I'm curious as to the length of the 401 studied.
Nontheless, on a corridorwidth basis, there should be unamious agreement that many mass transit systems are much more efficient at moving
people than some of the world's widest freeways.
TTC subway handles
~30,000 people per hour per direction (it has exceeded that on many overcapacity days). Officially, it's 28,000 ppphpd, but in practice, it can significantly exceed that on really bad days.
A freeway lane according Transport Canada / USDOT -- moves only
1700-2200 cars per hour (tailgating less than 2 seconds -- there's only 60 x 60 = 3600 seconds in 1 hour).
And when a freeway hits "congestion collapse" (aka really bad peak period) and crawls at human-reaction-time stop-and-go -- it actually can move less cars than when it's flowing well. So the freeway efficiency versus subway efficiency actually becomes a bigger delta during peak.
Either way, do the math...
You'll quickly agree that mass transit is really, really important for dense cities
Now that said, subways should be built where they need them the most, to be utilized efficiently.
That said, DRL will certainly generate induced demand (if you've been watching subway-boom cities of lore & recent), which will potentially interact with DRL's ability as a relief. So DRL simply relieves future demand (or it becomes even more dangerous like this (staff pushing people onto subways in certain Asian countries), it just won't necessarily make Yonge Subway less crowded than today. There will be operational/choice flexibility which will help people avoid congestion. The point is DRL is simply relief to future status quo, not relief to today's status quo. Meaning, Yonge likely won't ever be less crowded than today. In the ultra-long term, if Yonge North and DRL-Long+West is built, then even Sheppard+Scarborough will be heavily used because of the induced demand effect of a well-gridded subway system (which Toronto doesn't really have yet). Core subways must be built first though to properly utilize taxpayer efficiencies. So need DRL badly --- and beyond -- just to run rapidly in place on a rapidly-speeding-up treadmill.
To learn more about how well-designed mass transit can move an order of magnitude more people than freeways per lane:
-- Google "TTC 28000 people per hour" and you'll see lots of stats. Also, fully milking the 1min50sec headways of the upcoming CBTC system and maxing out the trainset count, can bring the peak capacity much closer to 35,000 ppphpd, potentially up to 40,000 ppphpd on the overcapacity days (We'll ideally need platform doors to get to 40,000 though).
-- Google "freeway capacity vehicles per hour". And look at the transportation ministry PDF files including USA too. Worldwide, you will get vehicle numbers ranging from ~1400 to ~2400 with the averages being extremely close to 2000 ideal and 1500 realistic per lane, and average passenger per vehicles being typically under 2.
