Correct me if I'm wrong, but is your conclusion basically a variation of "if you build it, they will come"?
I was under the impression that detailed studies of demand along the Eglinton corridor had been done, making use of the city's Official Plan for development and use of the area along the route, to determine whether there would be enough people wanting to use the line to justify it being a subway.
Now it is quite possible they cooked the numbers or the city's plan will not match what happens in reality or their accounting for drawing riders from other nearby routes is off, but I would think that methodology is slightly more valid than the assuming that the simple availability of a subway line will result in subway level demand.
If studies conclude that demand in the coming decades along Eglinton will remain within the levels serviceable by a grad-separated LRT line (with a central portion underground), then does it make fiscal sense to spend oodles more money to make it a full subway? Keep in mind that through the tunneled section, multi-car trains will be able to provide near-subway level capacity.
You are wrong.
I'm not assuming the simple availability of a subway line will result in a full subway line. If you don't build it, they won't come - they literally *can't* come unless you force them to go there by dumping feeder route riders onto Eglinton and hoping they don't stop taking transit...this was what has always filled all of our subway lines.
If you build it and you want them to come, it can be done, but there's more to it and you need to create the right conditions. If you cut back the line and only build 5km of it, you're not going to get as many rides as you would have if it was 15 or 30km long. If you permit extensive redevelopment, you'll get more rides. If you divert feeder bus routes to the line, you get more. If you lower the cost of parking or expand roads and highways, you get less. If you build bike lanes everywhere, you get less. If you offer good connections with other transit routes, you get more. If you mismanage the service and reduce the frequency, you get less. You don't just build a subway, you do or don't do a hundred other things.
You need to realize that calculating a "demand" for transit on paper and then trying to match infrastructure and service to this "demand" does not work and is not good planning. That's not what they study. "Demand" should not be confused with predictions of how many people will actually use the service once it's up and running. Improved transit can create rides out of thin air, and changing it can kill rides and physically prevent people from travelling.
There's no such thing as a "subway line's worth of riders" and even if you do settle on an entirely arbitrary number of riders, this completely ignores quality of service, local context, and any number of spin-offs like changes to the tax base. Eglinton is such a long route with so many intersecting and parallel routes that even a small amount of post-subway redevelopment, a small amount of modal growth, and a small amount of diverted riders would probably easily put the peak volumes above whatever arbitrary threshold you feel like choosing. If there is a "subway line's worth," why is there no "LRT line's worth"?
Of course, a study for an LRT line will conclude that "demand" fits an LRT line, just as a study for a subway line would conclude that "demand" fits a subway line. When the city has to fight an ideological battle to win people over to light rail, the last thing they'd do is tailor an LRT cost/ridership model to spit out numbers that support anything other than the LRT line as proposed. The only methodology involved here is choosing which buzzwords and phrases will be used to justify each choice. "Is warranted," "is not affordable," and so on are meaningless.
You're also wrong about the relative cost of a completely grade-separated LRT line with multi-car vehicles and a "full subway."
But we're not getting a grade-separated LRT line, we're getting a 1/3 grade-separated LRT line. A fully grade-separated multi-car LRT line can do pretty much whatever a subway does...it *is* a subway with somewhat different cars, and it doesn't really matter which of the two we build on Eglinton. Grade-separation vs waiting at red lights outside the tunnel is the real issue, not LRT vs subway.