Notably, the bus lanes from the southern end of North Washington Street Bridge to the Haymarket busway have all failed to do anything to make the buses faster. Instead, it seems the buses have gotten slower since the bus lane introduction in late 2019 and 2023. Bus lanes do work in other parts of the world, but it seems like the bus lanes between the bridge and Haymarket, and in other parts of Boston such as Allston, fail to to make any meaningful speed improvements to make the buses significantly faster.
The bridge and the route to Sullivan has been a relentless growth in traffic congestion. It's a complete disaster for the OL shuttle buses this week, with additional roadwork near the OL bridge and track replacements going on between Sullivan, Somerville, and Assembly. It's worse in this week's OL shutdown compared to the 2022 monthlong shutdown.
A look into the traffic data for the North Washington Street Bridge even though TransitMatters doesn't support the bus routes that go over the bridge. Is the bridge traffic actually worse than ever?
(Yes, I did try to rummage through dozens of gigabytes of data, filter out the N. Washington St. buses, put the data into a single format, filter it again, and finally vizualize it (I'm not the type to make pretty graphics, so if they look messy, so be it)
City Square to Congress & Hanover St. (pre Summer 2022), City Square to Surface Rd opp Public Market (after Summer 2022) (inbound):
The MBTA moved the bus stop and it's associated time waypoint due to the Government Center Garage demolition, so the data is not comparable if it overlaps between the two time periods.
Haymarket Station (Orange Line entrance) to City Square (outbound)
It's hard to tell a clear trend, aside from the 2022 traffic spike in the outbound direction while construction on the bridge came to a halt and the Government Center Garage demolition led to road closures and traffic diversions. The decreased signficance of the morning and even rush is noticable in the inbound direction, as well as a slight increase in weekend travel times (despite the haymarket waypoint being shifted closer to the bridge in mid-2022).
It's also worth noting that after the introduction of the southbound North Washington St bus lane in September 2019, there was NO change in travel times in the southbound direction in fall 2019. The bus lanes literally don't do anything here, even though they work everywhere else in the world.
Referring back to this 10 year ancient post to see what has changed in the past decade:
But that's a 92/93 bus problem. If the frequencies are too lousy (they are), increase them. It's a pretty crisp 11-minute trip to Sullivan from the Financial District on the 92, 22 mins. when the 92 gets extended at peak to loop around all of Assembly, and 15 minutes on the 93.
I'm not so sure where these travel times came from. Allocating 5 minutes to get between City Square and Haymarket in both directions, I get 17 - 18 minutes of travel time for the 93 and 15 - 16 minutes for the 92 to travel all the way to Post Office Square from Sullivan Square. Is the 11 and 15 minute travel times taken at 1am - 5am in the morning? Aside from that, the only other way those travel times would result is by using outdated scheduled runtimes the T had for these buses back in the 2014 - 2019 era, until the T updated scheduled runtimes in 2019 and 2022.
Buses take 2 - 3 minutes to get from Sullivan Station to the branching point where the 92/93 split off (highly doubt this can be reduced below 2 minutes at all given round-the-block turns/busy intersections), then 4 - 5 minutes on the 92 and 6 - 7 on the 93 to get to City Square.
Now, the 92 and the 93 share similarities with the North Shore buses in that 50% of the route consists of good runtime with frequent neighborhood stops and little traffic, and then literally the other 50% of the route is the stop-few and traffic slog to get to rapid transit. Past City Square, there's only a single stop on North Washington near North Station to get to Haymarket. I'm assuming travel times on this 0.88 mile (1.4km) stretch can't be reduced below 5 minutes, which is about 10.5 MPH (17 KMH). Today, buses lose an average of 4 - 5 minutes inbound and 1 - 4 minutes outbound on this stretch involving the bridge due to traffic, which means what takes 5 minutes during off peak hours takes 8 - 9 minutes involving traffic congestion.
This means that there were probably other factors at play that contributed to the lousy frequencies pre-COVID that led to even more terrible headways today. Buses lose minutes of cycle time to:
3 minutes of cycle time is lost during off peak early morning and late evening hours to loop around North St. to serve the Orange Line entrance of the station instead of terminating the routes at the Green Line entrance on Surface Rd (the 111 terminates there for increased frequency).
10 minutes of cycle time is lost during daytime hours to duplicate the Orange Line between Haymarket and Downtown Crossing. There is no need for the buses to make a 5 minute trip to Downtown Crossing in each direction during midday off peak hours or on weekends, (maybe rush hour that's fine). The Orange Line already does this trip in 2 minutes.
10 additional minutes of cycle time is lost by the 93 to make a loop around the Navy Yard. In 2023, after Eng came into play, the T decided to loop 70% of the 93's buses on this 10 minute detour, instead of 25% pre-COVID, resulting in the 93 losing 14% of it's daily round trips.
City Square to Schraffs Bldg (outbound)
Schraffs Bldg to City Square (inbound)
Today, it takes the 92 and the 93 a full 1 hour (60 minutes) to complete a full 6 mile roundtrip. The buses seem to only lose 6 minutes of cycle time to North Washington Street traffic, closer to 9 minutes during rush hour. Yet 20 minutes of cycle time is lost to making unnecessary loops and detours, some of which duplicate the Orange Line in the most congested part of downtown. This is the direct result of how, with combined with the MBTA's operator shortage, results in 30 minute headways rush hours only and every 50 minutes off peak for the 92 and the 93.
In Fall 2019, the T eliminated the Assembly Loop from the 92, which allowed weekday and Saturday frequencies to increase from every 40 minutes to every 30.
The only way F-Line's travel times would ever be realistic is chopping all of the aformentioned loops/detours/DTX OL duplication from the routes, including actually functional bus lanes that are planned for the bridge next year in March 2025 & functional bus lanes out to Haymarket's Green Line headhouse entrance on Surface Rd. This decreases travel times to 11 - 12 minutes for the 92 and 13 - 14 minutes for the 93, akin to F-Line's travel times, and buses completing a full cycle in 35 minutes each (70 if serving both routes in one cycle). It'd be barely enough to run KBR frequencies (14 - 15 minute headways) midday for both the 92 and the 93 using the 5 buses the T ran middays/Saturdays pre-COVID.
Under BNRD, I can't see T7 buses making F-Line's 15 minutes from Sullivan to Post Office Square, with it's lengthy detour to serve North Station and West End. The trip on the T7 from Sullivan to the end of the bridge would already take almost 12 minutes. That's only going to give 3 minutes to get to North Station with it's dedicated bike median limiting space for bus priority, go around the West End to get to Haymarket, and run all the way to Post Office Square to meet F-Line's 15 minute travel time benchmark from 2014; before the T7 continues on to the South Station traffic hellscape. At that point it'd take a full 19 minutes to Post Office Sq., with dwell times at North Station and Haymarket as well.
TL'DR, traffic on the bridge seems slightly worse in some situations, but not in all cases or clear cut. Most of the lousy bus headways across the bridge, while contributed to by the bridge's traffic congestion, largely stems from poor routing of the buses with lengthly loops, detours, duplication of the Orange Line downtown, and bus operator and funding shortages at the T.
If it helps get and keep folks in the correct lanes further back from the intersection, I have no particular objections. They're a bit big though - road signs should be sized to the distance/ speed at which they're meant to be read.