Today’s Headlines

When the tunnel first opened, Ks, Ls, Ms, often combined at WP to run downtown and back.
The math is inexorable; 5 routes x 2 minutes apart =10 min apart for each route barely adequate in rush hour, especially as more offices open downtown. If the promised increase of Ns in response to growing ridership actually happens, that makes gaps for the other trains longer. Over the years, shifting the J back to surface running on Market has been floated, but not implemented. Others have suggested running the K as a shuttle between WP and Balboa Park which should trigger abolishing the BART surcharge on Fastpasses.
Bottom line, getting trains less than 2 minutes apart is impractical, and anything less fails to move riders fast enough to prevent persistent delays.
Outbound service in PM rush suffers from ‘missing’ trains on one route or another resulting in crowded platforms and longer loading dwell when the absent train finally appears. For years, I and others have recommended dynamic reassignment of outbound trains to serve each route as scheduled. I have been told there would be massive resistance by train operators, and mgmt has never tried it.
MUNI needs to confront the physical limits of the tunnel with smarter operational policies, and double stopping is worse than doing nothing as it delays following trains even more.

“massive resistance” and “mgmt has never tried it”
sounds like an excerpt straight from the SFMTA employee handbook and mission statement. this is exactly the attitude that holds back SF becoming anywhere near that of a transit first city. when the agency itself has excuse after excuse and blunders through projects without a clue then this is the end result.

a basic efficient service REQUIRES tight headways. I am aware of the surface traffic sabotage issues. You mention 3 Ks then none–that is exactly what dynamic reassignment is meant to cure. I believe a rush hour supervisor/inspector should be re changing your 3 Ks WB into a K, L, M, respectively. The point is to continually tune the sequence so each route gets service at reasonable headways. Between shorter waits for each route and thus lower crowding on the platforms, outbound trains should move faster. The other necessity at Embarcadero is to have the WB operator board and enter the WB cab as the inbound is offloading pax. The inbound operator takes the train into the switchback area, talks to the outbound opr, on the PA and the train comes back to EMB to load up. The previous opr steps off with a few minutes to perhaps use the facilities, then becomes the boarding WB opr boarding the next unloading train. The point is to make time lost switching ends go away while keeping everyone moving.
As to moving the J to the surface–an obvious place to turn Js would be a loop at the museums in the Park. A very short spur into the park on 9th and over to the museums, then back to Balboa Park.
The current situation of the inmates running the asylum for their own convenience i FUBAR and must be replaced.

Double berthing has proven not to effectively work. There’s absolutely no rhyme or reason of its implementation between Civic Center and Montgomery stations.

Right. But my point was that that it would be VERY beneficial in Embarcadero Inbound since majority of trains end their trip there and stop for a longer period while they are searched for people still on.

Double berthing could be better if trains only stopped once. A few extra screens could direct passengers. Anyone who misses their train will have a 10 minute lesson while they wait for the next one, but a month later, few will still be confused.

Double berthing makes sense if you’re trapped in “traffic” and just want off the damn train now. I can attest that this is a common enough need. It does not make sense in terms of running an efficient transit service where trains don’t double berth because they they aren’t stuck in a 12-car queue between Montgomery and Embarcadero.

The years of effort that went into implementing it should have been spent preventing the problem in the first place.