The comments on Is Speed Obsolete? -- my post on Professor Patrick Condon's thesis that slow streetcars are better than rapid transit -- are a gold mine of perspectives and insights. I could spin a month of posts out of them.
Let's start with this one, from Adrian, in response to my claim that slow transit competes more with walking and cycling, while fast transit competes more with cars.
Streetcars don't compete with walking, they augment it. When you're out and about on foot you can easily hop on a streetcar to go a bit further than you're prepared to walk, and therefore access a few places that are otherwise out of your reach. Streetcars therefore make a walking-based lifestyle more attractive, and lead to more walking, not less.
This point of view depends on whether you think of the ideal trip as spontaneous or intentional. Local-stop transit (streetcar or bus) is relatively easy to use spontaneously, in the sense that if you see one coming you can decide whether to ride it. As such it lends itself to a pleasure-oriented exploration of the city, a kind of flâneurie, which is often a preferred mode of exploration among urban theorists. But most of us have intentional destinations, and deadlines for getting there. Do local-stop, mixed-traffic streetcars really "augment" walking usefully for a significant share of these more purposeful travelers?
First of all, let's note that Adrian's point isn't relevant to the difference between local-stop streetcars and local-stop buses. It's just as easy or difficult to "hop on" either kind of vehicle. Streetcars are more likely to have proof-of-payment systems that spare you (and everyone else on-board) from the delay involved in buying a ticket from the driver, but that's a cultural fact, not a technical one. Some bus systems do use proof-of-payment, and some streetcars don't. Streetcars may also be more legible, but if you're out in your own neighborhood it's not hard to remember what the key bus lines do. The legibility and marketing difference between streetcars and buses is also mostly cultural, and is not found to the same degree in Europe.
So Adrian's point about "streetcars" really refers to "local-stop transit," which is relatively slow especially in congested inner cities where it encounters many sources of delay. In Portland, for example, the Portland Streetcar and the Line 15 bus have about the same scheduled travel time, 13 minutes, from 23rd & Marshall to around 10th & Salmon midday, going by different but equally congestion-prone routes. (By the way, this suggests that if the bus used proof-of-payment fare collection, as the streetcar does, it would actually be faster.)
The Portland Streetcar's scheduled speed, within downtown and NW Portland, is around 6.5 miles per hour. That's just about twice an average walking speed. Cyclists will note that it's about half of a leisurely cycling speed, which is in the 10-15 mph range. In fact, 6.5 mph is around cycling's minimum speed, below which it's hard to maintain balance.
So in round numbers, the streetcar's in-vehicle travel time is about twice is fast as walking, and about half as fast as cycling.
But of course, that's just in-vehicle travel time. Transit also suffers from waiting time, which in Portland, with its 13-minute Streetcar frequency, is 6.5 minutes on average. I haven't run the exact numbers, but it looks like the streetcar, assuming this average waiting time, is faster than walking only for trips of over 0.8 miles. If you're a pessimist, of course, you'll use the maximum waiting time, and in that case it's faster to walk if you're going less than about 1.1 miles.
But what if you just walk down the streetcar line, and catch it if it overtakes you? Thanks to the relatively close stop-spacing (which has other major downsides) it's not hard to do that if you're reasonably fit -- just as it's not hard to do with most North American local-stop bus routes. In fact, as an impatient person myself, it's what I tend to do with North American local-stop buses or streetcars, unless I have real-time information assuring me that the thing is really coming.
By walking down the line, you have the benefit of the streetcar or bus if it comes without being stranded or delayed if it doesn't. But someone using this strategy still risks reaching his destination only in the time it would take to walk the whole way. After all, if you were sure the streetcar would come in time for it to be faster than walking, you'd just wait for it rather than walking down the line. So if you have an arrival deadline, the strategy doesn't allow you depart any later than you would if you were walking. If I have to get to an urgent meeting, or home to a sick child, a streetcar may speed up my trip but I can't count on it to do so, especially if it's the sort of short-but-too-far-to-walk trip that Condon praises.
And those calculations all assume that the Streetcar is perfectly on time, a big assumption given the conditions in which it operates.
So here's my main point:
Rapid transit is a far more viable "augmenter" of pedestrian trips because its travel speeds, and thus the trip-lengths for which it's suited, lie entirely outside the pedestrian's range, whereas the streetcar overlaps the pedestrian range substantially.
The rapid transit and pedestrian modes play entirely complementary roles, while streetcar and pedestrian modes have partly overlapping roles -- a less efficient arrangement. You'll walk further to a rapid transit station, but once you're there you can move at a high speed that makes that extra walk worthwhile. Driverless rapid transit, such as SkyTrain in Vancouver, is also extremely frequent and reliable, both factors that make a big difference to transit's usefulness if you're making an intentional, deadline-constrained trip.
Rapid transit's speed also exceeds typical cycling speed, by a large enough factor that it makes sense to cycle to the station. So rapid transit works with cycling to a degree that local stop transit, such as the Portland Streetcar, just doesn't.
Obviously, the usefulness of rapid transit requires a longer trip length, so rapid transit should be considered only for relatively long corridors. As several commenters have mentioned, the problem with Condon's view may be in the corridors to which he's applied it, including Vancouver's Broadway corridor, where he's presented it as an alternative to a SkyTrain extension. The Broadway corridor is 8 miles (14.5 km) from the University of British Columbia at the west end to Commercial Drive station in the east, and many people ride the full length to make connections to the suburbs further east. At such distances, rapid transit's speed advantage is substantial, and its ability to maintain reliability at very high frequencies is also crucial.
Most cities have urban corridors where average trip lengths are short, and there, a streetcar vs. local-bus debate may be in order. But as several comments noted, presenting slow streetcars as an alternative to rapid transit, as Condon does, suggests a belief that we can be forced to make short trips even though the land use patterns of our city (and, perhaps, the different commute destinations of different members of a family) require us to make long ones.
So I'll end with one of the most big-picture comments, one that I forwarded to Professor Condon and to which I hope he'll respond. It's from reader "micasa":