Fancy transit to the airport is one of those things that is popular, but not all that useful. Everybody thinks it's a good idea, but the actual market when push comes to shove is small:
- Travelling here for business? Your time is too valuable to figure out niche transit services, just spend the company's money on a taxi or Uber.
- Travelling here to meet family or friends? They probably don't live downtown, so you'd have to connect to another random transit service, plus they likely have a car and are picking you up.
- Travelling here in a group? That $11 fare adds up quickly; pretty soon it's comparable to taking a taxi, but with more hassle. If you have kids, you definitely want to have them in a cab rather than trying to keep them calm on a bus (and paying more than taxi for your trip).
- Travelling alone, on a holiday? You just spent four hours and hundreds of bucks cooped up in a plane; what's another 20 or 30 to save the hassle and just get to your hotel?
- Living here and travelling from the north suburbs? Why would you go out of your way south all the way to downtown to catch transit back past your house to the airport?
- Living here and travelling from the south suburbs? It's a 20-30 minute drive directly to the airport. By the time you shlep downtown to catch the bus, it's not really worth it.
- Living here and travelling from work downtown? You're a business traveller, just take a cab and expense it.
Yes, airports in major European cities get good transit use. That's because they are cities with dense populations, and high quality transit networks, so there are plenty of people who don't have a car or don't regularly drive but have enough money to travel. In fact, they're typically dense enough that it's feasible for public transit to be as fast as driving to the downtown.
On the Montreal case, there are a few things I agree with, but the service to Lionel-Groulx is not downtown service in any meaningful sense; it's a kilometre and a half from Concordia, which is about as far west as you could consider downtown (and Montreal's downtown is a long east-west one; the middle of downtown is at least two kilometres from Lionel-Groulx). The way the roads go, Lionel-Groulx is roughly as close a station to the airport as any other It could be compared to Calgary's 100, which goes to McKnight Westwinds in 22 minutes (only 4 minutes more than by car) and which in theory has like 6 or 7 stops but I'd be shocked if it actually stopped more than twice on any given trip. But the 100's headway also sucks, every 30 minutes.
I also disagree with the implicit suggestion that people want to get to an arbitrary point downtown, and that's it. Especially the City Hall LRT station, which is not the most salubrious area, and more importantly, isn't near any hotels. Who wants to drag their bag several blocks in the winter to get to their hotel? Here's the downtown hotels (cyan) and 300 stops (magenta). The City Hall stop is noticable because it's the one that's the furthest from any hotels.
The 300 has problems -- both the unacceptable headways and the lack of good information -- but stopping a few times downtown so it's near the hotels is not one of the problems IMO.
As a side note, here's two current transit maps; which one was made by a city that invests in and wants you to take transit? Which one would entice you as a visitor? Which city employs a professional graphic designer?
I suspect the point of the 300 is to provide better transit service on a corridor that actually uses it, while saying we provide transit to the airport because people don't actually take transit to the airport, and so they don't know or care if it's good, as long as it's been advertised. In a way, it's genius -- it provides service that gets real ridership without actually costing the money for near-empty buses to speed down the Deerfoot. I'm not sure that taking service off of our best corridor to serve a market that isn't likely to take transit is actually a good deal.
But if we were to go with the cover story that the point of the 300 is airport service, then a direct route does make more sense. Right now, the 300 is scheduled for 90 minute round trips, with 38 or so round trips a day. That's 57 service hours of transit. If the route went on Deerfoot (it could bypass traffic on the shoulder in the peaks), it could be scheduled as a 60 minute trip; roughly 17-18 minutes each way to and from the airport, 18 minutes or so for the downtown loop, and 5-8 minutes of break time at the airport. That reschedule would permit 20 minute headways from 5 AM to midnight (instead of the current 30 minutes) for the same cost, which is in the order of magnitude of $3M per year. For about 10% more, that could add hourly overnight service. For 20% more, half hourly overnight. For 50% more, 15 minute headways all day and 30 minutes overnight. Sure, that's an extra $1.5 million dollars a year, which is only, what, 1000 years or so, give or take a few centuries of improved service for the cost of an airport train.
For our city at this time, what we need is the basics executed well. Remove the hassle as much as possible, through clear wayfinding and good promotional materials. This is on both ends - there should be a visible, high quality shelter at the downtown locations with a countdown timer and ticket machine in it.