To your point, yes it should be considered as quick math showing the directional flow of passengers rather than what peak passenger demand should be expected at Yonge-Eglinton station. People will be alighting at Mt Pleasant, Leaside, Avenue, etc. before reaching Yonge (and I suspect in practice, many travel patterns will shift towards Cedarvale).
Both those graphs and the earlier EA show the peak point is eastbound approaching Cedarvale station. Though the numbers here seem about 25% higher; I wonder if they assumed faster travel times or something. Still, relatively comparable.
I think you need to compare this to the previous bus route capacity. Quickly looking, at the (future) peak point eastbound approaching Eglinton West were (last week) the following routes: 32, 63, 90, and 164.
The frequencies on these four routes were 3'13", 5", 10", and 22". That makes a total frequency of 1.52 minutes (39.4 buses an hour). That would be a peak capacity (at 51 riders/bus) of only about 2,000 per hour!
This all assumes that all the buses are at peak capacity for an entire hour.
Eastbound approaching Cedarvale station
| Bus Route | Frequency (min) | Buses/hour | Capacity/hour |
|---|
| 32 Eglinton West | 3.2 | 18.7 | 951 |
| 63 Ossington | 10 | 6 | 306 |
| 90 Vaughan | 10 | 6 | 306 |
| 164 Castlefield | 22 | 2.7 | 139 |
| Total | 1.80 | 33.4 | 1702 |
(including 63 Ossington may be overkill, as I doubt it's that busy approaching Cedarvale station!)
Also, we had routes 34, 51, 54, and 56 westbound approaching Eglinton
Westbound approaching Eglinton station
| Bus Route | Frequency (min) | Buses/hour | Capacity/hour |
|---|
| 34 Eglinton East | 6 | 10 | 510 |
| 51 Leslie | 24 | 2.5 | 128 |
| 54 Lawrence East | 6 | 10 | 510 |
| 56 Leaside | 15 | 4 | 204 |
| 88 South Leaside | 8 | 7.5 | 383 |
| Total | 1.76 | 34.0 | 1734 |
Though I do think that there may be a lot more people changing travel patterns, induced demand, and some modal shift. Double seems a lot, (you'd need triple, which is what you'd need to match the EA 2030s estimates is generous). I don't see how it would be an order of magnitude.
Your approach make sense, but I don't think the math/assumptions you are using to go from weekday ridership over the entire route in both directions to peak point ridership in one direction is quite right.