King County Metro has a very specific function.  Their voter base is spread throughout the county, and it is their job to provide the highest quality transit service to the most of these citizens that they can.  If you were in charge of the county and you started from scratch, with no buses and just a map and some data about the county, how would you design transit?  First, you’d look at where most people lived and worked, since you know that commuting is a very strong transit need.  You’d see that the largest job center was downtown Seattle, with Bellevue and Redmond as secondary job centers.  And you’d see that a third of your citizens live in Seattle, but two thirds live elsewhere in the county.  Then you’d consider your transit options.

Appropriate use of transit technologies (Matt Gangemi)

More after the jump.

This is a chart I made comparing the appropriateness of transit technologies.  It’s not based on data or studies, it’s my interpretation and you’re welcome to disagree in the comments.  I’m defining a technology as more appropriate if it reduces your travel time and is convenient.  For example, high speed rail will move you long distances quickly but would be infrequent, so wouldn’t be the best way to get from Ballard to downtown even if it ran between these two points.

 Now, back to the county.  Knowing most of your commuters are travelling more than a few miles, you’d be looking at light rail, heavy rail, frequent* buses, and scheduled buses.  The buses aren’t as good as rail, and traffic is a real problem, but you might look at your budget and just buy buses.  You’d get cheap scheduled buses most everywhere, and let them become frequent buses as lines merge.  You’d also try to get road agencies to give you bus-only lanes to help solve the traffic problem.  

And that would be a fine choice to make, overall.  Metro is acting rationally.  But the current system is only rational if you’re looking at the county level, with county average travel times and county average density.  This ignores the fact that there’s a dense city inside King County.  Take another look at the chart above.  When you get below 5 miles, a scheduled bus stops making sense.   The time and inconvenience spent is large compared to the distance travelled.  And as you get down toward 1/2 mile, you might as well walk.  Yes, multiple scheduled bus lines running the same path become frequent buses, but keep in mind that “Traffic” column – unless we run buses in their own right-of-way with real signal priority, even frequent buses are less appropriate than most other technologies at short distances.

 There’s a reason that most counties have just one form of transit and most cities have many forms.  I’ll be exploring these reasons and consider how Seattle would best be served by transit in future posts.

* Note: I’m considering “frequent” to be a bus that comes at least every 10 minutes – frequently enough that you don’t need to time your trip, just show up at a stop.

Update (1:30pm):

Here’s the updated, crowd-sourced version:

Appropriate use of transit technologies (Matt Gangemi)

115 Replies to “The Right Tool for the Job: King County”

  1. I can’t get a bigger view of the chart, but I like it. I would be interested in seeing “bus” broken-down further though, or exploring different busing on its own: dial-a-ride, dart, local, rapid-ride, express, etc. the way you have for rail.

  2. @Matt: You missed something on your chart that is particularly relevant to Seattle – water travel. No mention of ferries – good for high frequency, short distance (<10mi) travel. Water doesn't have hills like land does, and boats don't get stuck on the freeway.

    1. Yeah, I thought about ferries, but they’re hard to compare to other technologies. There’s rarely a question of: Should we build a ferry or a 4-mile cross-sound bridge to run a bus? There are many good comparisons, but in the end it doesn’t quite fit.

      1. Practically speaking, that’s true; however, your scenario wipes the map clean to begin with – you have the chance to answer that question however you choose. Just a thought.

  3. Yeah I definitely disagree with some of your chart. You think that bicycling is starts becoming inapprpriate at 2 miles? Yet heavy rail is appropriate at a 025 miles? I would shift the bike so 5 miles is green, 10 miles is yellow and 50 miles is red. I would also shift the light rail, heavy rail and frequent bus to match closer to scheduled bus. I’m much more likely to walk somewhere that is a 1/2 mile away than take a bus or train.

    My commute from Fremont to Downtown Seattle is about 3.75 miles. When the weather is bad, I take one of the frequent buses that comes by (the 26, 28, 5, or 16). When I have the free time and the weather is nice I’ll walk. Sometimes I’ll bike, but the 3 to 4 miles is hardly worth getting sweaty on a bike (or wearing special biking clothing), so I don’t do it as often.

    1. Let’s assume you bike around 10mph, which is probably high considering stoplights, etc. 5 miles then means 30 minutes, plus time to unlock your bike, clip on your bike bags, etc. Without traffic, a bus would probably beat you to your destination if it were frequent enough that you didn’t have to wait long. Same with light/heavy rail. Perhaps I should downgrade a scheduled bus though – I bet you’d beat it with a bike on average.

      Scheduled bus definately doesn’t match even a frequent bus at the lower range. You want to be able to hop on and off, or you’re better off walking.

      1. Even if it took me a bit longer by bike than by bus, I would prefer to bike because 1) I would get my exercise for the day while also taking care of my commute & 2) I wouldn’t have to pay bus fare. In my opinion, the sweet spot for a bike commute is 5 to 10 miles. Shorter than that and it’s hardly worth the trouble of bringing a change of clothes, showering when you get there, etc. Longer than that and it starts to take too long.

        For a commute under 5 miles, I would prefer to walk when I have enough free time (or maybe run?) for the same two reasons above. In both cases, I see bus as a backup for when I’m running late or the weather sucks.

      2. Dear Seattle: enough with the showers and changes of clothes!

        1-4 miles is the sweet spot “utility cycling.” Not just in Amsterdam, but in the dozens of cities where bike-share programs have dramatically altered the urban landscape in just the last five years.

        Cycle-share bikes have only two gears: low and lower. You can not get sweaty on the things. You also can’t go especially fast, so beyond 3 or 4 miles you’re going to want some real mass transit to get you there faster.

        Even better, the sturdy, low-center-of-gravity bike-share equipment can’t dart out into traffic or pull the kinds of dangerous stunts that make the road-warrior cyclists so irritating to the general public, which makes them extremely safe.

        But good luck ever getting them installed in Seattle. The Lycra cavalry has done a fantastic job of convincing the general public to support draconian helmet laws, under which neither cycle-sharing nor mass utility riding can really work. Seattle’s loss.

      3. Matt Gangemi, I would strongly encourage you add a modifier to the chart — “utility bicycle” or “basic bicycle” or something — to help distinguish between trips for which biking becomes a reasonable option for a large number of people and trips for which it is relevant only to the zealots.

      4. Done. For that I ran some simple calcs, assuming:
        10 minutes required per trip for “bike clothes” biking, average speed 15mph.
        1 minute required per trip for “work clothes” biking, average speed 8mph.

        The balance point between the two is a bit over 2.5 miles.

      5. Perfect …and I fully concur with your results: great for up to 4 or 5 miles, doable but “not ideal” for 5-10 miles.

        Biking less than 1 mile seems unlikely in the absence of a really great bike-share system (offering a bicycle essentially on-demand anywhere in the urban center). But they are certainly feasible at that distance.

      6. Matt – That completely depends on whether you can take the bus without transferring. As soon as there is no bus that runs straight to your destination, a bicycle becomes the must faster option.

    2. I’d put biking at:
      – Green: 1-3 miles (~10-22 minutes total) — a bit too far to quickly walk, but not far enough that I’d want a non-cruiser bike, saddlebags (instead of a shoulder-bag/backpack), and most can bike a mile without getting particularly sweaty and wanting to shower/change). Speed of a car or bus won’t make up for finding a parking spot or waiting for the bus. Depending on where you are, and how fast you ride, maybe you don’t worry about a helmet.
      – Yellow: 3-8 miles (~22-40 minutes total) — At 3 miles or so, available transit will start to take over. Further than that, and you’ve chosen a flat/down-hill, bike-friendly route, and maybe you’ve got a faster bike/saddlebags/etc. Probably filled up a water-bottle. In the cold, you’ve got gloves+a hat, when it’s warm you’re going to get sweaty, which may or may not be a problem, depending on where you’re going.
      – Red: 8+ miles — You’re exercising. Good for you, but you’re going to arrive at work later & sweatier than you would’ve if you’d just hopped on almost any bus.

      1. Bike shares are awesome. Bike shares got me back on bicycles for the first time in a decade.

        Bike shares are never going to happen in Seattle, largely thanks to psychotic road warriors like yourself and your opposition to utility-cycling-friendly infrastructure and the relaxing of helmet laws.

        But bike-sharing supplements walkable transit. It can help with the “last-mile problem” in certain de-centralized situations. It cannot replace core transit that is largely accessible by foot.

        Transit designed to require biking as part of every trip is a non-starter, so please shut up about it.

      2. Bike shares can help speed up Bus transit because stops are now 800 meters apart. If you are within 400 meters, you walk, if further, you bike.

        Bus systems which stop every 400 meters are slow. Doesn’t matter how many buses you have lined up behind each other. It’s station dwell time that kills mass transit and a local bus line is the perfect example.

        It’s an extremely “walkable” system using each mode to it’s best facility.

      3. Hey, you know what else is slow?

        People fiddling with the bus rack for 45 seconds after everyone else has already boarded. And then acting all superior because they “bike everywhere.”‘

        Bike share is awesome.
        And we’ll never have it in Seattle with people like you representing our city’s dominant bicycle culture.

      4. Transit designed to require bus riding as part of every trip is a non-starter, so please shut up about it.

      5. False equivalency.

        Nobody is advocating for transit that would require using a bus for inappropriate distances.

        You, on the other hand, have repeatedly suggested (in this and prior threads) that mass transit should be so widely spaced that a bicycle (rather than feet) would be required to access it.

      6. “mass transit should be so widely spaced that a bicycle (rather than feet) would be required to access it.”

        Since you are willing to walk 10 miles, that would be quite the system… Get on, then get off, walk to destination.

        That’s not what I said. Bike share with stations every 800Meters, it’s a mere 1/2 mile, less than 8 minutes of walking. It’s 4 city blocks instead of two.

    3. Your biking stuff is way off base. My kid bikes from Ballard to downtown Seattle, a distance of 8 miles. It’s flat, and it takes exactly the same amount of time as the bus.

      I ride from Bellevue to SLU, a distance of 16 miles. It takes 15 minutes longer than the bus, IF I drive to a Park and Ride first. However it’s not flat so I arrive needing a shower, which fortunately my employer provides. I used to ride from Bellevue to Pioneer Sq, which takes exactly the same time as the bus. If I take the neighborhood bus, then transfer at the P&R, then catch the SLUT from Westlake it takes LONGER than bicycling by 5 minutes, assuming the connections work reasonably and I didn’t “just miss” that next bus.

      But what’s really key is that if you had a “bike share” program, you could ride the last mile from your bus stop to your destination and eliminate those time killing bus stops every other block. There is a city in China, which included with your bus fare is a 30 minute bike share ride.

      1. Oh my average “no sweat” speed biking is 12mph, including stop signs & lights. Light workout at 15mph, working pretty hard at 17mph.

      2. Bicycling also depends a lot on topography in ways that other modes (even walking) don’t. My “sweat threshold” or whatever is totally different depending on direction. I live at the top of Cap Hill and sometimes commute to the U and sometimes downtown…And I can’t imagine getting sweaty going downhill in the morning.

        This really is more relevant, though, in terms of how we plan facilities and not that much in terms of time. Downhill I go fast without much effort and don’t really need a bike lane, but uphill it’s hard to go faster than traffic (although sometimes at rush hour I smugly ride uphill past traffic jams). The hills and available facilities totally affect everyone’s total decisions about appropriate mode. Riding around through downtown traffic on terribly bumpy roads is something many people don’t feel safe doing. Or going up the hill from the U-district to Capitol Hill.

        I guess the point of this post is all about trying to plan facilities etc that are appropriate for as many people as possible…but it’s still important to acknowledge that certain facilities will never meet certain people’s needs, you know?

      3. Sure, topography is key. And if you factor that in, a 1/2 mile up CapHill, or Queen Anne, or Beacon, or to the top of West Seattle requires either a shower, or tolerant co-workers.

        But since most workers are heading downhill, to the UW, Seattle, etc. The distance factor on this chart is still off.

      4. “It takes 15 minutes longer than the bus, IF I drive to a Park and Ride first. However it’s not flat so I arrive needing a shower, which fortunately my employer provides.”

        I think you probably need to count “shower time” as part of your trip time, then…

        Where I coming from:
        – I ride most nice days from Wallingford to Montlake (2.5mi), then put my bike on a 255 to Kirkland. This is the fastest non-driving way to work, reducing a 3-bus trip to a 1-bus trip. Fortunately, it’s all downhill, so I don’t get sweaty. The only bike-specific clothes I own are a pair of warm bike-gloves, and an under-helmet hat, without which this wouldn’t work in February. I have spent time planning this on Google Maps, and have tried multiple different routes in order to find the route that will give me minimal interaction with cars, given that Pacific St. & the Montlake Bridge aren’t particularly bike-friendly.
        – On the way back, this is uphill; I would not want to have ridden this distance as I would arrive at work sweaty & overheated.
        – My fiancee will not ride from our house to her SLU workplace, at a distance of 3.5 miles (almost entirely downhill), as the helmet will mess up her (long&curly) hair, and she will arrive generally disheveled. Also, she’s afraid of getting hurt en-route, which seems like a valid concern given bus-rider-safety vs. bike-rider-safety.
        – I’ve biked over the lake (~21mi) a couple of times, and I’ve finally found a route that doesn’t feature Tour-de-France-rated hills. I plan to do this more when it gets nicer, but it’s 50% longer than taking the bus (1h30m biking vs ~40m by bike+bus, or !1hr by bus alone). I also arrived somewhat sweaty, even in February, but I work in a fairly casual workplace, so that was OK, although I did feel a bit self-conscious. If a shower had been available, I would’ve certainly have rinsed off (again, requiring preparation).

      5. “I need to count shower time”

        Not really. I work at a “get a shower before work job” as opposed to “need a shower after work job.” So whether I take my shower before I leave home or after I arrive a work, it takes the same amount of time.

      6. So to be clear, the problem with the chart is that 10 miles is “red” when in fact I’d say 25 miles hits the red zone.

        On your fiancee not getting her hair messed up. Yep, I understand that, my wife won’t ride to work either, even though I volunteered to get a tandem and do most of the work. It’s the perception of beauty from cosmetics vs beauty from an athletic build problem.

        And even if your fiancee rode down Capital Hill, she would still have to lug that bike back up the hill. While I find that hill “invigorating” it took a couple of weeks of doing it before it didn’t make me winded. I have a friend who put his bike on the front of the bus for the ride back up until he got in shape.

        On the danger side, your fiancee is at most risk walking the last mile to SLU. Our cities injury rate of cars vs pedestrians is terrible. She would be safer on a bicycle.

      7. Gary, your kid’s Ballard-downtown ride is faster than the bus because the bus is lousy — specifically because it is a “scheduled bus” with long wait times and all of the lousy reliability metrics that invariably follow from a scheduled bus with long wait times on a medium-distance route with high demand and high turnover.

        Meaning that biking is faster than busing for the precise reasons Matt’s chart lays out — the bus service is inappropriate for the geography and fails spectacularly as a result! (RapidRide, at 15-minute headways most of the day, falls short of fixing this.)

        It is not because biking is inherently superior (or a reasonably expectation) at 5-20 mile distances!

      8. I don’t say this lightly, but everyone please ignore Gary on the narrow “scale-calibrating of bicycle-related transport” issue.

        He thinks everyone should have to ride a bike just to access public transit, even in the middle of the city.

        He thinks desiring fast and frequent public transit at all means you’re lazy.

        He’s basically the reason that mass utility cycling will never take off in Seattle and why cycling will always be the mode choice of a tiny minority here: for him, it’s 15-mile road-warrior sprints covered head-to-toe in Lycra, or nothing at all.

      9. No my kids Seattle to Ballard bike ride is faster because the bus has no dedicated guideway/right-of-way. And a zillion stops as it picks up and drops people off. As for convenience, it stops across the street from their domicile, which couldn’t be easier.

        I’m not advocating everyone wear lycra, you should wear wool, it’s better for the environment and more weather appropriate to Seattle. I’m just saying that a city bus which has to stop every other block is slower than a bicycle. In fact I pass them all the time. Their average speed is way less than 15mph.

      10. I get that, and everyone here agrees that lack of traffic priority, excessive stops, and a payment system that encourages slow egress at all those stops are a big part of the detriment of urban-use buses.*

        But as Matt and The Chart point out, all of those problems are compounded when you have to wait 15 or 30 minutes in addition to all the other drags on service speed. This sometimes makes the average 5-8 mile trip take twice as long as it would to bike the same distance.

        And Matt and The Chart illustrate that this disparity is a sign of a grossly inappropriate urban service pattern because — and this is the part that never seems to get through to you, Gary — the vast majority of people are never going to bike 5-8 miles on a regular basis.
        Because they are physically unable, or because they can’t get sweaty every single time they need to travel that far, or because they won’t bike in the dead of night, or for myriad other completely valid reasons.

        This doesn’t “prove” the superiority of bicycling relatively long distances. It proves the need for service patterns that don’t get the masses around at a tiny fraction of the speed of an athlete on two wheels.

        *(One thing I’ve observed Seattleites generally fail to grasp is that many of the slow-egress problems fix themselves when service is upgraded to “frequent bus.” Demand fluctuations between vehicles get ironed out; you’re less likely to have a single vehicle absorbing multiple very-slow-entries or filling to the gills so that people can’t squeeze by one another.

        Even more intriguing is the role inertia plays in people’s way of riding transit. Those who have been waiting a long time move toward the doors more slowly, board more slowly, pay more slowly, and settle in more slowly. When you show up and the bus comes, the boarding-riding-exiting process retainsm the momentum you brought with you to the stop. No, I’m not kidding; this is an observable phenomenon, and the time it saves adds up!)

      11. I would argue that that’s because the bus routes are no optimal, have too many stops, and there’s too much traffic slowing them down. There are exceptions based on local variables to every medium here. In your case, the bikes work great! But what if you wanted to get from say…Interbay to Discovery Park? Super short trip, but those HILLS!

      12. “what if you wanted to get from say…Interbay to Discovery Park? ” Electric assist. Or if you are me, what hills? There’s one and if you ride first to the bike trail at Nickerson, and then West, you circumvent most of it.

        As for everybody “not going to ride 5 to 8 miles” So what? I’m arguing that this chart is stupid because those who do ride a bike, can easily ride up to 10 miles. It’s not Red unless you are a cripple. Red is 25 miles, even an in-shape commuter rider is going to take near 2hrs. Its not the distance, it’s the time. I rode 34 miles last Monday to get to an errand and then to work, and then rode the 16 miles home. I wasn’t any more tired or sweaty than my normal route. It just took a long time. However a bus would have been even worse, as it would have meant 6 buses for the morning trip, AND a 2 mile walk. I could have driven it but I’m to cheap to pay for parking downtown and I need the exercise.

        As for making a Rapid Ride bus from Seattle to Ballard, where are you going to put it? The roads are full. That’s why the elevated Monorail system would have worked, it created a whole new right-of-way.

        And as gas costs rise, I bet we’ll see more people chose bicycles vs driving or even buses as they aren’t immune to rising fuel costs.

      13. As for everybody “not going to ride 5 to 8 miles” So what? I’m arguing that this chart is stupid because those who do ride a bike, can easily ride up to 10 miles.

        We get it. You like to bike. You’re good at it. You can go far. You’re a shining beacon of two-wheeled horizontal fury.

        The chart visually represents the reasonable average distances that the reasonable average modal user can be expected to travel on a given mode. Most people are not going to bike 5-10 miles multiple times a day, period. Insisting otherwise renders you irrelevant to the conversation.

        It’s not Red unless you are a cripple.

        And this renders you an asshole.

      14. We get it, you don’t like bikes. You want personal mass transit. It’s called a car. Works really well. Leaves exactly when you want, arrives without stopping to unload passengers on the way. Fine, drive, who cares if your blood pressure rises, you gain weight, are susceptible to type 2 diabetes, Medicare and Medicaid will cover you when you are old.

      15. Gary,

        Not once did d.p. mention that he wanted “personal mass transit”, or that buses are bad because they make multiple stops along the way. In fact, you’re the one who complained about the bus making a “zillion stops”.

        d.p.’s point — and Matt’s, for that matter — is that there’s a big difference between frequent bus service and scheduled bus service. Seattle has almost none of the former, and the latter is poorly suited to the situations in which we use it. The fact that the latter compares poorly with biking is not in any way an indictment of the former.

        If you’re suggesting that frequent bus service is equivalent to a car, then I just don’t know what to say. It’s different in the most important way — it takes up far less physical space per passenger, which makes it a key tool for urban mobility. Yes, frequent bus service is similar to cars in other ways; it allows you to travel between two points at a high speed, with a minimal wait time, and with minimal physical exertion. But that’s an asset, not a liability.

        Finally, bicycling’s health benefits are completely orthogonal to its usefulness as mobility. Yes, biking is great exercise. But the purpose of transit is not to get people in shape; it’s to get people to where they want to go. I don’t know if it was your intention, but I find it hard to take your comments seriously when your main argument is that anyone who doesn’t want to bike 16 miles to work each day is an overweight, diabetic cripple on Medicaid (all your words).

      16. You can absolutely ride at non-sweat speeds for 2 hours on a bike and travel 16 miles. But at the point where we decide it’s acceptable for our transit systems to take this long, we’ve failed.

      17. Gary, I can bike 10 miles if I feel like it. I can fucking run 10 miles if I feel like it.

        But I’m not going to do it as my primary method of getting around!

        That you consider this a character flaw worth berating me about in a public forum puts you about on par with people who stand on street corners shoving “Repent! Sinners!” pamphlets into strangers’ hands. Or PETA activists with cans of spraypaint. Or Larouchebags.

        Frankly, you’re probably the dickhead bicyclist who rides in the bus lane, forcing your pace upon dozens of people. You probably also “leapfrog” buses trying to pull out of stops, don’t you?

        Your “advice” is not welcome.

      18. And thank you, Aleks and Matt. Seriously.

        This tangent has made me weep for humanity. Matt makes a cogent visual argument for why we desperately need to reassign service at geographically appropriate levels.

        Gary’s takeaway? “What we have currently is worse than biking, so biking must be the only answer. This fits with my knee-jerk bicycle-centric agenda! I shall now harp on it!”

        Enjoy the Tour de France, Gary. While you’re there, you might want to make note of all of those svelte Parisian transit riders who do not have diabetes!

      19. Silly me, I forgot I was on the Seattle “I love Light Rail” blog.

        My ONE and only complaint was that the chart was drawn by someone who doesn’t use a bicycle for transportation. And thus appears to have no idea how it fits into the system. It’s best uses, or appropriateness. Bad data equals bad graphics.

        I don’t care whether everybody rides or not. Bicycles should be recognized by transit engineers for their best use. That once you chose to ride, you can easily beat a bus over a 10 mile in city ride. Over 10 miles and buses tend to be faster. That riding 1 to 2 miles on your own bike is hardly worth the extra effort of getting it out, and locking it at your destination. Plus the additional clothing for the weather. It’s faster to walk.

        And that putting 10 miles at “red” is wrong. That 20 miles is, “orange”, 25 “red” and 50, just about off the scale. Although there are a few regular commuters to Children’s Hospital who do ride that far.

        As for the health of Frenchmen, well they have national healthcare. One would hope that with preventive medicine that they would be healthier. But since the USA is #37 in health, it’s not hard to find a country that does it better than us.

        As for designing a transit system to incorporate health, should be one of its goals. There was a law that didn’t make it through to request the WDOT to consider health in it’s plans. That the USA is in general overweight, and out of shape is in part because our current transportation system encourages driving the worst choice for your health.

        But there are those who can’t seem to grasp that health of people, and health of a city go hand in hand. Reduced costs of healthcare allow people to spend more of their money on the things that make a city great, like sports, or theater. It’s all connected.

      20. The point I’m trying to make about fitness is about purpose versus benefits. The purpose of transit is to get people to where they want to go, i.e. to provide access. Transit also has many benefits, an important one of which is that transit users tend to be more physically active, and thus healthier, than non-users. But that isn’t its purpose.

        If you set out to design a system whose primary purpose is fitness, you’ll get a gym. I have nothing against gyms. But they’re not transit.

        I can think of tons of situations in which, given the choice between access or fitness, most people will pick the former. Here are just a few:

        – Shopping trips
        – Traveling with small children
        – Elderly or disabled users
        – Steep hills
        – Commuting to work site with no shower
        – Commuting to job that already involves manual labor
        – Bad weather

        There are people who would gladly bike 25 miles to work, uphill both ways, in the snow, just for the exercise. Maybe you’re one of them. Most people aren’t.

        I admire your holistic approach to problem-solving. But in this case, I don’t think your conclusions are correct. Of the world’s most successful transit systems, not one of them is designed primarily for bicyclists. If you designed such a system in the name of making everyone healthier, then everyone would just drive. As they say, the perfect is the enemy of the good. Let’s give people a transit system that they’ll use, rather than one that will make them better.

      21. “My ONE and only complaint was that the chart was drawn by someone who doesn’t use a bicycle for transportation.” I logged 56 rides last season, about half in work clothes and half in workout clothes (I don’t do spandex except on long rides) – and that ignored the first month of the season before I got the bike mapping app, and short trips to the store where it wasn’t worth logging. Most of those were commute trips, though many were riding with my son on a carrier to his school.

        So if that’s your one complaint, I guess you should be happy.

        “riding 1 to 2 miles on your own bike is hardly worth the extra effort of getting it out, and locking it at your destination” Completely disagree. My most useful bike trips are about 1/2 a mile away – far enough where I’d be tempted to drive rather than walk.

        “10 miles in red is wrong” I disagree. That’s over an hour ride at casual speeds.

      22. I support adding health to the state’s transportation goals. It isn’t the only or primary goal, however. We just have to make it easy for people to get their minimum 30 minutes of walking each day. Even if that were their only daily exercise it would do wonders for their health. We don’t need to turn everyone into cross country cyclists. With walkable neighborhoods and good transit, like in Paris, one could enjoy walking a lot more than the minimum.

      23. Matt, you fixed the chart for the first bit, by separating out the utility bicycle trips. The, no wool, no special clothes, ride a bike to the store trip. That’s great.

        Just fix the other one, the wear special clothes, ride a fast bicycle one. 10 miles is EASY, an hour+ is what I spend on the BUS if I took that (along with my P&R drive, if I bus to the P&R it’s the same amount of time as it is to bicycle). My commuting buddies complain that their ride in to downtown from West Seattle is too short. You have another color, “yellow” use it to mark it from 10 to 25.

        I’m glad you are using your bicycle for short trips.

        Most of my bicycle buddies have a “go to the store” bike outfitted with wide tires, racks, cruddy seats, and a basic AL frame, don’t bother to steal this bike for those trips.

      24. He still doesn’t get it.

        Psycho-cycling shouldn’t be on this list at all.

        Every other mode of transportation itemized above — including “feet” — is something that can be made available to most people, for most kinds of trips, in the appropriate geographic circumstances and over the appropriate distances that the color-coded chart has been designed to encapsulate.

        Psycho-cycling does not meet these criteria. It doesn’t matter if 10 or 25 or 1,000 miles is an “easy” ride for a handful, because it is not a reasonable option for even a fraction (never mind a majority) of the population.

        Even the most physically fit can’t make bike clothes and shower facilities appear from thin air.

      25. Driving a car should be on this chart. Why? Because 90% of all trips are done in a car. It’s the 800lb gorilla that’s missing.

        If you want to change that statistic, you need to understand why people make that choice.

  4. Walking more than a mile or biking more than 10 are inappropriate? That is a fair chunk of my trips.

    1. He’s clearly trying to create a chart that includes one standard deviation or so of normal and I think your 10 miles biking or walking each direction is somewhere above two sigma :-)

      1. 1/2 to 1 mile walking is yellow?? I walk 0.6 miles daily from the coffee shop to work. Given a preferred walking speed (wikipedia) of 3.1mph, 1/2 to 1 mile is a short 10 to 20 minute jaunt.

      2. I’ll buy that logic. There is data that 1/2 mile is about as far as people are willing to walk, so I made it yellow. But that doesn’t mean walking is a bad option even past 1/2 mile – especially when you consider most bus stops aren’t near where you want to start or end your journey.

      3. The willing to walk value depends on the population group and environment; individuals accustomed to shuttling between suburb and car and office will likely have a lower willing to walk limit than a student who needs to trek around the UW main campus. Should the suburbcarofficeparkers suddenly develop an interest in walking more (perhaps due to a medical report on the accumulated damages of their lifestyle to their body), they will likely not find good walking options in their environment, and therefore be less likely to acquire a higher willing to walk distance.

      4. Walking distance is also a function of the unreliability of transit. For instance the SLUT, at 15 minute intervals is on average slower than walking. If you just happen to arrive at a station just before the SLUT arrives, then it will be faster to ride. But that’s it.

      5. From a transit perspective, the key distinction is this: what modes should transit be competing with?

        It’s meaningless and pointless for transit to compete with walking. This is true for two reasons: first because walking is very compatible with cities (since the space overhead is 0), and second because walking is much more cost-efficient than transit, *for able-bodied people*, for the distances for which it makes sense.

        Therefore, it makes more sense for transit to compete with cars. Which means that rapid services generally make more sense than local services.

        So the next question is: how far must someone be willing to walk before we declare that the transit network is not able to meet their needs?

        The answer seems to be somewhere between 1/4 and 1/2 mile. Thus the numbers on Matt’s original chart.

        It’s reasonable to build transit services which effective serve trips as short as a mile, or maybe even half a mile. It’s crazy to build transit services which effectively serve trips as short as a quarter mile. That’s all.

      6. Totally agreed. However with a “bike share” you can ask people to ride 1 mile.

  5. The real question is “Are counties laid out over 100 years ago relevant and the right way to organize and manage urban society today?”

    The shape and size of the CPSRTA district would prove otherwise.

    And the history of bus service to Skykomish would testify as an expert witness.

    1. We could also ask whether state boundaries are appropriate. (No, they’re not. Ask the Portland/Vancouver metropolitan area.)

      Or even national boundaries.

      These things have a life of their own, unfortunately.

  6. A posting with the title of this one might also include a comparison.

    For almost twenty years, King County has been operating both transit and water quality services previously provided by the Municipality of Metropolitan Seattle.

    Does any of the readership have any assessment of the relative performance of both of these agencies?

    Mark Dublin

    1. The city of Seattle operated them in the pre-internet era – detailed performance/efficiency reports are not likely to be easily available.

      And my understanding is that the mid-80’s were kind of dark times for Seattle in general. High unemployment and high crime. City government really had its hands full – I don’t expect either agency was very healthy the day before their restructurings, making comparisons difficult.

  7. Nice try Matt, but too few variables make for a lousy outcome -( I’m defining a technology as more appropriate if it reduces your travel time and is convenient.)
    How about cost effectiveness as the 800# gorilla in the room? And your scale is too course to be of any value. Where does the average trip of 7 miles fit on the scale? Somewhere between 5 and 30 doesn’t really say much. That trip is covered nicely by streetcar/bus/light & heavy rail/Exp. bus – or just about everything except walking and planes. So what’s the point?

    1. I’d love to fit all variables in a single chart. But I’ve had to resort to gray areas enough with two – imagine trying to fit something as widely varying as light rail costs into that chart – or operations costs of buses.

      The way I see using the chart is to give you a list of what technologies are appropriate, then consider cost.

      That length of trip is better covered by light/heavy rail, but it’s perfectly acceptable to run buses as long as traffic isn’t a major concern. But that is the point – King County is making a fine decision given their constraints. It’s the lower end of the scale where their decisions don’t fit reality.

  8. So this chart shows what I’ve been saying all along:
    Heavy rail is the best, and when combined with Feet (for getting to the subway station), HSR for regional trips and Air travel for long-distance trips, it can effectively supplant all other forms of transit. It’s just a question of will (aka money).

    1. Yeah, but heavy rail isn’t worth it for low passenger volumes.

      I’ve seen a rather nice chart from some Japanese government which had two axes:
      (1) distance
      (2) *volume* (number of passengers)

      The space was then carved up by most suitable mode. Volume matters a lot.

      1. Any chance you can find a link to that graphic? I’d be interested in it.

        Volume is certainly important, but can get tricky. You have to figure in frequency, capacity, and speed.

  9. Don’t forget paratransit. That’s a decent chunk of the transit budget, and maybe mini-buses aren’t the automatic choice.

  10. Is it really true that the majority (>50%) of workers in King County work in DT Seattle? That seems astonishingly high to me, though I wouldn’t at all disagree that DT Seattle is by far the most important employment center.

    1. Most trips may be headed downtown, but a lot of workers have to get to a destination with a transfer downtown. I’ve been there.

      Metro’s one-way commuter routes don’t make sense if they are passing by employment centers whose employees are trapped downtown. The 122 is one example of a bus that could have gotten me to work if it had had some counter-peak directional service when I worked in SeaTac.

      It seems like every city center that has a lot of peak service to downtown, but very little from downtown to their local employment centers, is having an economic hard time. Tacoma. Federal Way. Lynnwood. Auburn. The one-way service makes it hard for these city centers to attract employees or shoppers.

    2. Good point [Steve]. I’m sure it’s the largest job center, but almost certainly doesn’t have over half of the jobs. (runs off to search) Looks like downtown Seattle has around 220,000 of the million jobs in the county. I’ll change the language in the post.

    3. I’ve pulled up the numbers before. Roughly speaking half of the jobs in King County are in Seattle. About half of the jobs in Seattle are in the DT or CBD.

  11. One idea that I think has enormous potential is the idea of a Gondola in Seattle but with more usefulness and relevance than the one in Portland which merely connects a university to another part of its campus – right?

    The proposed idea for a Gondola in Seattle would include linking tourist, work and residential areas – the Waterfront through to Capitol Hill via the Seattle Center and South Lake Union with stops at the Waterfront and the Center for tourists, in South Lake Union for work and Capitol Hill for residents (but tourists as well).

    However, looking at King County as a whole, I would certainly make full use of light rail on all main arterials throughout the county (including light rail out to North Bend, Snoqualmie and Issaquah), buses as arterial feeders and secondary roads and commuter rail through to Olympia (outside of King County I know but pressure could be applied) on the I-5. I would add stations to the North Sounder line at Ballard and Broad Street and possibly in Shoreline. I would possibly advocate for Sounders to run up to Bellingham or to work with Amtrak on increasing service up there. I would also add a Central Link stop on East marginal to serve Boeing Field and Boeing plants. A multimodal stop there could also serve Amtrak and Sounder.

    In all of this, I see King County as the visionary heart for the area and looking beyond its boundaries where necessary to act as an economic and transportation engine for the area.

    Of course, much of this is visionary but you did give us a blank slate. Though some of my proposals would end up being close to what exists today, I would stretch out what confines us to take in outlying counties.

    1. The Portland tram does technically connect the two parts of OHSU, but the bottom end also connects into their streetcar system while the top is a major job center. The tram is twice as fast as driving/busing up the hill (on a narrow winding two-lane road), so it’s actually a major improvement. The problem I see with a gondola in Seattle is that there isn’t anyplace like that, with two major close-but-separated destinations that don’t already have some kind of street between them.

      1. Our east-west connections are awful, and rush hour traffic more or less locks up many of our streets. Here’s an idea that connects a dense area with (future) light rail to a job center and two tourist attractions. A gondola on this route will be much faster than twice as fast, at least during commute hours.

        It’s a common (and wrong) assumption that gondolas are only the solution to certain specific problems. Though they’re great at skipping over geography or traffic, they could work just fine in a flat city.

      2. The best gondola locations in Seattle would probably run over water, but there are probably restrictions on doing that due to very tall ships.

      3. I will say, the biggest problem with gondolas is that they are best for point-to-point operations.

        If you’re going to have lots of intermediate stops, the technology just isn’t that good for it (most ski resorts just have everyone get off and board the next gondola), and you probably should go ahead and build rail.

        (If there’s a steep hill involved, that may mean rack rail, or it may mean funicular. San Francisco-style cable cars which attach and detach from the cable are too work-intensive and trouble-prone to use when permanently attached funiculars are available as an alternative.)

      4. This issue regarding intermediate stops, combined with the hill, means that the Seattle Center to Waterfront gondola actually makes more sense than the other one you proposed….

  12. You should include “grade separated light rail/Monorail” as an option. It doesn’t have the capacity of “heavy rail” but it’s faster than “Seattle style Light rail.”

    1. But not much, right? We’re grade separated everywhere but SoDo and in the RV, and we’re path separated there. That slows us down a bit, but it should keep us roughly in the same range. The line for grade separated should look almost identical to Seattle-style light rail.

      1. But, hey, you didn’t include “horseback riding” either. so I can live with drawing the line somewhere.

      2. Those were Seattle’s proposals for locally-funded streetcars. ST will design these lines as it does other Link lines, starting with an alternatives analysis of surface, grade-separated, and BRT. If ST forgets to do a grade-separated alternative, we can holler at them that the public deserves a comparison of the cost and benefits.

  13. Surprised to see HSR has any line at all in the half-mile range and maybe even the mile range. Also a bit surprised the scale isn’t more granular (with 25 and certainly 100 mile checkpoints) at longer distances, and I might have liked to see “car” on the list between heavy rail and scheduled bus.

    Another thing to consider is the willingness of people to use a given technology at distances longer than they really work for because of the experience of using it. This is really poor for buses, iffy for feet, very iffy for bicycles depending on how you count excersize purposes, iffy but irrelevant for airplanes, and OK for everything else.

  14. Why is funicular on this chart; is it any less esoteric than a ferry?
    Why is cable car in a separate category from streetcar? What are its limiting factors, and does anyone actually consider it as a transit solution?

    1. Hills. Streetcars can’t handle them, and if they’re steep enough even buses won’t stop on them.

      Funiculars and cable cars are absolutely useful technology, but you have to have the exact right need.

      1. Seems weird to me that cable car is best at 1/4 mile… That’s 5 blocks… I would say its optimal trip distance is a mile or two.

      2. I think the recent fight(s) over First Hill service provide a good reminder that there is a rational need for better-than-history’s-slowest-trolleybus service along very steep corridors of much less than a mile.

  15. As much as I like HSR, I definitely wouldn’t put it and airplane as the same for 1000 mile trips. The fastest HSR would take around five hours non-stop to go 1000 miles, whereas that’s an hour and a half or two on the airplane. Even when you include getting to the airport and everything, HSR is still slower for trips of more than 500 or 600 miles.

    1. Tell that to the many folks that travel between Boston, New York City, and Washington DC on Amtrak Acela service faster from downtown core to downtown core than they can fly between those cities.

      1. You’re both right. Most trips on ACELA are from Boston-NY or NY-DC, both of which are less than 225 miles. ACELA is still a good option for the SW Connecticut suburbs to DC, and an okay option for Boston-Philly, but beyond that the advantages of flying are unequivocal.

        On the other hand, ACELA is about as un-fast as “HSR” could possibly be, thus its advantages petering out at shorter distances. You could make a very good case for taking Shinkansen the 700 miles from Tokyo to Fukuoka rather than flying. Same goes for London-Marseille (750 miles).

      2. You could make a very good case for taking Shinkansen the 700 miles from Tokyo to Fukuoka rather than flying

        In practice few people do this. I have never been able to convince my wife to take the shinkansen from Tokyo to even Kobe or Hiroshima, because the price and time are just too long for her.

      3. I remember reading somewhere that in practice, the Shinkansen took over most trips up to distances of about 400 miles. So, most Tokyo-Osaka traffic (300 miles) is now by train, while Tokyo-Hiroshima (500 miles) is mostly by air.

    2. The flight from SEA to LAX is 954 miles, and takes over 2.5 hours. Add an hour or three for getting to and from airports and waiting in security lines, and an 5 hour train can compete. Of course, there are also likely to be at least a few stations along a 1,000 mile train trip to slow you down, so airplanes win. I’ll narrow the HSR line a bit in the next iteration.

      1. Seattle’s airport is quite far from town, isn’t it?

        LA’s airport… well, what is “town” in LA? It depends on what part of town you’re coming from, I guess.

        NYC is almost a worst-case scenario for plane travel, which makes it almost a best-case scenario for HSR. Chicago is nearly is bad as NY for plane travel (you can easily spend half an hour just WALKING BETWEEN TERMINALS in O’Hare, let alone trying to get to O’Hare from the city); I genuinely think Chicago-NY HSR would be viable (though I expect it would be used more heavily to go to the many intermediate points).

        Seattle-Los Angeles is somewhat less viable. And has fewer intermediate points to go to; the ‘problem section’ is Portland to Sacramento, which despite low population is expensive to build tracks in, thanks to the mountains.

      2. To be clear, I was listing SEA-LAX because it’s about 1,000 miles away, not because it’s the best place to put HSR. That said, if California’s HSR actually happens, then someday it might make sense to continue upward, hitting Sacramento and Portland on the way and continuing past us to Vancouver.

      3. It’s a bit unusual to have to change terminals at O’Hare save maybe going from the International Arrivals Terminal 5 to any of the others. But there is an efficient rail line that links all the terminals and remote parking lots.

        I found that it was easily more than 3/4 of a mile at LAX to get from United main terminal to the concourse where their United Express commuter service operates. That was a hike and no people movers or other conveyances.

      4. You have to count your travel time to and from the airports on each end. With Sea-Tac, the TSA operations are such that you need to allow significant time to get through. So, I make it a rule of thumb to get to Sea-Tac about 2 hours before flight particularly if I’m checking luggage. You also have to consider that with Sea-Tac’s current roadway configuration, driving there is a roll of the dice. Frequent traffic jams and mile long back-ups on the airport freeway are not rare. So your 25 minute drive from Bellevue could turn into 1+ hour of travel time. If you’re parking in an offsite even more time and expense.

        Chances are you could reach a train station quicker than the airport e.g. downtown Seattle, downtown Tacoma, or Union Station in Los Angeles. If the enroute travel time to go 1000 miles is 6 hours or less on HSR, then I think it’s quite comparable to flying block to block time wise, and I would guess a bit better for the environment.

    3. “The fastest HSR would take around five hours non-stop to go 1000 miles, whereas that’s an hour and a half or two on the airplane.”

      …plus an hour in security, arriving two hours early at the airport (actual recommendation), and an hour each way to the airport… it could easily be 5 hours without the actual flying, in a worst-case scenario.

      In cities where flying is like that, HSR is deeply, deeply popular. In NY, for instance, it’s typically an hour to any of the airports (assuming you start in Manhattan), and much less than that to Penn Station. (Likewise in DC, Dulles is way the hell in the middle of nowhere, although National Airport (like Union Station) is easy to get to.)

      NY also has routine takeoff delays due to airspace overcrowding. And then there’s security, of course.

      Frankly, if you’re starting from NYC, HSR becomes an attractive alternative to much farther out than if you’re starting from, say, Boston (with a relatively close-in airport which is relatively uncrowded, and had relatively fast security last time I was there).

      I guess this adds to what d.p. said: dp pointed out that faster HSR competes with longer-distance flight, and I pointed out that HSR competes with longer-distance flight when the airports suck.

  16. I will say that there are people who tkae the Shinkansen from Tokyo to Yokohama, which is less than 50 miles (maybe 25).

  17. 1. The use of technology terms is disappointing. What matters is service type, which comprises something like: length of line, stop spacing, average speed, frequency, span of service, reliability; maybe comfort as well. Setting aside the bus-rail distinction, which could be argued fruitlessly all day, look at distinctions within rail. Light rail technology works for service types from streetcar through standard “rapid transit” and beyond into exurbia. Heavy rail metros overlap the entire range except for street-running; and, of course, heavy rail commuter/passenger trains have different characteristics entirely. The same is true within buses — there’s a pretty wide range of service types, and you’ll find that within Seattle there’s a lot of diversity of service types.

    2. So you made up a chart based on arbitrary characteristics. Then you write, “Take another look at the chart above.” If you are going to refer to a chart when making an argument, the chart should actually be showing something other than more of your own opinion.

    1. 1. I await your method of displaying all of that graphically.

      2. The chart was just a method of starting a conversation – of showing what I’m talking about and why. I was very clear that it’s just my opinion.

  18. Re-imagining transit in King County?

    Ok. Rail backbones – 3 N/S lines on the west side of Lake Washington. The easternmost follows the current alignment of the 7 & 48S within city limits, the center runs the current Central Link path, and the westmost roughly follows the 120 corridor to the airport.

    The western line merges with Central at SODO station, and goes through downtown in an enlarged 4-track DSTT, before forking off to Ballard underground, following a mildly-winding path with stations serving the dense cluster of apartments just west of Aurora, heading straight east/southeast to hit LQA and then heading under the counterbalance.

    The Eastern line has a transfer to Central at Mt Baker before heading through the CD and merging with Central at Montlake, shares a 4-track tunnel as far as Brooklyn station, where it veers towards Ballard down the path of the 44.

    So then we have a rough ring-line around the city center on the high ridership paths. Central heads north out of the county, and either the east or west lines can be extended to Shoreline or Bothell if it seems to warrant it, but I’ll accept aggressive BRT for that section right now.

    Surface streetcar handles local transit in eastlake, westlake, and Capitol Hill. The currently proposed streetcars are implemented, with the Slut continuing out to the UW, and the FHSC continuing north along Broadway/10th until the apartments peter out around Roanoke. Of course, once you’ve gone that far, might as well connect to the SLUT at the north end and get that last mile to the U-district.

    South of the city, the eastern line heads through Renton, and continues south through the valley along the rough route of the 150, petering out somewhere in Kent or Auburn. The central or western line, doesn’t matter which, continues south to Federal Way.
    An Eastside line with widely spaced stops should come down from Kirkland, through DT Bellevue, hitting the business and apartment clusters at Factoria, Newcastle, and the Renton Highlands, before heading east through Renton, crossing the east Seattle line, and terminating at Seatac.

    East Link still gets built, from Seattle across I-90, crossing the eastside line in DT Bellevue, following the Bel-Red corridor into overlake, getting a new Willows station, and ending in downtown Redmond. That gives us a general X pattern on the east side.

    I’m not sure about the rest of the county. What kind of service Woodinville and Bothell should get is not something I want to figure out right now. I honestly want something like a frequent Bothell-Woodinville-Redmond-Sammamish-Issaquah line as kind of an outer-ring service, but I don’t know how those cities would react to it. They all have their islands of density – be it a strip of office parks or apartment buildings, and retail destinations in between. It seems like you could string them all together on a frequent line, with scheduled service forking off.

Comments are closed.