in favor of the MBTA please.
If the slow zones doesn't come back. I think the big difference between the Red Line shut down versus the Orange Line is planning. The Orange Line did a month long shutdown, but it seems it was only decided roughly a month ahead. The Northern portion was planned for months before it was cancelled. Yet the expansion to repair the entire line was had only a month to plan out.
Also along with time to plan, was the act of planning at all. The Orange Line found a bunch more issues as the repairs were executed as supported by previous MBTA statements and also personally derived by MBTA news reports alot of issues were not documented (or worse). Following the chatter of people like that redditor tracking the slow zones, it seems the MBTA spent a lot more time inspecting. This "caused" the Red Line August slow zone surge, but it also mean actually knowing all the issues and thus have the labor, preparation, and materials all laid to install without finding any new issues to also scramble to fix.
Now I hope I don't end up eating those words. Please this time finally see some breaks in favor of the MBTA please.
The OL shutdown was not a well planned maintenance break, it was a hasty response to the OL bridge fire and FTA investigation. No doubt they did what they could in that month, but as you said a good chunk of it was likely spent finding problems rather than solving them, and since solutions don't just appear overnight another good chunk of time was probably spent waiting. As you said, compare that with the recent RL closures and you can see that extended closures for maintenance can be effective, they just need planning.
The OL shutdown was not a well planned maintenance break, it was a hasty response to the OL bridge fire and FTA investigation. No doubt they did what they could in that month, but as you said a good chunk of it was likely spent finding problems rather than solving them, and since solutions don't just appear overnight another good chunk of time was probably spent waiting. As you said, compare that with the recent RL closures and you can see that extended closures for maintenance can be effective, they just need planning.
In some sense night time closures are best spent on thoroughly documenting the problems, and creating the work plan to fix them. Then supplies and crews can be properly queued up to make maximum use of a full shutdown period to do rapid but thorough repairs.
In some sense night time closures are best spent on thoroughly documenting the problems, and creating the work plan to fix them. Then supplies and crews can be properly queued up to make maximum use of a full shutdown period to do rapid but thorough repairs.
I mean, ideally you don't get so behind on repairs and maintenance that most necessary can be done during the night, but, hypothetically, if you find yourself running a transit system that's about 2 decades behind on essential work that is probably the best solution.
I'll agree that the shutdown did a fantastic job, but I have remaining questions about why we're not back to the best time recorded historically. What work remains to get us there, and what sort of timeline might we be looking at to get the designed speeds?
I'll agree that the shutdown did a fantastic job, but I have remaining questions about why we're not back to the best time recorded historically. What work remains to get us there, and what sort of timeline might we be looking at to get the designed speeds?
Red Line Transformation's re-signaling + the new vehicle order going to completion. The new CRRC cars have faster takeoffs and shorter stopping distance than the old mixed-lineage fleet, and the signals are being re-timed to take advantage of that. That'll enable speeds that haven't been seen on the Ashmont Branch since the pre-1988 trip stop signal system.
After getting my car's inspection sticker yesterday, I went to town to get a resident parking sticker. Rode the B-Line trolley & it was slower than cold molasses in January!!!! Why in the Sam Hill are trolleys & train still running so blasted slow, when the tracks are supposed to be repaired & slow zones lifted?!!!!
Red Line Transformation's re-signaling + the new vehicle order going to completion. The new CRRC cars have faster takeoffs and shorter stopping distance than the old mixed-lineage fleet, and the signals are being re-timed to take advantage of that. That'll enable speeds that haven't been seen on the Ashmont Branch since the pre-1988 trip stop signal system.
But this thing ran faster even back in 2021. What changed since then? Is it just the decaying trainsets, or is the infrastructure itself that's keeping the system slow?
I'll agree that the shutdown did a fantastic job, but I have remaining questions about why we're not back to the best time recorded historically. What work remains to get us there, and what sort of timeline might we be looking at to get the designed speeds?
But this thing ran faster even back in 2021. What changed since then? Is it just the decaying trainsets, or is the infrastructure itself that's keeping the system slow?
In what way did it run faster in 2021? I've played around with TM's historical trip time data a bit and it looks like the Ashmont branch is running as fast as it did in 2021 for Northbound. Yesterday's (10/31/23) median time for Ashmont->JFK was 8min 23sec, the same date in 2021 was 8min 42sec.
Looking at the year more generally, I don't see any point in the year where the weekly average or monthly trip time for the stretch would have come in below that ~8:20 mark by any significant amount. The single fastest day of the year looks to have been a median time of 8:09 on 3/28/21.
------
Southbound seems to be continuing to improve by day and so far today (11/1/23 9:45am), trips look to be within ~20sec of what typical 2021 mornings were like (from randomly glancing at a dozen or so). Which also fits with TM's dashboard that suggests there's a 15sec delay JFK->Savin Hill going SB as the only remaining slowness.
In what way did it run faster in 2021? I've played around with TM's historical trip time data a bit and it looks like the Ashmont branch is running as fast as it did in 2021 for Northbound. Yesterday's (10/31/23) median time for Ashmont->JFK was 8min 23sec, the same date in 2021 was 8min 42sec.
Looking at the year more generally, I don't see any point in the year where the weekly average or monthly trip time for the stretch would have come in below that ~8:20 mark by any significant amount. The single fastest day of the year looks to have been a median time of 8:09 on 3/28/21.
I'm basing 2021 purely on bigeman's analysis. I can't get the transitmatters dashboard to give me any data for most timeframes, but I'll agree that we do seem to be back to baseline here.
I'm basing 2021 purely on bigeman's analysis. I can't get the transitmatters dashboard to give me any data for most timeframes, but I'll agree that we do seem to be back to baseline here.
Travel times improved throughout the day yesterday. My analysis was accurate as of early afternoon, but the median came down quite a bit later in the day as the slowest 17 southbound trips of the day were the first 17 southbound trips of the day.
UPDATED ANALYSIS THROUGH END OF DAY YESTERDAY:
Yesterday, 10/31/23, the median round trip on the Ashmont Branch was 17 minutes.
In the days leading up to the shutdown, a round-trip was 33 minutes, which is 16 minutes slower than today.
As its worst, in late September, a round trip was 36 minutes, which is 19 minutes slower than today.
The last time the Ashmont Branch had round trips this fast was March 19, 2022.
Transitmatters has no data for any day in which round trips on the Ashmont Branch were faster*, dating back to 2018.
BOTTOM LINE: Yesterday saw Ashmont Branch round trips as fast as they have been on record, dating back to 2018.
Important note: I am rounding each day's median (single-direction) trip to the nearest minute for simplicity. This sacrifices a bit of precision, but being precise to the second is unneccessary here. I am purposefully aiming for a digestible summary, rather than minutia.
*Excluding some bad/misleading data that exists in the immediate aftermath of the 2019 derailment at JFK/UMass and the altered service/operational patterns in its wake.
In addition to the Ashmont branch improvements, the Braintree branch is also getting some love.
The long 10 mph slow zone from JFK/UMass to North Quincy has been eliminated, leaving only two 25 mph slow zones along this section that span 1,100 ft:
Northbound tracks from North Quincy to JFK/UMass still have one 10 mph slow zone (476') and three 25 mph slow zones (800' total).
A NEW POLL indicates MBTA riders are not happy with the transit authority’s service, with the slow-zone plagued subway and trolley system receiving the
EDIT: This is granular and long, so I am hiding it behind spoiler tags. It's Ashmont Branch round trip times.
MINUTIA
I do not think analyzing to the second is the best focus when looking at the big picture of performance. This entire post needs a big asterisk to note that a difference of seconds is marginal at best. That being said, I am indulging the desire for (possibly false) precision, to the second, because I love diving deeply into data, especially MBTA data and it seems like people have questions that can be answered in the minutia.
After spending some time in Excel with the data, here are some notes. I have cleaned the data and if anyone has adjacent inquiries, feel free to ask.
Summary:
Yesterday, on 10/31/23, the median round trip on the Ashmont Branch was 17:44.
The last time the Ashmont Branch had round trips this fast was 6/23/22, when the median round trip was 17:32, which was 0:12 shorter than yesterday.
September 2021 was the last month with an average trip time* (17:30) faster than yesterday's median (17:44) by 14 seconds.
Fastest on record:
The day with the fastest round trip on record, dating back to 2018, was 4/5/20, when the median round trip was 16:43, which was 1:01 shorter than yesterday.
The month with the fastest average trip time on record was April 2020, when the average of the daily median round trips was 17:04, which was 0:40 shorter than yesterday.
It is crucial to note that both of these records took place in the midst of the height of the COVID-19 pandemic and therefore had incredibly low ridership. On 4/5/20 there were 7,843 validations recorded on the Red Line, under 10% of the average daily ridership in 2023. This lowest-low ridership had the effect of minimizing dwell times and therefore improving trip times. The fact that yesterday was only 1:01 off of that lowest-low day and 0:40 off of the lowest-low month is impressive.
The last day a median trip took as short as x:00
Prior to the shutdown, on 10/13/23, the median round trip on the Ashmont Branch was 32:53, which was 15:09 longer than yesterday.
Prior to the shutdown, no day had a median round trip on the Ashmont Branch under ...
... 30:00 since 8/8/23, when the median round trip was 29:58, which was 12:14 longer than yesterday.
... 25:00 since 3/9/23, when the median round trip was 21:11, which was 3:27 longer than yesterday.
... 20:00 since 3/8/23, when the median round trip was 19:51, which was 2:07 longer than yesterday.
... 19:00 since 2/1/23, when the median round trip was 18:29, which was 0:45 longer than yesterday.
... 18:00 since 6/23/22, when the median round trip was 17:32, which was 0:12 shorter than yesterday.
This was also the most recent day with a shorter median Ashmont Branch round trip.
... 17:00 since 9/6/21, when the median round trip was 16:55, which was 0:49 shorter than yesterday.
Lookin back at monthly averages*
Yesterday, on 10/31/23, the median round trip on the Ashmont Branch was 17:44
Even with 10/31/23 included in the calculation, the average median daily round trip time on the Ashmont Branch ("average trip time*") in October was 30:32.
September '23 had the worst average trip time at 32:14.
June '23 was the last month with an average trip time under 30 minutes (29:08).
February '23 was the last month with an average trip time under 25 minutes or 20 minutes (19:52), before the great slow down of March 2023.
June '22 was the last month with an average trip time under 18 minutes (17:49).
September '21 was the last month with an average trip time (17:30) faster than yesterday's median (17:44).
Table of monthly averages:
Month
Mean*
October-2023
30:32
September-2023
32:14
August-2023
31:26
June-2023
29:08
May-2023
29:27
April-2023
30:01
March-2023
27:59
February-2023
19:52
January-2023
19:19
December-2022
21:36
November-2022
23:46
October-2022
23:31
September-2022
22:20
August-2022
21:41
July-2022
19:50
June-2022
17:49
May-2022
19:18
April-2022
18:05
March-2022
18:03
February-2022
19:10
January-2022
19:23
December-2021
18:34
November-2021
17:51
October-2021
19:13
September-2021
17:30
August-2021
17:24
July-2021
17:15
June-2021
17:13
May-2021
17:09
April-2021
17:12
March-2021
17:41
February-2021
17:37
January-2021
17:32
December-2020
17:29
November-2020
17:32
October-2020
17:20
September-2020
17:23
August-2020
17:39
July-2020
17:41
June-2020
17:19
May-2020
17:06
April-2020
17:04
March-2020
17:32
February-2020
17:39
January-2020
17:39
December-2019
18:02
November-2019
18:07
October-2019
17:48
September-2019
18:31
June-2019
18:12
May-2019
18:18
April-2019
19:00
March-2019
18:47
February-2019
19:01
January-2019
18:25
December-2018
18:20
November-2018
18:27
October-2018
18:11
September-2018
18:11
Note: Data is through 10/31/23
Note: I am excluding the preiod from 6/11/2019 through 9/7/2019. During this three month period, the Red Line had some unusual service/operational patterns while in recovery from the JFK/UMass derailment, leading to some wacky, wildly inaccurate data.
*This is an average of daily median round trip times. While we can think of this as a monthly average round trip time, technically this is a slightly different calculation.
EDIT FOR 11/1/23: As of 3:00 pm, the median for today (11/1/23) is even faster, at 16:51. If that holds, it would be within 8 seconds of the fastest ever, and by far the fastest of any day with ridership even close to today's ridership, with records dating back to 2018.
The OL shutdown was not a well planned maintenance break, it was a hasty response to the OL bridge fire and FTA investigation. No doubt they did what they could in that month, but as you said a good chunk of it was likely spent finding problems rather than solving them, and since solutions don't just appear overnight another good chunk of time was probably spent waiting. As you said, compare that with the recent RL closures and you can see that extended closures for maintenance can be effective, they just need planning.
On top of that, the Orange Line shutdown was conducted in 2022, and managed by the Baker/Poftak MBTA. This one was conducted in 2023, by the Healey/Eng MBTA (which also just fixed the GLX slow zones in two weeks). Kind of a big change there.
