Rush hour at University Street station

Sound Transit and Metro have released their 2017 ridership numbers, and they paint a rosy picture for our regional transit system amid a national decline in transit ridership (particularly among buses). The two agencies alone carried 155 million total passengers within King County; add estimated figures from Pierce and Snohomish counties and the number of total transit trips taken in 2017 increases to over 190 million. Leading the way is Link, which averaged 72,028 weekday riders and carried 23 million total passengers, an increase of 22 percent over 2016’s huge ridership. Sound Transit’s ridership grew by 10 percent overall, with only a small decline in ST Express ridership holding it back.

To put things into perspective, Link is now ~40 daily passengers away from surpassing the Minneapolis–St. Paul light rail system, which averages 72,064 riders on 23 miles of track. Even without the boost from the Northgate Link extension, ridership could come close to – or surpass – Denver’s RTD light rail system, which carries 75,900 daily riders over a sprawling 59 miles of track.

More numbers after the jump.

Link ridership by station in Q4 of 2016 and 2017

In the ridership reports for Q4 and December 2017, Sound Transit notes that declining ridership at SeaTac/Airport Station can be attributed to Angle Lake taking over as the terminus and preferred transfer point for A Line riders, leaving pure airport and SeaTac usage in its wake. Expect to see a similar drop for Angle Lake once (or if) the Federal Way extension opens, and perhaps even for University of Washington Station when Northgate Link opens. Q4 ridership itself was below ridership targets, but strong ridership earlier in the year puts Link over its budgeted 22.9 million boardings; as a bonus, the cost per boarding has dropped to under $4.00 and the farebox recovery reached 42 percent, well ahead of Sound Transit’s target of 35.6 percent. Sound Transit is forecasting another increase to 49.7 million riders systemwide for 2018, with Link carrying 25.2 million, based on natural growth and new development at stations.

Ridership figures from the December 2017 report, released earlier this month (Sound Transit)

Link took over as the largest mode in Sound Transit’s system ridership statistics about 22 months ago, and over that time has continued growing while ST Express has declined slightly by 0.5 percent. Some of this can be blamed on the closure of Eastside park and rides for East Link construction, which has lowered Route 550’s Q4 ridership by 10 percent and Route 545’s by 5 percent.

For Sounder, the North Line experienced a modest drop of 1.1 percent, while the South Line grew by 5.5 percent. The Sounder system now carries over 17,600 daily riders, good enough to leapfrog Miami’s Tri-Rail and Utah’s FrontRunner, and come close to the Virginia Railway Express system serving the other Washington. Tacoma Link also had a modest increase of 3.6 percent over 2016’s abysmal numbers, putting it about equal with the 2011 ridership but below the line’s peak of 1.024 million riders in 2012.

Metro is also reporting modest increases in their ridership, which includes RapidRide and the South Lake Union Streetcar. RapidRide ridership increased by 4.2 percent to 20.7 million, with the E Line leading the way at 17,200 daily riders on weekdays and followed closely by the D Line at 14,200 riders. Routes to South Lake Union are carrying 1.4 million weekday rides annually (an increase of 5.9 percent), while ridership on routes serving University of Washington Station increased by 3.3 percent to 520,000. Assuming continued growth – even with Link replacing some of the most popular routes – Metro could very well reach the all-time high of 130 million passengers set by Seattle Transit in 1944, helped by wartime gas rationing and a shiny new trolleybus system.

114 Replies to “2017 Regional Transit Ridership Grows By Leaps and Bounds”

  1. I think it needs to be noted that annual data for 2016 includes almost three months before U-Link opened and almost nine months before Angle Lake opened.

    The December-only growth in Link ridership was a disappointing 3.9 percent. Of course, December is an off month because of holidays and universities being closed — and maybe fewer Seahawks games. The January data would probably be more telling about natural growth.

      1. Well, in fairness, that’s true of every December. Any given December is likely to have considerably lower ridership that the preceding July because of seasonality, but a comparison between December in year “x” and December in year “x-1” should be as valid as any other.

  2. My prediction is that Northgate will push Link to at least 90,000 if not 100,000 during a weekday. Lots depends on bus restructuring for 520 and for North Seattle and points further north.

    1. Bus restructuring would be awesome. Eliminate the one-seat rides to downtown and use the saved service hours to increase frequency and/or increase local neighborhood feeder routes.

      1. That’s a good idea where there are people in the neighborhoods during the day. Over and over, however, Northeast Seattle has shown itself to be pretty resistant to mid-day service improvements. It’s peopled by two-striver families who both work, and the teen-age kids have their own cars. There is already frequent service on most routes in the areas that Link restructuring will affect. You’d be pushing on a string to provide further headway reductions there.

        Move the hours elsewhere for a couple of years on a trial basis and if ridership doesn’t respond, reduce the sales tax by a tenth of a percent and shrink the bus system. One of the selling points for Light Rail is that it carries more people for less per passenger. There may in fact be an upper level for ridership in the non-core areas of the region.

      2. Reducing the overall quantity of bus service would be unwise, as the region is constantly growing.

        And NE Seattle has been riding the 65.

      3. Good thing to plan for. But we’ve got considerable miles of track to lay before we can do very much of it.

        Mark

      4. asdf, Certainly. That’s why I said “move the hours elsewhere”.

        However, I do think that the scores of square miles of Sprawlsville in the Metro Service Area are pretty resistant to transit service of any kind except commuters. Yes, there are corridors of poorer people threaded through the SFH zones, but the fact is that there is nobody living in most of Sprawlsville who doesn’t have access to a car.

        Absent a nuclear war in the Middle East in the next decade or two, vast areas of the region will prove durably plangent against “local” service. And after the cars are all electric, they’ll be permanently a waste.

        Non-commute transit is wasted on most of suburbia except business arterials Let ’em take Lyft if commuters need to get home for illness.

      5. Northeast Seattle was restructured in 2016. There are no all-day routes to downtown east of Latona. The peak expresses are temporary until U-District and Roosevelt stations open. At that time there may be some minor switching of heads between UW and U-District stations, and changes related to RapidRide Roosevelt and 44, but everything else is already in place. There may be some rethinking of the silly 71 and 78 routes.

        North of Northgate will be restructured with Lynnwood Link and 130th Station. Lake City will be affected by both since it’s on the outer edge of Northgate’s service area.

        North central and Northwest Seattle will probably be restructured more than northeast Seattle, because they were outside U-Link’s service area. However, the 45 and 62 have already been restructured. The minor routing changes in RapidRide 62 are small. Unless Metro/SDOT change their mind about RR 62’s routing, which is possible.

        “that there is nobody living in most of Sprawlsville who doesn’t have access to a car”

        We’re not doing a major transit upgrade because people don’t have cars; we’re doing it because we want people to drive less for environmental, equality, quality-of-life, and economic-sustainability reasons. Metro has a good plan for the suburbs; you can critique it as you wish. But other countries that take comprehensive transit seriously have more frequent and local routes in the suburbs than we do, even in places we consider “too low ridership”. It’s part of universal access, and their ridership is also significantly higher because it’s more feasible to get around on transit.

      6. >> Northeast Seattle has shown itself to be pretty resistant to mid-day service improvements. It’s peopled by two-striver families who both work, and the teen-age kids have their own cars.

        How do you define “Northeast Seattle”? If you think of streets and avenues that have “NE” in them, you have a pretty wide variety of housing types. Everything from old apartment buildings in Northgate or Lake City to multi-million dollar homes in places like Windermere and Inverness. To say they are roughly the same is like saying Medina is similar to the C. D.

        >> Over and over, however, Northeast Seattle has shown itself to be pretty resistant to mid-day service improvements.

        What do you mean by that? That folks in the Northeast end aren’t interested in mid-day service improvements? Seriously? There are a lot of people who would love to see the 73 run every fifteen minutes, so folks in Pinehurst can get to to the UW at a reasonable hour, but they simply don’t have the service hours.

        Or are you saying that improved mid-day bus service hasn’t mattered? That is absurd. The 372 has seen a big increase in ridership, despite the fact that it doesn’t go downtown (unlike the 522, which hasn’t seen an increase). The 70 has also seen big increases as well.
        https://seattletransitblog.com/2016/08/09/the-u-link-restructure-by-the-numbers/. These are all buses that serve the Northeast end of town. You do realize the UW is in Northeast Seattle, right?

        >> Move the hours elsewhere for a couple of years on a trial basis and if ridership doesn’t respond, reduce the sales tax by a tenth of a percent and shrink the bus system.

        What an insane idea. People have repeatedly supported transit in this city, bus ridership is increasing, and you want to cut it? Why? The overwhelming number of trips are still taken by bus, and that will continue as long as we have a decent bus system. Why on earth would you want people to have a much worse transit system after we have just decided to spent billions more on it?

      7. @Mike — I agree with much of what you said in your last post, but for the life of me, I haven’t figured out what you mean by this:

        Northeast Seattle was restructured in 2016. There are no all-day routes to downtown east of Latona.

        Are you sure you mean Latona? The 41, 49 and 70 are all east of Latona. Maybe you meant 15th NE?. There are on all-day routes to downtown east of 15th NE*. The point being, much of Northeast Seattle no longer has all-day service to downtown. Places like Wedgewood, Ravenna, Matthews Beach, Sand Point, Windermere, Laurelhurst and even Montlake** no longer have all day direct service to downtown. This is a profound change in the way our bus system works. For as long as I remember (and I’m not young) you either don’t have bus service or you can find a bus that will take you directly downtown. It might not make a direct route, it might not be frequent, but you could find a bus that went downtown. Now a lot of people have to transfer. That’s a big change, and one that will happen more as our light rail system expands. Buses like the 41 and 71 will likely be truncated. So too may buses like the 522 and 512. There will be fewer buses going downtown, but there will still be buses like the 70, which goes downtown, despite serving some of the same neighborhoods as Link.

        * Actually the line can’t be drawn east of 15th NE either, because the 522 runs along Lake City Way. The area is a bit hard to describe textually, but the northern limit is Lake City Way, the southern limit is Thomas, while the eastern limit varies. North of the ship canal it is 15th NE; south it is 10th East. That is a pretty big swath of the city that no longer has all day direct bus service to downtown.

        ** Not that I would call Montlake Northeast Seattle.

      8. I forgot about the 49 and 70. I usually envision northeast Seattle as east of 15th Ave NE because that’s where average density drops off precipitously. The 41 and Lake City are outside the area I’m focusing on because U-Link doesn’t serve them and Northgate Link will only peripherally serve Lake City. The 522 is Sound Transit so not part of Metro’s restructures

      9. Ross, Yes, “seriously”. Show some data that streets other than “business arterials” which receive “frequent service” use it.

        Even Sweden which must be one of the most dedicated pro-transit nations in the world and provides wonderful “coverage” has seen ridership per bus mile on those coverage routes halve in the past two decades.

        The truth is that as the world ages the people who are afraid of, look down upon, or just can’t stand to be inconvenienced by another human being — and that is a LOT of humanity — will choose to drive whenever they are not faced with tear-out-your-hair congestion or thirty-cent-a-mile direct costs.

        Those are the people who live in the SFH suburbs. Now that the scourge of lead-induced murders of the ’70’s through the early ’90’s is over people who want to live near other people have moved into the cities.

        Yes there is not enough “city” for all the people who want to live in one, but bus service in Sprawlsville won’t help that until the SFH are torn down and density is built in their place.

      10. Ross, and as to NE Seattle (by which I do mean north of 45th and east if Roosevelt there are already frequent service buses on Roosevelt, 25th, 35th, Sandpoint Way, and NE 65th and 110th and 125th.

        None is used heavily except the “U-Trans” 372 which is a regional route.

        Except for an east-west route at 80th or 85th which don’t go through, that’s enough.

      11. And the 41 of course. I added 110th and 125th and forgot to add this bus to the heavily used group

      12. Used heavily off-peak

        Look, all I’m saying us “put the service where the riders are”, not where people vote for transit in order to clear the streets in front of their cars.

      13. As RossB said, people *are* riding the 65, 75, and 372 during the non-rush-hour hours. There is real ridership in this area. Originally after the restructure, the 65, 67, and 372 were supposed to run every half hour on Sundays. Within a year, Metro increased the Sunday service to every 20 minutes, due to higher than expected demand. You cannot just dismiss NE Seattle as the kind of sprawl that exists in Marysville or Black Diamond.

        And NE Seattle still has some glaring transit deficiencies that, if fixed, would attract more people to the system. Roosevelt to downtown still takes an excessive amount of time when the 76 isn’t running. Link will fix that. There is still no bus that takes a straight-line path over the 45th St. viaduct, forcing everybody into a grand detour of campus. 65th to Northgate Way is a huge gap with no east/west service. Northgate Way lacks a bus that goes straight through the Northgate area, forcing all east/west traffic into a detour through the transit center (with a transfer to boot), adding 20+ minutes to every trip between west of I-5 and east of I-5. And Lake City has no east/west service at all, which goes further west than I-5. There is still plenty of opportunity to improve the system beyond turning 10-minute routes into 8-minute routes.

      14. Ross, Define “a lot of people” who live in Pinehurst who would like to go to UW every 15 minutes.

        There aren’t “a lot if people” in Pinehurst who want to go anywhere in the middle of the day. There just aren’t that many of them to start with.

        Are there more than along 40th NE? Yes there are, but once Link reaches Northgate there will be no reason for the bus at all. The bus that comes down 15th NE from Mountlake Terrace will be completely adequate.

      15. There aren’t “a lot if people” in Pinehurst who want to go anywhere in the middle of the day. There just aren’t that many of them to start with.

        Absolute nonsense. Read the census maps. Pinehurst consists of two census blocks, each over 10,000 people per square mile. That is density that literally does not exist in Shoreline, Lake Forest Park, Kenmore, Bothell or Woodinville. As it turns out, the main census block for Pinehurst is even more densely populated than any of the three stops that will be added in West Seattle for Link.

        Of course these people take the bus in the middle of the day. They ride the 41 or the 347/348. It is not that they all want to go to Northgate or Lake City, but it is the only option. Just read what asdf2 wrote — heck, just look at a transit map: https://seattletransitmap.com/app/. Now look at a census map: https://arcg.is/0SHn54. Now imagine trying to get from one place to the other. Pinehurst or Lake City to Bitter Lake, Greenwood, Phinney Ridge, Fremont or Crown Hill.These are all places that you can get to fairly quickly by car (some of them very quickly) but they take forever on the bus.

        Yes there are, but once Link reaches Northgate there will be no reason for the bus at all.

        What bus? What are you even proposing?

        Look, all I’m saying us “put the service where the riders are”,

        That is where they are! Look, the 512 runs every every 15 minutes during the day. It is a fine bus that will eventually be replaced by multi-billion dollar rail line. Yet there are only about 1,000 people per day that ride that bus north of Lynnwood (2,000 if you count on and off). Now look at buses like 41, 372, 67. The 41 and 372 are top twenty buses. The 372 and 67 don’t even serve downtown. They are all popular because they are frequent. Even the lowest performing bus in that bunch — the 67, which has a nonsensical button hook that drags down its performance — still carries over 4,000 riders. That is twice as many as the bus that serves as the motivation for Everett Link!

        It really isn’t that complicated. Provide a frequent network that makes trips from one part of *the city* to another easy, and you will get lots of riders. That has always been the case, the world over. Of course you get lower returns as you leave the more densely populated parts — but as anyone can tell, that dividing line is obviously right at the city border (you can see the city border clearly on the census maps just by looking at where density drops off). Places like Pinehurst, Lake City and Bitter Lake will have much higher transit ridership if you provide a more frequent, more straightforward grid (as asdf2 said).


      16. Ross, and as to NE Seattle (by which I do mean north of 45th and east if Roosevelt there are already frequent service buses on Roosevelt, 25th, 35th, Sandpoint Way, and NE 65th and 110th and 125th.

        Except for an east-west route at 80th or 85th which don’t go through, that’s enough.

        Maybe that is the problem. I still have no idea what you are talking about. Your definition of Northeast Seattle is still not clear. Is Lake City Northeast? Northgate?

        Again, you can define it how you want, but right now, it isn’t clear at all what you mean, both in your definition or your proposal. Just to be clear, no one has proposed that we increase service on the 65 or 75. No one has suggested an east-west route at 80th or 85th.

      17. Yes, people are definitely riding the 65 more as frequency and reliability has improved. I’m seeing more people per bus even as there are vastly more buses on the route. If no major changes are made, I’d expect the 75 ridership to go up more with Northgate Link because it serves Northgate directly from Lake City. The only problem I’ve seen with Northgate transfers (other than how crowded the 41 has been) is traffic congestion at Northgate Way. However there will be a net parking decrease there and it appears Northgate way is now 25MPH, which seems to have evened out the worst of the backups. Both routes’ ridership numbers are limited somewhat by lack of density along Sand Point Way and 35th Ave.

      18. OK, Lake City is dense, yes. But between 55th and 110th or so east of 15th NE is yes closer together houses than Bothell, say, but populated by the same rich people who use transit only for commuting to the CBD.

        To get from Lake City to the U-District the buses have to go through this area, so the few folks there who ride during the day get great service, far beyond what the could support themselves.

        Sooooo, when the buses in the northern half of the area (north of 75th) divert to Roosevelt from HSS, the hours freed should not be wasted increasing frequency there, but moved elsewhere to see if they can attract new ridership.

        If not, then people need to admit that most Americans don’t want to be bus riders.

        I agree that they SHOULD be bus riders, but without that nuclear war in the Middle East driving gas to $15/gallon they’re not going to do it.

        It doesn’t even occur to them to try the bus for non-commute trips

        “I wouldn’t even know where to catch the bus.”

      19. The 372 picks up and drops off a lot of riders along 25th, especially south of 65th. There are a bunch of UW students riding the bus all day in this area. Enough so the buses are often packed.

        Re-routing the 372 to Roosevelt station without replacement service along 25th would be a huge mistake.

        Speaking of restructuring, the service hours used by the 41. between downtown and Northgate adds up to quite a lot. There are some express buses it would also make sense to eliminate once Northgate opens, but in terms of service hours I suspect these are fairly small compared to the hours from the 41.

        Another effect Northgate link will have is to make some trips from fairly far west be faster via Link than the straight downtown route. The most obvious case here is the 45 from 85th & Greenwood via Roosevelt Station vs the 5. It may also be faster to backtrack on the 62 to Roosevelt Station from Wallingford and Tangletown than to ride it to downtown via Fremont.

    2. I good guess would be talking all the buses that are north of the ship canal that aren’t rapid rides and imagine them as feeders to Link. depending on how comprehensive the restructuring is but could wildly exceed any estimate.

      1. That… isn’t a good guess. Link won’t serve SLU for a long time and won’t serve Westlake or Eastlake ever; most of the routes that do also go north of the ship canal. Even if you count the 40 and 70 as RR routes because of planned upgrades, the 62 isn’t going anywhere (local service along Dexter is necessary), and I don’t think the 5 or 28 are changing any time soon, either (maybe the 28 would feed Ballard Link farther in the future; the 5 isn’t particularly close to any Link station and serves stops along Aurora that the E Line skips).

      2. I can see a truncation of the various peak only expresses that use I-5, but the local routes which cover areas that Link doesn’t go, no.

      3. Actually asdf, the peak hour expresses will be needed, at the very least until an Aurora corridor Link route is a reality. With the truncations planned for Snohomish County and the rapid growth in housing there, there won’t be much capacity in the peak for Link at Roosevelt and U-District. Some may free up at Husky Stadium as people de-board for the hospital complex.

        But overall, folks in Northeast and North Seattle will have great Light Rail service for three or four years after Northgate Link opens and then will be fighting for standing room on crush-loaded trains.

      4. Wow. I don’t know how you make the 62 RapidRide without distorting the term “Rapid” beyond all meaning. The 40 will be comparatively easy — all it needs to avoid chronic inbound backups is a continuous red lane from Leary/39th to Westlake/Lenora, and all the necessary space is in general-purpose lanes today. Obviously that would take some political will. To get the 62 around chronic inbound backups would take some real creativity (there is some hope in that the worst backups are mostly inbound and mostly during peak hours).

        That still leaves the 5. What’s it going to feed, the overcrowded E Line? Getting to U District Station, either by 46th/45th (cutting off some unique coverage of Upper Fremont) or by 39th/40th (after serving the Phinney/43rd dogleg), would take about as long as going downtown. Getting to a future Ballard station is only a little better.

        Anyway, if the 40, 62, and 70 all get called RapidRide, then the conversions are pretty obvious: the 41, and a handful of North Seattle and Shoreline peak-only routes that don’t feed the 41 because it’s already full, which is… what we’ve been assuming all along anyway.

      5. RR 62 is straightened out around Greenlake. I assume it will include transit priority on Dexter, and repaving and stop-dieting of potholed NE 65th Street. I question whether it should really be a single route, but that can be argued later.

        ST2 Link really only serves the east half of Seattle. In 2025 all all the downtown routes west of Latona remain: 62, E, 5, 28, D, 40. In 2040 they still remain except the 28.

        The 41 is not replaced by RapidRide. In 2025 the 372 becomes RapidRide, and the 41’s 125th-5th tail is incorporated into the 75, In 2040 the 125th-5th tail is split into two very different routes.

      6. @Mike: The 41 and various other North Seattle and Shoreline routes aren’t to be RR conversions, they’re Link-feeder conversions/truncations (what Ian was talking about and I was referring to). I think we’re in violent agreement that we won’t vastly exceed Link ridership projections via north-end route truncations, because all the big ones are obvious and well-known to the planners making those projections.

        The chronic inbound backups on the 62 I’m talking about are on Dexter, southbound, approaching Mercer. A lot of other parts of the route suffer from slowdowns that could be solved with a bit of SDOT attention, but this backup, which often stretches as far north as Galer, is a tougher nut to crack unless we manage to reduce overall traffic volumes in SLU.

      7. @Mike The 62 will be RapidRide too. 40, 62, 70, 44, 48/7, 7 to Mt Baker, 120. Any others I missed?

        Huh? I’m looking here — https://seattletransitblog.com/2015/12/21/rapidride-the-corridors/ — and seeing nothing that remotely looks like the 62. OK, I do see on the long range plan by Metro (http://www.kcmetrovision.org/plan/service-map/) that there is something similar to the 62 for 2040. But that is a very long way off, and it is unlikely that the bus map will look anything like that 25 years from now.

        What Al said is right. We will definitely have Link-feeder conversions/truncations on buses like the 41. The route map will be significantly different. Maybe it will look like David’s vision (https://seattletransitblog.com/2013/08/19/your-bus-much-more-often-no-more-money-really/) or mine (https://seattletransitblog.com/2017/03/12/north-seattle-bus-routes-after-northgate-link/), or more likely, something a bit different. No matter how you cut it, the same sort of approach that was applied to the UW/Capitol Hill restructure will be applied to this restructure. Truncations will be applied and extra service will be added. How that shakes out will be interesting. For the last go round, most of the service improvements went to areas that also lost their direct connection to downtown. That makes sense from a political standpoint. If your bus runs twice as often, then the horrible transfer is less of a big deal. But that isn’t necessarily fair, nor is it the most productive approach. Maybe it would have made a lot more sense to double headways on, say, the 27 or 14 (at least to MLK) instead of boosting service on the 65.

        Neither of the transfers (at Northgate or 65th) will be as bad as the one at Husky Stadium. That transfer is horrible not only because you have to cross the street (twice sometimes) and enter a very deep station, but also because a lot of the previous buses avoided that part of town altogether. If you used to ride the 73 to downtown, it would turn on 42nd and be on the freeway several minutes before you get off the current bus and wait for the crosswalk. That means a large portion (likely a majority) had a faster, just as frequent, trip from the U-District to downtown before Link got north of the ship canal.

        In contrast, the transfers at 65th and Northgate will be no big deal. They also get several “extra” destinations. As good as Capitol Hill Station is (third biggest station already — look what happens when you build the subway in the city) it is only one station. With Northgate, you have several extra stops. While many will whine about both the transfer and the extra time stopping at each one, the destinations along the way are all in the same league as Capitol Hill (i. e. popular). That means that unlike the U-District, a lot of folks in Northgate will not only welcome the change, but welcome it enthusiastically.

        Which means the truncation riches could be spread throughout the network. Who knows what will happen? What is reasonable to assume, though, is that Metro won’t do anything radical, like convert the 28 to RapidRide, or truncate it at a Link station. Sorry, ain’t gonna happen. Even after the 5 is converted to RapidRide (which is *not* on the Metro plan for 2040) you will still have plenty of all day, regular buses running across the ship canal.

    3. I believe it. I rode the 41 once in the peak of peak when it runs every 4 minutes, and it was still packed.

  3. I’m curious if Capitol Hill Station has a difference between boardings and leaving the train. I would think exits would be higher as it’s higher on the Hill. People are more likely to walk downhill and take Link back uphill — so the station probably is more used than the boarding-only data suggests.

    The same effect will probably occur between U-District and UW stations.

    ST should occasionally publish some sort of station exit data to illustrate this.

    1. I walk downhill and take Link uphill. But Capitol Hill is also a major destination because of the college and nightclubs, and one of the major unexpected factors in Link ridership has been Capitol Hill to UW. I knew U-Link would be a welcome alternative to the non-express 49 and 43, but even I hadn’t realized how much I’d be taking Link between Capitol Hill and UW — sometimes six days a week. I’ve cut down on it now because the 49 works better for my particular commute, and its increase to 12 minutes has made it more frequent, and I avoid it peak hours when it bogs down. So it looks like Metro was right to make the 49 the most frequent route on Capitol Hill, especially during the gap betwen U-Link and Northgate Link. And I wonder whether Metro’s long-term plan to split the 49 will result in less mobility.

      One group of people who use the 49 is some thirty people going to a homeless shelter at St Mark’s Cathedral. I’ve encountered them twice, once when i was on an early morning walk and saw they waiting for the first bus inbound, and a few days ago one evening when I saw them get on at 4th & Pike. How will it be for them transferring, especially because I assume many of them can’t afford Link or don’t want the hassle of fare inspectors. The question of whether splitting the 49 would be detremental to them depends on what their real origin is. Do they board at 4th & Pike because they’re coming from downtown, or because it’s where the 49 is? Could they just as easily go to 12th & Jackson instead and take a north-south bus? I don’t know.

      1. Rather than adjust a major bus route, wouldn’t it just be easier to see to it that these people have pre-paid ORCA cards of their own?

        Mark

    2. That would be interesting data. I would also guess that exits are probably higher than entries. Both for the topography and also the nightlife – you can’t take Link home late at night.

    3. The full report has entry and exit data (for both directions) for each stop. For the last full report, it was 1,861 on, 6,852 off northbound, and 5,545 on 2,195 off southbound. So, yeah, you do see exactly what you mentioned. Lots more people take the train northbound and exit at Capitol Hill then take the train southbound. But there are also a fair number who take the train one direction to the UW as well. For whatever reason, more like to ride the train from the UW to Capitol Hill than the other way around.

      Failure to “log off” might be one reason, but surprisingly enough, that doesn’t show up at SeaTac. I would figure that a lot of people get off the train at SeaTac, and start walking to their plane. But the numbers match up fairly closely.

      Anyway, there are two pieces of data that I really wish they reported. First is trip to trip, the second is timing. Both of these should be in the system. Every trip has an origination, a destination, and a time. I wish they would report this, so that folks could crunch the data. I think you might find some interesting things. For example, how many people travel just within Rainier Valley. Or how many take the train from Capitol Hill to the airport (and when). All of this should be in the system, and if they just released (in raw form) folks could easily generate interesting reports.

      1. Sea-Tac has “barriers” that visually delineate the fare-paid area and have readers next to them, acting as de facto fare gates. It’s a good reminder to tap on/off. Capitol Hill doesn’t, and readers are scattered around off to the sides at each entrance. I see many, many riders not tap off at Capitol Hill on a daily basis, and while there are several reasons that this may occur, it’s a very frequent occurrence. (I’d actually expect to see many more paper ticket holders at Sea-Tac than at Capitol Hill; those people of course wouldn’t be “logging off.”)

  4. is 520,000 right that would mean 75% of all transit trips go through University of Washington

      1. The press release was a mess. I wish they would stick to one way of reporting the data. Personally I prefer weekday riders. Better yet both weekday and weekend riders. I find the yearly numbers to be confusing, and seem designed to make things look huge. Like the number of pretzels eaten each year by all Americans. Yes, I’m sure it is enormous, but on average, not that many pretzels are eaten each day by each American.

      2. @RossB. Agreed. The press release was a mess and certainly was driven by a certain narrative that ignored the agency’s rather dismal performance in relation to its own target figures.

        (Loved the pretzel analogy btw.)

  5. Sound Transit actually missed their annual targets for ridership on all modes except for Link.

    Additionally, we are still a long way off from the 105,000 weekday boardings figure that the agency had originally projected for the main light rail segment by 2010.

    1. I’d think it would be pretty hard to get 100,000 riders on the main light rail segment when one of the crucial parts hasn’t even been completed yet. Once Northgate opens, we’ll reach that target.

    2. I’m guessing you are referring to the 1996 Sound Move plan, right? It showed 107K for light rail.
      https://www.soundtransit.org/About-Sound-Transit/News-and-events/Reports/Sound-Move/Sound-Move-original-documents

      Obviously, the 2010 target operating date isn’t being met — although it’s not what ST built. The 2010 segment was supposed to go to 45th St and not stop at UW. It also included a First Hill Station and a 45th Street station (and a Convention Place and BAR station) but 2010. If the segment had been built as assumed, I would guess that First Hill and U-District/45th St boardings would be as high at Capitol Hill — but I would guess that the nearby stations might lose a few riders. So using 72K as a base, adding in two more assumed stations like Capitol Hill 8K boardings apiece would be 16K more boardings (assuming that the other stations would add no ridership because of station losses at today’s UW and Capitol Hill stations because of the additional assumed stations). Then double that because it would induce about the same number of people exiting at these two stations — so that would be about 32K more boardings. 32K plus 72K would be about 104K, which would put the number pretty close, albeit a decade late.

      The same report forecasted less than 12.6K Sounder riders in 2010, and we’re seeing 17.6K in 2017; and 54K Express riders and we’re seeing 64K. Sure it’s 2010 numbers so it’s a little late, but it’s actually not far off from the 1996 forecasts albeit seven years late. That’s 15K more weekday riders elsewhere from the ST plans.

      If any blame should be leveled, I don’t think it’s with the rider forecasts. They look mostly valid. The blame rests with the unrealistic completion dates and the unrealistic construction analysis and costs (and associated need to scale back the project) from 1996.

      1. The referenced figure is from the Central Link FEIS, Section 3: Transportation Impacts and Mitigation (pub. 10/22/1999).

        Yes, I’m well aware that the Central Link as described in the ballot measure has not been completed as of today. I’ve been here from the start.

      2. >> If the segment had been built as assumed, I would guess that First Hill and U-District/45th St boardings would be as high at Capitol Hill — but I would guess that the nearby stations might lose a few riders.

        Not necessarily. Someone could take the train one stop. They might take the train from the UW to First Hill, then have lunch a few blocks north, and take the train back via Capitol Hill. It is the difference between urban and suburban transit. SeaTac — despite being a solid destination in its own right — lost riders because it was no longer the terminus. But Westlake (the north end terminus) gained riders (by quite a bit). For every rider that used to get on Link at Westlake (including those coming from the UW and Capitol Hill) there are riders who simply take it on the new section. The numbers support this. Prior to the change (when Westlake was the northern terminus) there were 7,000 riders a day there. Now there are 6,000 riders who head south (a significant decrease). But there are 4,000 new riders who take it north. More stops means more combinations.

        The same is true for Northgate Link. When it is added, you will see a substantial increase in ridership to Capitol Hill, as folks from Northgate, Roosevelt and the U-District have an easier way to get there. With the Husky Stadium station it is harder to say. Right now, it is pretty much the only way to get downtown from various parts of Northeast Seattle. That likely inflates the numbers. It also isn’t a great stop (as has been noted repeatedly). Depending on where you are headed on campus, it might be easier to get there via the U-District station.

        On the other hand, there will still be buses that serve the UW station. The station will attract riders headed to the hospital from the north end. Likewise, there are events at Hec Ed fairly often, and that should be worth a thousand or so. Simply taking the train from the U-District one stop south is a reasonable thing to do. If the stop wasn’t so bad, I would expect to see a major increase in ridership for the UW station — I’m less confident given its obvious shortcomings.

        But if we had the full combination, then all stations would see a rise. First Hill to Husky Stadium (AKA UW Hospital) is a very common trip that is probably done by shuttle bus or car a lot right now. Likewise, U-District to Capitol Hill is done as often via the 49 as via Link.

        It isn’t that the estimates for the most important part of our line were too high, it is that they simply haven’t built it the way they said they were going to.

      3. That’s very true, Ross. I was merely trying to be a bit conservative in regards to current station use compared to the studies in the late 1990’s (Sound Move).

        I suspect that there will be some UW students who are nearer to the stadium to take Link to get to the U District.

        With Northgate less than 10 minutes away from the two UW stations via Link, I expect lots of short trips between the stations at all times of day. I wouldn’t be surprised if parking management at Northgate becomes a major challenge as many Downtown college students will park there and go to class — even evening classes. Easing the looming parking complexities at Northgate is another benefit of having a more frequent bus to Link at Northgate as opposed to a less frequent and less reliable bus to Downtown. I realize how wedded riders are to the bus system but people forget that Northgate is further from the DSTT than Rainier Beach is — as well as Link from Northgate won’t be slowed headed to Downtown like Link from Rainier Beach is today.

        By the way, thanks for the directional data! My hunch about more exits at Capitol Hill is proven!

      4. A corollary of wide station spacing is that it makes one-station trips more common, not because people are changing their travel patterns but because it’s technically one station rather than two or three. From Westlake to Broadway & John) to UW, Link wins hands down. People take the 49 because U-District Station isn’t open yet, or because both their origin and destination are north of Link (e.g., Capitol Hill library to University Library or farmer’s market), or because one end of their trip is several blocks away from Link. The 49 works for that trip pair because the distance from John Street to Campus Parkway is 15-20 minutes by bus. If it were 40 minutes or longer, the bus wouldn’t look so attractive.

      1. Thanks for bringing it up, Oran. Because from personal experience that everybody else is as sick of reading about as I am of writing about, I’ve got an way to at least cut the shortfall with slight policy change.

        Make possession of a monthly ORCA pass complete blanket immunity from a charge of fare evasion. Instead of State’s evidence over a missed tap on leaving followed by one actually mandated for boarding .

        And then advertise the Hell out of it. Online and in front of every TVM, a pic of a fare inspector-lot of them really are a lot cuter than anybody their age in Hollywood -holding up an ORCA card and beaming at it.

        Meantime, really no polyester off my seat-cushion. All-Day paper ticket costs me a couple cents. Less chance of it falling out of my shirt pocket. Billing time for a thirty-second e-mail, ST CEO’s desk and wages should click LINK’s fare recovery blackwards with concluding key-click.

        But on the bigger screen, for seniors (“n” in Spanish type font, por favor) income -adjust fares to wages. And replace reduced fare, and Social Security- with a public service job at decent wages and benefits.

        Same deal for young people. So when AARP puts membership age at 13…very-long-lived ever-Greatening-American political BASE. Who, remember, will also have increasing choice of fare-evasion-prosecution-free Uber and Lyfft.

        Mark

  6. To put things into perspective, Link is now ~40 daily passengers away from surpassing the Minneapolis–St. Paul light rail system

    That’s a good comparison because the metro area population is almost identical. What would be interesting to know is how much of the ridership in each case is new vs replacing previous trips via bus. One would hope Link would be much higher since this area has seen far more new development. That would show TOD is working. If we’re just creating sprawl depots, P&R lots and Express bus transfers, then not so much. My hunch is the vast majority of the new riders/boardings are DT/Cap Hill/UW shuttles that are mostly walk-on or “city buses”.

  7. It’s amazing how Link ridership can be so close to Denver’s much larger system. RTD is almost useless for those who live in the city – it’s a commuter system designed to get people downtown in the morning and back to the suburbs in the evening. Link by contrast has already changed the way people get around Seattle and the extension to Northgate (let alone ST3) will blow almost every other LRT system away.

    I wish this was built decades ago but I’m really glad we’re building a high quality system. Despite all its flaws (and there are many) the next generation will wonder how we lived without it.

    1. Denver’s much larger system. RTD is almost useless for those who live in the city – it’s a commuter system

      Which is pretty much what “the spine” is going to be south of SeaTac and north of Lynnwood (or Northgate really). Pretty much Sounder Light. East Link is a mongrel. It’ll see plenty of use between Redmond & Bellevue downtown serving Microsoft and Bel-Red (aka The Spring District). But DT Seattle will be pretty much an upscale 550. Anyone going beyond Bellevue DT is probably better off via an SR-520 bus; assuming they still have the option. And it does nothing for Bellevue/Kirkland’s primary corridor, I-405.

      1. Of course, anything beyond city limits will be for suburban commuters. My point is Link is also incredibly useful for city dwellers. One can ride Link regularly without ever leaving Seattle city limits.

        There’s nothing on RTD that compares to the utility of UW-Westlake or Capitol Hill-Pioneer Square, for example. I’ve spent a lot of time in Denver and I’ve never ridden their light rail. It’s just not useful for getting around the city to the same degree that Link is.

      2. Very upscale. East Link will be twice as fast as the 550, almost triple as fast during rush hour. I think there will be solid induced demand on East Link as people realize it’s easy to quickly get between Seattle and Bellevue throughout the day, particularly in the evening.

        East Link will also function as an east-west “spine,” as a bunch of east side bus routes will be anchored by east link. Some of these will be truncations of routes that currently go into Seattle, but several routes such as RR-B will not change but will be much more useful once they have a direction connection to Link.

        “anything beyond city limits will be for suburban commuters.” – right, because land use in King county immediately changes when you cross the city line. When I look up at the office and apartment towers in Bellevue, I think to myself, if only this looked more like Rainier Beach or Maple Leaf, maybe then East Link would be justified.

      3. East Link will be twice as fast as the 550

        According to ST website documents Link will be 20+ minutes from International Station to Bellevue TC. They’re pretty squishy on the actual travel time as you have to add/subtract from different station pairs since they don’t what to give that most obvious of metrics directly. The 550 is scheduled 21-28 minutes. Of course Link will be faster since it’s booting buses from the bus tunnel. No doubt a few billion $$$ buys you a nice ride and better reliability over a small fixed corridor but I question the regional value proposition.

    2. I’m also glad we are (finally) building a halfway decent system and wish it was built decades ago. I also agree that the system, unfortunately, is very flawed (but a lot better than nothing).

      “…the extension to Northgate (let alone ST3) will blow almost every other LRT system away.” First of all, I’m pretty sure ST3 will significantly depress our performance numbers in ridership per mile or per dollar spent. Second, it blows my mind how we all toot our horns over these ridership and firebox recovery numbers. For example, “2016’s huge ridership”. Huge?? Let’s put these numbers into a different perspective. Bangkok’s elevated rail line (BRT) is just about the same length as ours, but it carries OVER 10x the number of riders per day. 10x!!! Not 50% more or twice as much, over 10x(!!!!!!!!!!!!) as much. When it first opened (and was shorter than our current system), people were disappointed in its ridership of about 200,000 per day which only covered operating costs, not the construction loans, about 3x our current “huge” ridership. They were upset that it only (only!) covered operating costs. We’re happy about 40% firebox recovery. It was also mostly privately funded, I believe.

      We need to have much higher standards in this country. Sure, you could point out that BRT is heavy rail, not light rail. The two aren’t comparable because heavy rail is a better system. Of course they are comparable, and the fact that BRT and heavy rail are better is exactly my point. Why are we building junk by comparison?

      1. One of my biggest peeves is when people compare Seattle to cities like Bangkok. They simply aren’t in the same league by any measurable standard. If you compare Seattle to any North American city, we’re already punching above our weight. And our system isn’t even close to being fully built out.

      2. Bangkok’s system has nearly ten times the riders because it has nearly ten times the population we do in its metro area.

      3. Sleeknub, most respect to Thailand, and its people. But brief as I can be, here are the real transit-related difference between our two countries:

        One, compared to most countries in the world aside from Russia, we’ve always had unparalleled space for expansion into territory with a climate and other circumstances comfortable for everyday life.

        And two, especially since the Second World War- whose main effect on our country was to end the Great Depression- the average person has had enough money to buy at least one car, and often more.

        Reason for advocacy-efforts like Seattle Transit Blog is that the sheer number of private automobiles is rapidly becoming chief limit to our freedom to travel for any purpose, be it work, school, health and hundreds more reasons.

        And our chief barrier to re-adjusting transit so it can perform like the Bangkok elevated, is that we have to convince millions of people to relocate their homes, and lives, so that the line-haul transit we all really need will work at all.

        Luckily, our government doesn’t have the coercive power it would need to make necessary changes by force. But neither is force necessary. Conditions that originally made “Sprawl” desirable have now changed in the opposite direction. Ours to take advantage of.

        Worldwide, I think many of our products have an excellent reputation. Honestly, does anybody think the industrial machines we make class as “junk”?

        However for things like transit- which for reasons above, which many if not most think we don’t need- economic imperative is the Path of Least Resistance. Forty years ago- exponentially expanding dispersion. Which is now becoming the Path of Least Tolerablility.

        Curious about your own trade. If it’s got to do with industrial design, good idea to visit some of our trade schools. Lake Washington Technical Institute is in Kirkland. At all transit-related- spend some time on LINK and the rest of our system. Sound Transit Express buses worth riding. Same for Sounder.

        And most important, get as close as you can to the people who operate the system. Explain where you’re from. If you’ve got any transit connection at all, you’ll have an “in” here.

        But behind it all, here’s basic truth about our country: Every bit of our damage is self-inflicted, and so repairable the same way. Any help with this, much appreciated. Welcome to the United States.

        Mark Dublin

      4. Sure, Bangkok is a bigger city, but the BTS doesn’t even come close to covering the entire city. As I said, it is almost the exact same length as Central Link, so it is serving the same total area (roughly speaking) as Link. You could say that Bangkok in denser, and that explains the difference. That is also a key part of my complaint. My complaint is exactly about the form/structure of many of our cities. This includes density and transit infrastructure. This is all within our control. We can and should massively upzone around all stations (and other places as well). To a large degree, especially now, we control the league we are in. Also, not sure what exactly you mean by punching above our weight, but I would guess several cities in NA are doing substantially better than us (NYC, for example, SF, probably Toronto and Chicago, but I’m familiar with their systems) when looking at per capita numbers.

        As far as my use of the term “junk” to describe our system, I am referring to the design (slow, no express trains, at grade sections, low capacity – although these things alone aren’t enough for me to call it junk), but also to the build quality. I just rode Link from the airport to downtown, and the train (which is less than 10 years old), was jerking all over the place (much more than I recall several years ago). Is it really deteriorating that quickly? There are plenty of noticeably older systems without these problems.

        Finally, I’m not sure if “Welcome to the United States” wasn’t meant to be taken literally, but just to set the record straight I was born and raised in Seattle.

      5. lol sleeknub…

        NYC is far beyond the weight of Seattle. Have you ever left the state of Washington? To compare Seattle with NYC is hilarious.

        Maybe try Minneapolis, Denver, Portland, etc.

      6. Although it is in the midst of implementing a massive master plan, Bangkok is a laggard in transit compared to its world peers (not Seattle). When its rail plan is it built out it will indeed be impressive but transit isn’t just rail, it’s buses too and that’s where Bangkok falls apart. Bus ridership has been declining for decades and the traffic is legendary on a scale comparable to or worse than Los Angeles.

        Just like Seattle, Bangkok was going to be first in its region to build a rail rapid transit system in the 1970s but politics intervened. The city has been inflicted by auto addiction since as evidenced by wide roads and copious amounts of parking at commercial buildings, even downtown. The central city’s street network has decent walkability but the suburbs are practically cul-de-sac hell.

        You don’t need to go far to find a decent transit city that’s about our size in North America. Just go north to Vancouver. They too had record breaking ridership last year. “In 2017, ridership in Metro Vancouver reached a record-breaking 407 million boardings! That’s 5.7 per cent increase over 2016.” Four hundred million! That’s over twice our record ridership!

      7. barman, note how I said per capita. NYC’s subway carries roughly 3/4 of the city’s population each day. Seattle doesn’t even come close. And yes, I have been to NYC, SF, Vancouver, Bangkok, and many other cities with far superior transit to ours (including some little towns in Japan). If I hadn’t I probably wouldn’t know enough to complain about our system. We should be building a system for the next couple hundred years that anticipates substantial growth, not a dinky, low-capacity train.

        Oran, buses in Bangkok are often packed in my experience, but it’s been a few years since I have ridden them extensively. If they build out a good enough rail system then buses won’t be as important. My guess is that overall transit ridership (bus + train, etc.) is up, and to some degree the rail is replacing bus service. They also have some transit options that don’t exist here.

      8. Packed buses don’t necessary mean higher ridership or even good service quality. The buses are unreliable, not accessible, incomprehensible and dangerous. They are cheap but as soon as people can afford a motorcycle or car, they buy one. Which is why motor vehicle ownership continues to grow. Even with full build out of the rail master plan there remain swaths of the city, especially outside the walkable core that will depend on buses to get to rail.

        While rail is now well over a million daily riders, BMTA bus ridership dropped from 4 million in the 90s to under 2 million riders in 2006. In 2016, BMTA sold less than a miilion tickets each day. Yes, there are private buses and songthaews, etc. but they are all affected by the same conditions that are hurting bus ridership. The city has grown by 3 million in that timespan. So per capita overall transit ridership is down.

      9. Oran, indeed the buses are inaccessible (I assume you mean for disabled folks), and for that matter so is the entire city. I’m curious why you think they are incomprehensible and dangerous, however. Google Maps works just fine for finding directions, and the buses list where they go on the outside of the bus. I agree that they are a little incomprehensible from my perspective because I can’t read Thai very well, but the Thai folks I know don’t seem to have a problem (I just use Google Maps, as I said before).

      10. sleeknub,
        Bus accessiblity is slowly improving thanks to activists demanding low-floor buses with ramps in the new bus order.

        The system is incomprehensible because the route network is needlessly complicated, having grown organically over time without thought as to how they work as a single network. That is compounded by a lack of basic information such as service hours and levels. People can learn their local routes to work or school but it’s not a system one can easily master.

        Bangkok is overdue for a reimagining of its buses. There are too many routes overlapping with each other so much that it’s difficult to depict on a map. I went to school there for ten years and I can describe much of the routes in my area but the rest of the city is a mystery to me even though reading Thai is not a problem for me. Try asking your Thai friends how to get to a part of town that’s not in their home area. Unless they are super bus nerds they will struggle. There’s no official system map online or in print, only third-party maps.

        Google Maps is an improvement over having to ask for directions in a forum as many Bangkokians often do but it’s incomplete at best. I’m a spatial navigator and I prefer to see all my options and build a map of the network in my mind rather than following a set of directions. This is where the rail network excels through simpllicity. It has fewer lines with direct trajectories.

        As for safety, the buses are old and poorly maintained and the drivers don’t always put safety first. Perhaps it’s the media sensationalizing it but you rarely hear of people falling off buses and getting hit or bus fires or bus crashes in Seattle. Thailand’s traffic safety record in general is poor.

      11. Oran beat me to it. Comparing Seattle to Bangkok , New York City or Paris is silly. Different leagues.

        But Vancouver and Calgary have much higher ridership than Link. Those are similar cities (Seattle is somewhere in between in terms of density) but like most Canadian systems, they have extremely high ridership per mile.

        They do it using an approach that most U. S. cities ignore, especially when building light rail (as opposed to heavy rail): They tend to focus their transit efforts in the core of the city, with lots of stations, knowing that proximity and density are the two big factors when it comes to transit success. They try hard to make sure the buses and trains work together. There are suburban stations, to be sure, but not at the expense of good urban mobility. Not that Vancouver doesn’t build things out of order (like we did) but when the UBC extension is complete, that will pretty much be it for Vancouver. You can easily argue they spent too much extending it to the suburbs, but nothing like, say, Dallas did. Even so, there is nothing wrong with excess in the suburbs, as long as the core is served well (and Vancouver serves its core very well).

        In American cities, there is a tendency to do the opposite. I can think of several reasons:

        1) People are impressed by size. It is actually backwards, but people assume that a longer system is always better. A series of small, overlapping lines in the central core can be very effective at moving huge numbers of people, as long as it integrates well with outer bus service (or streetcar service, as is the case in Toronto). But a system that barely leaves the city seems less impressive than one that goes for miles.

        2) People think in terms of driving, but not in a good way. Not only does Tacoma to Everett seems great, but it seems complete. After all, it is what the freeway does, and the freeway delivers. If it wasn’t for traffic, the freeway would be great and more than adequate (so the thinking goes). Why focus on areas like Ballard to the UW, or the Central Area to Lower Queen Anne — that is a relatively short distance, and traffic isn’t that bad. But again, that is backwards. Not only is traffic bad (for those trips) but is is much worse, day or night, then on the freeway. It is very aggravating to go 20 MPH on the freeway. But 20 MPH in town is normal (and it is worse than that during rush hour). It is also very fast for a subway. There is no thought to the fact that a subway line to, say, Tacoma, will make lots of stops, and thus be no faster than the commuter rail (during peak hours) or the bus (at other times).

        3) Focus on commuting. Again, this is part of “car-centric” thinking. Driving to work sucks. But taking fast transit is great. The problem is, this is not what drives most successful subway systems. Go to Boston, D. C., Vancouver, or any other city that has a great transit system, and you will find lots and lots of people riding the trains (and buses) all day long. They aren’t even going downtown. The same is true with our system. The number of riders who are going from Capitol Hill to the UW is over 2,000 by now (more than many of the stations) and that is only one of the few that can be measured. Imagine if First Hill was added. Now consider how many will go from Capitol Hill to Roosevelt, the U-District or Northgate. They won’t all go during rush hour (of course) which is a good thing, and what drives every successful transit line. It explains, for example, why commuter transit service is never more popular than inner city service, anywhere.

        4) It is cheap. Generally speaking light rail to far flung locations are often built on the cheap. No big tunnels, no big bridges, just plunk the rail down and have at it. Many of these light rail lines run right through downtown without any tunnel (even the very successful C-Train does that). This is also why it is common to run next to the freeway.

        5) Suburban and urban political interests are often at odds. There is a (relatively modern) tendency for suburban voters to feel like too much is wasted on the inner city. This can lead to a “me too” approach by suburbs, as well as cities in general. If Shelbyville has it, well then, Springfield should as well.

        6) Some big cities just don’t have much density. Phoenix is like that. It just sprawls and sprawls, with very little of a core. This means that even if you focused on the inner center, you would have a hard time getting good ridership.

        7) People assume that rail is always the best choice, everywhere. Thus cities (or suburbs) that would be served just fine by better bus service (and new right of way) have unpopular and expensive light rail instead.

        All of these go together. It is much easier to sell a system that runs next to the freeway to the distant suburbs than it is to build a line that serves the urban core with a subway, and the suburbs with better bus service. There is very little consideration for ridership per mile or even just how much time you will save. It looks great on paper — it is exactly the kind of thing that people imagine — but it is not a great value. Your return on investment often turns out very bad, and you wonder whether it might have made more sense to build something different, even when you are way more successful than average (See what transit expert Jarrett Walker says about MAX in the section labeled UPDATE here: http://humantransit.org/2009/11/brisbane-bus-rapid-transit-soars.html). In contrast, places like Boston, Vancouver and D. C. have their regrets, but they never wonder if they built the wrong thing.

        Our system, like many, is a hybrid. We are spending way more than most light rail lines, and have leveraged an expensive downtown tunnel. It does serve the core better than most U. S. light rail systems, but that says more about the flaws with most U. S. light rail lines than it does our ability to build things well. Just look at the numbers. After dinking around for years and years, we finally build the piece that should have been built first (UW to downtown). We fail to build a First Hill station, it doesn’t include the U-District, the UW station is in the worst possible spot, we still haven’t figured out how to connect the 520 buses to it, yet that section represents a huge part of overall ridership. The two stations that serve that small section are second and third in our system (higher than many downtown stops) and eclipse the ridership of three suburban stations (combined) even though one of those is the airport. It is not hard to imagine that a “complete” U-District to downtown section (with a better UW station, good 520 access and a First Hill station) would carry more riders than all the miles of track south of downtown.

        The sad part is, after building things fairly well (with ST1 and ST2) we are focusing on repeating all of the other cities mistakes (with ST3).

    3. “Which is pretty much what “the spine” is going to be south of SeaTac and north of Lynnwood (or Northgate really)”

      That’s the point: Denver and Dallas don’t have anything comparable to Capitol Hill or the U-Distrct in their rail network,or the lesser Roosevelt or Northgate. Dallas has one pedestrian neighborhood on it that may be the size of Roosevelt. So practically no inner-city dweller can effectively use those networks, just like those around Aurora and Ballard can only marginally use the Spine for in-city trips. This was a big strategic issue from the beginning: should Link serve only the suburbs, or only the inner city, or should it serve the suburbs and a few highest-volume regional centers in the city (and Rainier Valley for social equity). The decision was the latter, and that makes Link useful for many trips in eastern Seattle, but not as much as an urban-focused network would. Still, as everybody who saw the 71/72/73X struggling under the limits of capacity and congestion knows, and 41 riders, and those who wished for a downtown – Capitol Hill – U-District express that never existed, know, Link is making and will make a major contribution to those high-volume corridors that account for a lot of Seattle’s movement.

      Re the Eastside, I was initially skeptical of going through Bel-Red rather than NE 8th Street and Crossroads. However, Bellevue’s and Redmond’s commitment to density in that axis has really made it a good route for Link, and the apartments make it more affordable than most of the Eastside’s single-family areas.

      Re north and south, I have always been ambivalent about Everett, Tacoma, and Federal Way. I don’t see them as necessary, but if the regional consensus is for them, I don’t think it’s worth standing in the way. A seamless light rail network throughout the region will improve overall mobility, even if part of it isn’t strictly “necessary”. And Lynnwood is unusually important because of the hundreds of express buses it will replace. And it will get more ridership than you may think, because of the strong ties between Snohomish County and north Seattle. A lot of people in Snoho work, school, or recreate in north Seattle, and people in north Seattle go to Snohomish County. There isn’t any equivalent from the Eastside or south end: there’s no major destination before downtown except the airport. Federal Way/Tacoma have a double whammy with not only a longer distance to downtown but little in south King County to go to. The north end is the opposite of that.

      1. Lynnwood is unusually important because of the hundreds of express buses it will replace.

        If an extension is primarily to shorten an existing express bus route it’s a terrible investment. How much did it cost to extend Link by one P&R south of SEA? The billions in capital expense and the high cost of operation can’t justify peak only service.

      2. It’s not just peak service. The 512 does run all day, and often gets stuck in traffic, both along I-5, and in downtown streets, even when it isn’t rush hour. In the northbound direction, unpredictable delays getting through downtown propagate across the entire line, all the way out to Everett.

        The lack of reliability, in turn, makes transfers to infrequent buses in Lynnwood very hit or miss. If you’re lucky, you wait two minutes if you’re not lucky, you wait an hour. Besides improving frequency, Link will be much more reliable. With Link, you can actually plan a departure from downtown to have a 3-5 minute for a connecting bus in Lynnwood. With a bus, you just can’t.

        Furthermore, the money that Community Transit saves by truncating commuter routes during the peak period will buy a lot of extra bus service all day long. Peak commuter service is so much more expensive than all-day local service to operate, due to extra buses for just two trips/day, labor rules requiring minimum work shifts (which means a driver who works 2 hours might have to get paid for 4), more than half of the daily miles traveled being out of service (travel to/from the bus base), and loads of traffic padding added to every single bus schedule.

      3. It probably should have been a strictly urban system (not necessarily just Seattle though), with a separate, higher speed, regional system. Each node on the regional system, like Seattle, Tacoma, and Everett, for example, would then have their own urban systems that feed the regional system. They could choose bus or rail to fit their circumstances.

        Alternatively, we could have built the proposed system with the capacity for fast express trains plus more local stations to the north and south. As it is now the trains are too slow to cover big distances like Everett to Seattle well, and they also don’t have enough local stations to be used extensively for short trips on the north end and in South King County.

      4. That thing where the HOV lanes end at the start of the express lanes is usually a mess, but especially reverse peak it seems to me.

        When the MAX orange line was built, it cut a corridor in half, just like Lynnwood Link will. There was a lot of peak service in the section between Milwaukie and Portland, with bus routes 31, 32 and 33 all going there during peak periods. Replacing all of that with MAX does now require that people transfer if they are going downtown from those corridors, but it also allowed for increasing service levels on a number of routes that had never had that level of service before. What used to be the 31 on King Road is now an extension of the 33 and gets much more frequent service. A section of Johnson Creek Road has bus service now in an area that hasn’t seen transit service since the interurban line stopped seeing passenger trains in 1958. While people have to transfer in Milwaukie to get from the highway 99E corridor to downtown Portland, they don’t have to transfer to get to the King Road area, which is also a popular trip.

        So, let’s suppose instead of heading to downtown Seattle, the 512 stops at Lynnwood Transit Center, then heads through the community college and winds up at Edmonds? This makes a non-transfer connection that hasn’t been done before. It’s a shorter route than downtown Seattle, so you wind up being able to run it more often and run it later at night. Edmonds winds up with a reasonably quick way to get to Link, and so does the community college, but you also get a direct route from Edmonds and the community college northward to Everett. Because those buses aren’t stranded in traffic during peak periods, you wind up being able to operate this service until 11 at night, vastly improving Edmonds connection to Seattle and Everett. It means you wind up over-serving Everett – Lynwood until development along that corridor matches demand, but traffic levels will probably grow into it at some point.

        I’m not saying it makes sense to do this, but it is sort of transposing what happened with the Orange Line into Lynnwood. There’s an awful lot of bus service hours that don’t have to be consumed with peak period service if you replace, say, 5 or so peak period buses with a single Link train, and it allows you to reconfigure bus routes to do things you couldn’t do when they were oriented around a single corridor.

      5. There’s an awful lot of bus service hours that don’t have to be consumed with peak period service if you replace, say, 5 or so peak period buses with a single Link train.

        Peak only train service is way more expensive than buses for the same reason. You may get some bus hours reassigned from agencies that don’t foot the bill but in the end it all comes out of the same tax payer pocket. With the over reach of ST3 we’re starting to see the inevitable “enough already” reaction; even from tax payers in Seattle.

      6. “If an extension is primarily to shorten an existing express bus route it’s a terrible investment.”

        You start with buses and when they reach their capacity limit or traffic congestion become too unacceptable you upgrade to trains. Those hundreds of buses in horrible traffic with collision blockages at least once a week are a good reason for trains. Not only is the 512 all day and standing room only, but CT has an all-day route to UW. That’s hundreds of barrels of gas per year that won’t have to be used or can be redeployed to local see I’ve in Snohomish County, leading to more ridership and less driving.

        There’s another untapped transit market to the rest of North Seattle — Northgate and Ballard and northeast Seattle and such — that has no effective transit except 2-hour transfers, so of course people drive. But with Link having stations at Northgate and Roosevelt and reliable frequent service to 45th, plus feeders from there, suddenly transit becomes a viable alternative.

        Everett is much less justified because unlike Lynnwood it’s not in the center of the population concentration and doesn’t have 80% of the buses Lynnwood has, plus Link to Lynnwood will bypass most of the congestion. Link to Everett is in large part to attract companies to locate in Everett and Pains Field.

      7. “Peak only train service is way more expensive than buses for the same reason.”

        Twice-daily peaks are the same issue with both buses and trains. They ecost everywhere in the world, even in countries that have 5-minute midday service on all subways and buses. So we aren’t going to eliminate them. Moving people during peaks is essential for the economy and people’s participation in society. So that means extra trains and buses peak hours. A corridor train like Link can serve .multiple origins/destinations simultaneously, giving everyone 3-minute service rather than some waiting for half-hourly buses. That’s efficiency and usability. It also gives people flexibility to change their mind about their destination en route — whereas with buses you have to go back to the original bus stop or somewhere else to catch the other bus. So with a train you can easily stop for an errand or turn around en route. On other words, it enables kinds of transit trips that aren’t feasible o. A bus-only network, and that’s worth some of the billions on capital costs.

      8. “It probably should have been a strictly urban system (not necessarily just Seattle though), with a separate, higher speed, regional system.”

        Yes, two systems would be more effective than one hybrid system that’s mediocre in both areas. However, ST didn’t want to double the capital costs, and the Sounder track is not suitable for Lynnwood.

      9. >> It’s not just peak service. The 512 does run all day, and often gets stuck in traffic, both along I-5, and in downtown streets, even when it isn’t rush hour.

        Are you really saying that the 512 is slow compared to an urban bus? Do you really think the 44 or 8, for example, are anywhere near as fast? Holy cow, the 44 and 8 average around single digits *in the middle of the day*. If the 512 did that, no one would ride it. Yet more people ride the 44 or the 8 than ride the 510, 511, 512 and 513 combined. .

        Besides, no one is suggesting we should have just built a subway from the U-District to downtown. I can’t think of anyone that has a subway that small. Maybe a streetcar, but not a subway (with major tunnels being built). It just isn’t worth the investment to build such a short rail line (even if it might carry more riders than our current line). We already had a bus tunnel, and it would have made more sense to just improve that before building a tiny rail line.

        For the north end, it made sense to go to Northgate, which is what we did. It also makes sense to improve the connection from the suburban buses to the terminus. That could have been done any number of ways:

        1) Build a bus-only freeway ramp to the Northgate station.

        2) End at NE 130th, with a similar ramp. That has the advantage of connecting better from Lake City and the 522 corridor.

        3) End at NE 145th, with a similar freeway ramp. Again, that allows for a good connection from the north end.

        There is a major drop off in density and destinations once you leave the city. Thus going past 145th is a bad return on the investment. You are better off putting money into bus only lanes on the freeway. Speaking of which, that is exactly what should have been done north of there (all the way to Everett) instead of extending rail. If we couldn’t find the paint to convert HOV-2 to HOV-3, we could just build new lanes. It would be expensive, but cheaper than what we built. It would also be better for the bulk of the riders. That is because they would be able to skip the stops that they aren’t that interested in. The 512 is fast, but not as fast as the 510 if you are coming from Everett. A bunch of overlapping express lines that start in the neighborhoods and get on the freeway (like the 41 does) would better serve suburban riders. Except instead of going downtown (like the 41) it would terminate at the end of the light rail line. This would mean someone going from say, Ash Way to 185th, would have a slower ride but everyone else comes out way ahead.

        With a relatively short line to Northgate, 130th or 145th, it would also make sense to have a split. Run a train from Ballard to the UW and again, folks in the northern suburbs come out ahead. That is because the toughest part of a trip from Lynnwood to Ballard (via transit) is that ride from the UW to Ballard. Yes, the bus gets stuck in traffic getting to the UW, but nothing like the traffic (and traffic lights) a bus encounters getting from the UW to Ballard, First Hill, Fremont — these are all places that are tough to reach from the suburbs, and will remain so, even after we spend a lot of money “serving” them.

        That is because we aren’t building the right thing. A high capacity, grade separated rail system for the inner city along with good bus intercepts for the suburbs is much better — for just both city and suburban dweller alike — than what we are building.

      10. >> You start with buses and when they reach their capacity limit or traffic congestion become too unacceptable you upgrade to trains.

        That, in a nutshell, is the problem we have in the transit thinking in this city. Mike is way smarter and more knowledgeable about transit than just about anyone — and that includes the folks on the ST board. Yet he casually drops a myth that most people in this city would just assume to be true.

        You don’t need to build a rail line to avoid congestion. In this, of all cities, we should know that. You can build busways. It is almost always much cheaper, because you can leverage what exists already. That is assuming that you can’t just change the HOV-2 to HOV-3 lanes. It is absurd to think that we are willing to spend billions on a transit system; not because it is ideal, but because we don’t want to inconvenience the HOV-2 drivers. That is a dysfunctional political system.

        But you are absolutely right about capacity. The thing is, we are nowhere near capacity on any of our bus lines, let alone our buses serving the northern suburbs. The 511 is the only bus that is even close to approaching capacity (for some of those runs) and that is because it runs only every 14 minutes, and serves three suburban stops. You could run buses from each stop every five minutes and have plenty of room (and a much better trip for the vast majority of riders).

        A corridor train like Link can serve multiple origins/destinations simultaneously, giving everyone 3-minute service rather than some waiting for half-hourly buses

        Yes, but that is only relevant within the urban core. That just aren’t that many people trying to get from Mountlake Terrace to Ash Way (or any other suburb to suburb connection). Just look at the numbers on the various routes.

        510, Northbound: 20 people from South Everett.
        511, Northbound: 18 from Mountlake Terrace,
        511, Northbound: 14 from Lynnwood.
        512 Northbound: 23 from Mountlake Terrace
        512 Northbound: 66 from Lynnwood
        512 Northbound: 32 from Ash Way
        512 Northbound: 14 from South Everett.

        The biggest number by far is Lynnwood, but even that is tiny, and could be handled just fine by Community Transit. The vast majority of riders are not making suburb to suburb stops (or even trips to Everett) but are headed into Seattle. In contrast, Capitol Hill has over 2,000 riders heading north during the day (and that is only the connection that is available to be measured).

        Of course there is a cost savings to running a bus like the 512 (that stops along the way) instead of serving each neighborhood individually. But that is a service cost as well. The vast majority of riders are delayed (unlike Capitol Hill, where huge numbers board and exit). Meanwhile, running trains is very expensive, unless you have the demand to justify it. You have that demand in the city (in large part because you have that all day stop to stop demand) but you don’t have that in the suburbs. Ridership on the 512 is simply not that high. That means your farebox recovery is much lower, while you have provided a weaker system. It is still cheaper to operate, but not when you consider the enormous sums involved in actually building it.

        Again, no one is saying we should have built a bus-only system. But the rail system is stretched way too far (in every direction) while failing to solve core urban mobility problems that will haunt urban and suburban riders alike.

      11. “Are you really saying that the 512 is slow compared to an urban bus? Do you really think the 44 or 8, for example, are anywhere near as fast?”

        City streets don’t have 20-minute traffic bottlenecks every day, or lanes closed because of collisions a couple times a week. Travel time on city streets is at least predictable, whereas on on I-5 and other congested freeways it’s completely unpredictable when a bottleneck will occur and whether you’ll be sitting in it for 10 minutes, 20 minutes, or an hour. I don’t travel to Lynnwood much but I used to take the 71/72/73X reverse-commute every day, and sometimes go from Northgate to downtown in the afternoon, or take the 512 from 45th to downtown. You never know when you’ll get caught in a bottleneck or when to avoid the freeway (other than generally between 3-7pm). My 71/72/73X’s got caught in 10-minute bottlenecks on I-5 or the Stewart exit at least twice a week, and major bottlenecks (30-45 minutes)-a couple times a month. And even when they don’t happen there, they happen somewhere and affect somebody — just listen to the traffic reports on the radio.

        Good travel time is also relative to the distance. The average commute is 25 minutes, and 30 minutes is the threshold where people feel it’s starting to get excessive and it has negative physical impacts and cuts into people’s time with their family and community activities. Obviously there’s some distance at which 30 minutes is unreasonable, but we know from driving without traffic that downtown to Lynnwood is 15 minutes and the U-District to Ballard is 10-15. So that should be our goal for transit. We can’t meet it 100% because of intermediate stops, but we can do a lot better job than we’re doing now. Link will be 26 minutes to Lynnwood, which is pretty good and a reasonable goal. That means you can get from Lynnwood to 45th to Ballard in 40-60 minutes depending on the quality of east-west transit (subway, RapidRide with transit priority, or the current situation). That makes more activities feasible than if it takes 60-90 minutes or if you have to drive because transit is so minimal.

      12. Peak only train service is way more expensive than buses for the same reason.

        Link is a light rail line that will operate most of the day. It winds up being cheaper than the bus service itvreplaces.

      13. “You don’t need to build a rail line to avoid congestion. In this, of all cities, we should know that. You can build busways. It is almost always much cheaper, because you can leverage what exists already. That is assuming that you can’t just change the HOV-2 to HOV-3 lanes.”

        It’s not politically feasible to get those approved; we’ve tried to for years. It’s either trains or nothing. I’m not willing to go with the status quo for the rest of my life, so it’s trains.

        “Link is a light rail line that will operate most of the day. It winds up being cheaper than the bus service it replaces”

        Link’s operating costs went below Metro’s several years ago for short inner-city trips such as downtown to Beacon Hill or Beacon to Othello. That’s why ST has not had to raise fares as much as Metro, and Link’s fare is lower than Metro for the trips I’ve described. However, this principle diminishes as you go further out to the low-density suburbs where riders per operating hour are fewer. At some point it will not be able to catch up to Metro’s or CT. I don’t know where that point is.

        There’s also the complication of local transit vs express. Link has aspects of both, so you have to decide which kind of bus you’re comparing it to. Express buses with long nonstop segments can’t serve as many kinds of trips as those with more stops, and they may have to deadhead to meet unidirectional demand. Both of those can raise the cost per rider of express buses or long-distance trips. Everyone would agree that Lynnwood-UW should be compared to an express bus, but what about Lynnwood to Shoreline? That’s like downtown to Northgate or Rainier Beach, so maybe it should be compared to a local bus. (The agency switch and forced transfer at the county border is an arbitrary administrative issue, which only distorts calculating the cost.)

      14. However, this principle diminishes as you go further out to the low-density suburbs where riders per operating hour are fewer.

        While it is true that the riders per operating hour are fewer in the suburbs, the ability to operate fewer peak period buses and replace with a single train results in a cheaper operation. That’s how the MAX orange line allowed for the reallocation of a bunch of service hours from one corridor to another. Unlike ST, TriMet is all one agency, so that is real savings of operating hours.

        MAX also has a lot of low density as it goes out into the far reaches (by Portland standards) of the suburbs. MAX costs about $0.52 per passenger-mile to operate while TriMet bus service averages around $3.50 per passenger-mile. Furthermore, due to the construction of the MAX lines, more direct routes are possible than with the buses, so miles traveled to meet the same service is somewhat less.

        Sure, I know those costs per passenger mile aren’t exactly indicative of what it would cost for the exact same service operated with buses, but it indicates that TriMet is saving money by operating MAX on these corridors – even those very light density corridors – as opposed to the buses MAX replaced. The further indication is the sheer allocation of service hours. So many hours got saved by the orange line that bus route 33 alone nearly doubled in frequency and the span is better. This would not have been possible if the orange line had been more expensive than the buses it replaced. Sure, a single train with only several people on it going from Beaverton to the Airport at 3:30 am costs more than a bus would, but a bunch of the rest of the trains make up for it to the point where it is cheaper to have the train.

        Upzoning in Shoreline and north Seattle seem like it has been far more aggressive than anyone has tried after MAX lines were built here. 9 years after the Green line was built there is finally a new apartment complex being built along it. 32 years after the Blue Line was built there is a plan to put a fairly good sized development in place of a huge surface lot next to the Lloyd Center MAX station.

        I certainly understand the annoyance of not having the centralized urban network that would be really desirable to have in Seattle. However, just because the optimal network isn’t what you are getting doesn’t mean there will be no cost savings. If TriMet can save money and realocate operating money from lots of bus hours saved into light rail hours + more bus hours, even with the terrible land use along its lines, ST shouldn’t have any trouble doing the same.

      15. I don’t know the Portland system. I seem to remember that operating cost per hour is less than Link? Keeping in mind that ST always lists Link operating cost as per vehicle meaning the minimum 2 car train is double the reported cost per hour.

        Link, outside of the Seattle core (Northgate to SODO) replaces very few bus lines. East Link for example, one. The idea that you can replace many buses with one train applies only to a short period during peak. Otherwise you’re turning 15-20 minute headways into 30-60 minute service which defeats the purpose of a light rail system (i.e. frequent service). If demand all day demand isn’t there light rail can not pencil out. In reality, for light rail to be effective local bus service has to increase to feed the capacity of the rail system.

      16. Mike: It wouldn’t double capital costs. There would be a regional, high capacity rail system that should be cheaper than Link (many fewer stops, for one thing), then each community would chose the best feeder/local system for their situation. In many cases these would be buses, which of course are cheaper. In Seattle we would have a serious local “subway” system (possibly light rail, but grade separated) that covers the entire city.

        Instead we are using one technology for everything, one that is very expensive and only really appropriate for dense urban areas. It doesn’t provide local service to anyone outside of Seattle, Tacoma, and Bellevue (sort of), but also is too slow for longer trips from Everett or Tacoma to Seattle.

      17. “he casually drops a myth that most people in this city would just assume to be true”

        Where we disagree is that I also count political factors as a given. You’re looking at ideals or what a fully-enlightened government and public would do. That’s more or less what Germany and neighboring countries do: they start with buses, and when it approaches capacity/reliability limits they upgrade to rail and build downtown tunnels. In the US there’s an additional factor: various entities and groups that don’t understand or reject optimal transit values and land-use values have approval/veto power over what kind of transit we can build. I’m interested in what we can get running on the ground soonest: something I can use, and that other carless people can use. So I consider both the geometric factors and the political factors as immoveable, and make my recommendations within them. You’re right that we’re upgrading to rail faster than physical capacity limits require, and is unnecessary from an ideal standpoint. But when you add the immovable political factors, the limit on bus capacity “politically lowers”, if that makes sense. The public is more willing to build train lines than to convert HOV2 to 3 or replace street parking with transit/BAT lanes. Those are long-term issues, and we can’t just do nothing until they’re resolved and we can build BRT everywhere, because it may take twenty or forty years for a critical mass of people to change their minds.

        I think of it as the “ideal” vs “pragmatic” approaches. Ideally we should have a city subway network like NY or DC (the most grid-like ones) and a separate regional train/bus network like German S-Bahn or Metra. But pragmatically, people want a hybrid system because of supposedly lower capital costs, and the majority of the population (3/4 is suburban) doesn’t consider the city subway network important or want to pay for it, and they have veto power.

        I think it’s important when we argue for a position, to think about and say whether it’s ideal or pragmatic. Because the two are often different and we should have different attitudes about them. Ideally I will always advocate for the 2-tier network above, and will lament that ST had unrealistic estimates in the 1990s that led to severely-delayed U-Link and U-District openings. But practically, I spent forty years navigating Putetopolis on infrequent/unreliable/slow buses and I want something better as soon as possible. And practically, it’s not worth lambasting ST and SDOT forever over those dodgy estimates and missing First Hill station and doubtful streetcars — things that were decided years ago and we’ve told the agencies/politicians of our dissatisfaction. So now is the time to (pragmatically) move on and try to affect future decisions, and (ideally) lament what could have been like the Romantic poets and psalmists did.

        PS. Thanks for the unexpected complement.

      18. Putting aside the very cheap and apparently difficult task of changing HOV 2 to HOV 3, I disagree that building busways would be difficult from a political standpoint. Imagine if you did this:

        “Hey everyone, we’ve run the numbers, and it turns out, we can do much better with a busway. Imagine this. A bus goes from *your neighborhood*, by your house or at the very least, a park and ride close to your house. Then it travels in it’s own lane, until it gets to the freeway. Then it travels in its own, brand new lane, next to the HOV 2 lane, until it exits a brand new ramp at Northgate. From there riders will be able to ride the train to the UW, Capitol Hill, downtown — all of that. You won’t have to make any stops once you get on the freeway until you get to Northgate. This will actually be significantly faster from your suburb to downtown, to say nothing of getting from your house to work much faster. Here, let me show you this chart showing various neighborhoods in Snohomish County, and how much faster it will be to get there.

        Oh, and we’ve talked with the folks in Seattle. They want to pay for a new train line from Ballard to the UW. That means if you work in Fremont, Wallingford or Ballard, you can get to work much faster.”

        I can’t imagine anyone preferring the ST3 plan over that. If you can’t sell that, than you have no business being a politician. If you don’t realize that plan is better than what we built, you should have asked someone who actually knows a thing or two about transit.

      19. Then why do we have Link to Tacoma, Everett, and West Seattle instead of BRT extensions if it’s so easy? Why did we fail to convince the ST board, the Pierce and Snohomish politicians, and the public? ST suggested alternatives and they all said, “No, no, light rail only! Light rail or bust! We don’t believe that BRT would be adequate no matter what you say!”

      20. That’s simple. King County, mostly Seattle, primarily North Seattle, voted as a block to increase taxes on the entire region. It’s also skewed by ST shady tactics like giving public money to PACs and misrepresenting the amount collected vs the benefit delivered (i.e. over promise and under deliver).

        ST killed Eastside Rail with it’s slanted reports. Then, when it wanted ROW for rail to Kirkland changed it’s tune. There is no accountability with the currently mandated board structure. It’s literally foxes guarding the hen house.

  8. Pierce Transit is also reporting growth in their ridership numbers, which might be contributing to the increase in Tacoma Link ridership, as those are closely paired.

  9. Great to see it making progress! Still a long way to go though — for context, Boston’s light rail line (green line) alone is almost 230,000 daily riders with 23 miles of track. The rest of the subway system there is over half a million.

    1. The Green Line in Boston also has four branches that split apart at the edge of Back Bay. It’s like having four different light rail lines. Many segments are also in street boulevards like MLK but the stations are not as widely-spaced. Of course, only the line out to Newton is mostly suburban so overall it’s a more concentrated core system.

      If ST had created a light rail system that branched north of Westlake to SLU/Fremont/Wallingford/Green Lake, the Central District, UW and Northgate as different corridors with more closely-spaced stations, we wouldn’t have gotten to that 230K number but I think 150K could have certainly been possible.

      One cool thing: The Green Line has different loading platforms at Park Street as well as a turn-around, which does help operations so that they can push through more riders than a DSTT station in Seattle can.

      1. @al S, I’m aware of that.

        It was context that I thought could have been added to the main article comparing seattle with denver and minneapolis by miles of track.

        Also, responding to previous comments about seattle’s system “blowing every other LRT system away.

        It’s great that seattle is starting. But I don’t want to let boosterism get in the way of seeing things objectively.

        Personally, I don’t understand why we’re building a quasi-regional system using light rail, and I’d love it if someone who knows more than me could help articulate that.

      2. The problem is that in ST3 we are building a hugely expensive second downtown tunnel to carry one train line. There was no reasonable alternatives reviewed such as terminating the West Seattle line at the ID and branching one of the mainline trains to SLU.

        That second tunnel won’t be useful for 50 years and those funds could have gone to expanding the system outward with branching or terminating lines at IDS or WLS.

      3. “I don’t understand why we’re building a quasi-regional system using light rail, and I’d love it if someone who knows more than me could help articulate that.”

        The ST board decided on light rail in the early 1990s because it’s compatible with street-running. Heavy rail and monorail require grade-separated intersections and some have electrified third rails. At the time ST envisioned a lot more surface track — e.g. from Mt Baker to SeaTac — to keep capital costs low like the existing light rails: Portland, San Jose, San Diego. ST praised light rail as versatile: it can run surface, elevated, and underground, so they thought it would meet all their needs.

        So Rainier Vley got surface, but as the other neighborhoods went through design, one by one they all said they wanted a tunnel or elevated, abnd people said they were willing to pay taxes for it. By ST2 everything was grade-separated by default, although there was later battle backsliding in the Spring District and Redmond to pay for downtown Bellevue’s tunnel.

      4. “The problem is that in ST3 we are building a hugely expensive second downtown tunnel to carry one train line.”

        The second time is a good down payment for the future the way the DSTT was in the 1980s. By prepaying the cost of the tunnel, you keep it out of future lines’ budgets, and that makes it more likely the public and state lawmakers will approve the future lines — as happened in ST1.

        We can have one tunnel that may or may not become overcrowded, or we can have two tunnels with plenty if capacity for our current and future needs. It’s also insurance so that if something happens to the first tunnel and it has to close for maintenance or safety, we’ll still have the other tunnel and, if the tracks allow, lines can be switched to it. Seattle and Pugetopolis are large enough and growing that two tunnels makes sense. San Francisco has two tunnels. Chicago has two tunnels and an elevated loop.

      5. There are a lot of reasons that “light rail” is chosen in the USA.

        You can build a high platform subway system, but then it would be really difficult to do certain things with it. For example, you could never run it on a freight railroad line (New Jersey Transit RiverLINE shares track with freight traffic, but not at the same time of day) due to the high platforms interfering with freight traffic. High platforms make street running a bit more difficult, and while none of the ST3 lines are doing that, it is not inconceivable that some extensions might want to appropriate auto lanes and do that.

        The only real advantage to high floors is that it allows for a really open interior layout with space for a lot of standees. Link is busy, but huge standee capacity still doesn’t seem to be that necessary yet.

        I know that there are some that will point to high platforms allowing for lots of doors, but the reality is that with 100% low floor light rail cars becoming a thing, you can accomplish almost the same huge numbers of doors with a low floor car.

        Even now, though, Boston is running an awful lot of capacity (232,000 daily passengers on 23 miles of track) with a light rail line, so even with the car designs of a few years ago ultimate capacity really isn’t that bad.

        You can get higher maximum speeds with a high floor car design than with a low floor car, but Alstom’s current car for the North American market is a 100% low floor car with a design operating speed in the 62 mph + range.

        You would not be able to board buses and light rail trains at the same platform if the light rail line were built as a high platform subway style system. While there are certainly issues to joint operation of buses and light rail trains in the downtown tunnel, there are also advantages that shouldn’t be ignored. There may be other situations, such as the new West Seattle bridge where light rail and buses will need to be at the same platform.

        Several decades ago, there was a prediction that the Los Angeles metro, which was then still under construction, would probably be the last subway style system built in North America. There are simply not enough capacity advantages for most cities to want such a system – and Los Angeles was the last of the cities that would see any advantage to building one.

      6. “There are simply not enough capacity advantages for most cities to want [a heavy rail city subway].”

        Europe found the same thing. Some cities built “pre-metros” or “light metros” which are similar to Link, expecting to upgrade them to heavy-rail subways when the population/ridership increased enough to require it. But they found that the light metros were so scalable that they could accommodate the latter crowds so they didn’t need heavy rail after all. Seattle is the size of these cities: they range from 200,000 to a million something. Larger cities like Los Angeles and Chicago probably need full metros, but Europe doesn’t have many of those. (The largest are London and Moscow at 7+ million and climbing. Berlin is 3 million like Pugetopolis.)

  10. The fact that station is still named University Street when there is also now a University of Washington Station is beyond ridiculous! Its the little details that reveal how f*cked up things really are!

    1. How is that ridiculous? One station is on University Street, the other is on the University of Washington Campus (I take more issue with this name…it really should be named Husky Stadium station or something similar. The walk from there to the vast majority of classrooms is very, very long).

      1. I think the main issue is that it confuses out-of-towners. You could easily rename the station to Seneca to avoid this.

    2. If you think that’s bad, you should check out the Chicago ‘L.’ There are 5 stations named Western, 2 of which are on the same line, not to mention duplicate Kedzie, California, and Pulaski stations.

      1. Yes but probably none of them is a large regional university that gets a lot of out-of-state students and visitors every year. I saw the multiple Western stations, but unless somebody tells me to meet them at one of them, I as a visitor have no reason to go there.

  11. Is it conceivable that LINK will break 80,000 riders per day (or more) before the next northern extension opens? It’ll be really interesting to see just how many riders this single line can attract before East Link or Northgate and Lynnwood open. It looks really impressive to me to see such high numbers for only one line, and the numbers continue to grow. Wouldn’t be awesome if it managed to hit 100,000 daily riders before Northgate opens? It would certainly be a vindication of the investment thus far made.

    1. The trends for October through December at 8.3 percent (the first quarter with Angle Lake opened in both years) show that double-digit annual growth could be doubtful unless a new station opens. Keep in mind that U-Link ridership heavily consists of college students, who pretty much switched in the first few weeks or at the latest by the fall school term; that market probably won’t show much growth unless parking at UW becomes more difficult. Even in that time period, it took a few months for the Angle Lake parking to fill up so the 8.3 percent growth could easily be unsustainable. I think 80K is possible in a summer month in 2018 but I think it would be hard to get it beyond 80K as an average annual weekday condition in 2018.

      Of course, a variety of factors could change that. One thing that can probably quickly grow Link ridership into double digits is feeder bus restructuring,, especially reducing through buses Downtown. Say if SR 520 riders were forced to transfer to Link. Or if 550 riders or 101 riders or RapidRide riders were forced to transfer to Link. That doesn’t look imminent, but the looming congestion crisis in Downtown Seattle could put something major like that on the table and boost Link ridership before Northgate opens. That could be quite notable if one considers that the AWV permanent ramp closures could really change congestion patterns on Downtown Seattle streets.

Comments are closed.