« streetcars: an inconvenient truth | Main | learning, again, from las vegas »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Chris Smith

I have the great good fortune to live in NW Portland and can access much of the downtown by either streetcar or bus. I choose one mode or the other based on a mixture of parameters:

1) Which comes closest to my destination (often a toss up)
2) Which vehicle is arriving next

But only the streetcar line is sprouting tall buildings and adding new destinations to my "20 minute neighborhood". The investment in streetcar improved my access but did NOT reduce my overall mobility, because I have many choices.

I wouldn't recommend that streetcar replace a bus where that bus was the only transit service. But in an environment with rich transit choices like Portland or SF, adding streetcar to the mix of choices has definite benefits.

Jarrett at HumanTransit.org

Chris.  No argument there.  But the given the high capital cost of streetcars, you will need answers to these questions:
1.  What features of the streetcar generated that development outcome?2.  Are there cheaper tire-based technologies that would achieve the same outcome?
3.  In what direction does this calculus seem to be trending?
This is matter for another post, but fundamentally, development outcomes are ultimately based on what developers think they can sell, and that depends on what current attitudes are about streetcars and buses.  In fact, given how long development takes, they're based on the attitudes toward streetcars and buses about 10 years ago.

So that goes to my other point (which I'll expand on in another post) that buses are improving fast and improving specifically in the direction of rail, which is at least a reason to wonder whether the public perception of the bus's inferiority is going to continue.  


That's not quite what you should mean by mobility, because just plain distance covered is not always a very useful measure. A slightly better measure would be something like WalkScore integrated over the area, to account for the density of places you'd actually want to go. And ultimately it comes down to various somewhat fuzzy definitions, and you start getting arguments like "oh well transit means denser development which means more destinations available with less travel" and so on. Still, if that's the case, then planners need to take a good hard look at Portland's system and figure out how much redevelopment potential there in the existing city versus an empty railyard on the edge of downtown.

Jarrett at HumanTransit.org

Anonymouse. Good clarification. I agree that purely measuring the area of the blobs on the map would not be a measure of mobility, but I don't think I said that exactly. I said "how many places can you get to in a fixed amount of time?" Any reduction of this map's message to a single number should of course be weighted by density of "places" at each possible destination.

Meanwhile, any citizen who can read maps can get a good sense of her mobility using this tool, and if she knows her city, the blobs on the map will convey her real mobility to her in a way that no single measure would do.

But by either definition, a zero increase in mobility is still zero.

Chris Smith

What features of the streetcar generated that development outcome?

Fundamentally when you put rails in the ground, a developer KNOWS you will be providing transit service there for a long time. Any rubber-tired alternative is by definition relocatable and therefore much less interesting to private capital.

Streetcar may look expensive, but a $100M investment in Portland was matched by $3B in private development. Name one other public sector investment that gets a 30-1 match in private investment?

Michael Druker

Say it takes me 20 minutes of walking to get from home to a store. Is my mobility hampered by it being cold and rainy? Well, I'm less likely to make the trip, and if I do, I'm less likely to enjoy it.

I would take a high-speed train over an airplane, even if it took a fair amount longer -- because it has no turbulence, causes no ear pain, and has more leg room and space to walk around.

It is not that distance / speed doesn't matter, it's that the conditions of a trip -- comfort of vehicle, pleasant or unpleasant interactions, real or perceived uncertainties, and so on -- not only change the experience, but (in addition to speed and cost) help determine how people will decide whether or how to make a trip.

That's why I take issue with your definition of mobility. Yes, speed and reach is important, but mobility should ultimately be concerned not with what trips people are theoretically able to make, but with those they are willing to make.

Jarrett at HumanTransit.org

Not saying those experiential values aren't important, but just suggesting that they're logically separable from mobility, and that the issues deserve separate dicussion.

Michael Druker

I suggest that you need to consider what the general usage of word "mobility" is, and why you think it is appropriate to apply it to the specific notion in your post.

Of course these issues can be logically separated. My concern is with the effects of language on the discussion.

Jarrett at HumanTransit.org

I'll take this up in another post, but briefly, why do the rails guarantee permanence? I can show you lots of abandoned railroad tracks in a lot of streets in a lot of cities. If you want permanence, you should look for great transit markets, not rails in the street.

Hawthorne Blvd.'s transit service (currently provided by frequent buses) is permanent, because the land use pattern and the available street add up to an excellent market that will always generate lots of riders. Certainly, you could make that permanence more VISIBLE to more people by upgrading the stops to something more architecturally substantial. But the real permanence is determined by the durability of the service market itself.

Look, Portland's current development culture, responding to prevailing attitudes of the moment (or the recent past) has decided that rails mean permanence, and has to talk up that view because, after all, they have investments riding on it. I'm not going to convince somebody who's already made that commitment; frankly, I'm not trying to.

I'm just suggesting that the mobility offered by a transit service is an independently assessable feature that some people might want to decide if they care about. Because streetcars are really unusual among our transit investments in offering absolute no improvement in mobility.

Cheers, Jarrett

Chris Smith

If you take any FTA money, you're committed to 20 years of service!

Jarrett at HumanTransit.org

I think the general usage of the word mobility in these contexts is pretty close to how I use it.  See http://www.merriam-webster.com/dictionary/mobility meaning #1:  "capable of moving".  I'm talking about how much moving a citizen can do, given that we always have some constraint on time.  A discussion within a particular topic will always sharpen definitions a bit, but I'm not inventing a new idea here.

Jarrett at HumanTransit.org

Yes, you are. I'm not sure everyone is going to agree that FTA policy is a sufficient basis for belief in permanence compared to underlying urban structure, especially since real estate investments need to hold value for more than 20 years. But again, I'm just suggesting some questions people might want to think about. If at the end of the day Portland really wants streetcars, you should have streetcars.


Sure, mobility and passenger experience are separable, but why would you want to consider them separately?

See, I just invented an awesome form of public transit. It's called the trebuchet. Just climb in the basket, have the operator point it toward your destination, and you will be flung quickly and efficiently wherever you want to go. Imagine the mobility a network of trebuchets would provide! And for much less money than traditional transit like subways or light rail.

Of course, nobody would use this system. But if mobility is just about moving people around as quickly as possible, it's fantastic.


Jarrett, I have to back you up here.

I do believe that perception matters, and that the subjective experience of travel is an important part of the total experience. I don't enjoy traveling on subway cars with smelly people, or loud people, for instance.

But the salient point that I believe you are highlighting with the exercise about the size of the 45-minute blot is that government decisions should be made based on quantitative criteria, not qualitative criteria. This is not Sim City and our public servants are responsible for spending our tax dollars in a prudent way. If public servants choose streetcars over buses and therefore make fewer mobility improvements, aren't they wasting our money?

Chris Smith

I think the important factor is that the developers who plunked down the $3B appear to believe there will be continued service (I know, I serve on the board of Portland Streetcar, Inc. with them).


I think rails do provide a sense of permanence, in that while they may be abandoned eventually, it's usually not a quick process, while a bus can be rerouted literally overnight. Rails are part of the capital infrastructure of the city, while bus routes are largely considered to be an operational detail of the transit agency. Rails also tend to guarantee some level of service: the transit agency built the expensive infrastructure, they might as well use it as much as possible, so with a streetcar line, you're guaranteed some reasonably frequent (20 minutes or less, usually) service 7 days a week. I guess this would be what the economists refer to as signalling: by spending all that money on rails and such, the transit agency is making an implicit promise that they are here to stay and that they're committed to providing a certain standard of service.


The "buses can be rerouted overnight" argument is a weak one when you look at agency practice.

In L.A., Metro gives the public a four-month notice between announcement of service changes and the changes themselves. The planning to these changes goes months before the announcement, so the service change can take about a year from conception to execution.

FTA and states require public notice and comment before all changes are made. So this isn't an instant process.

This is also true regardless of systems: If there's one thing passengers hate more than the way things are now, they hate change even more. Public hearings were intended for the riders to tell the agency what to do; most of the time, they tell their planners what not to do.

Also, if you have a system that needs constant route changes, you either have bad planning or bad planners.

Maybe a better sense of the word is "anchoring" rather than rerouting. There can be good reasons for buses to be rerouted, such as for planned disruptions (such as parades) or construction.

Anchoring means the service, such as a streetcar, can not be easily physically altered once in place.


Um it matters. But it doesn't *always* matter. *Not all* major investments need to improve mobility. Some can improve comfort, or reliability, or operating expenses, or user-friendliness, and *that's OK*.


Many of those abandonments of rail service took decades. Literally decades of unprofitability, during which the operators often deliberately sabotaged service quality in order to reduce the market.

Bustitution has happened more quickly than that, to be fair. With its attendant ridership losses.


Trouble is, although these are reasonable questions to ask, the answer to (2) is usually "No". If passenger volume is high, the tire-based techologies are more expensive when considering all costs. The streetcar improvements are designed to increase passenger volume due to their promotion of development.... if you had a bus line with the same development potential, it would probably end up being more expensive once the riders showed up.

Curitiba is planning to build rail lines because of high volume on its bus lines.

Unfortunately, precisely *none* of the improvements buses are receiving in current improvements give them the volume advantages of rail (bigger vehicles, longer trains, higher frequencies). If you're building a streetcar line for *potential expansion to high volumes*, you simply can't do the same with an otherwise comparable bus line. You can with a streetcar line: even if you're *currently* running it with short, uncoupled vehicles, you can upgrade it later without a total rebuild.

Yes, I know articulated buses exist and are being developed, but my God, the ride quality is *appalling*, and the lack of tracks means it *cannot* be improved -- it's fundamental to the design that the back end sways. Apparently Londoners, one of the few cities where people actually like the buses about as well as the trains, detest the "bendy buses".

On the other hand, for low-volume situations, streetcars are simply ridiculous and buses should be used -- because the capital expense can't be justified, and the operating expense advantages tilt in the bus direction. Just so you know I'm not a total train fanatic.


To put it another way, yes, expanding that bubble on the map matters -- but it is *totally unreasonable* to expect that every major transit investment will expand that bubble.

For a really good example, some of the most expensive transit investments ever, clearly good ideas, did not expand the "reachable within X minutes" bubble.

Some of these very expensive transit investments expanded capacity, and nothing else. (Grand Central Terminal. Yes, this allowed for reduced trip times in *future* investments, but the same is true of streetcars.) Others increased safety, while sometimes actually increasing trip times (certain types of signalling improvements and grade separations).


Where I come from each year's budget includes cuts to the bus service. Shortened routes, reduced service frequency, reduced hours, elimination of routes. Or at least the county executive does that and the county board reverses the most contentious bits of it. I'm sure if there were tracks laid down service cuts would be proposed for that, too, but maybe people would hope that would be the last rather than first thing cut. And one suspects some of the anti-rail feeling in town from the anti-all-transit people is because that that would be the hardest thing cut, after all the investment.


Jonathan, I have to disagree. I think qualitative criteria should be AT LEAST the equal to quantitative.

If someone has no choice but to use public transit (no car, physically unable to drive, etc.) then they will use public transit, be it bus, streetcar, light rail or subway. However, if a person does have a choice (they have a car) then the quality of the trip is most likely a major factor in mode choice. They may say "I'm not taking a bus, it's noisy and uncomfortable (vibrations, jerky stops and starts). But I will take a streetcar, it's quieter, has a smoother ride and I know where it's going (just follow the tracks)". If using strictly quantitative measures means running buses that few people use, isn't that wasting money as well?

Does that mean every transit route should be a streetcar? Of course not. A well balanced transportation system should be a mixture of modes, both buses and rail. But just because a streetcar doesn't appear to increase mobility doesn't mean it's not worthwhile and shouldn't be built.

Even if mobility is not increased by installing a streetcar that doesn't mean transit usage hasn't increased as the "I don't ride the bus" people start using the streetcar.

In the US, people who won't use a bus will use a train. That has nothing to do with quantity and everything to do with (perceived) quality. And no matter what the technical improvements over the last 20 years and over the next 20 years a bus is (and will continue to be) a bus. As long as it runs on rubber tires the ride will still be noisier, bumpier and shakier. That is physics and cannot be changed.

jack lecou

The "buses can be rerouted overnight" argument is a weak one when you look at agency practice.

Clearly a re-routing decision that can take place within a year or two IS practically overnight as far as developers, home buyers, etc., are concerned. And as compared to rail.

And people clearly do NOT look at agency practice, or their internal operations to make their judgements. While it's true that useful, well-designed bus routes are well nigh permanent, they will never be truly perceived as such, and thus simply do not send the same signals. That is something that has to be reckoned with.

jack lecou

To get back to the original argument, I'd second what Michael Druker and others have said about your measure of "mobility". Sure you can separate that from quality, but why?

If you decide that a bus (or maybe a fleet of cheap flat bed trucks) will provide the same "mobility" in terms of destinations available, that's fine. But then you still have to take the next step and ask what service people are more likely to use. Even if you define it out of "mobility", quality still matters. A lot.

And then, even in terms of "mobility", it's far from clear that street cars don't improve it on appropriate routes. Increased development/density along a route = more destinations in a given radius, and thus greater "mobility" by your measure.

And while you mention Hawthorne as an example of bus transit service providing a lot of development and density benefits, that's only half the argument. You also have to show that a street car on that route wouldn't produce even MORE.

jack lecou

Also, this,

I said several times that this wasn't the only criterion for transit investments, but if you're going to call something a transit project, you might expect someone to ask if it improves transit mobility.

...doesn't seem right. With "mobility" so narrowly defined, I would certainly NOT expect anyone to ask that. In most cases, certainly in Portland, viable street car routes are going to be replacing or augmenting an existing bus network. It's not the changes to the list of places people can get to that's important, it's the changes to HOW they get there.

Not only is improvement in "mobility" not the only criterion you should use, it's not a very relevant criterion at all.


Have you ridden the 14, Jarrett? It's crowded, noisy, uncomfortable, easily stacks up and is out of scale to development. Clearly there needs to be an increase in scale and capacity for the line, but buses can only do so much before maintenance and staffing costs become prohibitive.

And mobility is simply a function of scale, something addressed by focused and maintained development and an aggressive attempt at managing such scale. You see it in highways, where the desired scale was outstretched veins of development going out into the suburbs. Clearly in that case, mobility required both speed and few delays. For Portland's purposes, the scale they desire for Streetcar corridors requires focused short-trip generation and clustered development. A bus could certainly address this, but only on the side of conveyance with no sense of place. A streetcar, on the other hand, is the tree-lined lane of old, or the cobblestone heritage "village" portion of an old city... it gives a sense of place and offers additional features that fit the scale of intended development, like easily identified stations and a signature look and presence (the "screech").

There is no mistaking the purpose of the Portland Streetcar in particular, since it is being tracked mere blocks away by MAX. MAX serves the greater purpose of long-distance mobility and regional scale, but the Streetcar creates place-making and local circulation. Many of the corridors created for the new streetcar network have a component high capacity system nearby or one in planning, one reason they were selected as top priority lines.

The mobility argument is not applicable.


For the Portland Streetcar, however, that is irrelevant based on the frequency offered. Replacing that corridor with buses would not oversaturate the street, and arguably would provide a higher level of service by adding more frequency, despite the reduction in capacity in the vehicle. If a streetcar operated every 10 minutes and the bus operated an average 7 minutes, the bus is providing better service because of a reduction in average wait time. The other thing is that you can add service on parallel corridors a quarter mile or half mile apart, if you are worried about dumping too many people on a street. Perhaps by throwing all the resources on one street, you are neglecting surrounding areas.

Now obviously corridors like Eastside MAX where a two car train is arriving at a station every 5-6 minutes during rush hour cannot be expanded to bus (although it should be noted that Portland reached capacity for its light rail through downtown and has to build another light rail through downtown, and that it's still an open question as to whether an additional rail bridge will be needed to supplement the Steel Bridge).


But that's where the math needs to be done, and is sorely lacking. If, for the $25 million a mile we are going to spend to lay down track, we can provide a bus that runs twice as frequently? Perhaps we could use that money to subsidize bus fares along the transit corridor for X number of years, which could drive up ridership just as much, or spend it on architecturally permanent transit stations (like Cleveland did with the Health Line, or Kansas City did with the MAX), or whatever. Sure, streetcar will attract people who wouldn't take buses, but maybe giving the bus its own lane would do the same thing, or attract those who wouldn't take a streetcar because it is too slow, or feels too slow.

Michael Druker

"If a streetcar operated every 10 minutes and the bus operated an average 7 minutes, the bus is providing better service because of a reduction in average wait time."

No it is not.

(With what buses currently are like, and with what streetcars currently are like, very few people on the street would agree with your assessment.)


If a person won't take a bus because they don't like the EXPERIENCE of riding the bus, giving the bus it's own lane isn't going to entice them.

I've commuted by bus. I didn't particularly LIKE the experience; I found the bus extremely noisy, the stops and starts were jerky, the whole bus shook all the time and hitting potholes could sometimes nearly knock you out of your seat. But I chose to ride it. But most people who have the luxury of choosing will choose not to ride the bus.

Again, I'm not saying replace every bus line with a streetcar. I'm not saying replace the majority of bus lines with streetcars. But there are bus routes where the $25 million/mile to replace them with streetcars is worthwhile, even if it doesn't "increase mobility".

Yes, the math needs to be done, but there are things that cannot be quantified with dollars and cents measurements that need to be taken into account as well. We all do it every day. Why do people buy the brand name of something instead of the generic? Because they PERCEIVE that the name brand is worth the extra money. Of course, not everyone does, but they are in the minority. Go to anyone's kitchen pantry or freezer and see how many brand items you find versus generic/store brands.

Literally TRILLIONS of dollars are spent in our economy every year and millions of jobs depend on the fact that people will spend more money on what they PERCEIVE to be a better product. So why would you think transit would be exempt from that?

Americans PERCEIVE that trains are superior to buses. If their only option is a bus running on rubber tires on cement that opinion will probably not change. But, if you can get them on the streetcar or light rail, they might eventually say "You know, there's a bus stop a block from my house that will take me to the rail station and the ride's only 12 blocks, maybe I'll start taking that instead of driving my car to the station".

Michael Druker

There's another very simple way that mobility is necessarily tied with factors of experience. If a streetcar attracts more transit ridership than a bus, those additional riders increase demand for transit improvements in the entire system, which does include getting to more places and faster.


But you miss the point. YOU may choose not to ride the bus ever in your life, but there are doubtless many people who would drop their objections to riding the bus if the fare was reduced, or frequency increased, or technological improvements like next bus displays were implemented. All of these are cheaper than $25 million a mile, and the cost per new passenger of these improvements are much less than that of streetcars. In other words, we need to take the low hanging fruit - all buses must be GPS tracked, with next bus information available over the web and via cell phone, and at busy stop locations; a "high frequency" network of service needs to be developed; different classes of service need to be implemented on a street (including express and limited stop service), and fares should be made reasonable. Incidentally, Portland has done all except the latter, and their fares are slightly above average.


So people are going to wait longer for the streetcar rather than take the first bus that comes? that's not true, as you can see along market Street. If one is going to ride surface transit (and not Muni Metro down below), aside from tourists, most people take the first vehicle that comes along, that goes to their destination. This was easy to see when the F line didn't go to the Embarcadero, since the route of the F to Transbay Terminal was essentially the same as many of the local routes that use Market Street toward downtown.

Jarrett at HumanTransit.org
Capacity is very much part of mobility.  If you can't fit on the transit vehicle, then for you the service doesn't really exist.  I've acknowledged that at very low headways the capacity benefit is important.  But if we're at Portland's prevailing frequency, say every 10-15 minutes, we should be setting frequency by policy based on service experience we want to create.  A streetcar every 15 minutes may offer the same total capacity as a bus every 10 minutes, but 10-minute service is better mobility than 15-minute service, because waiting time is such a large issue.
I completely agree that station investments matter, though these usually generate some capacity improvement and/or improvements to pedestrian flow.  I'm not arguing for the abolition of aesthetic values.  I'm just pointing out that there's such a thing as mobility, which is what most transit is about, and streetcars don't seem to improve it much.
Jarrett at HumanTransit.org

Yes, I've ridden the 14 (Portland's Hawthorne Blvd. bus line) many times. Crowded? Yes, but of course that's the Yogi Berra fallacy: "Nobody goes there because it's too crowded." I think crowding usually means high ridership, no?

Again, if noise is your issue, consider an electric trolleybus. Meanwhile, there's also the issue of vibration, which cuts the other way. Tires cushion the vibrations arising from vehicle, while steel wheels transmit it to the street and the surrounding buildings.

Jarrett at HumanTransit.org

I think the thread of responses to this comment captures the fact that qualitative values are just not measurable against quantitative values, which means that we each have to weigh them for ourselves. Unfortunately, quantitative values often require is to think a little, or learn something, while qualitative values are instinctively evaluable by anyone.

For example, I can explain the concept of frequency to a non-transit rider in less than a minute, but it still requires an explanation, and motorists don't really get it until I compare it to the cycling of traffic signals. I think that's part of why frequencies are so often below what the market could really support, and why we get more talk about the quality of experience than of its quantifiable aspects. I talk a lot about travel time because it's an output of transit that a motorists can understand.

I wonder how much technology-first thinking is simply an expression of the fact that quantifiable features of transit sometimes require motorists to think a little. Transit would be so much simpler if our only choice was what kind of vehicle we want -- just like it is for a motorist.

Jarrett at HumanTransit.org

Thanks to everyone for the comments to date. Feel free to continue. The discussion has really helped me sharpen my own thinking about these issues, as I hope you'll see in some of my responses. I just arrived back in Portland for a week with family, so I'll be exploring these issues again and will return to them in a few days. Cheers, Jarrett


A streetcar every 15 minutes may offer the same total capacity as a bus every 10 minutes, but 10-minute service is better mobility than 15-minute service, because waiting time is such a large issue.

In L.A. and many other big cities, transit systems often have the equipment and resources to offer 10-minute or better bus service.

They just can't deliver on the schedules.

In one example from L.A., I am a frequent user of Line 16. During rush hours, the schedules promise headways of 2-3 minutes. In practice, I know that I can expect to wait 10 minutes during rush hour, and a pack of three bunched buses shows up at once. This shows Metro has the equipment to provide buses 3 minutes apart, but they end up bunching and delivering something approaching midday service.

Jarrett, as a professional, you have more insight here than anyone. What do riders prefer in surveys and after the fact? A higher promised frequency or a lower frequency but higher reliability service?

Jarrett at HumanTransit.org


I've never seen ridership respond to a frequency that was promised but not delivered. Reliability always matters. And in mixed traffic, reliability is really hard. If your boarding time is governed by fare transactions -- as is still the case on most buses -- it's even harder.

Unfortunately, reliability is also hard to measure. I wrote more about this, and the causes of it, here:


One key improvement needed is for operations to learn to manage the headway rather than the on-time performance of each vehicle. At high frequencies it's the actual headway that matters, not whether each vehicle is on schedule. Most rail operations managers are trained to think this way, but relatively few bus operations managers are, so this becomes one of those bus-rail differences that affects real mobility but is not intrinsic to the technologies -- it's just cultural and historic.

I hope someone is doing some rigorous research using freely available nextbus.com data to better understand patterns of bunching on both bus and rail. Anyone can query nextbus outputs in realtime and form a database of actual reliability of various services.

San Francisco's Muni Metro is an especially interesting example because every line has some surface operations with local stops in mixed traffic (streetcars, essentially) as well as long segments in subway. It should be MORE reliable than an all-surface streetcar, but individual lines still seem to routinely show large gaps and bunches (expressed on Nextbus as something like: Next car, 22 minutes. Following car 28 minutes.)

Now Muni Metro's performance has improved noticeable since the "Muni Meltdown" days of the mid-90s. But anyone who believes streetcars will remain reliable as loads and traffic increase will need to explain why individual Muni Metro lines -- each functioning like a streetcar on its unique surface segment -- seem to be bunched as often as not.


I hope someone is doing some rigorous research using freely available nextbus.com data to better understand patterns of bunching on both bus and rail. Anyone can query nextbus outputs in realtime and form a database of actual reliability of various services.

I've been observing this for years as a rider, many times on the short end of an unreliable line.

I've staked out a crude theory I call Wad's Law.

Reliability is inversely proportional to frequency.

I've tested my hypothesis on many systems, heard observations from cities I haven't visited, and have seen it broadly.

I think, once you have to run service more than every 15 minutes, services becomes gradually unreliable with every increase in frequency.

What's even weird is, this scales!

It even applies to small systems that may have one 30 minute route and several 60 minute services.

And this was on systems generally regarded as well-managed and well-operated.

I use L.A.'s Metro because its "petri dish" has several "specimens" to examine.

The Line 16 example is typical of L.A.'s many very-high-frequency bus lines.

The problem is neither one of poor management nor of poor worker quality.

The problem is really the chaos of the operating environment, and has some foundation in chaos theory. Passenger transportation has its own "butterfly effect."

Busier services tend to draw more butterflies, in a matter of speaking. More people have to board, and by definition, more people have to exit. More people board at more stops, with the same result. Longer routes have more stoplights. People board at different speeds. People can only board through the front door, and most buses are still designed with back doors that only allow one person to exit.

Then, there's a psychological effect of bus driver behavior. The higher the frequency is, the less a driver will try to maintain the headway since the lost time to the passenger is very low. Few complain about waiting 10 minutes for a theoretical 3-minute bus because the 33% on-time percentage does not affect them as much as a 7-minute loss.

This is stuff that can be controlled, but there still leaves a lot of chaos.


"A streetcar every 15 minutes may offer the same total capacity as a bus every 10 minutes, but 10-minute service is better mobility than 15-minute service, because waiting time is such a large issue."

If you were to argue that streetcars should not normally be used until peak frequency reached a critical level, unless there was an opportunity for off-street/segregated running (there often is), I would probably agree. It seems that from a mobility point of view, if you want to increase route capacity, increasing frequency should be done first, before increasing per-vehicle capacity. (The question of whether to increase route capacity or increase the number of available routes is a thornier and more location-specific one.)

Portland Streetcar is a *weird* example. But then it was an economic development tool, primarily; so signalling that the streetcar would not go away probably mattered more than anything else. The future incentives are now to make use of the existing "sunk costs" track by developing in such a way that increased service can be supported; that would not be the incentive for a frequent bus service.

Jarrett at HumanTransit.org

Portland Streetcar may be weird, but it's also the most frenetically marketed model in the urbanist world these days. Good comment.


Coming from Toronto, one of the few North American cities which still uses streetcars, I'd like to point out that HOW the mode of transit is used is just as important - if not more so than what is used.

Some here talk about how much nicer streetcars are compared to a bus, but Toronto's tram fleet dates back to the 70s and early 80s. While they have held up, many newer buses are much nicer. And this is not including bus rapid transit or commuter buses.

Also worth noting is how the mode of transit is operated. The Toronto region has several lines using several modes that serve several purposes. The Queen St streetcar operates like a bus in mixed traffic with frequent stops, however there is a portion of the line in its own right of way with few stops. Likewise the north-south subway lines operate at high speeds with stops every 0.5 to 1.25 miles until reaching the downtown. However the west-east subway line has a generally high stop spacing outside of downtown, with stops under a half mile outside of downtown and for much of its run. Though it is used for long haul trips from suburbia into downtown, it is best suited and most competitive for local trips - almost more like a LRT than full rapid transit.

Does this mean streetcars are a waste of money? Not exactly. While operation plays a bigger part than many wish to admit, mode still is important. Two clear advantages they have include capacity and acceleration. Most modern streetcars can carry far more passengers than most buses. Also since they are powered by electricity rather than fossil fuels, they tend to accelerate far quicker than buses. These features make streetcars far more ideal for high density areas and nearby areas to spur growth. Beyond this, buses are better.

Jarrett at HumanTransit.org

Unquestionably, if it turned out that either capacity or acceleration was the decisive advantage with streetcars in a particular situation, that would be a reason to build one.  Note, though, that acceleration benefits related to propulsion are independent of the rail vs tire question, because electric trolleybuses are also available (and battery-powered electric buses are a topic of intense research, just as with cars).  

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name and email address are required. Email address will not be displayed with the comment.)

the firm

Jarrett is now in ...

Related Posts Plugin for WordPress, Blogger...
Related Posts Plugin for WordPress, Blogger...