Midweek Roundup: humble bus

83 Replies to “Midweek Roundup: humble bus”

1. Some of Wilson’s housing goals from the Urbanist article:

   ““If we only plan for housing to match expected population growth, that’s a recipe for continued tight housing markets and unaffordable rents, with homeownership remaining out of reach for most,” Wilson told The Urbanist’s elections committee earlier this year in explaining why she supported going further than Harrell did. “We should be adding more neighborhood growth centers, expanding the definition of ‘near transit,’ granting social housing the same density bonus that other types of affordable housing get, eliminating parking minimums, fix the stacked flats bonus so it actually results in building stacked flats, exempt smaller projects from [Mandatory Housing Affordability] fees so they will pencil out, etc.”

   The first sentence is what I’ve been saying for two decades now: the growth targets need to include not only additional units for population growth, but addressing the backlog of some 150,000 units that has displaced existing people to the suburbs or homelessness or made them cost-burdened.

   1. One thing I’m curious about – is there induced demand for housing? The price of housing is one of the key reasons people cite for moving between cities, so it seems plausible to me that in the long run, the effect of building more housing may not control rents as much as predicted. Interested to know if there is research on this.

      1. Yes, there is definitely induced demand for housing — I think that is the point. There are plenty of people that want to live in Seattle but can’t afford it. But the demand isn’t infinite. There are only so many people that want to live in Seattle just as there are only so many people that want to live in Tokyo. But Tokyo has a lot more housing which is why it is cheaper to live there.

         (Oh, and way more people want to live in Tokyo. Japan itself has been shrinking but a lot of people have moved to the cities. About 2 million people moved to Tokyo in the last twenty years.)

      2. There is definitely induced demand, but in general it seems that the increased supply outweighs the increased demand. Induced demand is essentially the ‘housing causes gentrification’ theory.

         Here’s an article on why:

         https://cityobservatory.org/another-housing-myth-debunked-neighborhood-price-effects-of-new-apartments/

      3. I’ll add that induced demand exists when you make basically any improvement, including bike lanes, transit, and sidewalks.

         The reason why we mainly focus on cars when discussing it though is because the space inefficiencies of cars mean that it’ll never be possible to fully satisfy demand because of induced demand. This is especially true because expanding roads makes urban development worse and forces people to live further from their destinations, making the problem worse.

         However, for housing and alternative transportation modes, that is not true and we can get a lot of density that can actually overcome induced demand.

      4. Sunny,

         The problem that Mayor Wilson, or any city mayor is going to have with housing is it’s really not in their control. The Seattle housing market (population of 800,000 +) is part of the Greater Puget Sound housing market ( population of 4.5 million+) and this is part of the Left Coast housing market (population of 53 million+) Seattle now has more population that San Francisco! That’s a lot of growth in 30 years. Seattle should be proud of this.

         If California would have put the same effort as Seattle did adding more housing over the last 30 years, Seattle’s housing (and job) markets would look very different. Seattle really has been amazing at adding new housing! The reason it doesn’t look that way is other Left Coast cities didn’t help out much. It’s a lot to ask the Mayor of Seattle to change these regional market forces.

         To put this in some sort of numerical perspective…. the City of Seattle desperately needs 60,000 units of low income and senior housing. This would cost 30 billion dollars to build ($500K per unit). And yet Seattle is adding 20,000 new residents a year! The scale of growth in Seattle and raw numbers are really hard to even imagine.

         And don’t be tricked into believing zoning really has much to do with housing shortages in Seattle. I’m pro growth and pro liberal zoning, but I also understand that zoning doesn’t build housing. Capital, materials and labor build housing. Even if Mayor Wilson had a check for 10 billion dollars (that’s the bare minimum for a realistic social housing project) she couldn’t hire the labor to get the housing built before she was out of office.

         Seattle apartments cost $2000 a month or so to rent and there seems to be plenty of people who are willing to pay that. If somebody can’t afford that I feel bad for them, but I’m also happy for the 20,000 people who move to Seattle every year that see $2000 a month as a bargain. As long as a couple of thousand new residents show up every month willing to pay 2 grand a month for an apartment, how can anything change?

      5. Tacomee, what is your impression of the situation in the trades? My informal impression is that massive shortages are developing, partly due to Americans aversion to real work, and partly due to Trump scaring immigrants away.

         The “good news” w.r.t housing is that as a nation, we are now well below replacement fertility rate, and with low immigration looking at a falling population. I’m wondering if the smart money in the housing sector is already sensing this, and therefore will be reluctant to pour Capitol into what is a generational investment.

      6. And don’t be tricked into believing zoning really has much to do with housing shortages in Seattle.

         Yeah, sure, and birds aren’t real. There is so much science backing up what a ten-year old would understand it is really baffling how people can make such an absurd claim. Of course it has to do with zoning. That and the increased number of jobs in the area (which is just as real). In other words, very restricted supply along with a lot of demand.

      7. “I’ll add that induced demand exists when you make basically any improvement, including bike lanes, transit, and sidewalks.”

         It’s an admirable theoretical hope — but I don’t think it’s significantly true.

         That’s not to say that there is no value in offering better or safer multi-modal facilities. I’m just saying that it doesn’t always induce demand.

      8. Induced demand for housing and driving are similar but largely different things

         For driving one person can go from driving 5k miles a year to 10k to 20k miles a year. Or a family can go from sharing a car to all 4 members driving. The point is that the mileage driven per person can easily quintuple.

         Housing is not like that. In fact is is usually pretty consistent with the amount of square footage people want. People generally want like 700 square foot per capita (less for cities more for the suburbs). People don’t really demand huge houses besides a couple mansions. It’s why studios and one bedroom apartments still command such a high rent that’s usually sufficient.

         What can happen is that cheaper housing can have more people move from other states. But the opposite is also true more expensive housing can push people out into other states/regions as well.

         But bringing back to the main point this still means unlike building more lanes more homes still is helping more people. Whereas if you build more car lanes it can be easily eaten up just by driving more

      9. @NoneOfTheAbove

         Population stagnation on the national level doesn’t necessarily translate on the local level due to internal migration. For example Shanghai’s population continues to skyrocket despite China’s population stagnating

      10. Here’s where I’m coming from:

          I define generic demand is a term meant to describe the propensity to make a trip, regardless of mode or destination. So “induced” demand to me is making additional travel (more trips or longer trips) more desirable — usually as a result of a land use change.

          That’s a term different than making a change in a chosen mode because of some new facility. A new transit route or bicycle lane or sidewalk can get someone to more often consider making a trip without using a car (if their destination is reasonably reachable with that other mode) but that would be changing modes rather than changing overall demand.

          There’s a Walgreens six blocks from my home. I can drive, walk, bicycle or ride transit to go there. My demand to go there doesn’t change if the bus isn’t running or the street is blocked for traffic — although it may affect what mode I choose. Whether I make the trip or not is going to be more affected by the weather or what I need to buy rather than by the ease of getting there.

          Line 2 opens in 2026. That means I will be able to ride a 7 bus to Judkins Park and get to Bellevue on Link pretty easily without much waiting. So I’ll probably go to Bellevue Square more and SouthCenter less when it opens. But that is just a destination choice and not a choice whether or not to travel to a mall.

          The term “induced demand” evolved as a way to describe how major transportation investments can change the market to encourage more or more dense development (a land use change) at a macro scale. That’s an important aspect that should be highlighted. However, it seems to me to be only significant when the change is significant like a new light rail line or highway. For little things, demand is more governed by other factors — like weather, time of day or if I’m housing visitors that create more tripmaking.

      11. NoneOfTheAbove,

          Yeah, a lack of immigrant workers is already slowing down construction. It will take a couple of years, but because construction is around 5% of total employment is construction, over time our current immigrant policies may flatten out economic growth. Long term I doubt the US remains anti-immigrant however. Our economy needs them to keep growing! America has a love/hate relationship with new arrivals going back over 400 years. But because they keep the economy growing, it always doubles back to love.

          As far as native born Americans not wanting to work in construction, I’d blame our public education system and the Liberal mindset about work. Nobody really loves pouring concrete, but there’s actual jobs in Seattle doing this. With a little experience and a lot of overtime, you can make more money than a lot of college grads reading this. I think the Seattle Public Schools are slowly changing, but for decades they were completely anti-blue collar and are still anti-cop and anti-military. There is this deep educational bias in Lefty America. You see it on this blog all the time in regards to zoning. The idea that a few white collar people can change zoning and all this housing just magically appears! It has to work that way! Some Grad student wrote a paper on it! Pouring Concrete? No, thanks, I didn’t go to college to do that.

          Imagine social housing if the first 500 units were townhomes set aside for purchase at a discount for construction workers only? Because without construction workers how in hell does the rest of it get built?

   2. Tacomee,

      Thanks for your thoughts. There’s a lot of folks on this forum who like to opine about construction of housing, but I gather there are few who know much of anything about financing, permitting, or which end of a nail gun is which. W.r.t zoning constraints, I tend to agree with you, but I think most of this crowd just lumps all forms of land use controls under “zoning”. I certainly don’t, and my own personal experience is zoning has never constrained me, but a whole lot of other regs do, from tree preservation to critical areas to stormwater management, setbacks and impervious lot coverage. In my experience, NONE of those things are getting better with the recent state changes, and in fact in my local community it is getting more difficult, not less, to build an ADU on a SFH lot.

      1. NoneOfTheAbove,

         Sorry to hear your local community is trying usurp the more “pro-build” regulations the Washington State government approved. But I’m not surprised… NIMBYs always want a back door to jam things up in their “neighborhood” . The “Tree Police” are a real problem in PNW, that’s for sure.

         Albuquerque really has a strong infill/ADU game going on. The City has free plans even! Because how hard is it to build an ADU anyway? With these plans pre-approved by the city, any run-of-the-mill general contractor can print up the plans and bang out housing at a reasonable price for new construction.

         https://www.cabq.gov/planning/accessory-dwelling-unit/free-casita-construction-plans

      2. This may just be a nomenclature issue, but I’d say that you are listing things that are regulated in the zoning code. It’s common to use overlay zones for critical areas, and setbacks/lot coverage are definitely included in a typical city’s zoning code. Sometimes tree regulations and storm water ends up in building codes, but it’s more common to see them in zoning codes.

         There are definitely issues with zoning codes and other land use or building regulations making it functionally impossible to build anything, and city permitting offices fee structures rarely help, but in most cases that I’ve seen, this has basically nothing to do with recent state regulations.

      3. blumdrew,

         The way I see it is the State (which has generally avoided building codes and zoning in general as local issues) finally got involved in zoning trying to make it easier build housing and passed some new State level zoning.

         Now it up to the cities and counties to work with the State guidelines to allow more housing…. or not. My guess Spokane and Yakima take the State zoning to heart and allow more housing. Seattle? I’m a lot less hopeful, but maybe Mayor Wilson pushes reforms though?

         I know a guy who studied urban planning in college (his dad is a local builder). After graduation he got a job with a. small city in Northern California and became, in his words, an “urban jammer” . This city used all these environmental rules and super high fees to quash anybody but multi-millionaires from building anything. The guy just quit and is currently building houses with his in Eastern Washington and Idaho. This is the same struggle Seattle is going to have to really change the rules around building.

      4. Yes, I use “zoning” loosely to include regulations and permitting efficiencies that have the same effect. Both restrictive zoning, counterproductive regulations, and slow/expensive permitting processes discourage housing construction and limit its types.

         Even if zoning allows a certain footprint, parking minimums, setbacks, and FAR may eat up so much of the space that it’s not enough for a few units.

         Even if they get past that hurdle, it may take too many years to get a permit, uncertainties about whether it will be approved, expected permits being denied, design review creating delay, and too high a fee.

         However, the differences between these kinds of rules are a technical issue that the city experts can resolve, and lay experts can offer suggestions on specific changes to certain rules. The rest of us don’t need to bother about all those distinctions.

         The net result of all these rules has been a bifurcation of housing construction: single-family houses with ADUs, or large boxy ugly apartment buildings often spanning multiple lots, with little in between. That in between is what a lot of people and families want.

      5. “My guess Spokane and Yakima take the State zoning to heart and allow more housing.”

         Yes, sometimes smaller cities race ahead of the big city. Spokane and Yakima have done that, allowing their cities become more like northeastern cities. They’ve eliminated 50-70 year restrictions that should have never been enacted in the first place. Redmond has impressive policies with urban villages and in permitting Link. Shoreline has more BAT lane coverage on Aurora for RapidRide E than Seattle has. All those cities are doing things Seattle should be #1 on.

   3. There are warning signs that the housing market will continue to have difficulties. The Trump 2.0 policies are stupid; tariffs will increase prices on materials; immigration and ICE policies will lead to labor shortages. Those and Boeing difficulties may lead to recession. The Seattle Times just had a story about difficulties with the new social housing program. Even it gets some projects going, the basic double counting flaw of targeting higher rents on those tenants that can afford them to both handle maintenance and subsidize lower income tenants may hurt their finances. There units sitting vacant in some buildings with subsidized units as those market rate households have choice to live anywhere. All that said, the zoning changes will be good; it takes time for the market to respond. In the long run, we will have significant growth. The very low income households need subsidies; what is the source for that?

      1. The new social housing program is planning on buying existing apartment buildings? Can they even do this? I thought the House Our Neighbors crew were supposed to build social housing and buy current market housing? There’s nothing good about this…. leave the market rate housing alone please.

         Mayor Wilson needs to call in a couple of big construction companies and build some housing… get the ball moving in her first week on the job.

      2. tacomee, it’s hard to parse what you’re confused or concerned about, here – more so than usual.

         What’s wrong with the social housing developer buying existing unsubsidized buildings?

      3. Nathan Dickey,

         Well, there’s a shortage of market rate housing? That’s some of reason rent is so damn high? Buying some of it up for social isn’t going to help the housing market.

         Social housing developers should be “developing social housing?” Build some more housing! Everything the critics said about “House Our Neighbors” and Seattle’s social housing schemes in general is coming true. Zero planned construction, infighting from day one, board members quitting? So far the plan is buy an apartment building to two? Who gets to live in these rent controlled apartments? Board members? Friends of the Board?

         If you value social housing, I’d suggest emailing Tiffany McCoy and asking her what the Hell is going on. Read the article in the Seattle Times. It doesn’t sound good.

2. I’ll say it again – the #1 thing Sound Transit needs to do to make these service disruptions more palatable is to provide better shuttle bus options. Every planned disruption should have at least one “express” shuttle that makes no stops in the disrupted segment and runs on I-5, and another “local” shuttle that makes local stops. A timed transfer to an express bus may even increase trip speeds compared to Link. Just take a look at WMATA to see how it should be done – they often offer up to 4 different shuttle services, contracting local private bus operators too, and running in fleets of several buses leaving together to smoothly handle the large amount of people from a train.

   1. Yes. I’ve been saying this all the time. But people here act like Link is fast enough. It’s terribly slow. We need better express bus service and proper transfer timing to Link.

      Link best acts as a local service. A SODO to SeaTac direct express bus would be incredible and widely used. I-5 should have bus infrastructure to avoid traffic.

      The best transit system will have plenty of buses that branch out from Link and adequately serve surrounding regions with an express with further connections to local service. A 3 seat trip to anywhere you want (1-2 seat for higher density).

      1. The 194 was that express bus. It was much less used than Link.

      2. Because people probably didn’t know how to use it. It’s a lot easier just to stay on the train, even if you lose 10-15 mins. And most of the time, the bus gets stuck in traffic or runs late.

         In a developed system, I-5 would have fast lanes for buses and HOV exits, plus a convenient drop-off area near the airport terminals. Buses with extra space for luggage. With that advertised, it would be a no brainer to get off Link early.

      3. Link is 9 minutes slower than the 194 was from Westlake to SeaTac. But it’s more frequent (8-10 minutes instead of 15-30 minutes), immune to traffic congestion and accidents that may make you miss your flight, has higher capacity, more space for people to load and store their luggage easier, and gives a one-seat ride from many neighborhoods where people live.

      4. Just some perspective.

         To drive from Federal Way to SeaTac airport (10-11 miles), it takes about 20 minutes without traffic. Link beats it with 17 minutes, but of course including the potential hassle to reach the train station if you don’t live nearby.

         In comparison, it takes 20 minutes (roughly same distance of 10-12 miles) to drive from SeaTac to SODO. But the same trip takes 28 minutes long. If it only took 17 minutes like the Federal Way extension, it’d be a much better ride.

   2. Seattle seems to have rushed into light rail as the technology because it was popular in the western US, and Seattle felt obligated to join the club.

      The structural problem of two different agencies overlaid across the same city and region adds lowers the priority for easy transfers or strategic service offerings. In particular, ST Link expansion planning seems as much if not more about what non-riding elected officials and developers want it to be rather than about designing the system for rider productivity, ease and benefit. It’s like a bunch of children drawing on a regional map that they view as playtime rather than the daily lifeline that it should be. ST’s reliance on other agencies to provide drivers and maintenance staff I think perpetrates this mindset.

      While many large metro areas in the HS have multiple transit operators, most have a predominant one that operates most of the service in the core city. They choose system expansion to augment the system rather than to build it because it was promised it in a referendum map.

3. The talk about the Fremont Bridge brought up some ideas that are worth considering. The first is routing. The 5, 28 and E use Aurora to get downtown. The 40 and 62 do not. I would keep the 40 and E the same but other routes could change. In general it is a trade-off between serving (lower) Fremont and getting to downtown a lot faster.

   Consider the 62. There are lots of riders who take the 62 to Fremont. There are also plenty of riders who take the bus to Dexter (from the north). It is quite likely a significant number of those riders started from some place north of Fremont. They would lose out with an express. For some trips there are alternatives for getting downtown. For example, the 62 serves the Roosevelt neighborhood. Even if the 62 used Aurora it would still not beat Link. Thus not as many riders are hurt by the slow option to downtown (it is just those between the Aurora ramps and Roosevelt).

   In contrast the 5 goes along Phinney Ridge and Greenwood Avenue. There is no fast alternative for any of the riders to get downtown (at least not yet). Eventually the Pinehurst Station will be added and folks up by 130th will be able to transfer to Link but that is a relatively small subset of the riders. This creates some tough trade-offs. If the 5 was sent across the bridge it would make for a nice connection between upper and lower Fremont. It would complement the 28 quite well (especially if the 28 ran every fifteen minutes). Despite looking good on a map I don’t think it is worth it. There are too many riders on the 5 that would be worse off.

   That leaves the 28. The 28 should run every fifteen minutes. But it isn’t clear where it should go. Right now it offers an express to downtown from Fremont that is similar to the 5. It could instead go across the Fremont Bridge. But then what? If we left the 62 the same it would be doubling up a corridor that probably doesn’t need doubling up. The cheapest option would be to end at SPU. Not only is SPU a significant destination, it is the endpoint of buses coming from Queen Anne (currently the 4 and 13). This would double up service from SPU to Fremont (with the 31/32) but only for a short section. By looping around it connects western Fremont with SPU while the 31/32 connects eastern Fremont to this area. If the 28 ran every fifteen minutes and the two sets of buses were timed, it would give riders a bus running every 7.5 minutes from SPU to Fremont. This would be especially handy for those transferring (e. g. Queen Anne to Fremont). I’m sure a lot of riders would prefer a bus like the 4 continue to Fremont but that has many issues. You have to lay wire, find a layover, etc. You also run the risk of the 4 being backed up as it starts its route southbound. That seems like a prescription for bus bunching. The same thing could happen with the 28 but the 28 is a far less important bus. Riders would benefit from it running every fifteen minutes even if it was occasionally late (northbound). You would end up breaking the connection between the 28 and 131/132 but you still save money. Riders on the 28 would miss their one-seat ride to downtown. But they could easily catch a 40 in Fremont (by using the same bus stop). Assuming the bus followed the current routing from Holman Road to Fremont Avenue (and then went south) riders could walk a little ways and continue to catch the 5 (as an express to downtown). I think this is the best approach in terms of routing.

   1. The 62 is several routes in one so I wouldn’t mess with it except straightening out the Latona detour. It serves downtown to Fremont; downtown to SLU; Fremont to SLU; Fremont to Greenlake; Sand Point or eastern 65th to Roosevelt, Fremont, and SLU. Splitting it in the middle would break so many trips, and moving it to Aurora would break other trips that we probably haven’t realized as much yet.

      People in eastern 65th or Sand Point quickly learn that it’s much faster to transfer at Roosevelt to Link to go downtown rather than taking the one-seat ride. But the one-seat ride is still useful to get to Greenlake, Fremont, and SLU even if it takes 30-40 minutes. Metro’s snow instructions even recommend transferring to Link as a much faster way to get downtown.

      With the 5 and 28 the difference from the existing routes isn’t as much, so it’s not as much of an impact to move them.

      1. I wouldn’t mess with it except straightening out the Latona detour.

         I agree. The main thing is that I would have the 62 continue to cross over the Fremont Bridge. But I would do a couple tiny changes, including the one you mentioned. I would have the 62 overlap the 40 more from downtown to South Lake Union. That would make it faster while also giving more people the opportunity to take either bus if they are headed to Fremont.

      2. “I would have the 62 overlap the 40 more from downtown to South Lake Union.”

         If something else serves Dexter like the 5, that would be feasible.

   2. There are two different issues with the 5: access to Fremont’s center, and the long travel time from north of 85th to downtown. Improving access to Fremont would make travel time worse. So if you move the 5 to Fremont Ave, you could compensate by improving general transit alternatives from 80th on north.

   3. If the 5 went to Fremont, would bringing back the 5X during rush hour be a good enough solution for faster downtown commutes? If the actual time difference between Dexter and Aurora is only a couple minutes, maybe this is good enough.

      Also, riders starting from a point in between the Greenwood and Aurora could always just walk to the E for faster downtown service.

      1. Or potentially break the 5 into two routes at 85th? The south version (5) runs from 85th and Greenwood through Fremont to downtown; the north version (355) turns on 85th and takes I-5 to end at Roosevelt Station?

      2. It’s not faster downtown commutes, it’s every time you’re going a long distance on the 5, so including weekends, midday, evenings. Rush hour service addresses only a specific subset of trips (those with 9-5 jobs).

      3. If the actual time difference between Dexter and Aurora is only a couple minutes, maybe this is good enough.

         It is about a six minute difference as I write this (with very little traffic). I think this would be a considerable hit.

         Or potentially break the 5 into two routes at 85th? The south version (5) runs from 85th and Greenwood through Fremont to downtown; the north version (355) turns on 85th and takes I-5 to end at Roosevelt Station?

         I think the problem is that isn’t that fast to get to Roosevelt. As I write this, it takes 8 minutes in a car. So if you didn’t pick up anyone along the way it still takes 8 minutes before you have to then transfer to get downtown. I could see the attraction but long stretches without picking up anyone is costly. You would also force a lot of transfers for people who people who are just traveling along the corridor. Hard to say how many people there are (and whether they would prefer the express to Roosevelt).

         Where this approach definitely makes sense is up north, at 130th. It is a much shorter distance to the (Pinehurst) station. You can pick people up along the way and still get there fairly quickly. You are also much further north so the one-seat ride to downtown (from north of 130th) takes considerably longer. It would be very different if there was a station at 85th.

   4. When Metro cut service on the 28, I thought it might be better to keep the same frequency but just run it as a shuttle from Carkeek to Leary, supporting transfers to the 40, 44, and 45. I like this plan even more since it gives even more transfer opportunities for not much more cost. That said, I wonder how much of a challenge it would be for Metro to figure out how to disentangle the 28 from the 132?

      1. I’ve never understood the 28. Why does 8th Ave NW have express service when it’s mostly single-family, and the higher-density areas around it with many more destinations don’t have express service? It should be the other way round. It’s like burying the express where it’s the least useful.

      2. Why does 8th Ave NW have express service when it’s mostly single-family, and the higher-density areas around it with many more destinations don’t have express service?

         It basically just evolved that way. I don’t have maps for every change. There is a gap between 1988 and the first of Oran’s maps. But I have three different phases:

         Phase 1) The 26 and 28 went on Westlake. The 5 and 16 ran express via Aurora. The 17 ran on Nickerson and then Dexter. This was well before the 40.

         Phase 2) The 40 was introduced and was the only bus on Westlake. the 26 and 28 ran on Dexter. The 5 and 16 continued to run express via Aurora.

         Phase 3) The big shake-up with U-Link. The 62 replaced the 16 and went on Dexter. At that point it really didn’t make sense to spend a lot of service time sending the 28 via Dexter (or Westlake). Might as well save money and run express to downtown.

         This goes back to what Jack has been suggesting. Since the 62 has more riders, it should run express while the 28 should run on Dexter. There is something to be said for that (as long as the 28 runs as often as the 62). But I also think the opposite argument can be made. In other words, this isn’t a bug, it is a feature. Think of the number of one-seat rides to Dexter or Fremont from the north. There are probably quite a few. Now consider the number of one-seat rides if the 28 went that way. Not that many. The 62 also has more coverage at the south end of Stone Way. This is not an issue with the 28 as the 40 covers western Fremont. As I see it, they really don’t know what to do with the 28 once it reaches Fremont, so they sent it express to downtown. This is quite reasonable. I just think it would be better if it went to SPU.

      3. When Metro cut service on the 28, I thought it might be better to keep the same frequency but just run it as a shuttle from Carkeek to Leary, supporting transfers to the 40, 44, and 45.

         I think ending at Fremont has merit. You would have a good southern anchor while also connecting to the 31/32 and 62. I think layover and turnaround might be an issue though. I agree though, extending it to SPU (where we know the bus can turnaround and layover) would add a lot more for very little cost.

         I wonder how much of a challenge it would be for Metro to figure out how to disentangle the 28 from the 132

         It would take some work but doesn’t seem impossible. Most of the 131 buses just head back to Ryerson (according to the schedule). This is less than ideal and eats up some of the savings. But shuffling routes around is pretty common and this would be a fairly minor change (since the 28 only run every half hour). I could also see the 24 and 33 bumped up to run every fifteen minutes. At that point the 24 and 124 could be paired while the 132 gets paired with the 33. In general there are a lot more buses coming from the south than the north through downtown. That is why they added the Eastlake layover. As we eliminate the buses from I-90 (with East Link) that gives us more layover space as well.

      4. If Greenwood could get a second route, so it has a bus both to downtown via Aurora and downtown via Fremont/Dexter, that would solve the problem, while also dramatically boosting frequency of Greenwood-downtown service.

         The obvious place to take service from to pay for such a route would be by shifting the 28, having it jog over to Greenwood at 85th. The catch, of course, is that people along the 28th further south would be left without service on 8th, but there is other service they could walk to, for example, the D-line is only half a mile west of 8th, and maybe an 8th to downtown bus could still continue to run during the weekday peak hours.

         Ultimately, the hippocratic oath of transit makes this a difficult sell, but I do think more people would gain from such a change than lose. The winners would be people along Greenwood and lower Fremont, the losers would be only people along 8th, between maybe Market St. and 80th., which is mostly low-density houses. And the losers would still have the option of walking either west to the D-line, south to the 40, east to the 5 or new 28, or north/south to the 44, from point they could connect to another bus, or stay on the 44 to Link in the U-district and go downtown that way.

      5. The north-south routes in greater Ballard (17, 40, D, 28, 5, E) are pretty much perfectly spaced. It would be a bad idea to skip one just as it is a bad idea to have one run infrequently. Phinney Ridge has more people but it still doesn’t have that many. It seems hard to justify doubling up service there while leaving a huge gap between Phinney and 15th. I could see the 28 cutting over Market, which would make the transfer from the 5 to Fremont much better. That wouldn’t leave much of a gap but it also doesn’t get you that much.

      6. “The 26 and 28 went on Westlake. ”

         I forgot about that. There was probably a 28 local and a 28 peak express. The mandate was for 15-minute service on Westlake and Dexter, by whichever route(s). In the restructure that replaced the 28 local, the 28 express was probably made all-day.

      7. In the early 1990s, most Seattle corridors had a 30-minute headway local oriented to downtown Seattle along with an express one-way peak-only overlay with limited stops (e.g., routes 17 and 17X, 18 and 18X, 15 and 15X, 28 and 28X, 5, 5X, and 355, 6, 359, and 360, 16 and 16X, 26 and 26X, 71 and 76, 72 and 79, 73, 78, and 77, 74 and 74X. The 71 series varied by time of day with the reversible lanes. There were several patterns in the CBD; some express variants used the 5th and Cherry pattern. In a further complication, routes 16X and 355X ran in two directions. South and West Seattle also had express variants for routes 7, 39, 42, 56, 130, 132. Over time, there were consolidations that yielded shorter headway, improved crosstown service. Notable consolidations were Aurora in 1999 and Ambaum in 2004.

      8. Going back to the problem of connecting Lower Fremont to points north…I remember this time I was at a weekend activity at the Phinney Neighborhood Center and wanted to visit a place in lower Fremont during my hour-long lunch break. Between the walk time and wait time, getting there, getting food, and getting back, all in one hour was impossible, even though the place is only 10 minutes away by car, in a straight line.

         So, the walk time from 39th matters, but ultimately, the biggest problem with the trip is the bus’s poor frequency on evenings and weekends. The 5 used to run every 15 minutes 7 days/week until the pandemic service cuts. However, money can be found to restore its service back up to 2019 levels.

      9. In the early 1990s, Metro did have night time shuttles on routes 15, 28, and 16 with timed meets with routes 18, 16, and 6, respectively. Service is much better today. I doubt the three shuttle turnaround loops and layovers would be feasible today; there is too much traffic and demand for that curb space. Seattle is not sleepy any more.

      10. RossB and Mike,

          Yes, for the north Seattle routes, there were major changes in 1998-99, 2012, 2016, 2021, 2023, and 2024.

          1998-99
          Routes 15 and 18 improved to 20-minute headway and paired with routes 21, 22, and 56. They retained their express variants.
          Aurora restructure; 20-minute headway routes 16 and 358 replaced deleted routes 6, 16 shuttle, and 360. Route 16X was retained.
          Route 31 implemented; routes 30E and 30W deleted.
          Route 65 improved.
          This may have been when the reverse peak direction trips of routes 16X and 355X were deleted.

          2000
          There were reductions following I-695. See Route 37.

          2003, NKC
          This largely Shoreline project consolidated routes 78 and 370 into Route 373.
          Route 316 replaced Route 16X.
          The 5th/Cherry routes 317 and 377 were deleted.

          2012
          Lines C and D were implemented; in West Seattle, peak routes 55, 56, and 57 were implemented.
          Lines C and D were through routed until Seattle paid to split them in 2016.
          Route 40 was implemented; it consolidated Northgate patterns of routes 5 and 75; on Westlake, it replaced routes 17 and 30
          Routes 5 and 21 were improved to 15/15; all trips extended to SCC
          Routes 31 and 32 were coordinated; Route 30 was not possible with the Mercer Street project.
          QA and Magnolia routes were frozen in place.
          Routes 45 and 46 were deleted.
          Pay-on-entry fare collection implemented.

          2016, U Link
          Routes 26 and 26X and 28 and 28X were consolidated on the Aurora pathway.
          Route 62 was added to Dexter; Route 26 was extended north to replace Route 16
          Route 70 finally became full time.
          Routes 65 and 67 paired and improved.

          STB folks know the projects of 2021 and 2024.

          In 2020, when Seattle reduced its purchase of hours, they withdrew their subsidy of the split of lines C and D; Metro cut other Seattle routes to keep them split apart. Example: Routes 5 and 21 Saturday headway became 20 minutes from 15.

          In 2023, to correct for the operator shortage, several peak routes were suspended (e.g., routes 15, 16, 18, 29) and several all-day routes were reduced to hourly headway at off peak times (e.g., routes 20, 28, 79). Some peak routes had already been suspended for Covid.

          I suggest some changes, but not as much shifting between canal crossings.
          Shift several routes to 1st Avenue from 3rd Avenue to provide better service than the CC Streetcar would have.
          Split routes 5 and 21.
          Extend Route 5 to Lake City via the Pinehurst station. Extend another routes to SCC.
          Route 5 would use the Route 40 terminal on 6th Avenue South; Route 40 would be extended to First Hill via Yesler Way, 8th/9th avenues, and East Jefferson Street.
          Pair Route 21 with Magnolia routes.
          Route 62 needs some help. Minutes could be saved by truncating it at NE 74th Street and the east Green Lake streamlining; use the minutes to extend it to First Hill via the Route 40 pathway above. The southbound SDOT bus island on Dexter between Roy and Mercer streets should be deleted. It leads to the single general purpose lane getting jammed as the right turn pocket is not long enough.
          (Routes 40 and 62 would serve the south part of First Hill: Yesler Terrace, Harborview, Swedish, SU, and JJ).

      11. the walk time from 39th matters, but ultimately, the biggest problem with the trip is the bus’s poor frequency on evenings and weekends.

          I agree. It also depends a lot on which part of Fremont you are headed to. If you are headed to southwest Fremont (Indoor Sun Shoppe) it is a long walk. If you are headed to southeast Fremont (e. g. Fremont Brewing) it isn’t bad at all. You don’t use the stop Fremont Avenue, you take the last bus stop before Aurora (it is better southbound than northbound).

          The transfers vary as well. Southbound to pick up the 31/32 or 62 is not that bad. It is the transfer to the 40 that is bad. It is so bad that I can’t get Google Maps to suggest it as an option. At worst it just suggests walking over to the 28, catching that and then transferring to the 40* Fortunately I don’t think there are that many trips that are like that. Westlake doesn’t have many destinations between Fremont and South Lake Union. For something right at the south end of the lake (e. g. MOHAI) they recommend walking or going all the way to 3rd Avenue and backtracking.

          I don’t think there is an ideal solution given our geography. Buses going north/south are fast. Buses going east-west are slow in most places and nonexistent in others. There are ways of dealing with this, like:

          1) Having the buses “weave”. The 5 could get off Aurora and head over to Fairview. The 70 could do the opposite, following the route of the E initially (through Belltown) before heading east to Fairview and north to Eastlake. This eliminates backtracking with transfers but it makes the buses a lot slower to downtown.

          2) Create fast and frequent three-seat rides. In this case you could make the 8 much faster and a lot more frequent. It could be like the G, running every six minutes midday. It is about a 12 minute walk between 7th (with the 5/E) and Fairview (with the 70). Backtracking to downtown and making the transfer there takes about the same amount of time. The bus should take a couple minutes which means a travel time of about five minutes is reasonable. Likewise a Ballard-UW automated metro would help with this trip. A rider trying to go from Phinney Ridge to Eastlake would take the bus to the Fremont/Aurora station. Then ride to the UW and take the 70 south.

          That won’t happen for a while (if ever). But a fast 8 will speed up some transfers. I think a weave approach has too many drawbacks. I would, however, change the 62 (as others suggested). Have a southbound 62 turn on Harrison and go over to Westlake. Then it wold follow the C and 40 downtown. I think this would be about as fast when there isn’t traffic and faster when there is. This also has the advantage of a common market. Someone in South Lake Union can take either the 40 or 62 to get to Fremont. That still retains some awkward trips (Phinney Ridge to Eastlake) but improves some of them.

4. I agree that the KC Streetcar is likely one of the best applications of streetcars in the US. Its route is very straight and it connects many of the main destinations in Kansas City on one vector. Comparing ridership in KC (1,806,946 in 2024) to Seattle’s First Hill Streetcar (1,269,768 in 2024) could be misconstrued. The KC Streetcar is free while the FHS requires fare payment. The article refers to Kansas City’s “relatively anemic” bus system, which is an understatement when compared to Seattle (or Portland or SLC). Kansas City is a great place to visit, but it’s difficult to get around car-free if your destinations aren’t close to the streetcar line.

5. Here’s when I estimate Cross Lake to open:

   So, I did some calculations for ST2 openings…
   – Northgate Link began pre-revenue in September 2021, a month before opening.
   – Hilltop Link began pre-revenue in June 2023, a quarter of a year before opening.
   – East Link began pre-revenue in November 2023, five months before opening.
   – Lynnwood Link began pre-revenue in June 2024, two months before opening.
   – DT Redmond Link began pre-revenue in January 2025, a third of a year before opening.
   – Federal Way Link began pre-revenue in August 2025, a third of a year before opening.

   So…

   19 months in total / 6 openings = 3.16666666667 months from December 22, 2025 (the date Cross Lake began pre-revenue operations).

   The opening date would be sometime around March 27, 2026 (a Friday)

   1. My bad, DT Redmond and Federal Way are ST3 openings, not a ST2 openings, though it’s a estimation based on recent extensions.

      1. Well they both have their feet in both ST2 and ST3. The planning was approved in ST2.

         I’ll let others explain that.

      2. Yes, I know. The tunnel under Downtown Bellevue made DT Redmond Link be postponed to another vote (and maybe also the recession), and the southern extension from SeaTac was only to Star Lake and not to Federal Way Downtown (which was a study from Star Lake to Tacoma which included Federal Way to be exact). First the southern extension was truncated at Kent Des Moines, then at Angle Lake. Then another ballot measure which included an extension south of Angle Lake (ST3) was passed, and it’s what we know as Federal Way Link Extension which opened nearly three weeks ago when it feels like forever.

      3. ST2 funded to Redmond Tech and Star Lake. In the 2008 recession South King’s sales-tax revenue plummeted, so the project was truncated to Angle Lake. Later in the recovery it was re-extended to KDM, to open in 2023. ST3 superceded it and rolled it into the Federal Way extension, to open in 2024, and funded the rest of the way to Downtown Redmond.

      4. “The tunnel under Downtown Bellevue made DT Redmond Link be postponed to another vote (and maybe also the recession)”

         I don’t think it was either of those. The segment east of Redmond Tech wasn’t included in ST2 because it exceeded the budget size of the package. The Bellevue tunnel took money from Bel-Red (leading to the surface alignment) and the CID-Judkins Park segment and station (which North King agreed to pay for to free up money for the tunnel).

   2. I wouldn’t base a test year on average testing periods. Testing is testing everything — and a systems failure or two can delay things. WMATA and BART openings in the past decade had to be pushed back significantly. LAX Peoplemover is still not fully ready to open. Even Lynnwood was delayed a few weeks more right before opening.

      East Link is more complex, as it affects an 1 Line section between Lynnwood and CID with all of those communications as well as the split in the line south of CID. That’s on top of the long section that crosses Lake Washington as well as tying in the Eastside 2 Line.

      I’ve mentioned before, but the commencement of full simulation will probably provide service changes before riders can officially ride cross Lake Washington. Things like running 2 Line later or North Seattle getting trains every 4-5 minutes can begin earlier if ST wants.

      1. I did see somebody say (couldn’t remember who) what you said about North Seattle, the 2 Line could start running from Lynnwood to IDC (but not I-90) if ST wants, but that’s something that they could consider as it does show that the 2 Line is operating there, and running the 2 Line late is also feasible because of OMF East’s capacity. Dow said the testing is going great and he hopes to have it running by May 31st.

      2. Now I’m starting to doubt those two things will happen before the I-90 crossing since those components are still part of it.

   3. Crosslake pre-revenue service started Monday. It’s sporadic through the holidays, but will go full-time in early January. I haven’t seen any indication of passengers being allowed in Lynnwood-CID. I don’t know when that might start.

   4. ST will likely move the opening date forward; it’s just pinning down exactly what it can promise. Once it has a tentative date, it will have to confer with the host cities, and leave a six-week buffer for logistics. It will probably announce the date after all that is outlined.

6. In recent days (maybe the Urbanist podcast), there was wondering about where the full 2 Line ribbon cutting and station festivals would be.

   My armchair suggestions:

   Ribbon cutting at 23rd Ave S entrance to Judkins Park maybe with a street closure — unless Mercer Island decides to go all out to push for it at 77th or 80th Ave SE overpass with a street closure.

   Judkins Park street fair: either close 23rd Ave S or use the museum parking lot.

   Mercer Island street fair: top deck of park and ride lot at 80th Ave SE.

   Would the ribbon cutting for the first train be at one of these stations, or would it be at South Bellevue?

   Those are my guesses. Does anyone have recon on what’s being planned?

   1. If there’s an opening party, it will NEVER be at the station that exists and is the start of the extension, since this is the first extension to not have a new terminal, I would expect the ribbon cutting to be at Judkins Park (which is close to Mt Baker where the Central Link ribbon cutting was held). The party in Judkins Park could be either on Jimi Hendrix Park or on 23rd Ave S. The Mercer Island party could be either on Mercer Way or on the garage top floor. I was also wondering if they should host parties at all 2 Line stations from Mercer Island to Lynnwood City Center or just Mercer Island and Judkins Park (and wondering what they will do).

   2. I suspect the speeches will be at Mercer Island, because it’s a city that doesn’t have Link yet.

      1. That makes sense — except Mercer Island hasn’t been exactly embracing the project in the past.

         As an act of goodwill, ST should probably ask MI if they haven’t already. But they should have a Plan B just in case.

         And every other opening has been at the end station. This opening has no end station.

      2. That’s the problem, this has no end station that hasn’t been opened yet. Though I still think the speeches will be at Judkins Park because it’s easier to get to than Mercer Island. Though your point is sensible, they had speeches for link expansions inside the Seattle limits more than once, for example U Link.

      3. The cities are ST’s clients and indirectly its bosses, so ST can’t discriminate against them for being uncooperative. I still think it will likely be at Mercer Island because of these optics: Seattle already has Link, while Mercer Island doesn’t.

         On the other hand, logistics may make Mercer Island infeasible. In all the westside segments, there were multiple ways to get to the event via transit. In the 2 Line Starter Line, there were at least five busfuls of people on the 550 from Seattle, Eastsiders had other ways, and the 271 was another way from the U-District. But with Mercer Island, all that would have to be on the 550 (and marginally 554). ST would either have to add more 550 runs, or shuttles from various locations.

         Another factor is Mercer Island has a small P&R, while Judkins Park has none.

         But how will Eastsiders on transit get to Judkins Park when the 550 doesn’t stop there? They’ll look askance at going downtown and taking the 4 back.

         So maybe the solution in both cases is shuttles from South Bellevue P&R and one or two Seattle locations.

   3. The dreamer in me wished that Jimi Hendrix Park would have had a community gathering plaza with a makeshift stage across from the station for things like this — but it’s not been planned. Maybe it will in the future.

      Bring a spring festival I’d worry about crowds in a potentially muddy park.

      1. Same, I’m also hoping they can open Pinehurst on the same day. So it can be a 3 station opening and we get something for the opened segment at least.

   4. Another possibility for a Judkins Park fair is 20th Ave S between Massachusetts and State St. The street is already often closed for Hamlin Robinson school activity. State now has a pedestrian signal now too.

      1. I expect opening speeches at Judkins with the ribbon cutting at the east portal. Mercer has no rep on the board and actively sued ST more than once. I expect there will be a minor street fair there, but the big action ought to be at Judkins.

   5. I have no idea what they will do but I would vote for Downtown Bellevue. Sure, the station already exists. But with all due respect to the other stations, this is what the project is all about. You are running a train from Downtown Bellevue to Downtown Seattle. Everything else was optional. Kudos to ST for not skipping a bunch of stations (like they did with the main line) but that is the main project. It is the connection of two major urban centers. It is not about commuting to the big city from the sprawling suburbs (South Bellevue) or saving buses some time on the freeway (Mercer Island). Nor is it about another station in Seattle (Judkins Park). We will celebrate that with the Pinehurst and Graham Street opening. No, it is really about Downtown Bellevue. It has emerged to be a major center. Civic leaders should use this opportunity to show off the city — to make the claim that it is more than a soulless set of office skyscrapers and a mall. Have the usual razzmatazz nearby but encourage people to walk around and explore the area. Walking west on 6th, skirting the mall and then looping around the park is a worthy trip. Try and get food from nearby vendors so that people from Seattle get a feel for the culinary variety there.

      An opening should be about more than just showing off the new infrastructure but showing off new possibilities. Selling the East Side as a real destination (not just a place to commute from or work at) is best done at Downtown Bellevue.

      1. It’s fine they didn’t skip stations because they are well spaced. The 2 Line operates at a decent speed compared to the 1 Line.

         The 1 Line already has enough stations. If we really want such fine grained travel, a bus does the job better. A train stopping every 0.5 miles especially traveling over a long segment makes zero sense. It should be easier to transfer between Link and local buses.

      2. Unless a ton of people get off at each of these stations. Which is definitely not the case, especially along the Rainier Valley and North of UW to Lynnwood segments.

      3. “I would vote for Downtown Bellevue.”

         That’s a good point. Bellevue Downtown has the infrastructure to host it, and could easily do so again. The Crosslake gap and separate opening was never intended: it was a side effect of implementation issues. Bellevue is a principal city like the previous hosts, not a bedroom community that has positioned itself against jobs/density. Mercer Islanders identify with Bellevue and won’t contest it being host. It is part of “East Link”, which downtown Bellevue is the center of and primary purpose of.

      4. “it is really about Downtown Bellevue. It has emerged to be a major center. ”

         That’s another point. A large downtown Bellevue was planned in the 1950s. It was going to get Forward Thrust in 1980 but that failed. Highrises started appearing in the 1990s. Bellevue Square expanded at that time, and the other Bellevue Collection buildings were built. Now it has the last piece: high-capacity transit to the biggest city and the rest of the region, and the alignment is mostly away from freeways.

      5. A 0.5 mile stop spacing is not totally inappropriate with enough density. I think it’s appropriate downtown due to the density there (would be better if there were more housing there though, but that’s fixable), and that stop spacing would have been as appropriate or even better in Capitol Hill. The 1998 ST1 draft EIS had a potential station at E Roy St or E Aloha St and 10th Ave E (north Capitol Hill), a station at the current location at E John St and Broadway, and the abandoned First Hill station at Madison St and Summit Ave. This is probably the biggest missed opportunity in ST1 in my opinion, at least from a ridership perspective. The First Hill station would obviously have been a big deal for its walkshed and bus connections, but there’s also a lot of density along 10th Ave E in the form of old apartments that would be illegal to build today. I bet a northern station would have forced an upzone there.

         As Ross always points out: most of the ridership in successful systems consists of short trips in the densest locations, and it doesn’t matter how much you slow down the relatively smaller numbers of long-distance suburban riders. Worst-case you run express trains.

         Anyway, the Rainier Valley alignment might be obvious as the “biggest ST1 mistake,” because the train hitting a car (and having to walk over to Rainier Ave S) is a bigger rider experience problem than just having to walk a long way through the Hill, but it’s a shame to think how much better we could have done in some of the densest parts of the city.

      6. I could see the street fairs and ribbon cutting happening at different stations from each other.

         I would think the place to cut the ribbon would be at either South Bellevue or ID/C Station if it was going to be at an existing station.

         For one thing, trains before the ribbon cutting will be in service. It will be awkward to halt 2 Line trains in Downtown Bellevue for a ribbon cutting.

         I would think that the Eastside elected officials want it somewhere nearby. Seattle elected officials don’t seem as eager to get the PR. And there will likely be a separate event for Pinehurst if Seattle electeds want the PR.

         Regardless, it’s not of major concern in the long run. The region could even ignore the opening celebration if they wanted and it wouldn’t really affect anything. It’s just curiousity of how the desire for good press with an opening will be rolled out.

         Given the popularity of such events, maybe ST should create a “Linkfest” Saturday every year. These upcoming openings are the last ones for Link for at least several years — and community groups and others seem to gravitate to the forum.

   6. I think they could just do it in CID if not South Bellevue or Bellevue Downtown.
      Mercer Island and Judkin Park simply have no space to host the primary event and I doubt Mercer Island is capable of managing something like that.
      Last year there was a lunar new year event in Mercer island town center. MIPD was so underprepared that the small event pretty much paralyzed the entire local network north of I-90.

7. “How Kansas City Quietly Built One of the Best Streetcars in America”

   I went to KC this summer. It was really crowded on a game day. I think part of its success is because it is a free ride. It is a good transportation mode for people to do bar crawling along Main Street before and after the game. It connects destinations that visitors would like to go on one day. It is like having one streetcar line that connects Stadiums, Pike Market, Westlake, and Seattle Center. The fact that there is no other transit option make this line more special. The overall transit scene in Kansas City metro and the broader area surrounding it is very sad, so when locals see something like this they cannot stop talking about it.
   I think it is a good business decision to build this in Kansas City, but getting too excited about projects like this will lead to more projects like SLU streetcar which provides little benefit to commuters. If there is private real estate money into this or whatever automated shuttles, then go for it.

8. The Seahawks will not be in town the weekend of January 10-11. They will either be on the road as a wildcard, or getting the bye that is a reward for winning the conference regular season.

   I wonder if ST is planning to add that weekend to the tunnel closure list.

   I don’t see ST beginning full-schedule testing on the full 2 Line until all the tunnel upgrades are complete and tested.

Comments are closed.