Friday Roundtable: Graham Street Route Idea

113 Replies to “Friday Roundtable: Graham Street Route Idea”

1. Thank you for these concepts, both today’s politically ambitious major route restructure map, and having a weekly what-could-be column.

   As a frequent rider of route 60, I will have to play Corson’s advocate. Route 60 goes that way northeastbound partially because serving stops along the north end of Boeing Field is pretty pointless, except to have a faster path and basically skip Georgetown. But there is also a public college campus on the north side of Corson: the South Seattle College Georgetown Campus. A look at stop use data could show whether that might be an important destination for a new Graham St route.

   1. I should have clarified this in the post, but the Georgetown routing will stay the same (inbound on Corson, outbound on Ellis). Map only shows the Ellis routing for simplicity. Likewise, near Westwood Village, only inbound trips would stop on Barton. Outbound trips would use the same routing via Trenton (not shown no the map above). The validity of one-way route sections may warrant a separate discussion.

      It does not look like the college drives a ton of ridership. Here is the Route 60 ridership data from March – September 2024: https://seattletransitblog.com/wp-content/uploads/2025/05/DailyRidership-3.png

      1. Thanks for collecting all this data!

         One could still argue that the three-block-wide Georgetown couplet is ruining Georgetown ridership, partially by making route 60 unusable for Georgetown Campus riders in one direction.

         But the larger problem in the proposed route is missing Cleveland High School. That would probably be the second-largest trip generator, after the station, along the whole route.

         May I suggest Lucille St as a new path.? Yes there is a stub railroad track to sometimes have to triangle around, but I bet the route would end up faster, and be able to finally generate ridership to/from Georgetown Campus.

      2. The bus stop for Cleveland does well (although it isn’t the second biggest trip generator). But almost everyone by the high school takes the bus to and from the north (towards First Hill). A northbound bus picks up about 120 riders next to Cleveland. About 20 get off. The 120 riders would just switch from taking the 60 to taking the 107. They wouldn’t have the same frequency but if the 107 is bumped up to fifteen minutes (which is the plan) they would be OK. The 20 riders from the south would have to transfer or walk farther. You could easily make up for that with the 60 going east.

         One could still argue that the three-block-wide Georgetown couplet is ruining Georgetown ridership, partially by making route 60 unusable for Georgetown Campus riders in one direction.

         I think people take the 131/132 instead. The stop closest to the school gets about eighty riders (between the two routes) while less twenty use the 60. This brings up an interesting subject though, but I’m going to put that on another thread.

2. I really like this idea! A couple of notes/issues that would need to be addressed and my own alternative for addressing some of them. Full disclosure: I’m a Metro operator (and have driven the 60).

   1. This is very strong route on Graham until Rainier. East of Rainier, Graham essentially becomes a residential street, including roundabouts which would need to be removed to accommodate a 35 or 40ft coach.

   2. Graham East of Rainier is very low density and the area is already served by route 50 with it’s connections to Othello and Columbia City stations. Adding the 60 I think over serves this area and will have very weak ridership.

   3. There’s a couple tricky spots for a bus to maneuver in this proposal. Specifically the aforementioned Graham St. Between Rainier and Wilson Way. The turn from Wilson onto Orcas is very sharp. Also there would need to be a turn around loop and layover somewhere down near Seward Park. Given the character of the neighborhood this might be tricky to find.

   Additionally Graham at Swift Ave is very narrow and curvy. It’s also a very busy intersection given the I-5 access there. There isn’t a good alternative here. But it might require some redesign of the intersection to accommodate turning movements of buses.

   My alternative to address this: send the 60 down Rainer after Graham and have it turn off at Seward Park Ave and left on Henderson with a layover adjacent to the Rainier Beach H.S. sports field. This area is an incredibly strong anchor point for a route and connects with three other routes.
   Rapidride R is going to be routed away from this area and head to Rainier Beach station instead, and there isn’t a good alternative route to serve the area yet.

   1. Another option could be to restructure the 50. For instance, if the bus went Graham to Rainier to Othello, it could then take over route 50 from Rainier/Othello to somewhere east of I-5 (see below), while the 36 is extended a few blocks east to terminate at Rainier instead of Othello.

      I personally think route 50 is a mess, and the slow, detour-filled route between west Seattle and Rainier Valley isn’t even really any faster than just going downtown and taking Link, but offering it involves a lot of redundant service, and treats SODO like the job center it definitely isn’t. So, maybe I have the new 50 go to Beacon Hill station and not cross I-5 at all. Then, the west half of the 50 becomes a separate route, but ends downtown, rather than SODO.

      1. Interesting ideas. A couple of issues jump out:

         1. routing the 36 across the light rail tracks is probably a non starter. The overhead work would require a slow order for not only the bus but also the train. Both use different voltages and there is a low likelihood, but very dangerous risk if a bus dewired here and it’s poles made contact with the Link OCS. We do have these crossings with the streetcar, but the streetcar isn’t the regional backbone of our transit system and can accommodate slow orders through this work.

         I’m a trolley bus operator, and we have instructions to stop our bus in place and wait for a supervisor if we lose our poles under the streetcar OCS due to the voltage risk. I don’t think we would want a similar scenario on the Link tracks.

      2. The bus could just go off wire for that section. We’re talking about half a mile here, on buses that are supposed to have a good 20 miles of off-wire range.

         I don’t think trolley wire is a good reason why the 36 corridor shouldn’t have a connection with the #7. If you’re relying on the bus for everyday transportation, little stuff like this can make a big difference in travel time.

      3. The wire of a bus is not like the pantograph of Link or the streetcar, where the operator can easily raise and lower it. When a bus comes off wire, the operator needs to step outside the bus and poke the wire into place. That’s not acceptable for regular service.

      4. Current trolleybuses have off-wire capability but Metro is annoying reluctant to use it except in rare emergencies. It could use it for more common contingencies, or to bridge small gaps in the wire like on the 12, but it keeps refusing to. So we’ve got this expensive productivity feature we’re not using, and that increases noise and pollution and hill slowness and fossil-fuel use for passengers.

      5. If they’re in it going to use it, why are they paying for it? Just last year, Metro spent a fortune to replace the batteries on all of the trolley buses.

         In the meantime, I’ve said it on other posts and I’ll say it again. Trolley wire is great when it works, but it should not be used to artificially constrain the service network into something less optimal. In particular, the trolley network effectively says that every route that existed in 1930 must follow exactly the same route in 2030 as it did 100 years ago, with limited exceptions, such as two trolley routes being allowed to swap tails, and a few blocks of new wire being allowed, but only if the change is important enough to justify a capital expense of several million dollars per mile (which it usually isn’t), and even then, a trolley route can never be extended to cross Link tracks, so no bus down Beacon Ave. is ever allowed to connect to the 7, and no bus serving upper Queen Anne is ever allowed to cross the ship canal.

         Technology should be a tool for implementing routes, not a means to constrain them in a way that forces people into unnecessary detours and transfers. And I can’t help myself wondering if Metro is going to fall into the same trap with battery buses. For example, if a big, expensive facility is built to allow buses of some route to recharge at the end of the line, what happens if, ten years later, they want to extend the line another few miles to serve some new development? Now, because of the need to charge batteries, the bus can’t be extended (at reasonable cost), so this service must be implemented as a separate route, leading to forced transfers and possibly less efficient routing.

         This is a hidden cost of bus electrification, in that the cheapest solution nearly always involves hyper-optimizing the infrastructure around whatever the bus network is right now, in a way that makes it very difficult or expensive to do service restructures later, with the result, over time, becoming a bus network that is less efficient, or neglects key corridors.

      6. The wire of a bus is not like the pantograph of Link or the streetcar, where the operator can easily raise and lower it.

         Yes it is. The buses now have that capability. You just need to add the little things* on the wire to enable it. Here is a video of a bus doing just that: https://www.youtube.com/watch?v=9k1z10U3utc.

         *I have no idea what those ‘V’ shaped things are but they can’t be that expensive to install.

      7. The little guider things were used for the dual mode breda busses in the dstt.

   2. Chris,
      Thanks for the insight! I’ve always found it a bit odd that the east and south sides of the rainier/henderson/seward park ave area is only served by the ~1/3 of 7 buses going to the prentice loop. Prentice itself is lower ridership, but along Rainier, a huge new apartment building got built a few years ago and another 200+ units are planned or under construction, in addition to the Lake Washington apartments already there. Rainier is fronted by businesses and apartment buildings that entire way. it would be a big win to have regular service there, and also diversify routing at the south end of rainier valley.

      1. Yeah, the tail of the 7 is weird but then the situation is a bit odd. For almost the entire corridor there is very little to the east of Rainier Avenue. Then you have this island of development bounded by Henderson, Rainier (as it curves around) and Seward Avenue. South or southeast of that island there isn’t much again. Thus need a loop to serve the dense area and an extension to cover some areas to the south. What the 7 does is weird but it actually makes sense in that context.

   3. East of Rainier, Graham essentially becomes a residential street … There’s a couple tricky spots for a bus to maneuver in this proposal.

      I was thinking the same thing. As you mentioned, just going on Graham between Rainier and Seward Park Avenue is challenging. The street isn’t well designed for buses. If not for that you might be able to do a live loop like so: https://maps.app.goo.gl/Hk1aUTqtk2tnFHYT6. But the route might be too long for a live loop. It also doesn’t get you that much. Other than the elementary school there is not much there besides low-density housing.

      Another alternative would be to go up to Orcas via Rainier. That avoids the issue with the east part of Graham. Orcas is an arterial the whole way. That would be close to what Micheal wrote. Orcas seems like it is higher density east of Rainier but it still seems like it is mostly (if not exclusively) single family housing (the lots just seem smaller). You still have an issue with turning around (and laying over). The only option I see is to leverage the route of the 50 by making a series of left turns (left turns are actually easier if the angle is bad). So basically this: https://maps.app.goo.gl/eT3WbKVXuNq8KnuS8. If you could layover somewhere around there that would work.

      But I think your alternative would pick up a lot more riders and not cost a lot more. From Graham & Rainier the closest layover is Othello & MLK (where the 36 and 50 layover) but your suggestion is much more straightforward. Connecting Rainier Beach to Georgetown, South Park and Westwood Village sounds really good to me. It is worth noting that you can’t just go straight across (because Boeing is a big obstacle). Thus while the overall route is not especially straight it is about as straight as you can get. Every trip combination is plausible. Being able to take over the current tail of the 7 is a huge bonus.

   4. from the graham street infill station workshop the king county metro discussed an east-west bus route. They said it is likely the bus on graham will have to detour onto orcas east of rainier for the same reasons you mentioned.

      they mentioned having the bus continue east on graham until rainier then head north on rainier for a couple blocks before turning right and continuing east on orcas

   5. It could continue south on Rainier to Safeway and the Rainier Beach library. That will also need service when RapidRide R abandons the area, and is the thing I most worry about with the reroute. This would give south-of-Henderson a one-seat ride to a Link station (Graham).

      A tradeoff for either of these alternatives is it wouldn’t give access to Seward Park from more places. Currently it’s limited to just the 50 by its isolated location and the 50’s infrequency.

      1. Yeah, Chris mentioned this taking over the tail of the 7 once the 7 gets sent to Rainier Beach Station (and becomes the RapidRide R). This would be a good way to backfill service south of Henderson.

         A tradeoff for either of these alternatives is it wouldn’t give access to Seward Park from more places. Currently it’s limited to just the 50 by its isolated location and the 50’s infrequency.

         Yes, but Seward Park is a fairly low-density neighborhood. The ridership of the 50 reflects that. Between 38th and Myrtle (where it lays over, close to Othello Station) to Othello & Rainier it picks up over 200 riders. Then it picks up only 125 riders and drops off 65 before it is back to Rainier again. None of the stops get over 20 riders going either direction. In contrast about 38 riders get off an eastbound bus at Othello & Rainier (this is obviously a short trip). It is easy to make the case that the 50 should run more often but not because of Seward Park.

         I think there are two practical options for sending the 60 to Graham. The first is the proposal by Chris. The second is doglegging to Orcas and then using the existing loop as I described above (https://maps.app.goo.gl/aaErBuhNM8nFMH2n9). You would need to find layover space. It would mean that riders within that loop would only have service inbound or outbound but there aren’t that many people within either part of the loop. Both options (heading to Rainier Beach or looping after serving part of Orcas) would take about the same amount of time from what I can tell.

         I think the Rainier Beach option would get more riders but I think doglegging to Orcas would get you some as well. There is development along Rainier and a set of apartments at Orcas & Wilson. I wouldn’t call this a classic “ridership versus coverage” situation although this would add a tiny bit more coverage (along Orcas between Rainier and Wilson. Mostly it is a case of giving some riders (at Orcas & Wilson especially) a much better option for transit versus giving a lot more riders (in Rainier Beach) a one-seat ride to the west.

      2. “Seward Park is a fairly low-density neighborhood.”

         The issue isn’t the neighborhood, it’s the park. Seward Park is one of Seattle’s biggest park assets. The problem with Seattle’s “string of pearls” park vision is the parks are at the periphery and hard to get to on transit. It needs to be easy to get to the parks on transit, so that residents can get to parks more easily and visitors can access all the tourist attractions.

      3. Well, the waterfront is one of Seward Park’s biggest attractions, so it kind of has to be where it is.

3. “When the station opens, King County Metro will start operating a new route along Graham St to provide better east-west connections with Link.”

   Presumably, if a new Graham St bus route isn’t created, that would be a problem, because area residents won’t have a good east-west connection to the new Link station. So I have to ask, what are residents doing now? How do residents of the Graham St corridor currently get to bus stops on Rainier, MLK, and Beacon Ave?

   1. Graham Station is primarily for the denser area within walking distance of the station. Currently they have to walk 15-20 minutes south to Othello station, or significantly longer north to Columbia City station, or take the 106 and transfer. And the 106 only runs every 15 minutes, so it’s not always a good transfer.

      I’m not as concerned about people on Graham Street getting to Link without a route. Going west it’s just one block to the bottom of Beacon Hill. Going east it’s just four blocks to Rainier. Having a Graham Street route is more about having a more complete street grid: it’s really about people going longer distances to/from/through central Graham Street.

      I’m not sure how to characterize eastern Graham Street. Since Rainier is 4 blocks from the Link station, east of Rainier is 5-12 blocks from Link. After that you get to a steep hill, so it’s doubtful walking is as practical. Chris the Bus Driver says everything east of Rainier is low density, so maybe low priority. I’m not sure how to weigh all those factors with east Graham.

4. Nobody is bothered by splitting the 60’s east-west and north-south halves? I’ve been wondering whether it’s fair to southern riders’ travel patterns and access to retail/services (on Broadway), since the east-west half doesn’t have much of that (Westwood Village isn’t much). While they could transfer to Link or the north-south 60, if the routes are every 15 minutes, that’s a potential 15-minute wait.

   Metro has proposed splitting the 60 sometime in the future, so maybe the southern riders don’t need to go to the northern half as much as I thought. I think Metro’s concept would end at Othello station if I remember, although that was before Graham station was re-added.

   1. Route 60 is the last remaining all-day through route between First Hill and SE Seattle south of McClellan. The only through bus routes left would be Routes 7 and 106, which both run on Rainier only to Jackson. First Hill is where the nearest emergency room is to all of SE Seattle at Harborview.

      It’s a challenging topic, and one that is probably better avoided in a Graham Street discussion.

   2. From the west part of the 60, does the 60 even offer a time advantage to First Hill vs. the alternative of just riding a bus downtown and transferring. Yes, it is a transfer, but both legs run more often than the 60 does, and the downtown to First Hill route is every 6 minutes on the G, plus additional options on routes 2, 3, and 4, 11, and 12.

   3. I feel like if one were to split the 60 it could be two overlapping routes between Beacon Hill and Othello/Graham St.

      I do disagree with the author’s assertion that the 60 should be rerouted onto Broadway. The 60 is where it is to serve Harborview and it is a bit of a hike from Harborview to Broadway. If it were rerouted I would suggest Boren to Madison in First Hill, to also interline with the future Metro Connects SLU – SE Seattle bus via Boren.

      1. The 60 detours to 9th because when the 60 was created, the default was to remain on Broadway, but the First Hill community argued to have it detour to 9th to serve more of the neighborhood. They won, and it has been that way ever since. Since then congestion has gotten worse, so now at least it takes a noticeable amount of time to get from Broadway & Union to 9th & Madison. I imagine that pisses off Beacon Hill and southern riders, but they were never considered.

         Metro has since proposed long-term Broadway north-south restructures without the 9th detour, so it’s ready to get rid of it, at least in the vague future.

      2. I’m just not a fan of the idea of rerouting the 60 without some other alternative north south route through First Hill. That would leave no north south route between 5th and Broadway, which is a substantial distance and elevation gain.

      3. Metro’s long-term plan reroutes the 106 (currently MLK-Rainier-Jackson) to MLK-Rainier-Boren-SLU. Many STB commentators are eager for that. It would address the lack of northwest-southeast service between SLU, Pine/Bellevue, First Hill, and Little Saigon. It’s arguably close enough to 9th and east of Boren, maybe?

         (I live in the Capitol Hill part so I have to walk to First Hill or SLU, or go out of the way to the 8, or go further out of the way to a route to Little Saigon. The described route would be just three blocks away. I also have to walk to Swedish Cherry Hill or take a bus halfway and walk the rest of the way.)

      4. @ Henry:

         I agree with you. I think it’s routing is important for a few reasons:

         1. O’Dea High School
         2. 1101 Madison St MOB (which has lots of relocated specialty clinics including opthomology and colonoscopy centers)
         3. Virginia Mason
         4. Harborview’s front door

         Not only are these busy destinations, but they are also large employers.

         Plus Broadway is a slow slog and traffic often blocks things because turn pockets are not big enough in places.

         If anything, I would suggest looking at jogging the route at Seneca rather than Madison.

      5. The 60 is where it is to serve Harborview and it is a bit of a hike from Harborview to Broadway.

         It doesn’t seem that bad: https://maps.app.goo.gl/Tfg8hBPCngprfnidA. It is worth noting that the 3/4 goes right next to Harborview and you could always transfer. It is just a flawed approach to pick and choose destinations in an area with so many. It would make sense if the only area worth visiting was Harborview but that clearly isn’t the case anymore. Fifty years ago there wasn’t much on that part of Broadway (or east of Harborview in general). Now there is.

         Having a bus zig-zag in that manner hurts in two ways. First it makes that route slower which means that it (or other buses) run less often. Second it destroys the possibility of a spine on Broadway. If you are trying to get anywhere in the area (even Harborview) it makes things worse. Think of trying to get from Harborview to the UW. You can catch the 60 to Capitol Hill Station but the bus only runs every 12 minutes on weekdays (and 15 minutes weekends). You could try the streetcar but it is no better. Ideally you would be able to walk one place where a very frequent bus/streetcar combination (running every 6 minutes) could quickly get you to Capitol Hill Station. That would be true for anywhere on Broadway.

         The biggest weakness in our system is frequency. A bus (or streetcar) running every 12 minutes midday weekday is actually unusually good. Most of our buses run less often than that. It isn’t bad if it is the only bus you take. But it is not good at all if you make a transfer. A bus like the 60 (or the streetcar) is heavily dependent on transfers. It crosses paths with a lot of buses as well as Link. It sucks if you have to wait a long time to make that transfer. It is unrealistic to expect the bus to run every six minutes like the RapidRide G. But without spending any extra money we could combine it with the streetcar to provide the same combined headway.

         Of course I would like a bus on Boren. But that would cost money. We should first focus on providing the same basic level of coverage while dramatically increasing frequency.

      6. The 60 detours to 9th because when the 60 was created, the default was to remain on Broadway, but the First Hill community argued to have it detour to 9th to serve more of the neighborhood.

         That must have happened before 1988 (based on the maps). This makes sense given the lack of development east of there at the time. Google Earth allows you to see old aerial views. Before 2002 it is very grainy but 2002 is a good benchmark. The Boren/Jefferson/Broadway triangle was just a field. North of there were big parking lots on both sides of the street. Even up by Madison and Broadway there were a lot of parking lots and low-rise buildings. Now it is pretty much all skyscrapers and thick, dense buildings.

         But some of it is just a different approach. It is coverage versus ridership. In 1988 the all-day 9 went on Broadway and the 60 made a detour to cover more of the area. Now it is the 60/streetcar. Either way the area is covered. But it means that you have less frequency along the core. This in turn means less ridership. If the 60 and streetcar both ran every 6 minutes all-day long (like the RapidRide G) it wouldn’t matter. But they don’t. The only way to get good frequency like that is for them to share the same corridor.

      7. I’ve been riding Metro since 1979. I think the 60 was already running then from Broadway to Georgetown. At least, I don’t remember it not being there like I do the 8. The 60 has always had the 9th Avenue detour in my experience.

      8. The 60 has always had the 9th Avenue detour in my experience.

         That is quite likely. I haven’t been able to find a map between 1970 and 1988. That is a big gap. In 1970 the 60 didn’t exist. Nor did something similar exist with a different name. There was no bus from Beacon Hill to First Hill. So somewhere in that 28 years they added the bus and it was probably the 60.

         It is worth noting that the precursor to the 3/4 was called the 12 and it looks like it went on 9th (as it does now). But instead of only going a couple blocks (between Jefferson and James) it went all the way to Madison, Spring or Seneca (I can’t tell which). Thus it is quite possible that the 60 took over some of the functionality of the future 3/4.

      9. 1977 Seattle transit map. If you click back and forward, you’ll see some other 1977 area transit maps, and a 1950 Seattle transit map. The route 60 existed in 1977, but it just went from Broadway to Yesler.

         https://www.flickr.com/photos/tigerzombie/11960463075/in/photostream/

      10. When I started riding Metro, the 2, 3, and 4 and most other East Seattle routes were as they still were in the early 2010s. There were minor variations and splits but the routes generally went to the same places. Changes I saw over the years were:

          * 4S short extension to Rainier.

          * 7 split to 7/49.

          * 8 created. Previously no Denny Way service. Originally only to 15th weekdays. Frequency/span repeatedly increased. Extended to Mt Baker, then to Henderson Street (MLK), then truncated to Mt Baker again.

          * 9 created as U-District – Broadway – Rose Street (northern Rainier Beach) express. Originally all-day but infrequent. At various times reduced to peak only or suspended.

          * 13 split to 13/12. 10/12 through-routed for several years via 1st Ave.

          * 14 split to 14/47, creating a 1/14 through-route.

          * 43 split to 43/44.

          * 48 terminated at Columbia City. Extended to Henderson Street (MLK), turning west to Rainier. Truncated to Mt Baker station when it was created. Lower MLK then served by 8, later 38, now 106.

          * 60 extended from Georgetown in one or more stages to Westwood Village.

      11. East Seattle routes on the 1977 map I didn’t see, so they must have been gone by 1979:

          12 on Jefferson, James, 23rd. No, the 3 and 4 were there.

          4 on Madison and in Montlake. No. The 43 replaced it, moving to John.

          38 on north Empire Way (now MLK). I don’t I think there was any route there until the 8 extension.

          76 no. I have no idea what that route might have been.

      12. Thanks Sam. That map from 1977 helps fill the gap. So we have links to a streetcar maps from 1941 and bus maps from 1970, 1977, 1988 and 2015.

5. Thanks to Michael Smith for taking on this Friday column and coming up with a concept. We needed a Friday column to bridge the long gap between Wednesday’s and Sunday’s open threads, especially with those reaching 100 comments in two days (long comment threads are hard to navigate on mobile, and more people are reading on mobile now), and with some weeks having no other articles (which I hope will change).

   This column is meant to be a shorter and more lighthearted take than our usual articles, and to be more interactive (often posing a question rather than giving a recommendation), and we’re going to explore a few topic/format concepts here.

6. I see a proposed Graham only route doomed to poor performance.

   The core problem with a “Graham only” route is that the entire street is almost entirely residential in SE Seattle. The most active commercial destination is MacDonalds in front of the station site. It’s a problem with Route 50 in SE Seattle too — and ridership on Route 50 east of SODO remains pretty low partly for this reason.

   There are two main reasons that people in my part of town take buses:

   1. Making trips within the neighborhood. Directness is most important. No one likes to transfer once or maybe twice to reach a grocery or pharmacy just 1.5 miles away.

   2. Connecting to Link. Riders want to get to Link to access all the many great destinations in the region — from SeaTac to Downtown to UW to wherever.

   There’s a third consideration, which is wanting direct access to important places outside of the neighborhood but not on Link. First Hill medical offices and Southcenter are examples. (One big reason why I think Beacon Hill route splitting is a generally bad idea as evidenced by the added directness difficulty created by Mt Baker route splitting done a decade ago.)

   I see any service on Graham as needing to also serve larger commercial areas or it won’t get many riders. Then Metro and SDOT don’t want to pay for the service level and frequency gets cut. When Metro created Route 50 through the neighborhood over a decade ago, it was promised an early on as 15 minute service. The demand wasn’t there to justify that. It’s been 30 minutes most of the day until recently — and will return to 30 minutes in the next budget crisis. Frozen goods thaw waiting for a bus with 30-minute service, particularly when it also includes a bus transfer on top.

   I guess one could argue that Georgetown has commercial activity, but it’s pretty much a place with niche, trendy businesses rather than more essential and busy retail activity.

   The key principle I recommend is to design routes that will be heavily used rather than to design routes merely to provide coverage. That means that any route going to have to turn north or south and go for a bit to serve local destinations and attract riders. The more supermarkets and pharmacies the better. Hitting Seward Park helps — but not much (noting that it adds very little to current Route 50 ridership most days).

   I don’t think a through route on Graham is essential either. Not many people travel from one residence to another residence. I could see Graham west of MLK served by one route leg and east of MLK by another route leg. The nice thing about routes with jogs is that bus-bus transferring is easier if both bus routes share a stop. So I think new routes will be more productive traveling on MLK or Rainier a bit.

   What I think would be really telling would be a survey to all the addresses within walking distance of Graham Street. Ask them where, when and how often they need to go to destinations. That’s of course something that Metro should be doing.

   It’s always beautiful to see a city transit map cross-crossed by bus lines like a Tartan plaid. However, the reason that it’s not what Metro and other urban systems generally have is because pretty does not mean productive. Seattle is a city with almost exclusively north-south commercial corridors (Pike-Pine and Market Street in Ballard as exceptions for example). It makes productive east-west routes a challenge without having segments of routes that also run north-south.

   1. “ridership on Route 50 east of SODO remains pretty low partly for this reason”

      Ridership is suppressed by the 50’s infrequency. People don’t want to wait 30 minutes for a transfer or if they miss a bus, or work around its 30-minute pulses. I’ve sometimes wanted to go to Seward Park but didn’t because it was in one of the 50’s 30-minute periods. Neighborhood residents may be taking Metro Flex and demanding it more than they would otherwise because of this.

   2. “Seattle is a city with almost exclusively north-south commercial corridors”

      Especially in the southern half of the city, where hills and cliffs and the Duwamish Waterway and I-5 and the railroad tracks hinder east-west trips and even roads. This creates narrow north-south islands that are more oriented to downtown than to each other. It has kept the south end underdeveloped in retail and jobs because each island’s market shed is so small. That’s the ultimate long-term reason Graham Street doesn’t have a route yet, and crosstown routes that do exist like the 50 have relatively low ridership.

   3. I am a rider along the Graham St. corridor. I use the 36 and do most of my errands downtown because there are no convenient E/W routes. I would use a Graham St. route to go to Georgetown and Hillman City, commercial neighborhoods within a couple miles from my house that would take about an hour to get to on current transit routes. I would take the Graham St. route to Route 7 for access along all of Rainier’s commercial development – Columbia City, the supermarkets, etc., which now require either going all the way to the CID to connect with the 7 or multiple transfers, including a long wait for the 50. Graham is one of the few South Seattle streets with a full E/W connection. Given the current huge ridership on the 36 and 7, and in South Seattle neighborhoods in general, I have no doubt that a Graham St. route would succeed (assuming reasonably frequent service), and less niche-y commercial development would follow in the neighborhoods it serves.

7. I am more optimistic on the transit demand west of MLK given the hill there make walking less desirable.
   As for east of MLK, it is approx. 2400 ft between MLK and Rainier which is generally flat.
   So I wonder how much attraction does Graham St Station has to the area east of MLK. Will those living in the middle of MLK and Rainier corriors consider taking the new Graham route and then transfer Link rather than 1) walking to the new infill station 2) keep taking 7.

   In 2031, will it be faster to
   1) take new route to Graham @ MLK and transfer 1 Line
   2) take 7 all the way to Downtown or transfer at Mt Baker.

   I am guessing the new Graham St route’s frequency has to be real good so that people will give up walking. Of course the another reason to have a new station is that it expands the area where TOD can be developed, so existing demand probably won’t matter that much.

   1. It depends how much the 7’s speed/reliability has improved with the recent transit-lane segments, and how much it will improve with RapidRide R. If the 7 is as fast and robust as Link, people will take it to Mt Baker or Judkins Park stations and transfer to Link. If it’s not, they’ll walk to Graham station.

      Transferring at any of these stations has the downside that none of them are both 1 Line/2 Line stations, so that makes a transfer wait of 8-10 minutes, whereas if you take the 7 to Intl Dist station it’s 4-5 minutes (potentially an average wait time of 2-3 minutes, which gets into New York frequency and is something Pugetopolis has never had).

      1. I am also looking forward to seeing how good R Line will do. I especially don’t like this kind grid transfer because it always involves crossing street. It just make no sense to break turning point of a route in exchange for a little more frequency. If there is a huge demand between Seattle and Bellevue, nobody would break the U-shape ST 550X into three-seat-ride.

         I think the the whole Jarrett Walker grid philosophy highly depends on good grid and awesome frequency, but I don’t like that his firm is also selling this concept elsewhere and planning 30-45-min headway grid somewhere in this country.

         In a perfect world where you actually have enough data to do transit planning, service should be planned based on demand pattern rather than network grid. His approach does best at two thing which are 1) using least amount of money to plan service to cover the enough ground that needs to be covered and 2) make transit agency look good as they have more frequent service lines on the nice-looking system map they delivered.

         But I guess I would still pay the man some respect as he is dedicating his career to promote transit in his own term.

      2. “I think the the whole Jarrett Walker grid philosophy highly depends on good grid and awesome frequency, but I don’t like that his firm is also selling this concept elsewhere and planning 30-45-min headway grid somewhere in this country.”

         Grids partly depend on not having geographical barriers. Jarrett wrote an article saying Seattle has a lot of chokepoints, which increase ridership demand across them but make a street grid difficult: “No North American city has more chokepoints than Seattle. The city itself consists of three peninsulas with narrow water barriers between them. Further barriers are created by steep hills in most parts of the city. Nowhere in Seattle can you travel in a straight line for more than a few miles without going into the water or over a cliff. Seattle’s geographical isolation from its suburbs, of course, means it is also surrounded by chokepoints. There are only two bridges across Lake Washington to the east….”

         I was doubtful of the penninsulas analysis because I’m used to thinking of penninsulas from the other end (the tip is the least dense), but it makes sense.

         The restructure proposals in Jarrett’s blog have a core frequent network and peripheral coverage service. They always claim to offer more destination access to more people compared to the existing network, and to give denser areas more frequency than they have. I haven’t seen an exception. Maybe his company has poorer proposals he doesn’t mention on his website, but I’ve seen no evidence on that. Many STBers wish Metro and ST would hire Jarrett to redesign the Metro and Link networks, because it would probably be a better proposal than anything we’ve gotten from the agencies. Especially since Metro got much better in the mid 2010s but then backslid after 2020.

      3. Mike,
         Thanks for sharing this article. This is a great read to me about uniqueness of Seattle’s street network.

         I guess my concern is his transit planning ideology all seems to dependent of his graph theory rather than where demands are. The statement like grid is key is based on the geometric shape of city network without weighing in demand.

         When we looked at his proposal, we tend to judge based on how it looks on the map.

         Demand is always the challenging thing to study because it is hard to get a full picture using the data available and it always changes.
         I am sure his team looks into demand pattern for their projects, but I am not sure it is possible for an out-of-town consulting team to understand a place’s travel demand pattern within a short period of time to delivery a bus network redesign.

         That’s why I think it always makes more sense to have a well-defined objective for a smaller problem and make small network improvement to solve those specific problem.

         This is a conspiracy theory from me, but I think his bus network redesign work are popular because some transit agencies have budget shortfall and want to reduce its systemwide service hour without being caught.
         It would look bad to the public if you just cancel trips from specific routes, but if you do a redesign to “improve” the efficiency, it is more difficult to compare.

      4. A grid with 30 minute service is still better than a hodge podge of meandering routes running once an hour. Of course, more frequency is always better, but Jarrett Walker doesn’t get to control tax rates or operations costs, all he can do is make recommendations for how an agency should structure its service with the constraints it has.

         And, I would also be very weary about trying to design a bus route by enumerating the specific destinations that people want to, as that only really works when you have huge crowds of people all wanting to the same place. Outside of major hubs like downtown, demand is more scattered, and twists and turns to get closer to places where Metro imagines people are trying to do just makes the route harder to use to go anywhere else. Me personally, I feel like I sit through far more detours to places where I don’t want to go than have a detour going to where I do want to go, and even then, I could have easily just walked there from a bus stop on the main road that would have made the route more direct for everyone else.

      5. I agree that Seattle has choke points. Not only is the city three large peninsulas, but the hills and freeways create more islands than that.

         SE Seattle is pretty much an island south of I-90, east of I-5 and just south of Rainier Beach (along with Lake Washington). But even with that, there are elevation differences and limited street connections between 15th and Beacon Ave, and between Beacon Ave and MLK. There are other short-distance hills that block east-west movements between MLK and Rainier and between Rainier and Lake Washington. So it’s really two big islands — Beacon Hill and Rainier/MLK with hills that add further limits to through access.

         I remember staring at city maps decades ago when I have never been west of Missouri. I didn’t fully understand how terrain can affect things so dramatically. I used to wonder why streets didn’t just go through. After arriving on the west coast, it’s plain as day why street grids and bus networks do what they do because of elevation.

      6. A grid is one factor. It’s not the only factor. If all other things are equal, it’s better to grid than not to grid. North Seattle is a lot better off because of the east-west bus routes, and ridership proves it. But south Seattle is geographically challenged.

         Seattle has another problem in the distribution of retail/commercial areas. San Francisco and Vancouver BC have streets with commercial/mixed-use districts all along it or in a straight line, so a simple grid route can serve them all. Seattle has random small islands of retail in a sea of residential-only areas. The demand is to and between the retail areas, so routes have to bend to get to them.

         The 48 goes straight north from Mt Baker to the Montlake Bridge, but the retail is on Broadway and 15th and the 48 misses them. So the 48 is really only useful for the residents in between, whereas if it had retail all along it it would draw people from other areas too and ridership would be higher and car mode share lower. Ridership would be more even too, because on a residential street residents leave in the morning and come back in the afternoon/evening. On a mixed street residents leave in the morning and workers and clients come, then in the afternoon/evening the clients and workers leave and the residents return.

         Another grid route on paper is 15th Ave NE from UW to Mountlake Terrace. It misses Roosevelt, Greenlake, Northgate, and the Crest Cinema area. Those are all so-near-yet-too-far for people living on 15th.

         You’d really have to show one example where Jarrett recommended a grid route to the detriment of most people’s destination needs and travel patterns. If he suggests a questionable grid route, he’s probably be clear what the tradeoffs are: “This would serve these riders but not those riders.” He does that when he removes coverage from a street: “This would serve more people but it wouldn’t serve these people.” Removing coverage is always to increase frequency in a high-volume or high-potential corridor. It’s probably a grid route but not necessarily. He also talks about radial and orbital patterns, and having a mixture of more than one. Some of his networks do that.

         Restructures are usually revenue-neutral, a way to apply new hours more smartly, or a way to cope with recession reductions. Agencies would probably only hire Jarrett for the first two cases. I doubt agencies are sneakily reducing hours using a restructure to hide it.

      7. In San Francisco streets do just go straight through regardless of steep hills. That was the convention then, to impose the grid everywhere and ignore nature. Seattle was built up a few decades later when the convention had changed. Seattle also flattened some hilltops to avoid the issue.

      8. “ The restructure proposals in Jarrett’s blog have a core frequent network and peripheral coverage service. They always claim to offer more destination access to more people compared to the existing network, and to give denser areas more frequency than they have. I haven’t seen an exception. Maybe his company has poorer proposals he doesn’t mention on his website, but I’ve seen no evidence on that. Many STBers wish Metro and ST would hire Jarrett to redesign the Metro and Link networks, because it would probably be a better proposal than anything we’ve gotten from the agencies.”

         Restructures are fairly massive undertakings. It’s hard for many internal staff to have the time to take on the volume of research and public input required, plus it’s a way for internal staff to not have to face political fallout for making bus route changes. No internal staff wants to face a group of angry residents from a community where transit is a lifeline because they propose severing the route between their apartments and their essential services, for example. Consultants can be the “bad guys” and leave to never be seen again for a generation.

         Riders often don’t like bus route changes. They develop travel habits and even make housing, shopping and employment choices because of transit routing and schedules. Something as simple as moving a single bus stop can consume lots of time and energy. Most routes have multiple markets too — from choice commuters to shoppers without cars to special needs populations like blind or wheelchair-bound or young-aged citizens. There are sometimes routes that serve certain ethnic communities especially in non-English situations. But sometimes it just one little old lady who wants to protect their pet bus route and has the time and energy to fight a change even though it benefits no one else.

         Metro hasn’t seen fit to get a large volume of outside help in route restructuring. I’m not sure if that’s a good or bad thing. It’s one reason why I think Metro should tweak a restructuring after it’s been operating after 2-3 years. Sometimes tweaking is needed right away if the new route is hard to drive. It’s why I think Metro has enough on its plate for the next 13 months with East Link and FW Link restructures. The big route layouts are done — but getting every stop placed safely and getting every run time right remain. Plus ridership can shift once a new service is in place — and Metro staff may have to quickly adjust schedules for late buses, early buses, empty buses or overcrowded buses once they start running the new routes at new schedules.

      9. I guess my concern is his transit planning ideology all seems to dependent of his graph theory rather than where demands are. The statement like grid is key is based on the geometric shape of city network without weighing in demand.

         That is simply not true. Just to back up here, the grid (or graph) idea is not unique to Walker. It is just math. He just explained that (and other concepts) really well to an ignorant public. That is what his books are about. He is just explaining ideas, not coming up with new ones. A lot of the ideas (like the importance of frequency) are well known in the industry but ignored by the public. Many of the ideas are well researched. He has about 20 pages of footnotes in his book and many reference well-respected studies.

         The consultants don’t get paid to just build a grid. They get paid to provide a more cost effective network. The first thing they do is ask the agency what percentage of money should go into coverage and what percentage should go into ridership. Then they come up with ideas based on that. There are trade-offs at every turn. A grid is the most efficient way to have everywhere-to-everywhere coverage (again, that is just math). But some trips are more popular than others. There is a transfer penalty. It is real and it is well documented. But there is also a waiting penalty. There is a lack of speed penalty. There is a distance penalty. These different penalties work against each other just as coverage and ridership are often contradictory concepts. The consultants balance these when building a network. It is quite likely that the network they come up with will be more of a grid because that is one way to improve frequency (which in turn increases ridership). But that doesn’t mean that it would be a “near perfect grid” like Vancouver. We don’t have the kind of city or subway system that can support that. This is a shame because their bus system has the third highest ridership in North America (trailing only New York and Chicago).

      10. Jarrett defines his job as offering advice to agencies. The agencies decide their values (how to weigh the ridership/coverage tradeoff), and he makes suggestions within those constraints. So it’s all the agency’s responsibility to set the conditions and to follow or not follow his advice.

          Metro made BOLD RESTRUCTURE PROPOSALS between 2012 and 2019. It consolidated frequent corridors, reduced nearby redundant routes, split routes causing new transfers, and deleted the lowest-volume coverage routes. The C, D, E, and U-Link restructures had extensive improvements. Metro even did it during the 2014 cuts.

          It’s biggest failure then was being afraid to split the 2 or restructure the 12 amidst opposition. That caused cascading rollbacks of the 3, 4, 5, 11, 26, and 28 proposals. But those are small failures compared to the many successes.

          It also created Metro Connects in 2016 to offer a vision for the ST2 era and the not-yet-passed ST3 era, and to undergird an expansion levy that never happened. Metro Connects has many more bold restructures. Some are better than others, but at least it’s trying to think beyond inertia.

          However, in 2020 it apparently lost some good planners and the next ones were inexperienced and more timid. So it backslid on some of the changes it had proposed repeatedly before or that are in Metro Connects. The G restructure didn’t move the 2 to Pine, didn’t split the 2, didn’t create a Denny-Madison route to replace the 8 and 11, didn’t create a Broadway north-south route to replace the 49. didn’t extend the 106 on Boren to SLU — all of these are in Metro Connects — and didn’t consider eddiew’s suggestion to restore the all-day 43 and replace the 48 with a “reverse 43” (John, south 23rd), etc.

          It did move the 12 to Pine and move the 10 back to Pine, with both of them overlapping to 15th. That was a good innovation. But maybe not as good as the others above?

          The Northgate Link, Lynnwood Link, and East Link restructures were likewise not as good as the U-Link restructure, although they each had some positive innovations.

          So we’re left with a Metro that’s better in restructuring than before 2012, but not as good as between 2012 and 2019.

          The argument is that Metro should ask Jarrett for a second opinion on a Metro Connect network, ST should ask Jarrett for a second opinion on a Link network (and it should have done it before ST3 was finalized),and the governments collectively should ask Jarrett for a comprehensive local+regional network. That’s what’s been missing all along: a vision of a real passenger-first local+regional network like European cities or Vancouver have. That should be the starting point for creating Link and Metro visions that fulfill it. And fund it adequately so that it can go everywhere it needs to and run every 5-15 minutes full time. The politicians should prioritize this and outline steps to get there, and show they’re pursuing the steps as much as they can. And don’t get distracted with non-transit wishlist items like a light rail spine to Everett and Tacoma (wrong technology for such a long distance), redeveloping an obsolete Administration building for the county, or cowing to NIMBYs and large property owners.

      11. “It’s one reason why I think Metro should tweak a restructuring after it’s been operating after 2-3 years. ”

          I asked the Metro planners that after the G restructure. I said it should be reevaluated after a couple years to see how people voted with their feet and whether the waits between buses on the 10, 11, and 12 were excessively long. They said they will monitor it and make changes they think necessary. But what they think are necessary and what we think are necessary will likely be different. They said they’d add runs on the 10 and 12 if crowding became an issue. I said it wasn’t crowding I was concerned about, but passenger wait time. So they may make some changes, but likely less or different than what we’d want. But at least they’re monitoring the corridors.

      12. Many STBers wish Metro and ST would hire Jarrett to redesign the Metro and Link networks, because it would probably be a better proposal than anything we’ve gotten from the agencies.

          I am one of those people, for the reasons you mentioned in the other comment. I don’t think the planners are capable of making the changes that need to be made to provide a decent network. It isn’t just “making a grid” it is consolidating service. For a very simple example consider the 10 and 12. They branch at 15th & Pike. Every bus north of there has 20 minute service. That is the kind of frequency you expect for West Magnolia or South Mercer Island. Not the urban part of Capitol Hill. Why not just branch at 15th & Thomas? At least that would give you some bus stops with better frequency. You would lose a tiny bit of coverage very close to other buses. That isn’t really a “grid” issue. I’m not talking about sending the 10 south on 15th/14th to Beacon Hill. It is a consolidation issue. It is buses too close to other buses running infrequently.

          The same thing goes for buses that zig-zag instead of just going straight. Sometimes that is worth it but quite often it is just legacy routing that ignores ridership issues while focusing on coverage. Lack of consolidation combined with zig-zags leads to poor frequency even in areas where we are spending a bunch of money (like First and Capitol Hill).

          Restructures are fairly massive undertakings. It’s hard for many internal staff to have the time to take on the volume of research and public input required, plus it’s a way for internal staff to not have to face political fallout for making bus route changes. … Riders often don’t like bus route changes.

          Riders also need to push for better routing. I’m reading this book called “Better Buses Better Cities” by Steven Higashide. It focuses on the political aspect of getting a better network. I very much recommend it as it will sound woefully familiar. He mentions that agencies tend to neglect regular bus service while focusing on big projects. Or that agencies don’t cooperate very well. He mentions that change doesn’t always happen naturally — sometimes it takes a new organization to shake things up. In Miami a group called Transit Alliance did just that. They launched a campaign called “Where’s My Bus?” that highlighted problem’s with the systems reliability, route design, frequency and the fragmented nature of the various agencies. My guess is Miami was much worse than Seattle at the time which probably made it easier to make their case. I don’t think our system is terrible — I just think it is nowhere near as good as it should be and we seem to be getting worse.

          The best time to make big changes is when you have more money. Every change is bound to upset someone. As Higashide puts it, network design is “win-lose” although it can be “win-win-win-lose” meaning a lot more people come out ahead. In the case I mentioned a handful of people have to walk farther (a couple blocks) but hundreds of people go from waiting 20 minutes to waiting 10 minutes. Way more people are better off and the people who are worse off really aren’t in bad shape. But it is much easier to make changes of this nature when you have an influx of money. It softens the blow to those coming out behind while those coming out ahead come out way ahead.

          When Link got to the UW that is basically what happened. The combination of extra money for transit (via the Seattle Transit Benefit District or STBD) and the huge savings from truncating the buses that ran from the UW to downtown were put into big improvements in northeast Seattle. The transfer to get downtown was often really annoying. But the frequency improvements were huge. Snohomish County had a similar influx of extra service which they put into improving their network when Lynnwood Link opened.

          But Northgate Link came in the midst of a decline in funding as the new STBD had a lot less money. We are in the same boat with Lynnwood Link. But that is still no excuse for many of the ideas they came up with. For example the original plans had buses running on 80th and 85th — a clear violation of Metro policy and just common sense. They took ideas (run more east-buses) and took them to a bizarre extreme (e. g. running a frequent bus on 175th even though the Link station is on 185th). I don’t want to imply that restructures are easy but even the major changes just aren’t very good. Consider the corridor that connects Lake City and Bitter Lake. About half way in between will be Pinehurst Station (part of the Link network). The 75 currently runs on the corridor. As of now the plan is to keep it the same. It heads west from Lake City towards Bitter Lake. Just as it is about a half mile from the station it will suddenly turn south and head towards Northgate. Keep in mind, this is not the only bus — nor the fastest bus — from Lake City to Northgate. The bus just seems scared of the Pinehurst Station for some reason.

          So Metro will spend extra money running a bus from Lake City to Bitter Lake. The bus will be the 77. Look at this route for a second: https://kingcounty.gov/~/media/depts/metro/programs-projects/link-connections/lynnwood-link/pdf/routes/route77.pdf. It really doesn’t make sense from a macro perspective. The major destinations are Bitter Lake, Pinehurst Station, Lake City, Roosevelt, U-District and the UW. If you started at Bitter Lake you would ignore most of the route. If you are headed to Roosevelt or the U-District you would just get off the bus at Pinehurst Station and ride the train. Even if you are heading to somewhere in between the stations you would have plenty of time to take the train and catch some other bus heading that direction. It really doesn’t offer much to Bitter Lake riders other than a connection to Pinehurst Station and Lake City. In contrast if the 75 was extended to Bitter Lake it would connect those riders to Sand Point, Children’s Hospital, U-Village and much of the UW campus (that is difficult to access via Link). Oh, and the bus makes a hairpin turn. By its very nature this isn’t good. Even if everything works out well it means the bus is essentially doubling back on itself. But now look at the diagram again. Look at that little notch in the corner in Lake City Way. That is caused by the fact that it isn’t that easy to make that turn while still providing decent coverage in one of the most densely populated parts of the route. Thus it will make a bunch of extra turns one direction while missing out on key bus stops the other direction. From a macro and micro perspective it is just a really bad pairing. Why did they do this? Because they forgot that Sound Transit was abandoning Lake City Way so they just combined these disparate routes while ignoring the overlapping 75. WTF?!

          This is why I have lost faith in the Metro planners. I’m sure they still have some smart people. But as an organization they are just doing a poor job.

      13. The 77’s jog is getting fixed: https://www.seattle.gov/transportation/projects-and-programs/programs/transit-program/30th-ave-ne-paving-and-bus-improvements

      14. “a way for internal staff to not have to face political fallout for making bus route changes.”

          That’s where the elected leaders need to set the example and take the political heat and make it safe for the staff to be bold. The general manager and other managers need to support the staff on it too.

      15. John D,
          The SDOT 30th Avenue NE project does not fix the Route 77 jog, it is the minimum civil project that allows the jog. It remains a poor network design choice that imposes more walking transfers than necessary. The network should be changed.

      16. Thanks for the information John.

          What a fiasco. The city is spending money to support a route that is fundamentally bad. To be fair, most of the work will benefit the neighborhood anyway (e. g. sidewalk improvements). But the routing is still terrible. Yes, this avoids the ridiculous set of turns that would otherwise be necessary. But it means that most of Lake City is nowhere near a bus stop serving the closest Link Station (Pinehurst). The 77 really doesn’t serve Lake City — it just skirts it.

          Meanwhile, if you just want a one-seat ride to Roosevelt or the U-District you will be much worse off than you are now. The 522 stops along Lake City Way at 130th and 125th. This will only serve 125th. Anyone north of the existing bus stop on 125th has an extra five minute walk. It is over ten minutes to Fred Meyer and over ten minutes to some of the apartments south of Fred Meyer. In general they are worse off than they are now despite Sound Transit adding a train station specifically for them!

          Oh, and it isn’t clear if a northbound bus will be able to serve the stop at 120th. It will almost immediately have to move three lanes to the left to make that turn. So it is quite possible that folks at the south end of Lake City are out of luck as well.

          The worst part is that it is so easy to fix this. If they just ran the 77 north to Lake City it would serve all the stops on Lake City up to 130th southbound and 135th northbound (like the 61). It would give southbound riders the option of taking either the 61 or 77 between 130th and 115th. Meanwhile, the 75 could easily be modified to serve Bitter Lake. That would widen the coverage in Lake City which means a lot more people would be able to catch the quickest bus to Link. From that apartment just south of Fred Meyer you would have a five minute walk to the 75 along with the shorter walk to a bus headed to Roosevelt and the U-District. It would also be cheaper! You avoid the overlap on 125th with the 75 and 77. You could easily backfill service on 5th NE with a live loop bus from Northgate. Doing that would be much better than the proposal.

          This is why I have lost faith in the planners at Metro. The routing they have proposed is just worse for riders while also being more expensive.

      17. I’m reading some comments here that are along the lines of …”I don’t understand why Metro is doing routing A, but not doing routing B.” But, what’s missing is someone playing devil’s advocate. Explain to the reader what Metro might be possibly be thinking by wanting to go with one routing over another. Could it be a safety issue, ADA issue, money issue, equity issue, duplication of service issue? Something else?

      18. I’m reading some comments here that are along the lines of …”I don’t understand why Metro is doing routing A, but not doing routing B.” But, what’s missing is someone playing devil’s advocate.

          Go ahead. Advocate for the devil.

          Seriously though I often mention the drawbacks. When I write about the 10/12 branching later I specially mentioned the loss of coverage. Usually the drawbacks are fairly obvious. For example the 75 would go to Bitter Lake instead of Northgate. This means some riders along the route would not have a one-seat ride to Northgate. They would have to take the 61 or 348. But the number of people actually taking the 75 to get to the various shops in Northgate are pretty small — most just want to Link. Thus the argument is a weak one.

          Could it be a safety issue, ADA issue, money issue, equity issue, duplication of service issue? Something else?

          I always focus on service (money). Quite often my proposals are not revenue neutral — they actually save money. I also consider realistic layovers and turnarounds. It is rare that I suggest a new layover (because it can be tricky). I rarely propose running a bus on a street that doesn’t have a bus already. When I do I specifically mention it. Thus most proposals don’t have ADA issues or they are fairly minor. In the case of the 77 my proposal would require no work from the city while Metro’s proposal will require them to add new sidewalks and bus stops. So yeah, my proposal is cheaper both in service and capital spending.

          It is quite possible it is just a mistake. Planners make mistakes. The first proposal for the restructure after Lynnwood Link had no service on Lake City Way between Ravenna and Roosevelt. This was a clear mistake. My guess is they just forgot that Sound Transit was sending the 522 to 148th Station. They corrected it with the next proposal.

          But it wasn’t the only mistake. As I wrote they also had a bus on 80th (along with the bus on 85th). This is a clear mistake. So basically they made two very obvious mistakes. Given the obvious mistakes it is not surprising they would make a subtle mistake. If you just glance at the route it seems OK. It is only when you dig into the details and consider what this means in terms of coverage, trip combinations and overlap that you realize the routing is just a big mistake.

          We can theorize why they made the mistake. Maybe they wanted to avoid altering the 75. Maybe they just didn’t realize how poor the routing was. Does it really matter? The point is it is bad routing.

      19. Metro’s reasoning appears to be to have a 7-shaped route to avoid the cost of two routes. The Lake City-Roosevelt corridor is a late addition to Metro’s radar. It’s also constrained by the driver shortage.

          Early versions of Metro Connects or the north Seattle restructures had route 75 rerouted to Pinehurst station. (That abandoned the 5th Ave NE apartments.) I always wondered whether the 5th-125th corridor ware more or less productive than the Northgate Way-LCW corridor if we had to choose only one. 5th has apartments on the east side, but the west side is I-5 so no walkshed.

          Later Metro reverted the 75 to 5th and Northgate station even after Pinehurst station opened. We don’t know why. It could have gotten pushback from the apartments along 5th. But because it did that, another route had to connect Lake City to Pinehurst Station, namely a new route 77.

          Metro didn’t value the Lake City-Roosevelt corridor at all. Historically the 72 went north on 15th to 80th, and turned east to Ravenna Ave and Lake City. the 307 went express to Northgate, north via (5th?), east on 125th, to Lake City Way, Bothell, and Woodinville. When the 522 replaced it, it remained on Lake City Way but had no stops south of 120th. Lake City was not in the 522’s mandate but it served it anyway because the 307 had done so and it was on the way. When the 72 was deleted in 2016, Seattlites begged ST to add a 522 stop at 85th. ST grudgingly did, complaining that North Seattle wasn’t in the 522’s mandate and wasn’t paying for the route. In 2022, the 522 was truncated at the new Roosevelt station, thus providing a long-missing connection between the Roosevelt/northern U-District area and Lake City. But North Seattle still isn’t in the 522’s/Stride 3’s mandate, so at some point they’ll be truncated at Shoreline South station. (It has been delayed to avoid overcrowding the 1 Line until the 2 Line overlap starts. Transit fans say the oevrcrowding wouldn’t be in North Seattle but between Capitol Hill and Westlake, so it would be the same regardless of whether the 522/S3 terminates at Shoreline South or Roosevelt.)

          The future loss of the 522’s Roosevelt segment caused requests to pour into Metro asking for replacement Lake City-Roosevelt service. Metro took a long time to do anything about it, but finally did by extending the proposed 77 to Roosevelt station in a 7-shape.

      20. Early versions of Metro Connects or the north Seattle restructures had route 75 rerouted to Pinehurst station.

          Metro Connects had the 75 going to the Pinehurst Station but that was not an official proposal. The map was only drawn up to give people an idea of what transit might look like in the future.

          The first actual proposal to serve the station had the 65 going across. That seemed like a good idea. Most people seemed to like it but there was opposition from the north end of Lake City. They wanted their one-seat ride to Nathan Hale and Jane Addams. At the same time, the planners realized they forgot to serve Lake City Way. These changes were going to happen at roughly the same time. Thus it is quite likely the planners got lazy and just decided to combine the routes. There was never an official proposal to send the 75 to Pinehurst.

          Metro’s reasoning appears to be to have a 7-shaped route to avoid the cost of two routes. It’s also constrained by the driver shortage.

          Except this increases cost and makes the drivers shortage worse. If the 77 is sent to the Lake City Fred Meyer and the 75 is sent to Pinehurst you actually save money, even after backfilling service on Fifth. Instead of doubling up service on 125th you are doubling up service on Lake City Way. It is a shorter distance and there are fewer bus stops.

          Every restructure has drawbacks but consider the existing ridership. A 75 heading towards the UW picks up over 550 riders a day at the Northgate Transit Center. Obviously these riders are transferring from the station*. It then picks up 250 riders on Fifth Avenue. It also drops off 185 riders on Fifth as well. Obviously a lot of people were simply taking the bus along Fifth (nothing would change for them). More to the point the number of riders that boarded at the Link Station vastly outnumber those that boarded on Fifth. This sort of combination: https://maps.app.goo.gl/AwybhvpEgyGGC8Kk8 is just not that popular.

          The only good reason for the proposed routing is that it leaves the 75 alone. Spending extra money and inconveniencing hundreds if not thousands of riders a day to avoid changing the 75 is just a really bad idea.

          *Some of the riders who ride to Northgate Station may be walking across the bridge to the college. It is quite likely that transferring at Pinehurst would actually save them time. By the time the station is complete, trains will be running every five minutes all day long. It will be dramatically faster to ride the train than the bus between 130th and Northgate Station. From the platform to the bridge is about the same if not better via the train (than ground level).

      21. “Metro Connects had the 75 going to the Pinehurst Station but that was not an official proposal. The map was only drawn up to give people an idea of what transit might look like in the future.”

          The question is, why didn’t Metro make a proposal out of its published intention that everybody was expecting? Either it never intended to (in which case it shouldn’t have published it), or it changed its mind. If it did change its mind, it never explained why.

      22. “Every restructure has drawbacks but consider the existing ridership.”

          One issue is 75 riders transferring to the 40. The 75 used to go through Northgate to 24th and Ballard. Riders in Sand Point and Lake City are still going to Ballard, and have to transfer to the 40. But if Metro would extend the 40 to Lake City as some here have suggested, then the transfer could happen there.

          (The 61 is a new route from Lake City that appears to go to Ballard, but it doesn’t really because it peters out at 85th & 3rd NW, three miles from central Ballard. So a 75+61+40 trip would be a 3-seat ride.)

      23. If I were taking Metro’s perspective, I’d say they didn’t want too much churn. The 77 lets them solve two separate restructures (Stride and Pinehurst) by adding a single route.

      24. If I were taking Metro’s perspective, I’d say they didn’t want too much churn.

          Yes, I think that is quite possible. But that is just really bad reasoning. You are essentially screwing over riders in Lake City — that is much worse churn than having to deal with a new 75.

          One issue is 75 riders transferring to the 40.

          Yes, I didn’t mention transfers. Northgate is a transit center after all. Consider the possibilities:

          345/365: You are way better off with the bus going to Bitter Lake.

          40: You could take the 61 (from Lake City) or the 348 (from Pinehurst instead). At worse you end up using Link as the middle seat of a three-seat ride. That happens all the time.

          61 — Possible transfer point but quite likely most riders just transfer at Lake City if they are headed to Greenwood. Again they could take the 348 if they are headed to Greenwood from Pinehurst.

          348 — Unlikely transfer point.

          Consider who benefits from that combination. If you are at Sand Point it would make more sense to take the 62 and 44 if you are headed to Ballard. If you are at Pinehurst or Lake City you would take different buses to Northgate. The only people that would be significantly worse off would be those in between the main areas headed towards Northgate and there just aren’t that many riders there.

          Meanwhile a lot of transfers would be much better. I already mentioned the 345/365 being better. But there is also the north end of the 5 to consider. This would mean a straightforward two-seat ride from Sand Point to Shoreline College. It typically takes over an hour now: https://maps.app.goo.gl/AXQhshF5HniTJswT7. You would probably make the trip in a half hour.

          From a transfer perspective this is just better as well.

      25. The network itself could use improvement but I don’t blame Metro for punting there; when the network was moving through the final stages of approval a year and a half ago there was a lot of uncertainty around timelines and service hours. Metro was going through major issues with drivers/hours and Sound Transit has been struggling with project delivery timelines in recent years. Stride is still a ways out (2028+) and is another opportunity for a restructure

      26. I am trying to be sympathetic as well. I know one of the former planners. He knows a lot of people who still work there. I know Metro has some very smart people. But the Lynnwood restructure was a mess from the very beginning. I know it seems like a little thing to run a bus on 80th when we are already running a bus on 85th but that is just fundamentally wrong. This is a violation of the Metro Service Guide (page 23, Route Spacing and Duplication). It is not a very long document — I’m sure it is required reading by the planners. Yet they ignored it or simply forgot. It is also just common sense. You don’t run a bus five blocks away from another bus unless you have a ton of buses on that first corridor. That is not the case (the 45 only runs every fifteen minutes). This is just basic stuff. I’m quite sure it was suggested by someone who lacked experience but that points to a lack of leadership. It is one thing for a rookie to propose something like that. It is another thing for that to be an official proposal. The idea never should have left the office. It just shows a lack of oversight.

          The same goes for the lack of service on Lake City Way. It was clear they were caught flat-footed. The planners were unaware that Sound Transit was planning on moving the 522. I get that it is hard to coordinate between the two agencies but this is basic stuff. It was covered in the news, blogs like this and subscriber emails. Someone in the organization should have realized the obvious oversight before it went public.

          In both cases they corrected those obvious mistakes with the next round of changes. But that also messed up the process. Instead of having a pretty good plan that maybe led to some refinement they made changes late in the game and didn’t give people a chance to fix it. For example, sending the 65 to Bitter Lake was the initial choice. A lot of us expected the 75 to be sent there (in part because of Metro Connects but also because it makes sense geographically). But the 65 was a fine choice and it was met with almost universal enthusiasm. Not only on this blog but in various community meetings (where we discussed changes around the station). Yet suddenly it was scrapped. There was no opportunity to push for it — people had already pushed against it and it was too late. At that point it would have been quite reasonable to substitute the 75 in exchange. They could even ask for comments, asking for preference (65, 75 or something else running less often). But instead they decided to lump in their fix for the oversight with the east-west bus and just like that we had the poorly designed 77. After one round of big churns (do to the obvious flaws in the first proposal) they didn’t want to mess with things again.

          It is worth noting that various agencies have issues like this. SDOT ran into something similar in Fremont as there was a clear conflict between bike priority and people transferring from one bus to another in the area. While some are obviously disappointed in what SDOT decided to do, it was clear that SDOT explored all the options and gave people plenty of time to be heard. But that wasn’t the case with Metro. Instead of a bus running through the heart of Lake City to Pinehurst Station it will barely skirt it.

          I know these are tough times for the Metro planners. No one wants to restructure when you are facing financial woes. Even simple changes that are very easy to justify (like getting rid of the 73 or 20) are bound to be met with a lot of complaints. But they actually made those changes. Getting rid of the 73 did not come with a good substitute. Every one of those riders was worse off. In contrast if we moved the 75 it is quite likely that a majority of riders would prefer the new routing. They would welcome it as a much faster way to get to Link.

          For example imagine someone living in an apartments on 35th trying to get to Capitol Hill. Here are the options. It takes about 20 minutes to ride the 75 to Northgate Station. If the 75 went to Pinehurst Station it would shave 10 minutes off the trip. In their interest to avoid making a handful of riders upset they screwed over many riders on that same route!

          Not only that, but for a lot of riders things will get worse. This isn’t a case where we are just muddling along with the old system and adding a new bus. Quite the contrary. North of 125th, The 522 carried over 250 riders a day heading towards Roosevelt. Now those riders have to catch a different bus or walk a long way to the bus. From Roosevelt to the Fred Meyer you now have an extra ten minute walk. Instead of two buses heading to Link (the 61 and 522) they basically just have the 61. I really feel for someone who uses that bus to get to the U-District. Right now the weak point is bus service between Roosevelt and the U-District. The 522 gets you to Roosevelt very quickly but it is hassle to take the train just one stop, especially if you are headed a bit north or south of the station. The 45 isn’t that frequent and it is the only one running to the Ave right now. So you end up with trips like this. Not bad, but still a bit of a hassle. But now they will send the bus to the U-District. This isn’t free — Metro has to come up with the money to pay for extension to the U-District. But this would save you around ten minutes off the trip. Except it won’t serve you! You will have to walk ten minutes just to catch the bus now. Your other options involve taking the 61, then Link, then a long walk. Or maybe you take the 61, Link and the very bus that runs along Lake City Way (but inexplicably turns before serving the heart of it). Regardless of the choices, your life just got worse.

          Oh, and consider how this all came about from a planning perspective. Metro originally had no replacement for the 522. Now will run a bus on the corridor and spend extra money to send it to the U-District. Except now many of the existing riders of the 522 won’t be able to use it. Oh, and I get that it can’t go as far as 145th (those days are gone) but at least go as far as Fred Meyer.

          I usually defer to the professionals and assume they know what they are doing. I don’t think that is the case anymore with the Metro planning department. I think there are clear fundamental issues. This is basically three restructures in a row where they have made big mistakes.

      27. Stride is still a ways out (2028+) and is another opportunity for a restructure

          Stride 3 doesn’t really matter from a restructure standpoint. There are three changes in the area in the near future:

          1) 522 goes to 148th Station.
          2) Pinehurst Station gets built.
          3) Stride 3 replaces the 522.

          It was initially expected that there would be a big gap between the first two changes. Now that gap has shrunk. In any even Metro will change the routes to accommodate the first two changes. At that point Stride is basically meaningless from a restructure standpoint since it will follow the same pathway as the new 522. A lot of riders will be hard pressed to notice the change. The buses will be different (battery-electric). They will have all-door boarding and proof-of-payment. They will add a little bit of right-of-way improvement. But it is basically the same bus. From a routing perspective it would change nothing.

          Of course they could have another restructure but that is true anywhere. They could restructure buses in Magnolia next week if they wanted to. But having just shuffled around the buses it seems less likely that they would do so again unless there is a big change (like the county deciding to hire an outside consultant to modify our bus routes).

      28. “ It was clear they were caught flat-footed. The planners were unaware that Sound Transit was planning on moving the 522. I get that it is hard to coordinate between the two agencies but this is basic stuff.”

          I keep wondering when ST, Metro and SDOT will figure out a better way to coordinate. ST plans to even add two entirely new corridors through Seattle (West Seattle and SLU/ LQA/ Ballard) affecting tens of thousands of future riders. Every agency seems to be working on their part — but there isn’t a unified rider experience review happening.

          Seattle could dangle Move Seattle funds in front of the transit operators as a way to get better review from a special riders committee. Given the billions in proposed construction on top of the largest transit operating budgets, it seems prudent to require some sort of rider committee review.

          And to be clear, I’m not referring to an occasional special coordination meeting among staff from the agencies that is held without public notice or attendance. I’m instead referring to an actual working citizens committee with agendas, minutes, video and time for public comment.

      29. “The planners were unaware that Sound Transit was planning on moving the 522.”

          Maybe, but I doubt it. I think neither ST nor Metro considered the Lake City-Roosevelt corridor important because there were other surrounding routes. It was passenger activism that insisted it was an important corridor that forced Metro to change its mind.

      30. Metro and SDOT do coordinate; they just have different ideas about what route alignments are best. Metro acquiesced to ST’s G alternative. The Seattle Transit Benefit District usually adds runs to existing routes rather than creating new routes.

          Metro/ST relations have always been “We (ST) decide the Link stations and entrances and ST Express stops; you (Metro) work around it.” For years ST participated in Metro-led Link restructures: it gave Metro its ST Express proposals ahead of time, and took feedback and refinements through the Metro process. Then suddenly this year it stopped doing that: it withdrew the ST Express routes from the East Link proposals and never offered any for the Federal, Way restructure. It said instead it would consider those routes in an overall 2026 operating plan later. The first proposal on that is supposed to come out around now.

      31. I think neither ST nor Metro considered the Lake City-Roosevelt corridor important because there were other surrounding routes.

          Maybe, but with the 73 being cancelled at the same time that would have created a big service hole. It is worth noting that the stop at Lake City Way & 20th gets more riders on the 522 than any stop in Kenmore, Bothell or Lake Forest Park. Either way, creating a big service hole right where you get a lot of your riders seems like a really bad idea.

   2. I’ll be curious if transferring between Route 7 and Link at Mt Baker will shift to Judkins Park. Not only will that combo path be a few minutes faster but there won’t be a need to cross the street northbound.

      1. I’m sure some people will switch but it is hard to tell how many people are doing that right now. About 400 people on an northbound 7 get off the bus close to Mount Baker Station. But that is also where it makes sense to transfer to the 8 and 14. It is an option for transferring to the 48 but you can just make a same stop transfer further north for that. Anyway if you are headed to Beacon Hill or SoDo then you would make the same transfer as you do now (at Mount Baker Station). But for something further north (like Capitol Hill or UW) it seems a bit better to transfer at Judkins Park. Going the other direction it will be a matter of whatever train comes first I would imagine.

8. One out-of-the-box routing concept that I’ve toyed with is a north-south route that goes back and forth between the Lake Washington edge and commercial districts as well as a Link station or two on MLK or Rainier all the way between Montlake and Rainier Beach. There seem to be lots of tails serving certain areas but it’s not an easy north-south trip. It’s an area about a mile wide with some steep hills and narrow streets. The key would be to stay near LWB while deviating at strategic points.

   I don’t have a fixed alignment in mind. But here’s one example:

   UW Station to Montlake to jogging at Boyer to LWB. Then jog at Madison to take over the Route 8 path to Mt Baker Station, followed by going down Rainier to Genesee to take over the east tail of Route 50. Then at Othello, it would jog north to Graham and run west to Georgetown. Maybe it’s a wildly long route — but its ability to tie so many recreational destinations and Link would amazing.

   The route would have Link transfers possible at UW, Judkins Park and Othello and Graham as well as RR-G. It would connect neighborhoods to the long north-south commercial district between Bayview and Columbia City in the middle.

   I’ll call it Route 59 as a reference to Lake Washington (5 is L and 9 is W on a phone).

   There are certainly challenges. Route 50 east tail serves blind passengers and middle school students and some VA goers. These are why it’s hard to do major restructures.

9. The piece has the premise that Metro will serve South Graham Street. Per Porgy and Bess, that is not necessarily so. Metro is constrained (e.g., hours, operators, buses, and imagination). The recent Lynnwood Link ordinance provides poor service to the Pinehurst station, 2026.

   Routes 106 and 107 were revised poorly in fall 2016 and should be changed. The Georgetown deviation of Route 107 should be deleted; it has been mitigated by frequency improvements to Route 60. The extension of Route 106 via Rainier and Jackson is too duplicative of routes 7, 14, and 36; it could be shifted to 23rd and Yesler Way.

   Route 50 could be split. The west part could be replaced by a full time two way Route 56 oriented to SODO Link via South Lander Street. Routes 125 and 128 could also be changed. The east part could be replaced by Route 39 between Othello Link and the VAMC via Beacon Link and the 50 pathway.

   With the Judkins Link station, routes 7, 8, 9, 48, and 49 could be consolidated and more frequency provided.

   1. The piece has the premise that Metro will serve South Graham Street. Metro is constrained (e.g., hours, operators, buses, and imagination).

      Sure, but this looks like a fairly revenue neutral change. You save quite a bit by combining the 49 and 60 (and making it straighter). The 60 no longer overlaps the 107 and takes about as much time to get to Seward Park. The tail isn’t realistic (as discussed above) but in general it would work and we could keep roughly the same frequency across the board.

      I agree with many of the other suggestions. I’m not saying sending the 60 over to Graham is the ideal routing but it is quite reasonable.

      1. Routes 49 and 60 have different modes and turnaround loops and layovers are non trivial issues.

      2. There is trolley wire on Broadway and up to Beacon Hill. Thus the 49 would continue to be a trolley. It would turn around and layover at Beacon Hill station just like the 107. You might need to take away a bit of parking. Likewise you might need to add a tiny bit of wire. My guess is you wouldn’t. The bus could always end the route, go off wire, turn around and then go back on wire before shutting down and laying over. No change is trivial but this wouldn’t require much work at all.

         The trickier issue is Seward Park. You could turn around by using part of the 50 like so: https://maps.app.goo.gl/cmDo7sG8D2HvCJ8r7. But I don’t know if there is anywhere within that loop to layover. There may not be a comfort station anywhere (although there are a few shops in the area). If that doesn’t work then the bus could be sent to Rainier Beach (as Chris suggested).

      3. In the Rainier Beach alternative, Seward Park (the park) wouldn’t lose any existing service, so the alternative may be OK on those grounds.

      4. The current wiring doesn’t allow a trolley to cross Jackson on 12th Avenue, so Metro would have to add a little bit of wiring if the 49 is to continue to Beacon Hill. At the same time, they could also delete the turn wire that was used by the 9 when it was a trolley route.

   2. Yeah, they would basically do the same sort of minor work that has been done on Capitol Hill and the U-District recently. It is not that there is a lot of new wire. In this case there is wire on both sides of 12th. It is just that they have to set it up so that the buses can go a different connection. That is much easier.

10. This brings up an interesting area: Georgetown. You have the 60, 107, 124, 131/132 in the neighborhood. Ideally you would have some sort of grid but that just isn’t possible in the area. The buses are largely going north-south through there. This sets up a conundrum. If the buses keep going north they don’t connect with each other. If they go out of their way to connect (like the 107) then it is a time-consuming detour.

    One way to deal with the issue is with a “weave”. It used to be more common within our system. For example look at this old map that Oran Viriyincy made. Back then the 68 went north up 25th from the U-Village and then cut over to Roosevelt Way before continuing north. The 73 ran down 15th. That way someone could switch between one of the three north-south corridors with one transfer (at 75th and 15th) or by staying on the 68. We can see a similar example (although not as forced) with the current 65 and 372 (in Lake City).

    I could see buses doing the same thing in Georgetown. I think the easiest way to do that would be for 107 and 131/132 to cross paths using Lucile. So the 107 could do something like this and then continue north on the current pathway of the 131/132. The 131/132 would do the opposite like so. The 60 and 124 could continue to run on the same pathways through Georgetown. Thoughts?

    1. I’ve often felt that a weave works generally better in Seattle than a true grid. Many of Netro’s bus routes do weave.

       It also makes bus stop transferring easier. Two routes can often stop at the same stop or at worst just have one street to cross. In a true grid often two streets must often be crossed to transfer.

       In this proposal, perhaps a weave between a Graham St Route and Route 50 would be useful. Say the weave segment is on MLK. Or the weave segment could be on Rainier. That opens up direct access to neighborhood commercial destinations not convenient to infrequent routes primarily on residential east-west streets. It could even be beneficial before a Graham Station opens.

    2. I’ve pondered various variations on a Georgetown weave.

       I’d like to find a way to stop having buses use the 1st Ave Bridge. It is the destroyer of reliability. When the bridge opens for a boat, half the bridge has to be raised first, then the other half, then the reverse. The whole operation takes much longer than opening other bascule bridges. Route 131 does not serve any ride-generating stops between Georgetown and Roxbury. It may as well take the 14th Ave S Bridge, and Cloverdale through South Park, and also serve the Georgetown Campus along the way.

       I want route 107 to not loop-de-loop into and back out of the fading Georgetown saloon zone. If it were to give Cleveland-bound riders a straighter ride, ridership might improve. But if southern 107 riders want a one-seat ride to Georgetown, then let it get there on Lucille (with the triangling around the stub track when needed), and continue north to the SODO. It used to do that, but missed the high school.

       I would like there to be a bus serving 15th Ave S between Cleveland and BH Station every 10 minutes all day, and more frequently than that during the hour before school and the hour after school. Pass-ups are a problem during these peak hours. Having a single bus route seems like it would be more evenly-spaced than the 60/107 pairing.

       As I said in another thread, I would like riders along Graham to also have front-door access to Cleveland, the Georgetown campus, and maybe even the main campus of South Seattle College.

       However, my main bus is route 60, which I use primarily to get to and from BH Station and secondarily to get up the hill to businesses in White Center / Westwood / West Roxbury. I will miss my quick ride from BH Station to South Park. But if a route restructure has more winners than losers, so be it.

       At the same time, I don’t want to see the stops in front of Cleveland become transfer hangouts for anyone who does not have business on campus. That may be an intractable hurdle to having routes weave over each other on Lucille.

       1. I’d like to find a way to stop having buses use the 1st Ave Bridge. It is the destroyer of reliability.

          That would be fairly easy without losing any coverage. You would make the 131 longer and the 132 shorter. The 131 would loop around and cover all its current stops as well as some of the stops of the 132: https://maps.app.goo.gl/LReE31dPe1qm8GaK8. The 132 could just go over the South Park Bridge. At that point both buses could serve the south side of the Georgetown South Seattle Community College (but you would still need the city to add a crosswalk and bus stop there — https://maps.app.goo.gl/KjYe23Es5TAtAQzF7).

11. Brent brought up another issue as well which is connecting to the Georgetown campus of South Seattle Community College. None of the buses serve it well. The 60 and 124 get close but only going northbound. It is a tricky issue given the infrastructure in the area.

    There are some options but many would require (or at least be much better) if the city did some work along with Metro.

    The main entrance is on East Marginal Way here. It is a sprawling industrial campus as the main focus is on trades (I’ve actually been there and it is nicer than it looks on Google Maps). I could see either the 60 or 124 detouring slightly to serve it. The city would need to add a crosswalk there (with a beg button) to get from the other side of the street (next to the railroad tracks).

    Another approach would be to modify the 124 to run on Corliss both directions. Right now the bus goes north on East Marginal Way, Corson, Bailey, 13th and Airport Way. That would stay the same. Going the other direction the bus goes on Airport Way, 13th, Albro, Ellis, and East Marginal Way. So even now it doesn’t cover the same streets both directions. I would modify the southbound bus to go Airport Way, Corson, East Marginal Way (like so). This appears to be a faster way through the neighborhood as well. The problem is coverage in the north part of Corson. Without an additional crosswalk you would have some long walks. At a minimum you would add a bus stop on Corson just west of Airport Way to make for a shorter walk from the north. Another bus stop south of Bailey would cover the southern end (just barely). But ideally you add a crosswalk (with a beg button) at Orcas/Doris. Unlike the crosswalk next to the college it would be used by folks in the neighborhood (and not just people using the bus). A proper crossing there would be a big improvement in pedestrian and bike mobility in the area (no more of this crap).

    Without a crosswalk it would be worse for people in the area but you would be able to get much closer to the college both directions. The existing (northbound only) stop at Corson & Willow is quite good (especially if you are heading to that end of campus). A stop going the other way would be even better.

12. The few times I’ve been to Georgetown, what the trip planner suggested was Link to Beacon Hill and then a fairly slow trip on the 107.

    I’m wondering if this east-west route could eliminate the need for the 107 Georgetown appendage at all. I’m not convinced there’s that many people on the 107 corridor that really need to go to Georgetown.

    1. Trip planners often have less-than-optimal notions. If you’re coming from downtown, I’d take the 124. It runs on Airport Way, which is fast, because it was designed as a Boeing highway but that role has been superceded by I-5, so it has few cars.

    2. I use the transfer stop in Georgetown occasionally, when the 107 comes before the 60 southbound at Beacon Hill Station. But I will be happy to use the next stop north on the 107 when this ridership-killing detour goes away.

13. Once the Graham Street Station opens there will likely be a major restructuring of south end service. There are plenty of inefficient and obsolete routings to modify: the 14 tail to Mt. Baker neighborhood layover point, the 4 wandering through Judkins Park but never quite reaching Mt. Baker Station, the Prentice loop, and plenty of corridors frequent lacking service.

    Bus service on Graham Street in an east-west direction will be confined to the corridor between Beacon Avenue and Rainier Avenue. East of Rainier is low density and steep between Graham Hill Elementary and Wilson. Orcas may be the better street to connect to Seward Park. West of Beacon, Graham is a low density, narrow, congested, winding street that will need major civil engineering work before a bus is going to use that segment.

    The Graham Street Station will make everything between Othello Station and Columbia City Station walkable to Link. New construction will boost Rainier Valley density and Metro will need more east-west feed than the current 50 and 106 provide. A major restructuring will be needed from Jackson to Henderson when Graham Street is opened.

    1. You make it sound like Metro is eager to modify inefficient and obsolete routes. It was in the mid-to-late 2010s. It’s less certain now.

       Metro proposed deleting the 4S in one of the C/D/E restructures and in the 2014 cuts. The hours would have been folded into the 3S. The 3N would have been folded into the 4N and rerouted to Seattle Pacific. People complained that a blind services center at the end of the 4S would lose status. The proposals also would have split the 2. Splitting the 2 raised so much opposition in both cases that Metro retreated, and that caused cascading rollbacks of the 3, 4, 5, 26, and 28 proposals.

       Later the 3N -> 4N consolidation and reroute in Queen Anne was completed.

       Also later there was concern about the 4S being the only access from lower 23rd to First Hill.

       If the 4S’s tail is to remain, I think it should be extended to Judkins Park station.

       1. I agree. There have been much bigger changes that have resulted in very tiny modifications. My guess is they will add a route somewhere to serve the station but otherwise not do anything.

       2. A lot has changed since the mid 2010s. There’s much more TOD in the Judkins Park/north Rainier Valley area, including the 2 Line. I rode a jam-packed 7 on Thursday that took 30 minutes to get from 4th and Pine to I-90. The 2 Line will make that trip in 8-10 minutes. Fewer riders within the walkshed of Judkins Park will be using the 7 for trips to downtown. But the question is whether the new TOD will create enough riders connecting to other destinations to replace the people abandoning the 7?

          The same dynamic will play out along MLK, assuming TOD takes off in the next decade. Once it becomes an easy walk to Link for anyone living within the walksheds of the stations at Othello, Graham and Columbia City, will the new TOD replace the riders who abandon the 106 to downtown with riders to other destinations?

       3. “ A lot has changed since the mid 2010s. There’s much more TOD in the Judkins Park/north Rainier Valley area, including the 2 Line.”

          Unless someone has been on Rainier between I-90 and 23rd in the past year or two, they are not aware of how radical the area’s land use changed recently. The Census Bureau estimates that Seattle added 29,000 residents in the past two years, and this area is one that provides new housing to enable this increase.

          I’m curious if or how ground level retail will occur. The dense neighborhood, along with awesome rail and bus transit access, offers a marketplace that could accommodate almost anything. Once one or two anchors are opened, it could be a go-to neighborhood!

14. Even before Graham St Station opens, I would love to see route 60 re-routed to serve Lucille St, with the occasional re-route around the stub track when a train is sitting there. That would remove a lot of out-of-direction travel with few boardings.

    I’d also remove the 107 Georgetown loop-de-loop, as I’ve begged for in several posts and many comments.

    .

    I would also have the Graham St route serve Lucille / Corson / E Marginal / 14th / Cloverdale, but have it take over the 132’s weave through northwest Sputh Park, and then duck under the 1st Ave Bridge to head up Highland Park Way, on the way to South Seattle College, Westwood Village, or both. It would look like the proverbial milk run, but enable other routes to become faster, and connect the most major destinations in this latitude of South Seattle.

    It might help some other West Seattle routes.

    It could enable route 131 to avoid the 1st Ave Bridge, serve the Georgetown Campus more directly, and then follow the 60’s more useful path to Roxbury.

    It could remove the South Park zig-zag-limbo-wait-forever-at-the-1st-Ave-Bridge from route 132 by having 131/132 both continue on E Marginal Way / 14th, and then route 132 would continue straight on its current path through Boulevard Park, as envisioned in Metro’s Long Range Plan.

15. So here’s a question: Will opening Graham St Link Station initially take away riders from Metro, will it simply divert riders from nearby Link Stations or will the station Link boardings just evolve?

    I understand how it’s a catalyst for TOD in the long run. However I could see that Route 106 could have fewer riders in the short run. More residents will be within walking distance of Link and may just choose to start walking to Link rather than wait for a Metro bus..

    In addition, I don’t see many new riders induced from additional Graham St Metro connecting bus service. The area has good, frequent transit access already. I feel like addicting bus service there may simply take riders away from other routes.

    1. It will take riders from the 106 who currently use it just to get to Othello or Columbia City stations. That’s probably few people, because others probably just remain on the 106 to north Rainier or Jackson. There was a “Save Route 42” push that ultimately got Metro to extend the 106 from Mt Baker station to CID station in lieu of keeping the 42. The activists who pushed for it are probably riding the 106 now, and others are probably riding it too for trips between south MLK and north Rainier/Jackson, because if you don’t want to transfer to Link, you probably don’t want to transfer to the 7 either but instead have a one-seat ride.

       However, there could be an increase on the 106 with Judkins Park station, because it will give new access to the Eastside and a faster train trip to downtown and the north. That will be a draw for people all along the 106. That would probably dwarf any loss of riders using the 106 as a shuttle to Othello or Columbia City stations.

       The 106 is a sensible route south of Mt Baker station. The overlap with the 7 north of it could be justified (people differ on this). The entire 106 corridor from Renton to the valley and the extension to Jackson is all ethnically and economically similar, with a lot of overlapping trips all along it, and people having family ties along it, or going to churches and ethnic stores along it.

       My wishlist item would be, if the 106 is going to overlap with the 7, then it would have trolley wire on south MLK to Renton, so you aren’t taking a diesel bus in a trolley corridor wishing you could be on the trolley route but that would add a transfer.

       1. The Route 7/106 path/ segment between Judkins Park and the ID appears to me to be over-served with these two routes running on top of each other. (There is also the FHSC and Route 14 on Jackson too.) This segment also moves very slowly, even after the SDOT bus priority changes. The Judkins Park station opening in the next year will draw even more riders away.

          I don’t see Route 106 getting trolley wires. Battery-electric buses seem to be the trend, and stringing wire to Renton seems very expensive — along with the engineering challenges of wires a crossing Link on MLK.

          My preference is for RapidRide R to take over most of the Route 7 service, with that running to Capitol Hill (Route 9) and ending at Rainier Beach Link at the other end. Route 7 can continue but be reduced to just the Prentice loop trips (about 1/3 of the service today) and can continue to follow the Jackson path through Downtown. It may be possible to do RapidRide R with trolley wires but it would need to be studied. Rainier Beach Link would need a nearby layover spot for RapidRide buses too.

          And I wonder if RapidRide R could be better if they ran left and right door buses like RapidRide G. There are several places in the corridor where median stops might be preferable. I can’t necessarily say that it would be worth the cost, but I do think it’s worth mentioning.

    2. I don’t foresee many riders on Graham St using a bus to get to the new station, at least going downhill.

       But I can foresee a nontrivial amount of riders using that bus to get to Cleveland High, depending on the school boundaries.

       A smaller number will take the bus to get to businesses in Georgetown, on Beacon Ave, around the station, and on Rainier.

       1. The school boundaries for Cleveland are unusual. It isn’t an all-city draw but it doesn’t have the usual boundaries of most schools. They give preference to those who attended certain middle schools but you might be able to go there from other places depending on how much demand there is.

16. When you go to Uwajimaya from downtown, do you take Link or the 7/14/36? I’m trying to figure out if one is more convenient than the other.

    One answer might be the buses are more frequent but the Link station is further south.

    Another might be that the Link station is on the west side, while the main produce/meat/miso sections and main entrance are closer to the east side and doesn’t have stairs.

    Another might be that the tradeoffs are a wash.

17. I looked to see what exactly the 4 would need to do to serve Judkins Park station. I thought maybe it needs to go further south, but it already goes beyond the station. The problem is that ad Judkins Street it turns east from 23rd to MLK and goes down to a Plum-Walker Street loop. Whereas if it just stayed on 23rd, it would serve the station as well as the 48 does.

    So why does it detour east to MLK for a few blocks? Is there anything notable there? I looked to see where Lighthouse for the Blind is; it’s as 2501 S Plum Street. That’s closer to 23rd than to MLK.

    I think the 4 was extended south in the 1990s or 2000s. It had always said “to Judkins Park” since the 70s, but I think it was extended from Judkins Street to Walker Street for some reason, maybe for transfers to the 7 at Rainier. Still, it could remain on 23rd and do that, and still serve Judkins Park and its namesake station.

    1. The current 4 routing is not great. As you mentioned, it avoids Judkins Park station and it doesn’t go all of the way to Mount Baker station.

       Similar to Route 8, the MLK segment south of I-90 has pretty low ridership. I think Route 4 should follow the soon-to-be Route 8 path on 23rd -> Massachusetts-> MLK -> Mount Baker TC. Route 8 could also follow that route or it could stay on 23rd -> Rainier for a more direct path.

       Lighthouse for the Blind is at the corner of Plum and MLK, but the 4 and/or 8 could stop on MLK at Plum.

       1. You could send the 8 up the hill to Beacon Hill Link. It’s part of the Metro Connects proposal too.

    2. The 4 is very old. It was originally a streetcar route. It was one branch of the 12. It was quite similar in 1970 when it was a bus. It used to just end north of the freeway. It was renamed the 4 but was otherwise the same in 1988 when it was renamed the 4. Sometime between 1988 and 2015 (when Oran made this map) it was extended to Rainier. I get the extension. It connects to both MLK and Rainier that way (the 8 and 48).

       In my opinion it is outdated. It is running at coverage frequency (every half hour). Coverage routes should provide meaningful coverage. This does not. Riders can take other buses and at worse just transfer. Routes like this become part of a vicious cycle. Frequency is bad because of excess overlap. People want routes like this because they hate making a transfer to a bus that is infrequent. But with buses running every half hour it just doesn’t work. I would get rid of this tail.

       I would run the 3 from Madrona to Summit every fifteen minutes. I would run the 4 from Garfield (23rd & Jefferson) to Queen Anne every fifteen minutes. They would overlap for 7.5 minute service within the core. This is not that different than what they have now but much easier to understand.

       1. The last time a major central area restructure happened was with the U-Link opening in 2016. With so many land use changes and route changes and travel need changes in the past decade, one seems due in 2026-27 once Judkins Park opens.

          I think the riding public will have had enough experience to understand what’s better for them at this point in time, and could more objectively assess what routing and scheduling changes mean. (The 2016 effort had many unwilling to consider making any changes to the routes that they used since riders were generally more predisposed to directly going to Downtown and especially Westlake at rush hours then.)

       2. “The 4 is very old. It was originally a streetcar route. It was one branch of the 12. It was quite similar in 1970 when it was a bus. It used to just end north of the freeway. It was renamed the 4 but was otherwise the same in 1988 when it was renamed the 4.”

          In 1977 according to Sam’s map above, the 4 was on Madison-north 23rd. The 12 had branches from Jefferson to Madrona and Judkins Park.

          In 1979 when I started riding Metro, the 3 was the Madrona route, the 4 was the Judkins Park route, number 12 didn’t exist, and the 43 was a new route on John-north 23rd replacing the 4.

Comments are closed.