Looking forward to 2020 and beyond

79 Replies to “Looking forward to 2020 and beyond”

1. “Construction begins on Ballard Link, West Seattle Link”

   Sigh… if there were any justice in the world we’d start NOW

2. The “Type 2” light rail vehicles have distinctly worse aisle design. Computer models and newer NYC subway cars favor a kind of diamond shaped front/rear facing seating arrangement for maximum ingress/egress speed. In contrast, the new Link vehicles seem to favor a “sardine can” maximum occupancy model that will likely increase station dwell times.

   1. The worst part about the new cars is every single trainset will have 8 driver cabs, 6 of which will almost never be needed. Before anyone says it’s for maintenance, Portland cabs have diver controls under a panel that can be used if needed, but that almost never happens.

      1. An incredible waste of both money and capacity. Seattle should have gone with open-gangway trains at full platform length. The existing fleet of type 1 trains could be used to run 2 and 3-car trains during lower periods of demand.

      2. Chris I: tbh I dont know of any transit system with low floor open gangway,, but someone can correct me on that. And the other issue in pushing for that is cost to expand said maintaince facilities to accommodate said expansion, which might involve acquiring additional land around current and future ONFs. And considering the hoopla that happened with OMF South planning I honestly don’t want to deal with trying to deal with that.

3. Ack!

   I am in a wheelchair and my office is in the building at 800 5th. (Top of the steep downtown hill.)

   The best way I have found to get there from Link is to rake the 255 from International District for the few stops until I get to the top of the hill.

   Any chance some other bus will take over this short coverage if the 255 isn’t downtown any more?

   1. I do hope the Fifth/Sixth pathway isn’t abandoned, but I haven’t heard of any plans to put anything else there except the current peak-direction routes. Alternatively, you could use the elevators in the Fourth and Madison building and the library to get up the hill from Third?

   2. If the 252, 257, or 311 work for your location and hours, they serve the same downtown stops (and will remain in 2020), and are also quite faster than the 255.

      If not, you can get off the 255 at Evergreen Point or Clyde Hill freeway stations and just wait for a 545, which is frequent all-day (and also not changing in 202 except to remove the Capitol Hill stop). 545 southbound goes to the same stops as the 255, and 545 northbound stops are one block farther away than the 255, but in the downhill direction.

   3. It should be possible to use building elevators to get to 3rd by cutting through the king county administration building and courthouse. I did this when I had jury duty. There may be other buildings you can cut through also.

      The 545 down 5th may also an option for now, but will likely get restructured away, once Link runs from Redmond to downtown.

   4. I usually opt to take the 5th Ave buses if I’m traveling to that cluster of buildings, but a good alternative is using the underground concourse to Columbia Center and traveling down to 4th from there. Plenty of buses from ID-Chinatown to the Cherry Street stop.

4. There was a project that was going to be a nice relief in the period between 2016 and 2022 with no major project openings, and even provide some relief in the longish term period of “Seattle Squeeze.” A project that shouldn’t even be on this list, RapidRide G (aka, Madison BRT).

   Madison has been talked about a lot, but discussion really kicked off in 2015-2016, and partial funding was included in ST3. And as of its passage, service was supposed to begin in 2019, the year that just ended (I had to double and triple check that it indeed was going to open in 2019: https://web.archive.org/web/20161111205054/http://www.seattle.gov/transportation/MadisonBRT.htm).

   There were of course unforeseen issues with procuring the special buses needed, but it seems that no progress at all was made on construction of Madison BRT since inception! The SDOT website now says that “Construction is anticipated to begin as early as [Wow, so early!] Summer 2020,” and I’m thinking what have you been doing for the last third of a decade, in which the entire project was to be completed? It seems taken for granted that construction will go smoothly, but of course that doesn’t always work out. And don’t forget testing. So issues with RR G could further delay the line to 2023 or 2024 (like how First Hill Streetcar was delayed from 2014 to 2016).

   So what was going to be a 3 year project in 2016, now a 6 year project, could realistically end up being an 8 year project, like the ST3 BRT and earlier Link lines.

   1. Something stinky is going on at SDOT, a federal lawsuit has just been filed against them. They need a complete restructuring.

   2. Projects developed before 2016 should have been revisited upon passage of ST3. Both Center City Connector and RapidRide G were developed before ST3. Much of the energy going for those projects were zapped in favor of much more significant projects in Downtown Seattle since 2016. They can be reenergized if their ST3 relevance is established and their project definitions are perhaps adjusted to connect better to Link. Without doing that, it’s going to be slow-going to reality.

      To open the second tunnel in 2035, plans have to be set by 2025 including acquiring properties for station entrances. The next five years will be packed full of expensive and dramatic decisions on this, and this momentum will continue absorb the energy away from these other Downtown transit projects.

      1. How should they have been revisited? I can’t think of any Move Seattle RapidRide lines that are redundant with ST3 Link. Madison, Delridge, Roosevelt, 45th, 23rd, Rainier, and the 40 are all still needed, and Link needs local shadows anyway because of its wide stop spacing. The CCC is its own issue and this argument is not going to sway its supporters; they believe 1st Avenue is a distinct corridor and we’ll be glad for one-seat rides from Pike Place to MOHAI and Pike Place to 12th & Jackson and the tourist satisfaction it will bring. In any case, Ballard and West Seattle Link are at least 10-15 years away, and maybe much longer or never, and we need something in the meantime.

      2. I agree with Mike. If ST3 included a station on First Hill, or the Metro 8 subway, or some combination of both (a Metro 8 subway that bent towards First Hill) then I would agree with you, Al. But there is nothing in ST3 that makes those projects any different. If anything, the station right on Madison will improve ridership up the hill (although I think the distance between those two stations is exaggerated).

      3. My comment is about political support and not project justification. Simply put, there is not much political enthusiasm for either RapidRide G nor the Center City Connector. As stand-alone projects, they disrupt the streets where they are planned so places like the medical buildings aren’t working hard to make the projects happen, and riders that would use them don’t perceive significant travel time savings and bicyclists don’t perceive much utility in either project. There simply is not a compelling advocacy out there beyond general transit advocates — and many of those advocates will want to spend time the much bigger ambitions that ST3 represents.

      4. The CCC would die immediately if it lost political support. That’s what’s holding it up: Durkan’s vision of it, and whatever business support it has (if any).

         Everyone wants the G to succeed; it fills a critical hole in the transit network. If politicians are quiet about it it’s because there’s not much to say — it just needs to be completed. We achieved center lanes in the middle third: the highest level of BRT the region has ever had. SDOT’s travel time estimate is 10 minutes from end to end (1st Ave – 28th Ave). If that succeeds it will transform transit in First Hill and the CD. Along with the 2 restructure (Pine-12th-Union), which is to replace the 11 and 49. Madison has also gotten a huge amount of densification in the past decade — on top of the thousands of medical jobs and patients, and elderly residents in highrise towers, and the earlier development in Madison Valley.

         Move Seattle failed most of its goals because its budget was overoptimistic, and SDOT has since tried to reprioritize what it can do now and the next step beyond that. That’s in the revised Move Seattle schedule. That shows the political priorities and SDOT’s priorities.

      5. What is this “energy” that is zapped by building light rail in the city? I can see how SDOT would be busy with negotiations with ST for construction, but RR G is an SDOT thing, not an ST thing. ST’s only involvement should be delivering the ST3 funds for the project. SDOT can build stuff while ST builds other stuff.

      6. The “energy” I’m talking about is the overall desire to improve transit in Downtown Seattle — as ST3 details continue to develop and as interests get more excited about the second light rail tunnel (like where station entrances go, how tunneling will be done and where 10-year staging areas are). For example, what will the expanded Westlake Station look like? Should an extra $100M for Downtown transit go to make the station better and more accessible like having an entrance north of Olive Street nearer to the Amazon HQ, or go to the CCC (and I’m fully expecting ST to run out of money to build-out suitable Downtown stations like we want so the City is likely going to have to come up with the funding)? Or should we build the Denny Link Station with great South Lake Union access, or get a more frequent streetcar or Westlake Avenue? These kinds of design and funding decisions loom in our future in the next few years, and they haven’t been seriously in the collective consciousness to date.

         Also, keep in mind that well over 100,000 new riders (150,000 according to ST) will be on Link (compared just about 80,000 today) in just five more years. Northgate Link is about 20 months away if the current schedules hold. Getting to and from Link is going to be significantly more politically popular in the very new future than in the very recent past.

         As a consequence of this, I predict that any interest in advocating the Central City Connector or RapidRide G will be less enthusiastic in the next few years than it has in the past. It could even take just one more funding shortfall or significant dispute or scandal to stall out either project indefinitely.

      7. Again, I’m talking about SDOT. They get their own tax revenues and have their own staff. ST is responsible for finding their own money (which may or may not include a funding deal with SDOT, but that hasn’t happened yet). They’re supposed to be the ones getting Madison BRT done, from their own budget process. They can’t just say “we don’t need to work on Madison BRT because East Link is more exciting,” and if they did then they need new staff.

         You’re basically arguing that all transit money is the same so it is being moved to ST3 projects, or that money spent on projects with less enthusiasm shouldn’t be spent, against the will of seattle taxpayers in Madison Valley who voted for funding in 2015 in hopes of BRT in 2019. To the contrary, Seattle has specific responsibilities to its taxpayers and voters, many of whom want BRT and won’t be helped at all by the more exciting projects.

         These voters are also the most important factor in getting future ST measures passed, and their support in large margins is necessary. It’s not wise to anger them, or make them feel like their money is squandered.

      8. @Al (not Alex) — That is a bizarre theory. The projects are largely independent and complementary. The second transit tunnel will be almost exactly where the first one is, except that it will have a station at Madison. It is weird to think that suddenly a line on Madison doesn’t make sense because we are adding a single station on Madison. For that matter, while I think the streetcar project is a waste of money, Link has nothing to do with it. The new line will have a new station at 5th, not 1st.

         I suppose it is possible that the city will run out of money for something Link related as well as something related to these projects, but that can happen anytime, for any reason. I see no reason to think that planning involving downtown stations (which shouldn’t be too exciting) will somehow steal the thunder and thus doom the other projects. That just doesn’t make any sense — that is like saying that East Link is in trouble, because Northgate Link will be a really big deal (or vice-versa).

   3. Yeah, Alex, it is frustrating. The Roosevelt RapidRide (now RapidRide J) is similar. It started a long time ago, yet it seems to be taking forever to complete. My guess is that money was the big issue (and the mismanagement surrounding it).

      1. The RapidRide concept needs to be abandoned altogether. Most of the money is spent planning to move curbs around, fighting challenges to curbs being moved around, moving the curbs, and then re-moving the curbs when the roads are rebuilt for separate projects (like what is currently happening on the C-line in West Seattle).

         The key improvements of rapid ride–signal/lane priority and increased frequency–can be implemented without expensive branding. The roads can be restructured whenever the scheduled maintenance is occurring.

5. A new push for a big business tax in Seattle + New Link lines opening in the next few years = This will be the decade Amazon spreads out even more. Link isn’t just opening up new options for commuters. It’s opening up new options for employers.

   1. Not just new lines, but also new stations. BAR in particular seems like a good place for someone to build out. If it’s not going to be a transfer point between Sounder and Link (which I think it should be) I think BAR’s main value will be as a bus transfer point as there really isn’t much to walk to – as I remember there’s the big grocery warehouse, a gas station, a firing range, a few other businesses along East Marginal, and that’s about it.

   2. As long as they’re in station areas that are as walkable as downtown Bellevue, they don’t need to be in downtown Seattle. Lynnwood, Northgate, and Federal Way are also looking for employers. Amazon is still as big in Seattle as it was; it just reversed on a building it hadn’t occupied yet. Downtown/SLU/First Hill will still be several times larger than any other office district even if it doesn’t grow. And it was doing fine in the 2000s before Amazon started expanding, so it doesn’t need to be as large as it is.

6. Instead we just get excuses about how they are going as fast as they can.

   1. So? Maybe they are. Are you a construction engineer?

7. RapidRide letters:

   G Madison (maGison), 2022)
   H Delridge (delriHdge), 2021
   — Northgate Link —
   I Renton-Kent-Auburn (kInt), 2023 (1 year before Federal Way Link)
   — East Link —
   — Federal Way Link —
   — Lynnwood Link —
   — 405 Stride — (same year as Lynnwood Link)
   — 522 Stride — (same year as Lynnwood Link)
   J Roosevelt/Eastlake (rooseJelt, eastJake), 2024 (3 years after Northgate Link)
   K Kirkland/Bellevue (Kirkland), 2025 (1 year after East Link)
   R Rainier/Jackson (Rainier), 2024

   The years may slip depending on budgets, of course.

   1. “The years may slip depending on budgets, of course.”

      Or they may slip simply due to lack of progress. Case in point: Sound Transit’s Q3 financials show a considerable underspend on capital projects through three quarters for 2019 (79% of YTD budget). The underspend issue is particularly noticeable for the Lynnwood Link project which, imho, could easily have its revenue service date slip to Q1 2025. And this is despite a favorable economic environment for the year and a drier than normal spring and fall.

   2. Existing letters: highwAy 99 (or federAl way to the Airport and tukwilA), Bth street and 15Bth to crossBroads and reBmond, west C-attle, ballarD, aurEra to the county linE, buFien-one FiFty Fourth-tukFila-southcenFer-sounFer-renFon-lanFing. Hey, they aren’t much further stretches than yours.

   3. Yes, I didn’t post my A-F because I’d done it before. seAtAc, federal wAy, tukwilA; Bellevue; C-attle; ballarD; Eurora, aurorE; F*ing renton.

8. Don’t count all of the Rapid Ride lines as BRT. That’s just not what they are, at least in the current plans. Portions of them may be BRT. But we’re not getting 116 miles of BRT in the next decade, I wish we were but we’re just not. That said, the bus corridor upgrades will still be substantial and very helpful!

   1. I agree. I was going to point this out, but figured it wasn’t worth fighting over semantics. Most of the routes would be considered “Not BRT” using objective standards (like those used by ITDP). The only one I would bet on would be Madison BRT, but even that could get screwed up (SDOT doesn’t exactly have a great track record).

      1. My understanding is that by the BRT standard, projects have to be completely isolated from turning movements, meaning bus lanes on the side of arterials are out unless turns across the bus lane, even to access homes and businesses, are banned – if you can’t get a completely dedicated right-of-way (at which point you might as well run rail), the best approach is to run in the median of a two-way road, as on Madison, or on the side of one-way roads. So even Swift on 99, which I would consider the gold standard of BRT in the region, wouldn’t qualify as BRT at all, and building real BRT would require more investment and taking away space from cars than the region is likely to have the appetite for in most places, at least in the near term. Madison and *maybe* Stride (and potentially only 405 Stride at that) are the two bits of real BRT in the plans.

         If the region were up to the task of building real BRT, there are certainly opportunities – I’d love to see Aurora’s 50s-vintage Jersey barrier median be replaced with a busway – but who knows when they’d even show up on the radar. Of Metro’s original six RapidRide lines (which they stated they wanted to eventually upgrade to BRT Bronze status when they did their Metro Connects 2025/2040 planning some years back, though my analysis suggests even Madison would only *barely* be Bronze) only the F line faces major obstacles to becoming true BRT, with its many turns east of I-5 and with 156th/154th topping out at three lanes wide with no parking to take away for most of its run west of 99. Delridge is pretty wide for most of its length though it does run into pinch points further south, especially in places where BRT would leave it. The I is okay but could have problems finding room for stations on the Talbot Rd segment in Renton (if that were kept from the 169) and might have to accept leaving the busway to get to and from Sounder stations. Roosevelt-Eastlake is okay but a lot depends on whether or not they want to deviate off the Roosevelt/11th couplet in the U-District; the Campus Parkway-University Bridge interchange is surprisingly difficult to work with for this purpose. The K would likely require widening streets and/or taking away bike lanes north of downtown Bellevue and possibly on 145th Pl in Lake Hills. Finally, Rainier is five lanes wide for most of its length but I don’t even want to know how a BRT busway would interact with the streetcar tracks on Jackson, a deviation even the 9 makes, and if it were to continue up Boren rather than go downtown like the 7 and current RR R proposal, it doesn’t have a full-width left turn lane south of Marion so there would be challenges in inserting stations near Harborview and Yesler Terrace.

      2. It’s not BRT but I’m exhausted by the controversy. Swift is the closest to BRT in the region: it has full BAT lanes (at least the blue line does), limited stops, a local overlay, and guaranteed frequency (sort of), so it’s really something more effective than a regular bus route, the “missing middle” between local and express. It’s a “poor man’s” light rail as I think of it.

         Rapid Ride has hardly any lane priority and is the only route in its corridor so it has to stop every ten blocks, and in some places it stops excessively every five blocks (73rd-105th, Shoreline). In Shoreline’s case it zoned urban villages at all the RapidRide stops. Seattle didn’t, and is afraid of taking street parking for BAT lanes.

         There are objective scales of BRT, and RapidRide is below the bottom of all of them. However, the middle third of Madison will at least get us somewhere and be an example. And Metro says ridership keeps growing on RapidRide so the C, D, and E are in Seattle’s top ten routes in spite of their limitations. Metro also says the suburban communities have always been satisfied with their RapidRide lines; Seattle just has higher expectations.

      3. Swift is the closest to BRT in the region

         I’m not so sure. Right-of-way isn’t everything. The stop spacing on Swift is often too large, frequency is low, ridership is not that high, etc. My guess is that the E Line comes out the best, although it still wouldn’t be considered “Real BRT”.

         My understanding is that by the BRT standard, projects have to be completely isolated from turning movements …

         That is just one part of the scorecard. It gets pretty complicated (https://www.itdp.org/library/standards-and-guides/the-bus-rapid-transit-standard/the-scorecard/). Besides, that is just one agency’s measurements — who is to say that they have the best scorecard. You could do the same thing with light rail (or heavy rail) systems. You get credit for going fast and having really busy stations with good pedestrian access and connections with other transit. You get deductions for breakdowns, low frequency, and failure to cover the most densely populated places. Very little in the U. S. (outside the older systems) would grade well in buses or trains, I’m afraid, which explains why we trail most of the world in transit use per capita. Its frustrating — It is like being the Knicks. We should be better; we spend a lot of money; but we always seem to be at about the bottom of the league.

9. Bruce writes re Eastside service: “The updated Eastside network will take a bit of time to get used to, but should prove handy until the next big change in 2025 with the arrival of the RapidRide K Line.” It seems that 2021 and 2023 will also have opportunity to improve the bus network. In 2021, Link will have four-car trains; does that not imply less resistance to restructuring routes 252, 257, 268, 311, and 545 to meet Link at the UW station and avoid the congestion of the I-5 general purpose lanes? Does the extension of Link to Roosevelt and Northgate imply that routes 542, 555, and 556 could be restructured? In 2023, East Link will serve MI, Bellevue, and Overlake; that should prompt significant bus service changes to the local network and I-90 services.

   1. There will definitely be big changes from East Link, but I think he was referring to the local Kirkland / Bellevue routes, and not so much expresses to Seattle.

      What seems most likely to me is that…
      *545 will be canceled in favor of East Link
      *544 will get a boost with some (but certainly not all) of the 545’s service hours
      *542 and even 541 may get more trips as well
      Peak I-90 bus service ends at MI or S. Bellevue, probably split between the two
      *554 is changed to go to downtown Bellevue, replacing the 550 on Bellevue Way and connecting to Link at S. Bellevue.
      *555 will be deleted
      *556 will probably be deleted, even though I strongly favor running it from UW to Bellevue, Bellevue Way, and Issaquah replacing the 554 and 520 portion of the 271. Taking link from Bellevue to UW is taking riders along 3 sides of a rectangle, and is slower than even most of today’s slow 271 trips.

      1. If you change the 554, what bus takes you to Issaquah?

      2. AlexKven, re 2023 and I-90, MI is far superior to South Bellevue due to speed and congestion on I-90. In 2021, ST could have a simpler and more frequent SR-520 network. how about Route 542SLU, Bear Creek to SLU via UW station and I-5 or Eastlake (full time and frequent); Route 545W, Redmod to U District, (weekday and frequent); Route 556W, Issaquah to UW via BTC; and, Route 522, Woodinville to SLU via Roosevelt. There is no longer a ST need to serve downtown Seattle, Green Lake, or Northgate. Feed Link. if the rest of the west is completed, the SLU service could be sped up, but it would miss the U District.

      3. Mostly agree, but, I think post east Link, the 541 will be dead, and it’s budget shifted to the 542.

         The reason for the 541 being a separate route to begin with is to mitigate the loss of parking at OTC during eastlink construction.

      4. For Issaquah, the most efficient choice depends on the time of day. Peak hours, a Seattle bound commuter has a strong preference for Mercer Island, but off peak, when the HOV lane doesn’t matter, the savings is only a minute or so. By contrast, Mercer Island is a much bigger detour bfor people going to Bellevue than South Bellevue is for people going to Seattle. It’s also slightly less service hours.

         At the end of the day, I think peak hours, you run buses to both places, but off peak service connects to Link at South Bellevue. Most likely, that means a 554 to South Bellevue P&R and a 216/218 to Mercer Island.

         What I’m undecided on is whether the Issaquah to South Bellevue bus should continue in to downtown Bellevue, or just terminate at South Bellevue and force a transfer to Link. My gut says it’s better to truncate and force a transfer, in exchange for more frequent service.

      5. For Issaquah, the most efficient choice depends on the time of day. Peak hours, a Seattle bound commuter has a strong preference for Mercer Island, but off peak, when the HOV lane doesn’t matter, the savings is only a minute or so. By contrast, Mercer Island is a much bigger detour for people going to Bellevue than South Bellevue is for people going to Seattle. It’s also slightly less service hours.

         That is the problem for Issaquah as well as much of the East Side: some of the people are headed to Seattle, some are headed to Bellevue (and Redmond). The same problem exists for Kirkland, and it weakens the system.

         Anyway, I think it makes sense to terminate I-90 (AKA Issaquah) buses at Mercer Island during rush hour. But there should also be express buses from Eastgate to downtown Bellevue. The buses wouldn’t start at Eastgate, but they would start at some other location (BCC, or a park and ride in Factoria, for example). There may even be a mix — but there should be very good frequency from Eastgate to downtown Bellevue (via the freeway). This is what makes sense during rush hour — both directions. It is possible that a bus would deadhead back to a Factoria park and ride, but the key segment (Bellevue to Eastgate) would be bidirectional (and frequent) during peak.

         Outside peak it gets trickier. The cheapest thing to do is just send them all to South Bellevue. But that become messy. For example, let’s say I live in Issaquah, and stay late at work in downtown Seattle. I have to know the system, otherwise I’m stuck in Mercer Island. I get off the train, looking for my bus, only to find that the last bus has left. I have to wait (ten minutes perhaps, at least six) for the next train, to transfer to my bus. That’s not good.

         The problem doesn’t exist going the other direction. In the morning, you suddenly find your bus going to South Bellevue. Big deal — either way you transfer at Link. This means that it is possible that while you don’t send all the I-90 buses to Mercer Island, you do send all the buses from Mercer Island. That is actually pretty easy to understand for everyone — not just the folks who work in downtown Seattle. If you work in
         Bellevue, your eye is on that express bus to Eastgate. If you missed the last one, you have to take a train to Mercer Island, and transfer there. (Or slog your way on the equivalent of the 271, and catch your bus in Eastgate).

         Then you have folks from Renton/New Castle. They have the opposite problem. The easiest thing to do is simply send all the buses to downtown Bellevue. The bus is on the 405 HOT lane, and will get there very quickly. But that is terrible for people trying to get to downtown Seattle. So you send the bus to South Bellevue. That does nothing for the folks who happen to work in Issaquah, while also slowing down those headed to downtown Bellevue (and gaining little in service savings). It is a tough problem, but I’ll leave that for another day.

      6. asdf2,

         That all basically makes sense. I believe (from memory, not totally sure) that the 554 is going up to downtown Bellevue via Bellevue Way, which I actually think makes sense even though we want people to transfer, mainly because it’s replacing existing ST service on Bellevue Way (the 550). I think a lot of people will transfer because Link will prove to be faster, but if the bus is already going there anyway, there’s no harm in letting people ride along.

         @RossB,

         I think it’s confusing to have a bus that goes to S. Bellevue sometimes and MI other times. I think it’s easier to have people remember their bus route and take it, and look up the schedule. If the 554 goes to S. Bellevue and the 214/216 go to Mercer Island, but the 554 runs all day, I think more transit nervous people might gravitate to the 554 (but catch a 216/214 if it comes first), but always take a 554 on the way back (which is fine, because the 554 will always go to S. Bellevue, even in peak).

         And I don’t think the fact that some people go to Seattle and others to Bellevue makes the network weaker. I think it makes it stronger. It’s harder to serve everyone when we have a rigid set of express buses. That means one of the groups has to lose (and it’s the Bellevue commuters). But because there is a demand for a wide range of commutes via transit, that’s what makes it so we get trains with stops that people can transfer at (or gridded routes in general. The 48 isn’t so frequent because that many people want to go from Mt. Baker to UW, but because lots of people can take it along with another bus to go lots of different places). Having the 554 drop people off at Link instead of Seattle means that both Seattle and Bellevue commuters can win (whether or not it goes to S. Bellevue or Mercer Island, since that difference is only a few minutes, and beats the worst part of traffic however you go). And when it gets easier for both sets of commuters to take transit to work, much more Bellevue commuters will try the new 554 and Link, and not look back.

      7. Why not run the 21X routes to Mercer island and the 554 to South Bellevue? The peak oriented KCM express routes are optimized to serve Seattle, and the all day STX route optimized to serve Bellevue. Added benefit is the 554 previews the future Issaquah Link, because in theory a core mission of STX is building ridership on future HTC corridors.

         Serving Bellevue Way itself between south Bellevue and downtown Bellevue is important, but it’s gridlock during rush hour so that’s the job for a local KCM route, not an express. Running the 554 there is just lighting service hours on fire. A 40′ will do just fine.

      8. Agree. The transfer at South Bellevue to continue east on I-90 needs to be reliable and quick. The bus’s punctuality shouldn’t be subject to traffic jams on Bellevue Way.

         For service on Bellevue Way, a KC Metro route makes the most sense. Ideally, you’d want a route that won’t affect very many people if it gets stuck in traffic. Perhaps the 249 is acceptable, given the alternate option of just walking to East Main St. station and catching Link there.

      9. I think it’s confusing to have a bus that goes to S. Bellevue sometimes and MI other times.

         I agree, they would be different numbers. It would be similar to lots of buses in our system. During rush hour, I can catch the 76, which is an express from northeast Seattle to downtown. Outside of rush hour, I need to take another bus, and transfer.

         And I don’t think the fact that some people go to Seattle and others to Bellevue makes the network weaker. I think it makes it stronger.

         Service is spread between the two destinations, which means that frequency is reduced. It wouldn’t be an issue if East Link curved further east, and served Eastgate. Then all the buses in the area could serve Eastgate, making it a major transit hub. But that isn’t the case, and the geography of the stations make it very difficult for one bus to serve the two destinations. As mentioned, serving Mercer Island hurts riders headed to downtown Bellevue. Serving South Bellevue during rush hour hurts riders headed to downtown Seattle (and still isn’t great for those headed to downtown Bellevue). There is no single stop that works for both, and that is the problem.

         I just realized, though, that my idea won’t work. There is no good way to have an express from Eastgate to downtown Bellevue. Once you exit at the Eastgate stop (https://goo.gl/maps/rQ7AJNeLyJ4TmL6K6), it is too late to go northbound on to 405.

         That changes the dynamic. I think there will be three types of buses:

         1) Bus from Issaquah/Sammamish to Mercer Island (via I-90).
         2) Bus from Issaquah/Sammamish to South Bellevue (via I-90).
         3) Bus from Eastgate to South Bellevue (e. g. https://goo.gl/maps/U8qw8c1T4CTZ8zAx6).

         Ridership to downtown Seattle is much higher than ridership to Bellevue. The second bus should only run during rush hour. Folks from Bellevue who get off at South Bellevue and miss their express will use the third bus, then catch the first bus at Eastgate.

         It is also possible that the second bus doesn’t exist. The 556 does not have many riders, and that is with a trip directly to downtown Bellevue. If the other bus runs more often, then folks will just transfer in Mercer Island (they will spend more time on the train, but less time waiting). If you run the 556 a lot, then it is quite possible that it won’t pick up many riders. While it would be a bit of a detour to go all the way to Mercer Island and then back, increased frequency may be enough to make up for it. It is extra five minutes on the train, but that isn’t the end of the world compared to waiting a half hour for a bus.

      10. Assuming that there’s only one all-day express bus from Issaquah, it’s a tradeoff between a larger detour for smaller numbers of people vs. a smaller detour for larger number of people. Using Google driving directions as a guide, Eastgate Freeway Station->South Bellevue via Mercer Island is about a 6 minute detour, whereas Eastgate Freeway Station->Mercer Island via South Bellevue P&R is about a 1-minute detour.

          Yes, there will be more people headed to Seattle than Bellevue, but 6X more is a lot. And, of course, the transit service doesn’t just react to ridership demand, it also influences it. Today, almost nobody does the 554->550 transfer because it takes so *** long, and you might get stuck at the bus stop for a long time. But, if it’s a frequent bus connecting to Link, more people might do it, especially when the Link train also continues to Redmond, adding more potential destinations to the mix.

          At some point, we need to move away from mindset that transit is exclusively about getting downtown, and every second saved getting downtown trumps every other consideration. Other trips need to be considered too.

          Of course, Mercer Island has the advantage of allowing the bus to stay in the HOV lane, but that only really matters during peak hours. That’s why I envision the 200-series buses going to Mercer Island during peak hours and the 554 going to South Bellevue P&R all-day.

      11. I don’t think an extra express bus from Eastgate to Downtown Bellevue will be needed. Getting from Eastgate to Downtown Bellevue will be served by RapidRide K, as the post a few days ago explains.

      12. Don’t forget that a bus would only have about a half-mile to merge across several lanes to use the HOV stop at Eastgate and use ramps to and from South Bellevue Way. It would appear that any buses in the HOV lane would have to go to Mercer Island for operational safety reasons.

      13. Rapidride K isn’t slated to go anywhere near Eastgate.

      14. The K’s south end is Eastgate P&R.

      15. Assuming that there’s only one all-day express bus from Issaquah, it’s a tradeoff between a larger detour for smaller numbers of people vs. a smaller detour for larger number of people.

          Right, and in general, you want a smaller detour for larger number of people. Otherwise you are sending buses looping through the V. A. parking lot, or waiting to turn left into TIBS (while having the audacity to call the bus “Rapid”).

          As I wrote, ridership between Issaquah and Mercer Island is so low that it may not exist during rush hour. The 556 doesn’t have that many riders; only about 200 people a day get off in Bellevue (from Issaquah/Eastgate). That works out to less than 25 people per bus. At rush hour! This is for a bus that goes right to downtown Bellevue (no transfer required).

          It is worth pointing out that even when there is no traffic, it takes about the same amount of time to serve Mercer Island as it does South Bellevue. So that means that riders to Bellevue benefit about the same as riders to downtown Seattle lose. But let’s say it takes an extra minute to get to Mercer Island. Thus truncating at Mercer Island means it takes an extra 6 minutes to get to Bellevue. Truncating at South Bellevue means an extra 4 minutes to get to Seattle. Given the ratio of riders, it isn’t close: truncate at Mercer Island.

          You also can’t ignore rush hour service and the complexity of sending buses to two different places. I think it is safe to assume that Issaquah will have express service to Mercer Island. Given that, we run the risk of stranding people in Mercer Island, as the system switches to using South Bellevue. Riders have to become savvy (“if I catch the train at 6:23, then I get off at Mercer Island, but if I catch the train at 6:29, I get off at South Bellevue”). Of course the opposite occurs if you have rush hour buses headed to South Bellevue. But there are a lot more people going to Seattle than Bellevue (meaning that a lot more people have to be savvy). We have what — six express buses to downtown Seattle from Issaquah, and only one to Bellevue?

          Ideally you have everything: an express to Mercer Island, an express to South Bellevue, along with local service serving places like Factoria. But you can’t afford that. You end up having poor frequency. In the middle of the day, the biggest concern is frequency. The easiest way to have good frequency is to just serve the most popular destination — Mercer Island.

   2. In 2021, Link will have four-car trains; does that not imply less resistance to restructuring routes 252, 257, 268, 311, and 545 to meet Link at the UW station and avoid the congestion of the I-5 general purpose lanes?

      A little bit, but I would imagine there are a couple changes that would be a bigger deal. The first is essentially an HOV lane all the way from 520 to the station. The second is increasing frequency (from 6 to 3 during rush hour). It is all good and well to talk averages, but it is really irritating when you miss your train, and have to wait six minutes for a new one (when you used to just take the bus all the way downtown). I don’t think that will happen for a while (I honestly don’t know when they will ramp up the all day frequency, let alone the peak frequency — I asked below).

10. One thing that I am still not clear on is this: are there plans to get rid of the 542 once East Link opens? I would be very concerned if they did try to do this, as it would force riders to go way out of their way to get to UW. I get it saves a lot of service hours, but I don’t think it would be anywhere near worth it.

    1. It hasn’t been explicitly stated, but it’s exceedingly likely the 542 will remain, and perhaps might even get some frequency boosts once the 545’s hours are portioned out.

    2. I highly doubt it, as riding Link all the way from Redmond to UW would take at least 45 minutes, compared to 20 minutes or less on the 542. If anything, would expect the 542 to gain frequency, especially on weekends, with the 545 removed.

       There might be some brief oddball periods (e.g. midnight-1AM) where you have to ride Link all the way around, but certainly not during the daytime.

       1. That’s what I figured. I just wasn’t sure if there had been an official announcement yet.

          Regarding the late night/early morning periods, I think that those link rides would be useful. Currently the last 542 leaves UW at 10:45pm, having run every half hour since 7:10ish. So even though Link will be longer, I think there will be a group of people who take it at night (I can see myself as one of them if I am still at UW in 2023).

       2. From what I have heard, the 542 may not be favored outside of peak periods because although Link will take more time, the frequency and predictability will be extremely desirable. If the non peak average wait for Link is 7 minutes and the average wait for a 542 is 15 minutes, on a cold day, I have heard from people the preference will be for Link. 40-45 minutes, subtract 7 subtract train bias, subtract desire to be warm might tip the scales.

       3. At the end of the day, time is time. When answering questions like what time you need to leave to home to arrive at the destination at time XXX, how the time is distributed between bus and train doesn’t matter.

          Bottom line: an extra 30 minutes of travel time is a lot. And, outside rush hour and Husky games, the freeway is wide open and the 542 is consistently quick and reliable.

    3. No ST Express routes have been announced, and probably won’t be until a year before opening. It will probably be in conjunction with a Metro restructure that will have a series of public hearings for input. The closest we have now are some ST2 planning scenarios from 2016, and Metro’s 2025 plan, both from early 2016.

       If service is split between different stations at different parts of the day, they will certainly have different route numbers. AJ’s suggestion of sending the base all-day service (554) to South Bellevue or Bellevue and the peak-only service (21x) to Mercer Island makes the most sense.

       Metro’s 2025 plan suggests rerouting the 554 to South Bellevue, Bellevue Way (to replace the 550’s stops?), Bellevue TC, and 405 to the U-District. It has two Express routes: North Bend-Issaquah-MI, and Redmond-Sammamish-Issaquah-MI. Those would the the successors to the 21x. It’s unclear whether they’d terminate at Mercer Island or South Bellevue, or whether the’d be all-day or peak-only.

11. Great post! I enjoyed reading this.

12. So when is the SR 520 work done. That seems like a very significant point, as it will considerably change bus service from the UW to the East Side. Buses will be a lot faster, and thus the transfer penalty a lot smaller.

    1. https://www.wsdot.wa.gov/Projects/SR520/I5toLakeWa/default.htm

       As west as Montlake: 2024, when it would be wise to restructure buses to rely very heavily on the 520 Bridge from the Eastside to UW. Also, the 520 to I-5 express lanes to Mercer is delayed from 2023 to 2024 because of I-976.

       Portage bay bridge replacement (Montlake to I-5) will take place from 2023 to 2029, when there shouldn’t be any buses running there anyway because they should be rerouted to UW Station.

       1. Is that the bridge the 70 and one other bus go over? Or am I thinking of the wrong bridge?

       2. Hayden, you’re thinking of the University Bridge (the bascule/draw bridge). The Portage Bay Bridge carries SR-520 from Montlake to Capitol Hill, but is not a floating bridge like the Evergreen Point Bridge to the east.

       3. Cool (thanks for the info). Because of that, I think by 2024 it makes sense to truncate all SR 520 buses at the UW. Link will be running every 4 to 5 minutes and it will be faster to get to the UW than downtown. There will be a lot of other improvements to the network, which will help leverage the change. For example, let’s say the 311 runs to the UW, and does so relatively frequently during rush hour. Let’s also say that Woodinville doesn’t have an express to Bellevue. Riders from Woodinville headed to downtown Bellevue would catch the 311, then transfer to the 405 BRT.

          I would also like to see a bus that goes from UW Bothell to UW’s main campus, via 405 and 520 (making all the freeway stops). This would be an all day bus, connecting to local buses. So someone who lives in Woodinville and misses the express back home can take this bus to 195th, then take a bus from there. The biggest benefit, though, would be people going between campuses, or to Totem Lake.

13. Another question: When do the trains start getting frequent? Right now they run every 10 minutes during the day, will that change to 6 minutes once Link gets to Northgate?

    1. The current line will continue to run every 6 peak / 10 off-peak until East Link, and then each line will run every 8 peak / 10 off-peak. So downtown to Northgate will be more frequent across the board, and single line stations will actually have reduced frequency at peak, but all trains will be 4 cars.

       1. I do hope that they reduce those back down for 6 peak/ 10 off peak for the single lines. At least for one of them. I think 3 peak and 5 or 6 off peak seems like it may be needed.

       2. Oof, that is disappointing. It is crazy to finally have the key part of the system built (U-District to downtown) and then run the trains so infrequently. Ten minutes is just not that good. There are lots of buses that run that often, while the 3/4 runs more often. There are various other corridors (like Third and The Ave) that run a lot more often, even if they are out of sync. When RapidRide G starts (on Madison) it will run every six minutes. Sad to think that our key transit corridor (UW to downtown) will lag others.

          It is also nuts that for some sections (Rainier Valley, etc.) frequency will go down during rush hour. That is just weird — a degradation when the system is supposed to be improving.

          In general the low frequency will discourage transfers for a lot of the network (which in turn make it tougher to push forward with truncations). At least the north end will eventually get good frequency, even if they have to wait until East Link gets service.

       3. 6-minute peaks was only temporary for U-Link to address its additional capacity needs until Northgate Link starts, ST seems to think 8-minute peaks are an adequate baseline. It will be hard to go back to them because there’s a difference between waiting five minutes or less (as you do in London) vs waiting almost ten minutes (as you do in systems that don’t have that frequency). But… ridership demand has forced ST to increase capacity before, first with 1-to-2 and 2-to-3 car trains, and with dropping the 8-minute peaks to 7 minutes and 6 minutes, so there’s a possibility that what ST planned for earlier will be superceded and we will get more frequency. But there’s no guarantee of it, and a status quo interia against it.

       4. ST seems to think 8-minute peaks are an adequate baseline.

          Which is ridiculous. It isn’t about capacity, it is about service. Here are some peak headways, from various lines in North America:

          SkyTrain — 2 or 3 minutes (depending on line)
          BART — 2 minutes (for core)
          New York — 2 to 5 minutes (depending on line)
          Montreal — 2 to 5 minutes (depending on line)
          Toronto — 2 to 5 minutes (depending on line)
          CTrain (Calgary) — 3 to 5 minutes

          So, even in Calgary (freakin’ Calgary) you can expect a lot more frequent service. This reminds me of when ST thought they didn’t need a schedule for Link. I’m sorry, but ten minutes just isn’t frequent and frankly, neither is eight. If you get down the stairs in Toronto and see the train doors about to close, you just shrug it off. In Seattle you sprint, and try to pry your way on.

    2. per RossB, off-peak Link service frequency improvements would be an excellent cause for STB to pursue. today and later.

14. All the new RapidRide lines listed here are BRT. Are you sure? Current RapidRide lines are not BRT, not by a long shot.

    1. BRT (as used by King County Metro) is a political term meaning express-ish red buses.

15. Pence: BRT has a wide range. LRT has as well. both vary; the degree of grade separation and frequency provided are the key variables. Route 99B in Vancouver is BRT; it is very frequent, but has little exclusive ROW. the high end is demonstrated in Latin America.

    1. Sounds like how we looked at “Coding Standards”, back in my programming days.

       As seen on a company bulletin board:
       “That’s what I love about standards here, there are so many to choose from!”

Comments are closed.