Graham Street Station and the Failure of the Enterprise Initiative

167 Replies to “Graham Street Station and the Failure of the Enterprise Initiative”

1. The other problem with the Graham St. Station is that it was yoked together, for purposes of cost-cutting analysis, with Boeing Access Road Station, so they would either both be built, or both be not built. Boeing Access Road Station had a much higher cost, with much less ridership, thereby dragging down the overall productivity numbers compared to Graham, alone.

   I can’t help but wondering if the real motivation here is that the board reps from Tacoma and Federal Way simply don’t want an extra 30 seconds added to their travel times to downtown Seattle. So, they design the studies to cast Graham St. station in the worst-possible light to maximize the odds that it gets cut.

   1. You can’t blame them. It’s very slow from Federal Way to downtown… It rarely beats driving even during peak hours when you have to account for transfers and wait time.

   2. Also a new station would probably add 2 mins if we’re being honest. Time to accelerate, and other traffic light situations possibly. Not much but when someone’s trip is already over an hour… It’s just worse.

   3. I don’t think they’re yoked together. The board discussions sound like they’re considering them separately. The issue is how much their subarea wants them relative to the other things they want.

      FW-Westlake and Tacoma Dome-Westlake travel times will be almost an hour or over an hour respectively, which is already at the edge of tolerability, and two infill stations won’t make much difference. ST recognizes this and has committed to keeping the faster express buses running between these cities.

      Graham station’s fate revolves around how much North King wants an early deliverable there, at the tradeoff of making West Seattle and Ballard Link take longer. North King transit fans are divided on that. North King ST boardmembers are heavily on the side of the Link extensions.

      BAR station’s fate revolves around how much South King wants it. Tukwila does, and says it wants it a lot, but has trouble articulating a plausable reason why. So I think boardmembers want to honor Tukwila’s wish, but it’s a non-priority for them compared to the much larger Link issues.

      The travel time of Federal Way-Westlake and Tacoma Dome-Westlake is based mainly on their distance. Federal Way is almost as far as Everett, and Tacoma is further. And there aren’t any large destination clusters in between like UW or North Seattle.

   4. The longer travel time between the south stations and downtown Seattle is solved by retaining a consolidated bus serve that merges routes 574, 590, and 594 into one with short headway and serves the Federal Way Link station. The fall 2026 SIP has only the truncated Route 574 serve Federal Way station so does not have the advantage of very short waits from and to the south.

2. Zero reason NOT to build a central platform. It should always be a center platform.

   Stations should be as small and easy to access as possible for bus transfers and pedestrians.

   1. The reason to not build a central platform is cost. Cost is clearly a major constraint that Sound Transit is dealing with, and the whole point of the Enterprise Initiative was to find ways to reduce costs. It is almost certainly more expensive to realign tracks and move catenary than it is to close travel or turn lanes for a side platform set-up similar to other stations in the Rainier Valley.

      1. Yes, this. There is no reason to spend a lot extra for a center platform here. The benefit is minimal. In contrast, center platforms add significant value if riders are reversing directions (like CID) or there may be some confusion as to which way the trains are headed (like every underground station). Neither are issues in this case. It is quite obvious which way the train is going — the same direction as all that traffic you had to pass.

   2. I thought that too but Andrew gives a good reason why not in Graham’s case. Moving the tracks and catenary would require shutting down the Link segment for months. That would have a huge impact on passengers, and I don’t see it’s worth it just so a few Graham station users can turn around without leaving the platform, which most of them won’t do. That’s on top of the cost of moving the track and catenary, which could go to lowering Link’s cost-deficit elsewhere. Rainier Valley has three adjacent surface stations, where you can see from your train seat which one has the most convenient layout for turning around. It doesn’t have to be Graham station.

      1. Okay I guess I do agree in this case. If the station was built or tracks were already laid, then we should aim for cost.

         But I guess I’m trying to say we should’ve designed for all stations to be center platform to begin with. No clue why we didn’t build center platform for several stations in the current 1 and 2 Lines that should’ve been center.

         1. But I guess I’m trying to say we should’ve designed for all stations to be center platform to begin with.

            There is very little reason to. It is quite common in Europe for the tram stops to be on the outside. Center platforms are ideal if you are reversing directions or otherwise switching trains. They are also helpful if the station is underground and riders might get confused as to which platform to use. But in this case it really doesn’t matter.

         2. But what’s the reason to not use center platforms?

            Center platforms are just easier to use overall. You don’t have to cross multiple tracks, you don’t have to be confused where you’re going especially in a Downtown station… At minimum Chinatown and East Main should’ve been center platform..those are going to key transfer points between 1/2/3/4 lines.

            Center platforms also usually help keep the station design compact and efficient instead of giant and confusing. S Bellevue is one of my favorite stations. Super easy to connect to buses and parking, drop off people, and center platform without a multi floor crazy elevator / long flight of stairs.

         3. Because here Link is operating as a streetcar or a tram-train, where outside platforms are much more common. When Link is fast, frequent (i.e. where L1 & 2 are interlined), or not at-grade (elevated, trench, or underground), it should be center platform for safety, ease of access, and to minimize vertical conveyance. But in the RV, Link is slow, infrequent, and at-grade, and therefore it is better to keep the tracks straight & in a tight footprint, so the platforms are outside.

            Take East Link: Spring District is in a trench and is a center platform, whereas Bel-Rde is at-grade and are outside platforms. These are both the correction design.

         4. For an at grade station, side platforms avoids the need to cross the tracks to reach the train in at least one direction.

         5. @AJ – Spring District isn’t a center platform.

            (Agree with your points, however.)

   3. There are four advantages of center platforms:

      1. Transferring to an opposite-direction line. E.g., Eastside to SeaTac at CID. Inapplicable to Graham.

      2. You accidentally went to the wrong platform. Unlikely at Graham because you can see the surrounding landscape. E.g., The McDonald’s is on the right in front of you, when it should be on the left behind you.

      3. You change your mind en route and decide to go somewhere else instead, or your trip is taking longer than expected so you don’t have time to finish your original trip. You have to turn around somewhere, but it doesn’t have to be at Graham, it can be at any Rainier Valley station that has the best turnaround layout.

      4. You accidentally miss your stop and have to turn around. Then you have to go through Graham’s turnaround layout or go on to the next station, which may not be any better.

      Center platforms are more important at underground stations, where you can’t see what’s around the station or the transfer paths. And it’s not so much one station with side platforms, as in having several in a row, or all stations like that. All the stations from CID to Westlake are side platform.

      The best time to design center-platform surface stations is before the track is laid. But Rainier Valley was laid out for side platforms.

      1. You forgot one:

         5. Center platforms can not only be narrower than two side platforms, they can also require fewer vertical devices to be ADA compliant. I’d take two elevators for one center platform than one elevator for each of two side platforms. (Of course, this is not an issue at Graham but it is in other places.)

         1. This isn’t exactly accurate. Requirements for capacity on platforms are strictly a function of the state building code (WSBC). In the WSBC, Section 3116 in its entirety deals with fixed guideway and passenger rail systems. It covers occupant loads on platforms, which determines both sizing and egress requirements (stair width, elevator capacity, etc.).

            Occupant load typically accounts for a train simultaneously at the platform on each track and is equivalent to the train load of each train plus passengers waiting to board. Loads for both are based on the peak period ridership as predicted for a new system or updated for an existing system. In effect this means the center platform must be equivalent in size to the combined areas of two side platforms (potentially a bit less if you can show there would be fewer passengers waiting on one side than on the other, since the side platforms would need to account for the maximum load in that direction at any point in time and the center platform could account for peak waiting on one side/direction, off peak waiting on the other, but that would likely be a minimal difference).

   4. I’m not a designer, but I believe that it may be easier to spread one side of the tracks to allow for a center platform than to try to create two side platforms by taking left turn pockets.

      The phasing appears much easier to me. By moving the street first, there is a much wider construction zone to work with. Going with two side platforms could have meant that all of MLK would need to be closed for an extended period of time.

      Of course, the assigned designers probably looked at side platform options and uncovered challenges about constructibility. I don’t know but I think early on there was some side platform ideas put forth.

      And a center platform does provide a great median refuge when crossing the street. That appears to be a safety bonus.

      Ultimately, it’s the signal timing and phasing that also matters. The wider design slightly increases the pedestrian crossing time to cross all of MLK. I’m not sure if the designers considered this.

      This corner is one that gets lots of turning because there aren’t alternative streets close by — especially towards Beacon Hill. Adding more pedestrians seems like it increases accident potential. It’s why I earlier suggested that the platforms should be at least a block from the corner with station-specific crossing lights for pedestrians. Of course, ST responded that they didn’t want to make people walk an extra 200 feet even though it would be much safer for those people.

      1. Going with two side platforms could have meant that all of MLK would need to be closed for an extended period of time.

         Only if you consider the left turn pockets sacrosanct for some reason.

         That’s apparently true in this instance, at least, to ST, but it certainly doesn’t have to be so.

         Northbound is easy; just take the Chevron and make a mini-cloverleaf out of its space. Southbound isn’t quite so easy, but there is a lot behind McDonalds that could become a clover-leaf back to eastbound Graham. The “street” between McDonalds and the U-Haul storage building is wide enough to take one lane and still give the building access to the back side for emergencies.

         Think outside the box of crappy sprawl development at the corner and it’s easy to see side platforms where the left-turn bays are today.

         1. And, if there is a station added at Graham, that Chevron, the storage building and even the Mickey D’s are all toast anyway. Just build the replacements around the mini-clovrleaves.

         2. “Only if you consider the left turn pockets sacrosanct for some reason.”

            Nope that’s not it.

            Construction zones are simply larger and take more of a buffer. So setting up two takes more space. There needs to be room to store various things and fencing to protect the site and room for workers to walk around.

            Left turn pockets are also not wide enough for a side platform. ST prefers having side platforms 16 feet wide. Left turn pockets run about 11 feet wide. So merely closing a left turn pocket wouldn’t provide nearly enough room.

         3. Who needs a platform sixteen feet wide for a single direction platform at what is essentialky a trolley stop? That’s plain stupid.

            The MAX platforms at Delta Park — at one time quite a busy station and still a major transfer point — are about ten feet wide, if that. ST has an Edifice Complex and needs to get over it.

         4. Also, using the left-turn bays can be done in sequence. Sure, you’d need a protection buffer for the workers. The bike lane on the side being built might have to be closed for a couple of months while the pad is poured and the protective barriers erected. After that the platform is protected space for whatever needs to happen.

            Stuff happens inside street envelopes all the time and traffic adjusts.

   5. Why would Sound Transit make a big thing out of a center platform for this station but not for actual major transfer stations as part of existing or ST3 with reverse direction transfers like at CID, SoDo, East Main?

      1. Because Sound Transit is inconsistent. At different times it’s swayed by different arguments. At first it wouldn’t have any center stations except at SeaTac. Then it changed its mind for Capitol Hill and North Seattle. Then it went back to side platforms for most of the Eastside.

      2. ST doesn’t like to admit that past decisions could be suboptimal.

         I even said here (on STB) in 2017 that ST should have pursued a change order to redesign and make East Main a center platform station as soon as ST3 passed. It could have been completed well before the eventual 2 Line opening.

         (The observation didn’t get much traction even here on STB at the time. )

         And even with all the plan revision discussions, no Board member or executive staff has been publicly acknowledging the design mistake even to this day. Us riders talk about it; the Board members don’t seem to care.

         So instead we still have an official ST plan to require the walkway gates to get to the East Main southbound platform to be locked more than unlocked. There will be a train crossing every 120 seconds for 4 minute frequencies or 90 seconds for 3 minute frequencies. With time needed well before trains arrive, plus the time as trains operate slowly near the station, the gates will need to be locked for at least 60-70 seconds.

         The visionaries among us get that this situation will be a literal train wreck with human carnage.

3. Excellent review. Well done.

   1. Agreed. This is outstanding. It should be a must-read for everyone on the Sound Transit board.

4. Before this I was feeling the Enterprise Initiative has done a decent job getting cost analyses & tradeoffs into broader public conversation. But I this is a good post & argument that it hasn’t been universal for all projects and a reminder that I the relative low bar ST generally has is influencing that feeling.

   I hadn’t heard or read any details about the Graham infill station analysis until this and what you laid out sounds like the right call to me. As you point out in your framing of “parting thoughts,” the Enterprise Initiative is limited by the ST Board’s openness & willingness to do better transit vs being able to point to a transit project. I think reporting after the retreat where the 3 scenarios came up, I read staff stating that they tried to stay within what was deemed politically viable/considerable of the Board. This likely plays out in the no left turn option requires Seattle’s buy-in and Seattle process makes that hard so weight to avoid that showing up.

   I agree it’s unacceptable behavior & I do think/wish more planning staff would push the right questions and conversations into public debate more often.

   1. >” This likely plays out in the no left turn option requires Seattle’s buy-in and Seattle process makes that hard so weight to avoid that showing up.”

      In 2024 and 2025, it was pretty clear Seattle’s political leadership was not going to play ball on ROW reallocation. I don’t think that’s the case at all anymore, and Mayor Wilson can score an easy if not universally popular win here on both her transit and safe streets agendas by unilaterally (in terms of ST, ideally with at least some support from City Council) by signaling SDOT will be moving towards a MLK road diet anyway so ST might as well use some of the space for the station.

5. This situation is an example of how cutting corners has major consequences later. Even though the station wasn’t built originally, it was in the EIS. Then rather than leave room for a future station the final design acted like it would never happen.

   Had provisions been made to someday add it, it would be an easy thing to do. Scissor tracks, platform edges and those kinds of minor changes could have made a difference. But they didn’t.

   It’s why any time some station design gets adjusted because any feature is axed, hard questions should arise. Summarily doing things like eliminating escalators rather than merely deferring them is something that ST routinely does to save money because they then don’t leave placeholder space to add it in the future.

   1. MAX uses 2 tunnels that were built 20 years before there was a MAX line. The green line uses underpass space proceed at Tacoma Street when the overpass was rebuilt 25 years prior to MAX.

      It helps to have a plan so that years of infrastructure changes can accommodate those plans when it comes time to put them in concrete.

      1. Wasn’t the whole I-205 transitway reserved when ODOT built I-205? I think that made things significantly cheaper for TriMet when building both the Red and Green lines. TriMet’s capital planning isn’t without issue, but the first 4 lines were very cost effective because of re-utilization of existing rights of way. It feels like ST hasn’t really taken that approach writ large.

         1. That’s what a field trip guide told us during the Rail~volution conference in Portland. They said part of the 205 right of way was reserved for future rail.

            But I think Glenn said ridership has been disappointing, and it’s not that useful because the stations are so far from destinations. I think the Clackamas Town Center station is on the other side of the freeway from the shopping center? So that raises the question of whether it was a good corridor choice, and whether the reserved right of way should have been used or the MAX line routed somewhere else.

         2. The station is on the same side of I-205 as the Mall. There is another strip-type center across Sunnyside Road and a Kaiser-Permanente Hospital across the freeway. But the station is “next to” the Mall, albeit a long block away.

         3. The Green Line performed the worst post Covid of all the MAX lines to my memory, but I am under the impression this is more work pattern related than anything else. Clackamas Town Center is well-connected to the transit center (maybe a quarter mile walk on sidewalks), and Lents is a genuine urban center with good bones that’s pretty well served by the station. But other than that, it’s not great for station areas, and travel times are relatively slow.

            This is made worse by the fact that the southern portions of downtown Portland have done better post-Covid than the norther portions. PSU and OHSU are still very solid trip draws to the south, while the offices in the Lloyd Center area are pretty dead and Old Town is rough. Since the Green Line takes 21 minutes to go from the first Lloyd Center stop to the last PSU stop, most people have faster trips via buses than the Green Line. From Lents, the 14 takes 33 minutes compared to the Green Line’s 38, and things are more dramatic from Division where the FX2 is 10+ minutes faster to PSU (and way more to OHSU).

         4. The transitway that was already graded and tunneled went from Foster Road to Sandy Blvd. South of there they had to move buildings and otherwise create a new right of way. North of there they had enough space in the highway median for a bridge to get to the airport, but it wasn’t a big deal since the adjacent property was already vacant and being part of the airport shopping center plan.

            MAX dismal ridership due to its location along highways is a completely different matter. Metro (our equivalent of the PSRC) has a really awful transportation plan. MAX ridership along the corridors since Metro assumed that role reflect this.

            However awful the plan may be, TriMet, the city of Portland and ODOT have at least planned ahead with things like reserving right of way.

   2. Yes, this does seem like a big issue with ST deferrals, but I’m not sure I have enough comparative evidence for how common these kinds of provisions are. In my survey of other light rail systems’ infill stations, there were at least a few in SLC that had left room for a future center platform, but I’m not sure about crossovers and other provisions.

   3. I think ST in the 2000s was trying to kill any possibility of Graham station being revived, so it designed the alignment without forward-compatibility for it.

   4. I’m not fan of WSLE as rail but if it is proceeding forward at least futureproof it for allowing for a potential infill station where its currently being cut.

      1. Not possible. The new alignment crossing 35th Avenue SW is two blocks north and in a TBM tunnel. Avalon Station was to have been in a Retained Cut which will be bypassed.

6. The Enterprise Initiative has not looked under the hood of any of the ST3 projects. That has been disappointing. To date, the changes are much less than during the Sound Move reset, 1999 to 2001.

   1. Yeah that’s a fundamental problem. The entire management and Board is incapable of giving anything major a substantive and critical thumbs down. They won’t rank projects based on maximum benefit. Even now, they say “deferred” rather than eliminated even though that means at least a 25-30 year wait before opening date. They’ll long be retired and possibly dead when the construction contract is finally completed and may even when it’s awarded.

      When hard decisions are finally needed, they’ll make future riders sacrifice before anyone else.

   2. Yes. There were a number of potential changes to TDLE that would have provided a better design yet likely had substantial dollar savings. No substitute changes were even considered, as far as I am aware.

   3. A core belief should be challenged. The ST board is using faith-based planning to implement the regional Link spine between Everett and Tacoma. Subarea equity should not lead to Link equity; Link should be applied where it is the most cost-effective mode; subarea equity is about the proportional allocation of funds. The regional spine should consist of Link, bus, and Sounder; it should be a network. A spine needs a brain and nerves and supporting organs.

      (Is there a Wizard of Oz analogy here? Does ST lack a brain, a heart, or courage? Or does it just fail to recognize its potential to change ST3 for the better through the Enterprise Initiative? Where is our Dorothy to lead them? To date, Mayor Wilson has been spouting TCC points: build it all. The Chair’s package is a challenge. Is Councilmember Strauss our Dorothy. Click those red shoes).

      In ST3, the Ballard line is a good use of Link, as it serves several dense urban centers with street grids and two-way all-day demand for transit and where surface transit is slowed by traffic congestion. The TDLE and the 4 Line are not good uses of Link as they are oriented to P&R (e.g., Tacoma Dome and South Kirkland P&R) or are in freeway envelopes (e.g., I-5, I-90) They could be better served by bus. Issaquah could be better connected with both the 2 Line and downtown Bellevue via well-designed bus routes; they could have better speed and frequency earlier at much less cost. Link would be great service for Tacoma, the city of density. But is the Tacoma Dome in a pedestrian place? If you can look out your LRV window and see a freeway, perhaps you placed the line in the wrong place.

      So, in the chair’s package for ST3, the best Link line is tossed out and the weak Link lines are advanced.

      1. [ad hom against ST]

7. It seems taking over the left-turn lane to build side-platforms should be pretty simple. During the first outreach meeting I asked one of the designers. He said that the current tracks are not completely straight/flat. Therefore, the tracks would need to be redone. He also said that Sound Transit would like to add a crossover switch to the north to allow for operational flexibility.
   I’m in favor to add a crossover. For transparency, that should be a separate project/consideration.
   I believe Sound Transit is too rigid with their track requirements. Adding a side-platform should minimize both cost and disruption as seen in Portland. Left-turning cars could get redirected to 39th Ave S. or the road may get pushed east a bit. There is already extra space in the south-bound lane.
   Again, I had hoped Sound Transit would spend more time looking for cost reduction with over-engineered projects rather than simply delaying projects.

8. I don’t think we need a Graham Street Station. I say that as a South Seattle (I live a block from Graham Street) resident for the past 49 years who uses Link regularly . One negative for Link Light Rail along MLK Way is that it’s at street level. If you are driving a car trying to cross MLK, the congestion at Othello or Alaska (where there are current Link stations) can be ridiculous because the traffic light can stay red a very long time. Often, when it should be your turn to go, a train enters or approaches the station and the light stays red. In my experience, this is much less problematic at Graham and Orcas, MLK crossings , where there is no station, because the trains don’t stop, triggering a red light for cross traffic sometimes while they are in the station.

   I also oppose a Graham station because so many areas that pay Sound Transit voter-approved taxes still have none of the service they were promised (think Ballard, West Seattle, Everett, Tacoma, etc.). It’s better at this point to extend the system to areas that don’t have it at all than to build additional stations so close to existing stations. The last thing we need is a voter rebellion, which could endanger Sound Transit’s funding even more.

   1. As a south end resident, I’m also becoming convinced that it would be better to do nothing than to try and move forward with the station as currently planned by ST. It appears that ST has locked in its south-of-Graham plan in order to avoid very costly utility relocation projects. It’s unfortunate that the north of Graham locations will require moving utilities because–above ground–that’s where there’s plenty of space to build whatever type of station is best for the community.

      Is the best solution to ban left turns from MLK to Graham so an inferior station design can be built? I’d say that asking drivers to move their left turns to Orcas and Othello won’t work. Orcas is a smaller scale neighborhood street that would be overwhelmed with the added traffic, and it should be obvious that diverting more left turns to Othello is impossible.

      Ultimately, the solution to the numerous light rail problems in RV is to elevate the tracks. Why spend all this money on a project that will create more problems and delay the eventual solution?

      1. Ultimately, the solution to the numerous light rail problems in RV is to elevate the tracks.

         Again, it is much cheaper to go underground. But that is irrelevant. Sound Transit has no interest in improving the line. None. Of course it would be better for riders. It would also be much better for people in Rainier Valley. Does ST care? Of course not. They are primarily interested in adding as many miles of track as possible. That is the one consistency. It has never been about functionality. It has always been about distance. Skip First Hill? Sure. Who cares about stations — it all about length. Everett Link, Tacoma Link — even Ballard Link was designed based on that idea. Ballard to UW is a much better value but Ballard Link is longer. Therefore it must be better. Now, at the first whiff of trouble, what does ST do? Get rid of stations of course. West Seattle Link only has three station despite several over 4 miles of new track. Now it will only have two. Meanwhile, they are ready to eliminate a station with Ballard Link as well.

         So somehow you think ST will improve the existing light rail line just so the trains can be a bit faster, more consistent and frequent? Dream on. If ST is unwilling to add a simple infill station in Rainier Valley they sure as hell aren’t willing to spend over a billion elevating or burying the tracks. If they were they would be saying so but all we have is the typical whispers as they shortchange urban neighborhoods while they focus on track length.

      2. “They are primarily interested in adding as many miles of track as possible.”

         Not track miles but the number of cities served. With special priority for Everett and Tacoma because they’re the largest cities in their county.

         1. “With special priority for Everett and Tacoma because they’re the largest cities in their county.”

            Because they’re crucial for the regional spine.

            I really see that the ST3 light rail projects are replacing communities with abundances of express/peak buses. Ballard has the D Line, the 17X, and (formerly) the 29 cross the Ballard Bridge. West Seattle (oh my gosh) has the C Line, the H Line, the 21 (and it’s express variant), the 50, the 56, the 57, and the 125 ALL CROSSING THE WEST SEATTLE BRIDGE…

            What is up with Sound Transit having Ballard and West Seattle Link over bridges that cross bodies of water? I guess that was intentional.

            ANYWAYS, EVERETT HAS THE SWIFT GREEN LINE, FUTURE SWIFT SILVER LINE, 201, 202, 510, 512, 513 ALL RUN ON THE LIGHT RAIL ALIGNMENT!!!

            Tacoma Dome Link has the 500, 501, 574, (soon to be gone) 586, 590, 592, 594, and the 595 all run on the alignment.

            Issaquah Link has the 203, 218, 271, 554, and the 556.

         2. It is still based on this idea that the key is to go a really long distance. It is the whole notion of a spine. Just think about how they sell it: Trains from Everett to Tacoma! Fantastic!

            The implication is that it serves everywhere in between. We know these areas well — we can imagine going along the spine. We just take it for granted that it will somehow work for all the areas in between, let alone Everett and Tacoma itself.

            No one bothered to consider where exactly it would stop along the way. Because somehow that was less important than going from Everett to Tacoma. Not because Everett and Tacoma are so big (they aren’t) but because it is just assumed that a light rail line automatically serves everything in between. It is a sophomoric notion but not too surprising given the general ignorance of transit in the region (and U. S. in general). That sounds rude, but I was no different when all of this started. Your average American knows freeways really well — they can describe various aspects of them in great detail. Why it is stupid to have ramps on the left side. Why cloverleafs are quite effective. How traffic in one area spills to another. But mass transit? We have no idea. Thus a really big line — from Tacoma to Everett! — sounds like it will make all the difference in the world.

            And yet when the dust settles the vast majority of users will be in the areas close to the middle, wondering why there aren’t more stations (like most subway lines in the world).

         3. Not track miles but the number of cities served.

            That is even stupider. Mass transit is for connecting neighborhoods, not cities. If those neighborhoods are in different cities, so be it. But city borders (like county borders) are arbitrary.

      3. I would strongly contest that drivers can’t just change behavior. But I want to reemphasize the point that if we start from the position that the station should be built, then all of the additional cost that stems from preserving car access exactly as is should be framed as a cost we are paying to benefit drivers. I don’t know exactly how much more this project is supposed to cost as a result of expanding the ROW and other related decisions, but it is a major cost driver. Even at 50% of total costs, that’s still well over $50 million spent on one intersection, all so some drivers may not have to wait an extra light cycle at peak times a few blocks south. That is a profoundly bad investment.

         Why invest in safety improvements on MLK when we just need to elevate or bury the line anyways? Because that is a speculative project that isn’t in any specific plan and has no political backing. At ST’s pace, it’s 2075 at best. Let’s just wait 50 years to have a station that will demonstrably improve the lives of nearby and far away transit riders.

   2. I don’t think we need a Graham Street Station.

      Well sure. We don’t need anything in ST3. So what? Graham Street Station is a better value than any other project that is left. Consider:

      1) Ridership per dollar. Before recent cost updates, Graham Street was second only to Ballard in ridership per dollar (https://www.thetransportpolitic.com/2016/04/06/youve-got-50-billion-for-transit-now-how-should-you-spend-it/). The cost for Ballard Link has risen much faster. Thus it is highly likely it has the best ridership per dollar.

      2. Ridership time saved per dollar. This is harder to measure. But it is reasonable to assume that because it is much better in terms of ridership per dollar it also performs just as well in total time saved per dollar.

      3. Equity. The stop serves an historically low income, redlined area.

      One negative for Link Light Rail along MLK Way is that it’s at street level. If you are driving a car trying to cross MLK, the congestion at Othello or Alaska (where there are current Link stations) can be ridiculous because the traffic light can stay red a very long time.

      That is irrelevant. Traffic is bad in the city. Yeah, no kidding. It is a city. Try driving in Chicago or New York.

      I also oppose a Graham station because so many areas that pay Sound Transit voter-approved taxes still have none of the service they were promised (think Ballard, West Seattle, Everett, Tacoma, etc.).

      The same is true with Graham. All of the future riders have paid taxes as well. So maybe we should look at the total cost per rider. Which project will benefit the most riders for the money. I already did that — Graham Street is on top.

      It’s better at this point to extend the system to areas that don’t have it at all than to build additional stations so close to existing stations.

      Again, Graham Street doesn’t have a station. Not at all. It also isn’t close to the other stations. It is 1.6 miles between Columbia City and Othello Station! That is a huge distance. Go look at a real subway line, like in New York or Chicago. Hell, just go to Vancouver. It is hard to find stations that far apart while it is quite common to find stations that close together. The only reason *not* to build this station is if the cost is too high. The author addresses that. But even at full cost it is a tiny amount compared to everything else we are considering. $200 million is away too much to pay for a surface station. But it would be a bargain everywhere else. West Seattle Link is going to cost billions per station. Billions! This is just a much better value.

      The last thing we need is a voter rebellion, which could endanger Sound Transit’s funding even more.

      Right, and screwing over a low-income area for no apparent reason is a good way to get that rebellion. ST appears to be ignoring cost-benefit and just shooting from the hip with these projects. This should be an “easy win” — a small project that everyone understands is a good value. Instead it is being deferred forever so that we can start large projects that aren’t fully flushed out yet. West Seattle Link would not connect West Seattle to downtown — that part is still being worked out. So if things go as planned we will spend billions of dollars serving two stations in West Seattle while we then figure out how to run the line downtown. All the while screwing over low-income people in Rainier Valley. That is a recipe for a rebellion alright. Might be a good time to get in the pitchfork business.

   3. I disagree that Graham Street Station isn’t needed, but the purpose here is not to litigate that. It’s to look at how ST makes decisions, and if the decision making process (especially as it relates to the Enterprise Initiative) passes muster. I think it doesn’t, and Graham Street is just one example of that. This kind of decisioning is a big part of why the cost overruns exist in the first place – even before overruns these were some of the most expensive light rail projects ever conceived – and if we want to build out more transit, we have an interest in making sure we have some measure of affordability and practicality in our decision making processes.

      1. Agreed. I think that is one of the big reasons why this is such a good post and mandatory reading for the board.

         The planning process is flawed. We know it is flawed which is why they started the Enterprise Initiative in the first place. Yet Sound Transit seems to be ignoring the principles of that initiative. This is a major problem and yet the board seems ready to rush to the next project and worse yet, make irreversible mistakes along the way. If they skip Avalon it won’t be an infill station. It will be gone, forever.

         Graham Street is just the canary in the coal mine. The fact that it is dead should be a major wake-up call for the board. Things need to change.

      2. Oh, and Graham Street is just one example. Consider the Kubly/Reed plan: https://www.theurbanist.org/op-ed-reconnect-and-automate-ballard-to-west-seattle-rail-to-save-st3/. This is a serious proposal made by two men who have major transit and governmental experience. Yet Sound Transit has refused to take the idea seriously. Their response is not based on sound engineering or even bureaucratic concerns — it is based on bullshit. Worries about capacity even though the line would have *more* capacity than the current plans. Fear of spending a little extra time reworking the EIS despite the fact that you are simply shrinking the stations and running a different train set. It is clear that they simply are too lazy to consider alternatives and just want to keep going on the same path, regardless of what happens or what permanent flaws will occur. It is the complete opposite of the Enterprise Initiative goals.

9. NOOOOO we might be cooked if Graham St is deferred yet again after multiple attempts to build it. We should be getting infills on MLK to remove bus service along it because while 106 ridership is good on MLK, 1 Line ridership is bad. Making the 1 Line the only option on MLK would be sensible (though we should add Graham St as we can remove the 106 off MLK with that station because all key communities on MLK would be served). The Graham St Station also covers the huge gap between Columbia City and Othello.

   PLEASE DON’T CUT THE STATION!!!

   1. Mr. Doo,
      (a) 1 Line ridership in the RV is not “bad”. Rainier Beach Station is a laggard, but Mt. Baker, Columbia City and Othello are well used.
      (b) The distances between the light rail stations are long enough that there needs to be a parallel bus route along the route. Large housing projects have been built at station nodes, but we need to have small-scale affordable housing built outside of the light rail station footprints in order to encourage more businesses and an active community. If anything, there needs to be more bus service along MLK, not less.

      1. The distances between the light rail stations are long enough that there needs to be a parallel bus route along the route.

         Right. But if they could add enough infill stations, you wouldn’t need the bus. Or maybe you get by with a coverage route (not the relatively frequent 106). From a ridership-walking standpoint the worst gap (by far) is between Othello and Columbia City. This station helps reduce that gap. Graham also has the highest ridership of a bus stop that isn’t adjacent to a Link station.

         1. “But if they could add enough infill stations, you wouldn’t need the bus.”

            That’s my point.

         2. Light rail shouldn’t replace a bus…. Then it would have been fine to just have a bus in the first place? Light rail is supposed to be faster.

         3. No one wants a bus on rail for $20B.

         4. “Light rail shouldn’t replace a bus…. Then it would have been fine to just have a bus in the first place?

            Blame Sound Transit, we have to pay their consequences of adding at-grade rail on MLK.

            “Light rail is supposed to be faster.”

            What are you saying?

            “No one wants a bus on rail for $20B.”

            Who said that?

         5. Light rail shouldn’t replace a bus

            That is absurd. One of the big arguments for light rail is that it replaces buses. This saves the agency a huge amount of money on operations. The same is true for subway lines. It may not be the *only* reason you have rail but it is definitely a big one.

            Consider trips from Northgate or the UW to downtown. The buses were often faster than the train for those trips. But by forcing riders onto the train it allowed Metro to provide bus service to other places. At the same time, riders from those areas have a lot more connections via the train (e. g. Northgate to the UW is dramatically faster now).

            The same thing is true for Rainier Valley. Of course a bus is just as fast if you are traveling in Rainier Valley. But if you are trying to get to Beacon Hill, the train saves a lot of time. If you are trying to get downtown the train saves a lot of time. They could run express buses to the airport that might be faster but they would need to run them from several different locations. More to the point, what’s done is done. We aren’t going to replace Link with express buses to the airport. It makes sense to leverage it, and that means maximizing the areas that provide a lot of value, like service in Rainier Valley.

      2. I just think it’s a waste of resources to have two transit lines running on the same corridor when you know one is better.

         (a) I’ve seen the statistics, and you might be right. Rainier Beach is in a dead-end location and lacks TOD. Rainier Beach also lacks regional connectivity and integration.
         (b) Ok? Why not just improve Metro FLEX marketing so MLK residents can ride FLEX to their local light rail station or just have residents walk a few blocks to Rainier Ave to ride the R Line? You know, there’s lots of options though if you continue to run a 15 minute bus against a 6 minute light rail in the future of course the 15 minute bus is going to lose and those large scale developments are in favor of light rail, not in favor of buses.

         1. If we want to go the infill route and eliminate bus service along MLK, we are going to have to add another infill station between Mt. Baker and Columbia City and a station between Rainier Beach and Othello–in addition to the Graham Street Station. Those stations will ultimately add X number of minutes to the 1 Line schedule and increase the number of accidents at the stations. Bad and bad.

         2. “I just think it’s a waste of resources to have two transit lines running on the same corridor when you know one is better.”

            They’re for different purposes. Look at South King County. Link is the quasi-express service to regional destinations/transfers (airport, colleges, Seattle, Eastside). RapidRide A is for trips to the areas in between on Pacific Highway (supermarket, restaurant, adjacent suburb).

            Link could be full-stop on MLK if it weren’t also trying to go all the way to Federal Way and Tacoma. That’s what a local tram in Europe would be like. But that’s not the role Link was designed to fill.

            And the cost of the 106 overlay is so small that it’s not every worth worrying five seconds about.

         3. “They’re for different purposes. Look at South King County. Link is the quasi-express service to regional destinations/transfers (airport, colleges, Seattle, Eastside). RapidRide A is for trips to the areas in between (supermarket, restaurant, adjacent suburb).”

            Though in this case, the 106 runs on the same street as the 1 Line unlike how the 1 Line runs on I-5 and the A Line runs on 99. Though it makes sense, they interact like shuttles.

            That used to also be true with the 174 and the 194, the 174 was the local between Downtown Seattle and Federal Way via Georgetown, Tukwila, SeaTac, and Pacific Hwy. Meanwhile, the 194 used to be the bus route from Downtown Seattle to Federal Way via I-5, Tukwila, SeaTac, KDM, and Star Lake. That’s basically the relationship the 1 Line and the A Line have today though reflected throughout different generations… As they say, history repeats itself!

            “Link could be full-stop on MLK if it weren’t also trying to go all the way to Federal Way and Tacoma.

            That’s what I’m trying to do, isolate the MLK segment as a full-stop neighborhood tram.

            “That’s what a local tram in Europe would be like. But that’s not the role Link was designed to fill.”

            Well that’s what I was trying to do.

            “And the cost of the 106 overlay is so small that it’s not every worth worrying five seconds about.”

            Not to mention the ridership on MLK is the highest on the 106, it outwits Renton Ave.

            This overlapping shouldn’t be a big deal anyway given the minimal 1 Line stops on MLK and how the 106 is better in terms of being a local Rainier Valley/MLK route and the 1 Line being a regional tramway with stopping in the Rainier Valley being the elephant in the room.

         4. Doesn’t even make sense. Link stations can never have the same stop density as a bus. A bus can stop in front of stores, apartments, and other community centers. Link is meant to get you a bit further than that, usually paired with parking, TOD or bus transfers.

            If the A Line or 106 is deleted, it will force people to walk 10-20 mins from a Link station to their home .. versus 106/107 can drop them nearby. It’s an accessibility issue

            Light rail is never a replacement to bus service. And I’ve repeated this many times. Unfortunately we’re seeing light rail justify the replacement of essential bus services in South and East King County.

         5. All good express bus and light rail train services should come with a parallel bus route. Since there is express service, it doesn’t hurt for the route to make a deviation. That means one or two seat rides are seamless from all local stops.

         6. “Doesn’t even make sense. Link stations can never have the same stop density as a bus. A bus can stop in front of stores, apartments, and other community centers. Link is meant to get you a bit further than that, usually paired with parking, TOD or bus transfers.”

            Are you saying Link stations are less dense than a bus stop? Should be the opposite.

            “If the A Line or 106 is deleted, it will force people to walk 10-20 mins from a Link station to their home .. versus 106/107 can drop them nearby. It’s an accessibility issue”

            And an equity issue too.

            “Light rail is never a replacement to bus service. And I’ve repeated this many times. Unfortunately we’re seeing light rail justify the replacement of essential bus services in South and East King County.”

            Tell that to Sound Transit and their planners who want truncated buses to light rail stations.

         7. “All good express bus and light rail train services should come with a parallel bus route. Since there is express service, it doesn’t hurt for the route to make a deviation. That means one or two seat rides are seamless from all local stops.”

            I can see why you said this, you want to reinstate the 550 and 554.

         8. “That’s what I’m trying to do, isolate the MLK segment as a full-stop neighborhood tram.”

            Rerouting Link to a Georgetown bypass corridor and downgrading MLK to a typical tram route is a far larger issue than we can get into here. First you need the Georgetown bypass. And ST removed it from it’s long-range plan in 2014. The beneficiaries of it, South King and Pierce, didn’t lift one finger to save it or pay for it. They didn’t even say one word on behalf of it at the board meetings I attended. So it’s gone, dead, flown the coop, kicking up the daisies, it’s an ex-parrot.

         9. ” Link stations can never have the same stop density as a bus.”

            A surface light rail or subway can be full-stop (0.5 mile) to replace a bus corridor. But not one designed like Link, which has a different role, and has responsibilities to get people to the airport and Federal Way.

            Seattle could have a half-dozen local tram lines like many European cities do. But it chose not to go that way. And its track record with streetcars left many with the impression that it couldn’t be trusted to make them as exclusive-lane and signal-priority as MLK, or without excessive stations (every two blocks). The theoretical tram corridors are largely agreed on: Westlake/Fremont/Ballard, Eastlake/UDistrict, Aurora, Jackson-Rainier to Mt Baker (North Rainer). All of these have been elevated to current/future RapidRide, the alternative to trams.

         10. “I can see why you said this, you want to reinstate the 550 and 554.”

             SKR is mostly concerned about the Renton-Kent area, and South King County in general. It has different issues than the 550 and 554 corridors. Relevant routes are 101/102, 162, other theoretical express routes, and filling in local-bus frequency and coverage.

         11. Link is not a tram. I’m for building trams in high density areas but that is a different fish to fry.

             My point stands, a bus must stay in most scenarios. People rely on the bus. Link is just a bonus.

             In NYC we still see buses despite them having a subway.

         12. While Renton/Kent are my concerns since that is where I live, I also voice opinions about transit around King County.

             I believe the 545 shouldn’t be deleted. The 550/554 is debatable, and Metro service is sufficient.

             I also have strong opinions about transit to Federal Way, Tacoma, and Rainier Valley… As well as Bellevue, Bothell, and Kirkland. I also care about transit to community colleges and university campuses as well as general regional connections from different subareas. In particular I call out the lack of appropriate connections from the Eastside to SeaTac and South King County.

         13. Maybe you can write an article about your ideal level of service in those areas. What express-bus frequency to where? What local-bus frequency in what major corridors?

             Why should the 545 be preserved? What about the 542? ST is preserving the 545, so it’s doing what you want.

             You can address these in an open thread or Page 2 article.

         14. I actually mean restore the 544 to replace the 545.

         15. “I actually mean restore the 544 to replace the 545.”

             Now that’s even worse.

             I can see you prioritize redundant peak buses.

         16. If we want to go the infill route and eliminate bus service along MLK, we are going to have to add another infill station between Mt. Baker and Columbia City and a station between Rainier Beach and Othello–in addition to the Graham Street Station. Those stations will ultimately add X number of minutes to the 1 Line schedule and increase the number of accidents at the stations. Bad and bad.

             First of all, I see no reason why infill stations would increase the number of accidents. If anything it would be the opposite.

             Second, we want more stations, not fewer. That is how you get the network effect. Just look at Beacon Hill ridership by direction. A high proportion of riders are heading south, not north. It stands to reason that a lot of those riders are heading to Rainier Valley. By adding the Graham Street Station (and similar stations) you increase ridership at Beacon Hill. What is true of Beacon Hill is true of Capitol Hill. And the UW. And all the downtown stations. There is a real network effect in the city, which is why you build a metro in the first place. If all you are trying to do is provide fast trips from the suburbs into the city you sure as hell don’t build a metro. You run express buses and commuter rail.

             Look at subway systems around the world. The highest performing systems are those with a lot of stations. Of course long distance trips on the train take a while. But a lot more people save a lot more time by being able to walk to the train.

             That doesn’t mean you can’t have buses along the corridor. It just means that buses there aren’t as important. They don’t have to be as frequent or connect to downtown.

         17. Ross, so you agree?

         18. Ross, so you agree?

             With what? It isn’t clear what we are arguing about anymore. We have definitely strayed from the original subject. Here are my thoughts:

             1) We should add more stations in Rainier Valley.
             2) Hopefully enough that we don’t need extra coverage along the line.
             3) That doesn’t rule out having a bus line as well. It is common in cities like Paris and New York to have *both* a train line and bus line with frequent stops.
             4) But having better coverage allows more flexibility. Right now, if you sent the 106 to First Hill a lot of people would complain about having to transfer to get downtown. With more Link stations, fewer people would be hurt by that alteration.

   2. “We should be getting infills on MLK to remove bus service along it because while 106 ridership is good on MLK, 1 Line ridership is bad.”

      Link was never intended to be a full-stop service to replace local routes. It’s a limited-stop service to connect regional centers together, with stations at strategic points in between. The only neighborhood approaching full-stop is downtown Seattle, because it has an extraordinary amount of density, transfers and destinations — because it’s downtown and the center of the network. Downtown Bellevue may approach full-stop too depending on how you look at it, and by no coincidence it’s the second-largest job/activity center.

      Full-stop routes have stops every 0.25 mile (local bus), 0.3-0.5 mile (RapidRide like BRT), or 0.5 mile (NYC subway). At that level you don’t need a local overlay, or it’s debatable for a distance like New York.

      Limited-stop routes have stops every 1-2 miles. Link, Swift, and some ST Express routes are in this category (512, 522, 550). These need a local overlay because you can’t walk to a station from halfway between unless your name is asdf2. The spacing can be stretched if there’s little in between (industrial area, only single-family houses, highways, lake, etc). But typically it’s 1-2 miles.

      Mt Baker – Columbia City: 1.4 miles
      Columbia City – Othello: 1.9 miles
      Columbia City – Graham: 1.3 miles
      Graham – Othello: 0.6 mile
      Othello – Rainier Beach: 1.3 miles

      It wouldn’t be fair to redesign Link to make it full-stop in only one non-downtown neighborhood. It should be all of them or none. The decision for none was made in the 1990s, and all the existing Link is based on it.

      If any area should be full-stop, Ross has suggested the Capitol Hill area:
      – First Hill (Madison St)
      – Pine & Bellevue
      – Capitol Hill station
      – 15th & Thomas
      – 23rd & Aloha
      – 520

      That would replicate the 43 and make it arguably redundant, because you could walk to a station from almost any part of it. It’s one of the densest and highest-ridership areas outside downtown, with a wide range of destinations that draw people from everywhere 24 hours, and Seattle Central college. In retrospect this station distribution would have been ideal, but nobody mentioned it at the time. The discussion was about one more station in another alternative (Bwy & Roy) and losing First Hill station, not about four more stations beyond that.

      1. “106 ridership is good on MLK, 1 Line ridership is bad.”

         Precisely an indicator to remove 1 Line stops, rather than adding new ones. The ridership is from SeaTac and nearby P&Rs.

         1. No, the statistic is wrong. Ridership on both the 1 Line to Rainier Valley stations and the 106 is good. It’s less than Capitol Hill or North Seattle but it’s not bad. Both corridors are justified, strong, and have future growth potential, even though they’re on the same street.

         2. It’s an anecdote but whenever I ride the 1 Line, the MLK stations are heavily underused. Most people get off at TIBS or SeaTac. Rainier Beach should be dropped unless there’s a plan to use it as a transfer hub for SE King County and the SR 167 corridor… Which there isn’t. Boeing Access is also a waste. I’m ok with Graham St , Othello, Columbia City. Those stations seem to pull some riders. Mt .Baker is usually not so good but it has TOD so can’t justify it. I just think a bypass or at least some way to speed up MLK is more important.

         3. “whenever I ride the 1 Line, the MLK stations are heavily underused.”

            Define underused. What’s your threshold for good ridership, and why?

         4. “It’s less than Capitol Hill or North Seattle but it’s not bad.”

            Yes, that’s my point. Sorry for the hyperbole btw.

            “Both corridors are justified, strong, and have future growth potential, even though they’re on the same street.”

            In my POV, the 106 is better in terms of being a local route picking up residents of priority along the corridor who live there while the 1 Line is better in terms of being an express, and a regional connector.

         5. I don’t know why people think Link ridership in Rainier Valley is poor. It gets about 10,000 riders a day (as of the last report from Sound Transit). That is more than the four stations that make up Lynnwood Link. There is also a strong network effect. You can see that a lot of people are going both directions on Link in Rainier Valley (https://seattletransitblog.com/2025/08/25/ridership-patterns-for-link-1-line/). In contrast, almost all the trips with the Lynnwood Link stations are the same direction. This means that by adding the station in Rainier Valley, you likely increase ridership all over — including Rainier Valley. In contrast, a station at Ash Way will likely just result in a shift of riders from Lynnwood Station (while the stations in Mountlake Terrace and Shoreline remain the same).

            As for the 106 it does OK where it overlaps Link. A northbound bus has about 850 boardings and 1,000 alightings on MLK. The pattern is interesting. It is easy to assume that a lot of people are coming from the south and transferring to Link. That doesn’t appear to be happening. A lot more people get off the bus in the main part of Rainier Beach (close to the school) than by the Link station. The 106 isn’t acting like much of a feeder along MLK, either. For example there is a big gap between Othello and Columbia City. About 400 people board between there. But at Edmunds (the obvious transfer spot) there are only 90 alightings. Some of those riders are not transferring but taking trips from say, the Safeway in Rainier Beach to the apartments close to the station. Overall, Link appears to have little influence on the 106. A lot of those riders have the option of taking Link but prefer the 106. Maybe that would change if there were more stations on Link but maybe not. Thus the 106 is not so much a shadow but a bus that goes other places (Rainier/Jackson to the north and Rainier Beach/Skyway/Renton to the south). That means it would likely be retained even if you add a lot more stations in Rainier Valley.

         6. “I don’t know why people think Link ridership in Rainier Valley is poor.”

            Yes I agree. Its ridership is still healthy even after a Covid dip. And a healthy number of people do board it to go in both directions.

            I will also add that it gets used all day. Unlike Lynnwood Link, which has crowd surges at peak commute hours, the RV stations get riders all day long. So there won’t be some sort of crowd that appears for a particular train arrival (except when Franklin High dismisses students).

            Because there are news items about occasional acts of violence somewhere in SE Seattle and the demographic mix looks different than most of the region, people who don’t live in SE Seattle can have “warped” conclusions and opinions about the segment. Over time, I’ve read things like it takes too long when the street running speeds only add about 5-7 minutes given the distance (the area is geographically bigger than people sometimes think), or it doesn’t get riders because the riders are less likely to look like them so they’re more “invisible” or that it feels unsafe because a higher number of immigrants from other parts of the world board on the train dressed differently.

            While there are many people who celebrate diversity as a general concept, there remains a tribal aspect buried in our psyches that sees people that don’t look like us as either more dangerous or more invisible. And it will show up as a factor in various observations.

            It’s a reason why actual boarding data is important.

         7. “I don’t know why people think Link ridership in Rainier Valley is poor.”

            Because of the redundancy with the 106 and the lack of development along MLK though that’s going to change.

            Maybe Rainier Beach is to blame why people think that.

      2. Sadly, King County deleted (almost) Route 43 in March 2016; it only exists as an in-service deadhead for Route 44.

      3. At least a fair crowd getting on and off a car for a given station. Not 1-2 people.

         1. I guess it’s still better than 2 Line ridership, but there’s an issue of really bad crowding on the 2 Line because people can’t wait 4 more minutes for the 1 Line. But if we add more cars to the 2 Line, then it’s a waste after it deviates onto I-90…

            The 2 Line is another failure because again of how slow it is. Genuinely doesn’t make sense why it’s so slow given it’s freeway running segments.

         2. The 2 Line’s slowness is a mystery. I experienced it again on Thursday between South Bellevue and CID. We should probably ask ST why it’s slow, what’s the possibility of it reaching typical Link speeds, and whether ST intended for it to be this slow.

         3. It’s still faster than the 1 Line. I can’t even convince my friends or family to ride it. Whenever I bring them on board, their immediate complaint is speed. It’s not a surprise people still choose to drive except to games where Link is clearly advantageous.

         4. On top of that, the 1 Line is largely unaccessible to areas east of I-5. Federal Way, Star Lake, Angle Lake, and TIBS is not worth going to either on transit or by car for many of us.

         5. “The 2 Line’s slowness is a mystery. I experienced it again on Thursday between South Bellevue and CID. We should probably ask ST why it’s slow, what’s the possibility of it reaching typical Link speeds, and whether ST intended for it to be this slow.”

            My guess is that it has to interline with 1 Line trains at C/ID. The best way to time the interline is to operate the 2 Line slowly between South Bellevue and Judkins Park. A 2 Line train can’t wait for an extended amount of time at Judkins Park because riders can’t cross the tracks while the train is in the station, so dispatch will slow the train before it approaches JPS.

            Has anyone noticed eastbound 2 Line trains creeping across the I-90 segment? That would blow a hole in my interline theory.

         6. It’s still faster than the 1 Line. I can’t even convince my friends or family to ride it. Whenever I bring them on board, their immediate complaint is speed.

            That is because you are comparing freeway speeds with a metro. Let’s say you are trying to get from Rego Park in Queens to Midtown Manhattan. It is noon. Of course a cab is faster. It takes about 20 minutes (unless traffic is really bad). That’s because the freeway (which includes the Queens–Midtown Tunnel) gets you right there. In contrast it takes about 35 minutes via the subway. Does that mean we should get rid of a lot of the stations on the subway and just have the train run express all the time? No! That would be terrible. It would be profoundly stupid. Subway lines have stops. Lots of stops. New York City has by far the best transit system in the United States and that’s because of all those stops.

            Of course it doesn’t work that well for long distance travel. For that you can add express bus service. That is exactly what they did from that part of Queens to Midtown Manhattan. It is why ST will continue to run express buses from Tacoma to Seattle even though they finished Federal Way Link. Subway lines — even those with huge distances between stations — are really not well suited for long distance travel, especially if they are competing with freeways. They are well suited for trips that aren’t that long but would otherwise take a while by bus. For example Graham Street to Capitol Hill. The more stations you add the more combinations you have as well (a network effect). This is where a subway makes sense — not long distance trips where everyone is going to the same place.

         7. “My guess is that it has to interline with 1 Line trains at C/ID.”

            The slowdown is in segments throughout East Link. It happens both directions, even when trains have already left CID and are heading to Redmond. Slowdowns I remember:

            – From CID to Judkins Park
            – Maybe between Judkins Park and the shore
            – Downtown Bellevue tunnel
            – Wilburton area?
            – Bel-Red surface segments
            – Overlake Village segment
            – It was slow initially in the Marymoor Village segment but I’m not sure if it still is.

            Of these, only the Bel-Red surface segment is understandable, to reduce the risk of collisions. But still it seems slower than MLK. The floating bridge segment might also make sense, due to the unique risks with the bridge. I don’t remember if the floating-bridge segment is slow, but I’m inclined to not count that segment if it is. The other segments are grade-separated, so why can’t it run full-speed there like it does in other grade-separated segments?

         8. “A 2 Line train can’t wait for an extended amount of time at Judkins Park”

            It can stop before the CID entrance, as most merging trains at a “Y” would do if they come simultaneously. The only thing it would impact is a following 2 Line train. The next train doesn’t come for 6-10 minutes. If it does because trains are off-schedule, the signal system will tell it that track segment is occupied and it must stop in the previous segment.

         9. “Has anyone noticed eastbound 2 Line trains creeping across the I-90 segment? That would blow a hole in my interline theory.”

            No, the trains I usually ride across I-90 are pretty quick.

      4. Link was never intended to be a full-stop service to replace local routes.

         That doesn’t mean it can’t be. Nor does it mean that it was designed well. The very wide spacing is one of its biggest flaws. It is why it doesn’t carry that many riders despite the huge amount of money we’ve spent on it. It is why the overall transit system in Seattle just isn’t that good.

         The only neighborhood approaching full-stop is downtown Seattle, because it has an extraordinary amount of density, transfers and destinations

         It is also because Sound Transit didn’t build those stations. King County did. Oh, and ST got rid of one of the stations (Convention Center). There are fewer transit stations downtown then when we just had buses there. There are other gaps that are worse. There is only one station between downtown and the UW. This too is an area with amount of density, transfers and destinations and yet it has few stations. ST had other priorities.

         It wouldn’t be fair to redesign Link to make it full-stop in only one non-downtown neighborhood. It should be all of them or none.

         But cost has to be a consideration. Adding First Hill to the existing line (as was the original plan) would cost a fortune. Same thing goes with adding a station next to the HUB inside the UW. This would take billions. But surface stations are at best a few million and at worst a couple hundred million (as we’ve seen).

         As I wrote up above, it doesn’t have to be one extreme or the other. You reach a tipping point and frequent bus service to downtown along the corridor is no longer a priority. You want to cover the gaps with, well, coverage service. In other words less frequent buses that don’t go downtown. Some of the gaps are huge right now. But in some cases it doesn’t effect many riders. For example, look at that gap between Columbia City and Mount Baker. It is huge. But much of the area in between abuts a green belt. This means it gets very few people from the west. Some of the riders can easily walk to the 7 — a frequent bus that should be very frequent. You have relatively few people that would have to walk a long distance to transit if they ended the 106. Thus adding a coverage route (with a bus running every half hour) would be quite reasonable. There is also the possibility of routes that follow the corridor for a while and then deviate. This is similar to how the 101 and Swift Blue operate. Swift Blue is designed to serve every major stop along that part of SR-99. The 101 serves the coverage areas. But then the 101 turns and leaves the corridor, connecting riders to stops along 128th SW. The 101 only runs every half hour — even during peak. It is a coverage bus, even though it overlaps an area with frequent service (Swift Blue runs every ten minutes).

10. https://seattletransitridership.com/

    Look at the ridership levels for the 106 along MLK. Several of those stops generate over 200 on/offs a day. Replace that FLEX?

    The 106 serves over 5,000 riders a day, mostly along MLK. The 106 is a useful and productive resource–even at 15-minute headways.

    1. “Look at the ridership levels for the 106 along MLK. Several of those stops generate over 200 on/offs a day. Replace that FLEX?”

       Replace that with light rail.

       “The 106 serves over 5,000 riders a day, mostly along MLK. The 106 is a useful and productive resource–even at 15-minute headways.”

       I notice that too.

11. ““With special priority for Everett and Tacoma because they’re the largest cities in their county.”

    “Because they’re crucial for the regional spine.”

    There’s a lot to disentangle here. There should be a high-speed service connecting Everett, Seattle, and Tacoma. Something faster than Link. People want their trips to take around half an hour. Everett to downtown Seattle, and Tacoma Dome to downtown Seattle, are over an hour on Link. That’s not a way to connect a region together robustly. The right technology would be heavy rail at 85 mph. A two-level network: a fast one (S-Bahn) with extra-limited stops for the long regional distances, and a traditional city subway network (metro) for the densest/most walkable areas (Seattle). They might share a station at UW because it’s such a major regional draw.

    ST is trying to do everything with with a hybrid in-between technology. It’s mediocre at both of them. Its speed is lackluster to Everett, and verging on intolerable for Tacoma. At the same time it doesn’t have those “43 replacement” stops on Capitol Hill that could replace a high-volume bus corridor, or even half the “106 replacement” stops on MLK.

    Our network is less efficient, serves people less well, and leaves some trips out, because it’s not two-level and doing both things excellently.

    For Everett it’s a wash compared to ST Express, and the 10-minute Paine Field detour is Snohomish/Everett’s own fault.

    For Federal Way and Tacoma Link is not fast enough to replace the express buses, and Link takes an hour for 25-35 miles, which is poor.

    Most trips from most parts of Snohomish or Pierce County are only moderately helped with Link if at all. It’s not “the savior of the Spine”. And Pierce really needs to address its local frequency, which makes Tacoma Dome station (and thus Central Link) hardly a viable option without a car. Sound Transit could have done a lot for intra-Pierce service if the Pierce delegation had prioritized it.

    “I really see that the ST3 light rail projects are replacing communities with abundances of express/peak buses. Ballard has the D Line, the 17X, and (formerly) the 29 cross the Ballard Bridge.”

    Link doesn’t really address the 17 or 28’s area. It’s for the Ballard village area and its immediate fringe.

    “West Seattle (oh my gosh) has the C Line, the H Line, the 21 (and it’s express variant), the 50, the 56, the 57, and the 125 ALL CROSSING THE WEST SEATTLE BRIDGE…”

    Look at West Seattle again. It’s a large area with lots of steep hills and cliffs. You can’t walk between California Ave, 35th, Delridge, and 16th. So Link can’t replace all of the C, H, 21, and 125. It could only replace one of them. And because it doesn’t go north-south, it doesn’t replace any of them. The fact that they meet together at the bridge is largely irrelevant.

    The longer the distance, the more it makes sense to transfer to a faster service. West Seattle is right in that in-between stage where buses fanning out from the bridge is reasonable, and arguably better than West Seattle Link. West Seattle Link is a solution in search of a problem.

    “ANYWAYS, EVERETT HAS THE SWIFT GREEN LINE, FUTURE SWIFT SILVER LINE, 201, 202, 510, 512, 513 ALL RUN ON THE LIGHT RAIL ALIGNMENT!!!”

    Only the 201-202 and 510-513 could possibly be replaced with Link. Swift Green doesn’t parallel it; it shares one or two adjacent stations at most with it. Swift Silver, where will it go? It’s a line on 128th Street?

    1. I still think we can get a hybrid system working, if ST was more creative.

       Automated trains with top speeds over 100 mph. Select trips “skip” past dense corridors for the express aspect. Other trips run along the existing Link corridors to provide local connections. Sort of like an open BRT but for trains.

       Hit the big cities and destinations along the “spine” and make sure downtown Seattle, UW, the airport, and other key stops are served adequately. No transfers needed at all, fully automated.

       Lower frequency for express segments, but multiple express lines converge into the high density segments (like DSTT) to deliver the frequency and service in downtown.

       1. 100 mph is unrealistic for the Link alignment given its turn radii and inclines. The newer Link trains can go up to 65 mph. A purpose-designed light rail corridor can probably get up to 85 mph. The alignment would have to be designed with that goal from the beginning.

          The Link alignment is designed for trains up to 55 mph. In some segments it may be able to go up to 60 or 65. But not 100.

          So you’d have to build a new trackway for your 100 mph vision. The track and stations couldn’t go everywhere Link currently does. Quad-tracking for express/local overlap is possible (viz. NYC), but Link’s budget was never scaled that high.

          1. Yes, this is more so a suggestion for future expansion, since you mentioned adding high speed service along I-5. I’m suggesting it can be integrated with the existing Link corridors.

             100+ mph is a top speed for express segments. But it would obviously operate much slower if running on existing track or merging with Link for a brief duration.

          2. If we invest in a separate HSR track, then Sounder could use it like an S-Bahn at 100mph to serve Everett, Seatac, Seattle, and Tacoma. That would create a worldclass network. Our current Link tracks are trying too many things at the same time.

       2. In South King, the “express” alignment already exists in Sounder, which completely skips the RV. If you want better end-to-end speeds, invest in electrification. Even if the top speeds doesn’t go up much, the improvement in time spent entering & exiting stations is significant.

          1. That’s ignoring off-peak travel needs! Three-quarters of people’s trips are non-work trips. A significant number of work commutes are outside peak hours. Especially in lower-income areas like Kent and Auburn.

             Sounder is not a complete express solution. It’s a partial supplement peak hours.

          2. So? Steadily expand the span of service of Sounder over time and run ST Express routes when Sounder is not in service.

       3. I still think we can get a hybrid system working, if ST was more creative.

          Yeah, sure. Theoretically you can build a really good hybrid system from scratch*. You could build a double-tracked line with good station spacing inside the city but extending out to distant cities. They trains can be automated and really fast. The problem is, it is a bad value. That is why cities just haven’t done it. They end up getting cheap. BART shortchanges both San Fransisco and East Bay. DART shortchanges Dallas. Denver RTD shortchanges urban Denver. They are so focused on quantity (i. e. distance) that they ignore quality. Link is no different. Assume for a second that ST3 is built to completion someday (after we raise more money). It will likely be the most expensive metro in the world per capita. It will have two separate train tunnels through downtown. But it won’t include a station in First Hill. WTF? That is because completing the spine was always a higher priority than providing a high quality metro inside the city. The problem with that idea, of course, is that everyone loses. That is one of the key elements of this comment (https://seattletransitblog.com/2013/02/14/news-roundup-geeks/#comment-292594) which compared Link with DC Metro. The riders from the suburbs are out of luck as well.

          I can’t emphasize this enough. If you build a really good urban subway then it is pretty easy to make it work for the suburbs. You can give the suburban rider “the keys to the city” by just running buses to the outer edges of the line or directly into town. But it is really hard to do the opposite. Think of someone trying to get from Lynnwood to Fremont. I used to know a guy who did this trip every day (for work). They have rail for the first part of their journey. Great! But getting to Fremont takes a while. About a half hour, counting the waiting (https://maps.app.goo.gl/2VfXJXxMMJ9jwvGN8). Now imagine they ran a train from Ballard to the UW. The second leg of the trip takes about ten minutes. That is a huge time savings for everyone, including my coworker. It is the difference between driving (which he did) and taking transit.

          But at least the train goes to Lynnwood, right? Yeah, sure. But if the train ended at Northgate and they added HOV ramps to the station it wouldn’t be much worse. A lot of riders have to take a bus to the station anyway. So they spend more time on the bus but less time on the train — not much difference. The big difference is in travel around the city. That is where it is especially slow and it is where there are a lot more destinations. If you shortchange the city you end up with a system that doesn’t work very well for anyone.

          *To be clear, I’m not talking about S-Bahn, which leverages old tracks. Thus the phrase “from scratch”. If we owned the tracks then it would be quite reasonable to be running trains through downtown and then south, branching to serve Renton and Kent/Auburn/Tacoma, etc. It would be more frequent inside the core than outside it. It would have standard (frequent) metro stop spacing inside the core and wider stop spacing in the suburbs (like an S-Bahn). It would work well for all sorts of trips but alas, we don’t own the tracks.

          1. SF does have a hybrid system – they have Muni and BART. So Seattle had a hybrid system with the DSTT prior the buses departing.

             BART’s alignment and mode are fine for regional rail; the Bay Area’s problems are mostly around land use (Mission street should be as dense & vertical as Market street, etc.). SF’s problem is lack of investment in Muni, going decades between the Market Street subway and the Central Subway. If the T opened in the 90s and the Central Subway in the early 2000s, a subway down Geary could be under construction today.

             What Seattle needs is it’s version of Muni. It could build out the Monorail (try again!), it could bring back a bus tunnel (https://seattletransitblog.com/2015/02/18/westside-seattle-transit-tunnel/), or it could build something new (a standalone Ballard Link line). All of these idea are plausible, all could be funded by ST3/4, and none should use Link.

          2. BART’s alignment and mode are fine for regional rail

             Right. The problem is that it is nuts to spend as much money as they did on regional rail. Not when you lack a metro for the urban core. BART is a hybrid system but it fails miserably as a metro, despite the very high cost and miles of track. Oh, it’s fast. It hits all the highlights. If you are trying to get from Downtown Berkeley to Embarcadero it is great. But if you are trying to get to get between *most* of the urban areas it sucks. There are only a handful of stations in the East Bay. There is only one line in San Fransisco. Yes, I know Muni complements it to some degree but the combination still leaves much of San Fransisco proper with slow trains or buses. This despite the fact that it is an extremely dense city and not that large (physically speaking). And again, East Bay is much, much worse.

             They just spent their money on the wrong thing. They were focused too much on going out to distant cities, and not enough on providing a real metro for the area. They thought BART could do both but obviously it can’t. So they rely on buses. Buses in San Fransisco. Buses in Oakland. Buses, buses, buses. I like buses as much as the next guy but it is clear these buses are doing the work that the trains should do. The opposite is true as well. It wouldn’t be that big a deal if they ran buses from these various distant places to either the metro and/or the city. Many riders would be better off. They wouldn’t have to spend so much time as their bus slogged through the city streets. This would be the traditional approach but instead they experimented with something new and different and it failed.

    2. Because it’s automated, smaller faster trains and station designs can be used. That means all day frequency is fine even if ridership isn’t super high for certain trip types. The main thing is prioritizing trains making higher ridership trip configurations, and delivering adequate frequency to the most pressing trips + deliver high quality local bus connections.

    3. Then use automated trams for the urban village high density stop setups. Like MLK or Ballard. Make it easy to transfer to the Link system, or downtown bus/subway network.

    4. ST is trying to do everything with with a hybrid in-between technology. It’s mediocre at both of them.

       Exactly. It was a bad idea, really. It is the BART experiment and once more it has proven to be a failure (given the cost). We would have been much better off taking the traditional approach:

       1) A metro inside the urban core with lots of lines and stops.
       2) Commuter rail on lines that already exist.
       3) Express bus service from the suburbs to the metro endpoints and/or the center of the city.

       Instead we wasted huge sums of money on a half-ass system. It is a metro (even though it uses light rail). It provides some key elements like a normal metro. But there are way too few stops. There should be more in Rainier Valley — there should be more everywhere. The best aspect of it is that we’ve got excellent endpoints. It may have cost way too much to get to Federal Way and Lynnwood but at least buses can serve those stops really well. They can get in the HOV lanes miles away from the station and travel in them right to the very station. We should have a lot more express buses but of course we are saving up our money for extensions of the metro that add very little value. Our commuter line is pretty good, actually, considering we don’t own the tracks. So really, everything is pretty good except for the metro itself. By far the biggest weakness is just that we were too focused on running a metro line from Everett to Tacoma instead of building a good one within Seattle.

       1. Ross makes great points about the wrong technology applications. The thing is that the Link technology was mostly appropriate for Sound Moves. ST2 stretched that out to its outer practicality. ST3 just went ahead assuming Link technology and never questioned whether it was the best technology to extend things much, much further out. (Everett Link is longer than Lynnwood Link, Northgate Link and U-Link combined, for example.)

          1. ST only put one rail technology option on the table for ST3, and its subsequent EIS and preliminary design studies. The now-laughable argument at the time was that the system could be built quicker with an abbreviated screening and narrower environmental process — like fully by 2040! Well we now see what a crock that argument was.

          2. Sound Move was intended to be part of an Everett-Tacoma-Redmond Spine. If ST chose non-scalable technology then that wouldn’t support Everett in 30-45 minutes, that was a bad choice then.

             We have evidence that that was the case. The earliest plans for the initial segment would have been surface all the way from CID, around the northeast edge of Beacon Hill, down MLK, on Southcenter Blvd, and Tukwila International Blvd. That’s a heluva long way for 35 mph maximum as part of a much longer Spine. Would it have been surface in the middle of Pacific Highway all the way to Tacoma?

             Did anyone at ST calculate the travel time of surface all the way from CID to Tacoma Dome? Link is already expected to take 75-ish minutes to Tacoma Dome. A full-surface alignment would take two hours or close to it. That’s far too long for trips between Tacoma and Seattle! Both for the geographical distance (35 miles), and for the size of Tacoma and Seattle.

             In the end Tukwila was elevated because the City of Tukwila balked at tearing up Tukwila Intl Blvd again after it had just beautified it. And after that a precedent emerged for all segments to be grade-separated. And where it’s in a freeway ROW it can be on the surface but avoid level crossings, so it’s as good as grade-separated. (Parts of Federal Way Link and Lynnwood Link are in that situation.) But the original vision was along 99, so moving it to I-5 was a change. And the moves didn’t start to speed up future travel time to FW and Tacoma, but to avoid conflict with the City of Tukwila.

12. Jack, what’s your analysis of Link’s current ridership to/from MLK stations? Robust? Disappointing? Reasonably good? What threshold would you define as “good”, and why?

    1. As Ross has pointed out, in recent years, the ST SIP has included less data. They used to include stop level data for all three modes. I have not looked for it lately and did not find it when I looked in the 2026 SIP and 2025 SIP. I could dig more. Perhaps one could treat the entire SE Seattle service as a deviation from the direct pathway between downtown Seattle and SeaTac via the freeway. It clearly is worth the cost in minutes to attract the boardings and alightings in SE Seattle.

       This string included discussion of improving the MLK alignment. Another approach would add a bypass line via the Duwamish. That would make Link more attractive for the South King and Tacoma markets. It would also take billions. A friend suggested during the ST2 discussions that there were little used or abandoned freight lines that could go between the SODO busway and the Tukwila Link alignment.

13. “we should frame any additional spending above the minimum viable station that removes car access as spending done to preserve car access”

    Yes, yes, yes! Apply this to the whole Link system, and it will be clear that we are spending billions of “transit” money on making things easier for cars.

14. “At least a fair crowd getting on and off a car for a given station. Not 1-2 people.”

    It’s puzzling why Mt Baker, Columbia City, and Othello don’t have six or more people getting on every train most of the day. We’d need an expert to analyze why. Is the transit network still not good enough to draw them, or to make people who would ride it move to the MLK area? Are they being split half and half on Link and the 106?

    Rainier Beach station is more understandable because there’s not as much in walking distance. It’s being held back partly by its distance from downtown, and partly by the city intentionally slow-walking growth to minimize displacement and give more of a chance for destinations meeting that communities needs to emerge. (An institution of higher education has been suggested.)

    But it’s not a big concern. All the Rainier Valley stations have urban villages around them, and will grow in the next ten years and beyond. It’s only five miles further to the next big destination, SeaTac, and they’re on the way.

    1. Given the housing crisis, the lack of robust TOD is still concerning. There are a handful of midrise buildings around each station, which is great, but there should be much more. The failure here mostly sits with Seattle’s zoning & impact fees. North Rainer urban village has a good pipeline, likely responding to Judkins stations as much asM Mt Baker, but the growth around the other RV stations (plus Beacon Hill) remains anemic and is a policy failure.

    2. The last 1 Line ridership info is from February 26, and February is usually one of the lowest ridership months, so it’s difficult to draw sweeping conclusions about Rainier Valley ridership from that data. But, Mt. Baker had an average of 2,723 boardings every day; Columbia City had 2,595, Othello had 3,079.

       Let’s look at Mt. Baker (because it’s approximately at the median). Its boardings in February were 2,723; divide by 220 meaningful* trips per day = a little more than 12 boardings per trip. That isn’t fantastic ridership, but it means that each trip over the course of a day gains and spills about 25 riders per trip (every boarding is matched by an alighting somewhere else). Of course, peak direction trips during peak hours would generate significantly more boardings than off peak trips.

       * “meaningful” = trips that aren’t deadheads/partial trips or early/late coverage trips

15. While P&Rs may appear to have the large crowds, you have to look at the total population’s travel needs throughout the whole day and evening, and people walking up or busing up to Link. The people driving to P&Rs and going downtown get Link — but the people around them who live at neighborhood centers or going elsewhere don’t. A 500 or 1000 stall P&R can only fit a fraction of the traveling population, so most people aren’t driving to them. Or they couldn’t fit in them because the P&R is full. P&Rs are an inefficient way to collect people because they take so much space per person — it looks like a lot of people but it’s not really.

    That may suggest stations away from P&R, or in other neighborhoods. We can have one P&R as a cachement for a large area, preferably outside the city center (like South Bellevue for cars, Bellevue TC for walk-ups and bus transfers) rather than right at the city center (Lynnwood has both together).

    The ultimate problem in South King County is a combination of insufficient transit, and land use that doesn’t make it easy to walk from a transit hub to destinations, or for a feeder bus to get a lot of walk-ups. This is because of the excessively large parking lots and stroads between the transit stops and the destinations/housing, separation of residential and non-residential uses that force people to travel excessively far for every errand, and low density to the extent that low-density blocks exist, sometimes for a half mile in a row before there’s anything more. All this together is what makes South King County or the outer Eastside or 90% of Snohomish and Pierce Counties hard to serve. The solution is to fix all these. That requires the city governments to want to do it and lead the way, and not put roadblocks in the way to hinder it. And people need to vote for more transit funding, so that there’s more service hours to fund those additional bus runs. Without that Metro and Sound Transit can only do so much.

    1. P&R are highly efficient at unique passengers per month. They do not scale for daily riders, but for someone looking to access Link once a week or just for ad hoc travel, a P&R is much more compelling than running a bus for a rider that isn’t there for 99% of the bus’ platform hours. This is why P&R are a essential part of serving the lower density parts of the region.

       As Mike says, the key is grow nodes where there is both good density around stations and improve the pedestrian access so people can travel to stations by means other than cars. In somewhere like the Rainer Valley, the street grid and sidewalks exists so the city just needs to zone appropriately, but in South King there is a need for public investment to go with appropriate zoning. Better sidewalks and street crossing are good, but when you have “stroads,” investing in the regional bike/trail network is just as important.

    2. “P&R are highly efficient at unique passengers per month.”

       You’d have to know the number of unique walk-up and bus-up riders per month to compare it. Since P&Rs on average can only fit a quarter of the passengers at a station, three-quarters are coming by some other means. They can’t park in the P&R between 7am and 5pm because it’s full, so few people can use the P&R per day. The evening and weekend spaces aren’t used much in most cases, because P&Rs tend to be in areas where ridership is heavily peak-focused. The occasional really big ballgames aren’t enough to turn that around. I take the position that if a P&R serves only a small minority of riders on a weekday, the total proportion of unique riders at all times is probably similar.

       Look at South Bellevue P&R, where I struggle to see more than six cars on a weekend. Yet that’s the number of people getting off each train, and there are six people at the bus bays.

16. “the 1 Line is largely unaccessible to areas east of I-5.”

    South King County is too large an area for one north-south Link line to serve all of it. It’s not Link’s fault the Kent, Auburn, and Renton residential areas (east of their downtowns) are so many miles east of Link. That’s what makes Link not travel-time favorable for a trip west to it then north.

    Link is also limited because it’s so far west. The center of the population is around Kent Station and Auburn Station. That would have been the best place for Link to get full-time ridership from both the city centers, east, and west. But Sounder was a distraction: it addresses traveling peak hours but not the rest of the time, and not as much in the reverse-peak direction. And it does make sense to serve the airport, and it’s geographically impossible to go both southwest to the airport and straight south to Kent and Auburn. And there was that political desire to get Link to Federal Way and Tacoma by the shortest route, which also contradicted having the Link line in the Kent Valley.

    Federal Way, Star Lake, Angle Lake, and TIBS is not worth going to either on transit or by car for many of us.”

    And why is that? it’s the lack of destinations there. And when you get to Federal Way, you have to walk past large parking lots and stroads to get to the destinations that exist. And most of the destinations are in one-story detached buildings, so there’s only one of them rather than the three or four that could be in the space. And many of them are generic national chains, some of which have branches closer to you. All that reduces the reasons to go there. So the solution is to reverse these factors. Make it a city center or village that people really want to come to, meet a wide range of needs, and make it convenient and pleasant to walk from the transit stop to the destinations.

    1. There are walkable street grids is south King County developed before WWII; Link does not serve them. Burien was a streetcar suburb. Renton had transit. Kent and Auburn were served by the Interurban (Tacoma-Seattle).

       The TDLE will fall short of downtown Tacoma. Who thought that was sound transit?

    2. “And most of the destinations are in one-story detached buildings.”

       Mike, you just described the Graham Street station area.

       “But, Sam, you don’t understand. Many of the destinations near Federal Way Station are generic national chains!”

       Again, same with the Graham Street station area. McDonald’s, U-Haul, Starbucks, Chevron. All one-story. All national chains.

       “But, Sam, the Graham Street station area about see a number of multifamily developments that will transform the area!!!”

       And, same with the Federal Way station area. It’s about to see number of multifamily developments are in the pipeline, and will transform the area.

    3. “Federal Way, Star Lake, Angle Lake, and TIBS is not worth going to either on transit or by car for many of us.”

       Who cares? They are worth going to for the people who live and work in those areas, and that’s all that matters.

    4. “They are worth going to for the people who live and work in those areas”

       The issue was why would people NOT in those areas go to them? Specifically, people in eastern South King County going to Federal Way, Star Lake, or KDM to shop and/or take Link. It’s up to those station areas to offer something unique or better than what their potential clients have in their home areas if they want their money.

       1. “The issue was why would people NOT in those areas go to them?”

          Because they’re next to a noisy freeway and currently lack development.

          “It’s up to those station areas to offer something unique or better than what their potential clients have in their home areas if they want their money.”

          The only exception here is Star Lake.

17. The political failure here sits mostly with Seattle, not Sound Transit. It would be helpful if we had a Robert Moses who would tell SDOT that we are going to take 2 lanes from MLK, too bad so sad, the station would be built quickly & cheaper, but that is not the power dynamics here.

    The Seattle mayor/council need to be in the lead here and offer up the ROW to Sound Transit to have the station built quickly. Also, the lots on all 4 corners will be redeveloped. Because the city has complete control over zoning, it should be straightforward to trade greater density (height, FAR) in return for land; it’s plausible this can be done for zero cash cost. Take the SE corner – 38th street is a glorified parking lot; that street and the sort alley to the east can be fully vacated, creating land that can be traded for the west strip of the Starbucks lot. The city can play a role in facilitating lot assembly such that the alley can be fully vacated because every building facing that alley will be demolished and replaced with a single, midrise building facing the new station.

    The fact that Mayor Wilson & her team are not advocating for a cheap, simple Graham station underscore how unserious & amateur they are. It is littler different than the Issaquah mayor who insists on Link because bus in his mind buses can’t possibly provide a compelling solution.

    1. I don’t disagree that Seattle shares some of this political failure, but I would not absolve Sound Transit of anything. I have no doubts that the city put more constraints than they should have during the planning process for Graham Street, and I would broadly agree that the City should be approaching Sound Transit ~now with a plan to get the station actually built. But I think the constraints that Sound Transit has created for itself during the selection of a preferred alternative are still Sound Transit’s fault, and they do not indicate that the City had any substantial role in those constraints. The report reads as though Sound Transit is interested in preserving car access, not that the city forced them to.

       And I would contest that something like the land swap you are describing could be done in any reasonable amount of time, or that it could even reduce costs. ST will have to pay eminent domain to the current owner to begin construction, so they would need to find a developer willing to participate in the land swap (plus get the City on board) and that would meet their statutory requirements relating to affordability in the vicinity of stations before beginning anything. Not insurmountable, but not something I would suggest for a project that needs to be simpler.

       I don’t really agree with the Mayor’s approach on a lot of things – I think she’s too timid on a lot – but I don’t think I’d put many of the failures noted here on her and her team. She just hasn’t been in office very long, and these are all decisions that predate her term. Based on the options ST has presented, the City really cannot do very much to actually reduce the costs by keeping the station in the MLK ROW.

       1. I agree. It is not up to the mayor or city to offer anything. It is up to ST to propose something and then respond accordingly. If SDOT opposes the proposal, this should be made public. Instead there is a big lack of transparency here. For all we know ST just assumed there was going to be a problem before even bothering to check.

          1. Ridiculous. You think LA or NYC city leadership just sit around and wait for LA Metro or MTA to tell them where to put stations? Seattle needs to grow up, hire real expertise, and lead.

          2. Yes, Seattle wants the Graham Street station. That much is clear.

             But SDOT isn’t going to do their own engineering study at the same time ST does. That would be a huge waste of money. Even if they did, it is likely ST would just ignore it, like they’ve ignored everything else others have recommended (Kirkland comes to mind).

    2. It’s kind of a chicken or egg situation. The region has generally expected cities to suggest station vicinities but then let ST choose the exact location and design what they want.

       So the decision to build Graham Street is a decision made by Mayor Murray. After that, ST has held the cards if the regional card game is played as expected.

       And the station is still officially “planned” and the design would still be fully funded. So any surrounding building and land owners as well as the City will be on notice that the property may be taken one day once final design is identified.

       There are things Mayor Wilson could do to change things. For example, Wilson could insist even now that any needed land acquisition must be funded on Thursday (in addition to the other pre-planning funds).

       The City could also separately fund studying more alternatives like Bellevue did by using Mayor Wilson’s leadership and council budget action. (I have to wonder what a grade-separated station would cost.) The City could probably even built the station and then given it or sell it to ST.

       But the ball is in ST’s court as now structured. Seattle’s role is as a member of the ST Board until plans and permits need consideration . ST is just standing there dribbling away and there’s no shot clock. The City isn’t even out on the court.

       1. In this metaphor, Seattle controls half the rule book but treats it like a fixed gospel. Even something like fire code can be thoughtfully challenged.

       2. The region has generally expected cities to suggest station vicinities but then let ST choose the exact location and design what they want.

          Exactly. And there is nothing wrong with that. Obviously the two agencies have to cooperate. The ST engineers need access to utility maps and similar things. But ST has a staff dedicated to doing this sort of work. They do the actual engineering. They are the ones saying “if we build the station over here it will cost a lot more than over there” or “we can’t build it there because the turn would be too tight”. Then the city can push back and say “Sorry, that won’t work — we really need it over there — how much extra would that cost?”. In this case the city basically said “north or south of Graham is fine”. Obviously it would be better to have it oriented to the north (for better coverage) but if pushes comes to shove, a southern orientation will still work (especially if it is much cheaper). But when ST came up with their proposal they went with a very expensive and yet inferior proposal for no apparent reason.

          There is either a very big communication problem — nothing new with ST — or a major lack of transparency. If the city did push back because of traffic considerations then we should know. The electorate needs to know who is responsible for the decision so they can petition for a change.

18. I haven’t seen the full list of ST3 amendment propoals yet that we were expecting today. If you’ve seen them or have any information on them, reply to the comment thread at the end of the Strauss article.

19. redundant station with low ridership. delay is the right choice untill the budget is balanced

    1. How is it redundant? There is no nearby station. Ridership is not low; it is higher *per dollar* than most ST3 projects if not all of them. And what the hell does it mean to “balance the budget”. This is clearly within their budget — it isn’t that expensive.

       Oh, and you’ve also completely, 100% missed the whole point of this article. But feel free to just jump right on the blog, read the headline, misunderstand what the point is and spread misinformation. No sense actually spending time reading when you can share such wisdom to the masses.

       1. The 106 + transfer at any other station isn’t difficult for Graham St., and the ridership isn’t convincing. It isn’t worth slowing down the overall trip along the spine.. especially when funding is already a major challenge.

          It would be a waste too if we ever do elevate or bury the MLK segment. I would wait to add more stations until we have a better plan in MLK.

          In my opinion, ST4 should fund Renton Landing to Ballard (new track from Rainier Valley to Renton wouldn’t be expensive to add and it can be expanded to the Kent Valley in the future), as well as elevating MLK and rebuilding all the stations between Rainier Valley and Columbia City. In addition, it should fund extending the 4 Line from S Kirkland across SR 520 to the UW and extend to Ballard. Then expand express bus service for the rest of the region.

          1. The 106 + transfer at any other station isn’t difficult for Graham St

             So you are saying if a bus can serve it, there is no need to run a train? Using that logic, we won’t build anything more. We wouldn’t have built most of what we built already.

             It is really bizarre how people just don’t understand the fundamentals of a mass transit system or choose to ignore them. It isn’t about long distance travel. It is about the network effect. Do you think everyone in New York rides the subway to Midtown Manhattan? Get real. They ride it everywhere. Every hour of the day you can see people getting on and off the train at every stop. The more stops you have — the more coverage you have — the more this happens.

             Maybe this is tough to understand if you live in the suburbs. I get it. Why would you take a train from Star Lake to Federal Way. Or Lynnwood to Mountlake Terrace. I used to live in Lynnwood and I had no interest at all in going to any of the Link stations north of Northgate. I didn’t have a car but if I did taking transit for trips like that would seem even less likely. But that doesn’t apply here. This is not a distant, suburban extension. This is an urban infill station. This makes all the difference in the world. Proximity is a very important factor when it comes to transit. Within a half hour you can get to the UW, Capitol Hill or downtown. It is a much quicker trip to Beacon Hill or the various other Rainier Valley stops. Another key element in transit ridership is frequency. Link is just a lot more frequent than the 106. So not only does Link go to different places (like Beacon Hill and the UW) but it goes to places like downtown a lot more often.

             There isn’t a huge amount of development next to the station but that is likely to change. They’ll change the zoning and you’ll see a lot of new apartments and retail go in. Since it is in the city (and not by the freeway) a high proportion of people living there will take transit. A lot of those people will appreciate the nearby stop.

             Then there is cost. This is way more expensive than it should be. But it is still much cheaper *per station* than anything else we are thinking about adding. Tacoma Dome Link will cost close to a billion per station. Everett Link is even more. Issaquah/Kirkland Link is about 1.5 billion per station. West Seattle Link is somewhere around $2 billion per station (maybe more). So while this project is way more expensive than it should be, it is still a bargain compared to everything else. Hopefully they can get the price down with the type of work that the author of this post suggested, making this an even better value.

       2. Also Othello is quite close to Graham St. Although it is backtracking… Columbia City isn’t even that far.

          1. Othello is a 15 min walk from Graham St.

             Columbia City is a 23 minute walk.

             Graham St / MLK has a… McDonalds, Chevron, U haul, Starbucks and a bunch of parking lots… Literally nothing right next to it anyways. It’s not worth a station.

          2. The Aki Kurose Middle School attendance zone stretches from north of Columbia City to the southern City limits with MLK being the western edge.

             https://www.seattleschools.org/wp-content/uploads/2023/07/SPSD-map-B-AAMS.pdf

             The school is about 600 feet from the planned platform.

             There is also an active state DSHS office about 500 feet away.

          3. Some stations are must-haves, and some stations are nice-to-haves.

          4. Nothing that a bus can’t serve… I’d expect a dense amount of housing or shopping to justify a station,… Or at least some regional connectivity objective. Graham St has none of it.

          5. I guess there’s no pleasing those who think Link should only go from their doorstep to their top 3 destinations.

          6. It should go to every city and major destination..not every doorstep or random street along the road. That’s what a bus is for.

          7. It should go to every city and major destination..not every doorstep or random street along the road. That’s what a bus is for.

             You have it completely backwards. A mass transit system — whether it is called a subway, metro, or even light rail — serves neighborhoods. It is not meant to connect distant cities. For that you have regional rail or buses. Graham is basically a different neighborhood than Othello just like First Hill is different than Capitol Hill or Belltown. Without this station it is 1.6 miles or about 2.6 kilometers between stations — that is huge by world standards. This helps shrink that gap to a kilometer (from Othello to Graham Street). That is still wider than average for New York or Paris. There will still be a pretty big gap between Graham Street and Columbia City — about a mile!

          8. “Graham St / MLK has a… McDonalds, Chevron, U haul, Starbucks and a bunch of parking lots”

             You must have missed the large amount of medun-density mixed-income housing in the immediate neighborhood. NewHolly replaced a public housing campus from the 1940s.

          9. “Graham St / MLK has a… McDonalds, Chevron, U haul, Starbucks and a bunch of parking lots… Literally nothing right next to it anyways. It’s not worth a station.”

             If we were to judge the need for any future station based solely on what’s existing just across the street today, we wouldn’t want ST to build any other Link extension than SLU/ Ballard.

20. Boardmember Balducci’s amendment two today to add a unit to find innovations and ways to do things better should be staffed by those here on the STB. Mike Orr, Ross Bleakney, others, I hope that you will inquire about working at ST in this capacity. Having regular transit riders both here and in other places in the Sound Transit universe can only help that agency.

    PS- She’s right in saying that the deficit shouldn’t have gotten this far before the board intervened. That’s the problem with the agency having a blank check and no guardrails, for politicians avoid the unpopular. For instance, they prefer largely hidden taxes like sales taxes over property, vehicle, and income taxes.

21. Well put, Andrew. Maybe you should be on ST’s innovations team, too; many of us believe that transit agencies need talented people from outside of their isolated ivory towers who actually ride and know transit, i.e., more than one route or mode and perhaps different systems.
    I agree, any public comments are an exercise in futility. I had mixed feelings when the ST board thought about taking more public comments yesterday after the first hour’s worth. With another hour potentially coming, they reluctantly backed off. Why have public comments if they’ve already made up their minds? Ms. Balducci’s amendment-and her other observations-were breaths of fresh air in a stagnant environment. IMO, once this or any transit agency that lacks elected boards (e.g., Sound Transit, Community Transit) has made its choice, the purpose becomes a post-hoc justification for the decision made than an unbiased look at the universe of options. It was obvious at yesterday’s meeting that Sound Transit was again making decisions by exhaustion that attempt to please as many people as possible while heading off any potential controversy first and reasonable choices that may save money. I sensed the hope that the state legislature will bail them out by allowing 75-year bonds to effectively stick future generations with the bill and to avoid hard decisions. I heard talk of putting together ST-4 to cover ST-3’s overruns, much like I read was done with ST-3 vs. ST-2. I heard talk of local tax options to finance the overruns, the one option that made sense. I heard amendment after amendment yesterday to add back something into the plan, and repeated patronizing + attempts to pacify and to support the maker of the amendment, the Ballard extension being the only one turned away. For that, an ill-informed Seattle Councilmember Strauss repeatedly said there was no more buses to Ballard, with his first comment suggesting that the last trip was at 5:45 p.m. Obviously, like many, acceptable transit is only trains, and he is unaware that BRT is also a form of “high capacity transit,” with KC Metro’s “D” line running virtually 24 hours/day from just outside where he was seated to and from Ballard. That repeatedly destroyed his credibility. Another ding to his credibility was stating that Everett Mayor Franklin could still board a bus back to Everett. Until 6:30 pm, yes. After that, it’s only Link…to Lynnwood, then down an escalator, a Sea-Tac like walk, but out in the open, to a desolate bus bay to await a bus to downtown Everett. And, Everett spans 10 miles north to south, further than from Ballard to the ST boardroom. Much of Everett is not served by ST buses, especially outside of peak weekday hours. It’s part of the so-called “spine,” yet it has had far inferior ST express bus service, including no airport service, than Bellevue, South King County, and Pierce County while residents have been paying the same taxes…probably due to all $ going for the dogleg to Boeing and the overly expensive Sounder North.

Comments are closed.