The MTA runs 15 Fs, 15 Es, 10 Vs and 10 Rs per hour, totaling 50, as many as any other line. I ride the V or R almost every day at rush hour and there are often seats, so it’s not a big issue. Between those two trains, I can make just about whatever connection I need. Now, ideally, I think a new pair of tracks connecting the QB local and the Astoria line somewhere between Queens Plaza and 39th Ave would make the system work much better for intra Queens and Queens-Brooklyn travel. Basically, the Queens routings of the V and W could be switched and the G could go to Astoria if there were room.

Even if there were more tunnels, you couldn’t fit any more Es or Fs on the QB express. The location of the 63rd St tunnel doesn’t keep the F train from being packed to the gils as it crosses the East River, so it’s not like it killed ridership.

The mezzanines are a luxury that probably could have been dispensed with in many places, but I think you would have to have built most of the flying junctions to have the level of service that exists today. They were planning for a much larger system that took decades for the built-in capacity to be utilized. Whoever designed the junction of the 2/5 and 3/4 would probably have done things differently if they could see how that turned out.

The QB line doesn’t really have 6 tracks. The 63rd Street tunnel is far away from where people want to go, and due to track-sharing on QB can’t run more than 15 tph unless the E becomes local. The 60th Street tunnel involves track-sharing, again.

Besides, the IND made mistakes not just in the routing, but also in infrastructure design. Subway stations don’t need full-length mezzanines, and tracks don’t need flying junctions at every intersection. Those extra features are what blew the budget on the First System, ensuring the Second System would never be built.

The QB line has access to 6 tracks headed into Manhattan, and has to share two of those with the NW.

I agree that the G should have provided a connection to Atlantic-Pacific and the QB line should have provided easier connections to QBP, and the variety of stops along 49th, 50th and 51st Sts. They could have routed the G line down Greene Ave and State St, which would have connected it to the A-P complex. That probably would have been wiser.

However, my point it that the IND was following pre-established notions of where the lines should run. In retrospect, they were wrong, but these notions were not just the ideas of the IND. The IRT planned a subway down Lafayette Ave in Brooklyn that also would have bypassed the A-P station (although it would pass through Nevins.) The IRT el on 53rd St was replaced in the same location. There were precedents of the IND. The IND designers did not sit around coming up with inconvenient subways, they simply built other people’s ideas.

The F didn’t really provide more service to the LES than the 2nd Avenue el, which ran on 1st south of 23rd.

The problem with the QB line, if anything, is that it is only two-tracked when entering Manhattan. Even in the 1930s Queens had a clear shortage of transit capacity into Manhattan: it had the same population as the Bronx but only half as many two-track crossings into Manhattan, 3 instead of 6. A better-designed IND would have had the G connect to Atlantic-Pacific and QBP, and had a fully four-tracked QB line connecting to QBP and running along 50th to connect to the existing IRT and BMT subways.

And the reason SAS wasn’t built when it was first promised is that the IND First System was overdesigned and overbuilt, leading to cost overruns.

The problem with CPW isn’t just that there’s a park on one side. It’s that the buildings are just residential, and relatively low-density at that, whereas Columbus is more mixed-use.