TMC

The MTA considers cross-platform transfers to have 0 transfer penalty, even at current frequencies. Even with long transfers, passengers weigh the transfer time 1.75 times as much as they do the time spent riding on the trains. So, a two-seat ride that is quicker, or less than 1.75 times as much as a comparable one-seat ride is preferable.

The issue is that it's not necessary, you gain capacity by not having Broadway share with the . It's really just extra money. Tracks can handle 30 TPH, you want to split the TPH such that they are either equal or multiples of the other (a split of 15/15 TPH works, 6/6/12 TPH works too, 3/8 TPH does not work). Typically, you want to avoid having more than two routes interlining on a piece of track on a subway system, to keep branch frequency high. However, I do have a 5/5/5 TPH split between Lefferts, Far Rockaway, and Rockaway Park for the , killing off the shuttle. Reverse-branching reduces capacity, due to requiring many different merges in quick succession, so realistically, a 30 TPH line becomes a 20-24 TPH line, even though the signals can handle 30 TPH (they can actually handle up to 40 TPH, but that's the limit, and it's not worth purchasing the extra equipment to delete 30 seconds off of headways, when there's no need for that kind of capacity).

There's an issue where people measure de-interlining proposals against the current status quo when really, de-interlining changes the entire status quo completely. I take issue when people speak mostly about de-interlining as a speed improvement because that's a side effect of it, not the main benefit. The main benefits of de-interlining are reliability, frequency, and capacity: - Reliability is the biggest benefit from a passenger's perspective. The argument about one-seat rides completely falls apart, when you realize why one-seat rides are so important nowadays. It's not because they are the objectively quickest way between two points, but rather, in an unreliable system, people perceive transfers as another chance for an incident to occur. Schedules also are not very consistent, meaning it's hard for a passenger to be confident in knowing exactly when to expect your train. De-Interlining changes this by allowing schedules to be extremely consistent, so instead of a pattern like 2323223222323233, for instance, it will always be 2323232323232323232323 at the platform, each train running equal headways, or multiples of each other's headways. Obviously, this doesn't solve every issue with reliability, but I consider reverse-branching to be the root cause, as it allows delays to propagate around the system, as well as causing crowding on lines where the average crowding is actually pretty low. People have adapted to this system, so of course people would prefer one-seat rides when the chance of random long wait times, signal problems causing issues rippling across the network, etc. is pretty high. Commuters also add insane amounts of buffer time to avoid being late. I recall a NYT article who interviewed a commuter who gave herself 90 minutes for what should have been a 30 minute trip. My take is that one-seat rides would be much less relevant to non-CBD bound riders if reliability was seriously improved through de-interlining and other more minor fixes. - Frequency goes hand in hand with reliability and capacity, but the gist is that frequency induces ridership (to a point). About the , it runs abysmally low frequencies, and the 4th Avenue Line should be stronger in terms of ridership numbers, but it isn't. Put together the pieces. - Capacity is an issue. It's easy to say "build new lines" when we have capacity issues, but the current plans wouldn't even add capacity where necessary. Reverse-branching, even with CBTC and switch improvements, caps capacity, because after a certain point, reliability issues happen. Lexington Ave Express manages to get 24-27 TPH, though I believe it's timetabled for 29 TPH. 7th Ave Express achieves 24 TPH. They were both chronically overcrowded pre-pandemic (and I'm optimistic about post-pandemic ridership, if the system was run well), along with other Midtown trunks, and some lines in the outer boroughs. Organization before Electronics before Concrete, de-interlining is an organizational issue. Building new lines should happen, but it shouldn't be our first attack towards a capacity issue.

You can look at stringlines (which are maybe 85% accurate), know what current TPH figures are, ridership. Stringlines are here: https://pvibien.com/stringline.htm (Broadway does not run well at all) TPH is supposed to be about 25 TPH on the local north of 42nd, 17 TPH south of 42nd, 11 TPH express north of 42nd, 19 TPH south of 42nd. But these are all uneven TPH counts that contribute to unreliability and inconsistent scheduling. Broadway is the least used of all the Midtown trunks, which is a really bad sign that something is wrong because it is the most direct B Division hit on Midtown's peak job density (6th Avenue is also pretty direct, but gets somewhat better service).

There are tons of things they have to consider aside from rider feedback.

So true. If "all riders can agree on this" was the metric used for transit planning, nothing would improve substantially for everyone as a whole, but maybe a few individuals.

Simple, ignore complaints from the communities, if the community boards can’t be dissolved (they should be)

CBTC is not a true fix without de-interlining, it doesn’t allow capacity improvements, and definitely doesn’t improve reliability. Switch replacements should be routine, but they aren’t a main contributor to capacity issues, nor are they the root cause of reliability problems. Not to say these are bad, but planning should prioritize reliability and capacity, not the “convenience” of riders (which is a false idea, more reliable service is always more convenient, less reliable service is always less convenient, simple).

I don’t like this, it keeps reverse-branching on Broadway, and the shouldn’t touch Queens Blvd.

It isn’t flawed, and I wouldn’t call the fact that Queens riders get consigned to local service “abysmal”. There really isn’t a need to run 60 TPH along 2nd Avenue, especially at current costs and even globally average costs. If you’re going to build a quad-track line, you better be slamming it to capacity, or else it’s not worth the spending. Midtown absolutely needs more capacity, and that’s where the issues with 2nd Ave really are, not the lack of quad-tracking. From Queens, local trains make the same stops as s down to 34th Street, which is where job density starts to drop off in Midtown. For the Financial District, which is a much smaller job center (and might as well be dying as one), transferring to an express is doable.

We do, actually, from tons of different angles.

Doesn’t need to, they should spend more time circulating providing better off-peak service, rather than sitting in yards.

You’re trying to solve the issue with the by doing way too much. What I would do is this: : Coney Island via Brighton Local (15 TPH) : Brighton Beach via Brighton Express (15 TPH) : Broad Street to Jamaica Center (10 TPH) : Broad Street to Metropolitan Avenue (10 TPH) : 96th Street to Coney Island via Broadway Express/4th Ave Express/Sea Beach (10 TPH) : 96th Street to Coney Island via Broadway Express/4th Ave Express/West End (10 TPH) : Astoria to Bay Ridge via Broadway Local/4th Ave Local (10 TPH) : Astoria to Whitehall Street via Broadway Local (10 TPH) Any transfer penalty would be equivalent to zero or close to zero, and reliability along the 4th Ave Line would be drastically improved.

Nassau shouldn't interact with the rest of the system, being a very weak stub line, in a system where demand trends toward Midtown.

To be on the safe side, I'd plan 20 TPH on both the Broadway Express and the Broadway Local all the way from 57th Street to Canal Street. 3 minutes vs. 2.4 minutes is not a noticeable difference to the majority of passengers, and most of the crowding occurs on the from Queensboro Plaza (probably people ditching the ), which currently runs 15 TPH last I checked. I believe the MTA has very conservative assumptions about the limitations of termini across the system, so I had someone else comb through the math using a POV video of the N. Capacity=Entry Time+Dwell Time+Platform Clearing Time, and based on that, they figured out that Ditmars can turn 20 TPH with marginally better operations than today, without any changes to the interlocking, or recalibration of signaling.