Metro route 3. Photo by Tim Bond.

Metro recently released a summary of community feedback on its proposal to move a short segment of routes 3 and 4 from James St to Yesler Wy.  As we’ve come to expect with proposals to change the oldest parts of Metro’s network, the feedback was deeply muddled.  Metro’s Magic 8-Ball said: “Reply Hazy, Try Again.”

Online survey respondents favored the change, 53 to 40 percent.  Most of the few people who contacted Metro by email or phone opposed the change.  Stakeholder organizations were split along geographic lines; First Hill Improvement Association and WHEEL (which operates a women’s shelter near 8th and James) opposed the change, while Yesler Terrace Community Council supported it.  Supporters cited better reliability and improved service to Yesler Terrace, while opponents concentrated on potential difficulties accessing services on James for seniors and people with limited mobility.

In keeping with this split feedback, Metro plans to study a variety of options using both James and Yesler.  The agency will study transit priority measures on James, to see if there is any way to speed up buses despite the very high volume of I-5 traffic.  Previous studies have found bus lanes on James infeasible because the volume of I-5 car traffic trying to use James would create gridlock on other streets (including 9th Avenue, which the current routes use), but Metro will have another look.  At the same time, the agency will continue designing trolley overhead and other infrastructure along the Yesler route.  Finally, the agency will look at putting another (presumably less frequent) route onto James to provide access while moving routes 3 and 4 to Yesler.

By its nature, this feedback process could not include any voice representing the over 5,000 net new residents (including about 1,100 low-income residents) who will come to Yesler Terrace once redevelopment ($) is complete.  Redevelopment will turn Yesler Terrace into one of the city’s densest areas, and no comparable development is proposed for the area around James Street.  Yesler Terrace and downtown are currently connected only by infrequent route 27, which is obviously insufficient to serve the new population.  As a regular route 3 rider, I think the combination of reliability improvements and Yesler Terrace redevelopment makes the move to Yesler the obvious best option for routes 3 and 4.  Community feedback regarding access to the James/5th and James/8th stops, though, may warrant moving a low-ridership coverage route (the 27?) to James to serve those stops, despite the delays for riders that will certainly result.

58 Replies to “No Clear Consensus on Routes 3/4”

  1. The writing on the wall seems to indicating that the 27 is going to be the one moved to James, and Leschi will become the sacrificial lamb to satisfy people going to 8th/James who can’t walk two blocks. Most likely, I see the “jog over” happening at 23rd.

    That said, those headed from Leschi to First Hill, rather than downtown itself, will be better off with the change, so it’s not all bad.

    Hopefully, SDOT will get transit signal priority for all the extra left turns that will now be made around 23rd Ave.

    1. Why would the 27 jog at 23rd? That would leave a mile of the current route without service. Yes, it’s only three blocks to Jackson, but Yesler has had service since it was the “Skid Road”.

      If this is done, the 27 should follow the same route along 8th and 9th Avenues as will the 3/4.

      1. I grant that would mean that buses going both ways on 8th and 9th south of Jefferson would be both “outbound” and “inbound”, but that would not be unique to that location. It used to be true for the 4 between its Queen Anne tail and Boston once the 13 started running.

      2. I was envisioning the 3/4 staying on Yesterday to 23rd, so the 27 would need to cover the 3/4’s current path through First Hill to maintain coverage.

        Sounds like my assumption is not correct. That what Metro is asking for is basically a two block detour to preserve as many of the current 3/4 stops as possible, just avoiding James St. underneath I-5.

        So, the result becomes a tradeoff. Making the change speeds up some rush hour trips when traffic is really bad, at the cost of adding several minutes to everybody’s trip during non rush hour.

        Maybe the right solution is to keep everything as is, but add a new peak only version (e.g. 4X), which skips some stops and takes Yesler for a faster trip.

      3. For what it’s worth, the stop-by-stop data Bruce Nourish collected in 2011 suggests that Yesler will be faster even outside of rush hour, because it has fewer traffic lights and avoids the congestion-prone block of 9th between Jefferson and James.

      4. I agree with Richard, jogging the 27 at 23rd would be a bad idea. That would create an even bigger hole for an area that is under-served (the C. D. should have more frequent service as well as a more comprehensive grid). But if you alter the 27 to serve James, then it should turn on 12th or Broadway.

        Either way, though, folks lose out. Imagine you are standing at 23rd and Yesler. You have four buses that run by your bus stop. You are in the middle of a densely populated area (by Seattle’s standards). All you want to do is go downtown, which is only about a mile due west. Even now, things aren’t great. The 27 runs every half hour, as does the 4 and 14. So you might be in for a long wait.

        But with this change, things get worse. The 27 was infrequent, but at least it was a straight shot. It ran right on Yesler into town (no turns, no freeway traffic). Now the 27 will deviate to serve James. You pick up a connection to First Hill (as mentioned) but you had that anyway, with the 4.

        I understand what Metro is doing, and it makes sense. Run an infrequent bus to an area (James) that is hard to serve, and has service only a few blocks away. But the problem is that the 27 shouldn’t be infrequent. It is the only service on Yesler east of 14th (and Yesler west of 14th it is problematic, since the streetcar initially goes the other direction). Switching the 3 and 4 to Yesler will help things for Yesler, but only starting west of 9th. Much of Yesler remains without frequent service, and if you switch up the 27 service will be more indirect than it is today.

        I really think this should be part of a bigger restructure. There are two major things that will happen fairly soon, and both will change bus service substantially in the C. D. One is Madison BRT, the other is Judkins Park Station. With this change (which I think is warranted) they should look at the area in general and come up with routes that serve the high density areas (everything east of MLK) more directly and more frequently.

      5. RossB, you don’t need frequent E/W service on Yesler if you have it on both Jackson (just three blocks south) and Jefferson. I think the best move for Metro in this area is to increase frequency on both the 3S and 14. When Judkins Park station opens, many of the riders you’re thinking about will use the 48 and the planned replacement for the south part of the 8 to access it.

      6. I was envisioning the 3/4 staying on Yesterday to 23rd, so the 27 would need to cover the 3/4’s current path through First Hill to maintain coverage.

        Ah, that makes a lot more sense. Not only your comment, but the idea in general. It isn’t that different than my idea, listed below. The main thing is, buses follow a more regular, straight route. If Yesler is used as a bypass for the 3 and 4, the 4, especially does a lot of winding. It goes on Yesler, then turns up to Jefferson for a few blocks, then turns south again on 23rd. It just begs the question: Why doesn’t the bus just go straight on Yesler and turn on 23rd — what is the point of that jog? The only reasonable answer is history.

        So something like this makes a lot more sense: The 3 is a bit faster, the 4 is a lot faster, and Leschi loses out, with a less direct route to downtown (as you mentioned). Most of Yesler has much faster service, but now Jefferson takes a big hit. That last part doesn’t sound very good — I think Cherry Hill should have more frequent service.

        Here is a possibility: On the surface, this looks much the same. In some ways, it is worse, as folks in Leschi are forced out of their way (first north, then back to Jefferson/James). You also create a hole, on Dell (people will need to walk a couple blocks to the nearest bus stop). But here is the catch: you can save a lot of service. The 3 never runs past Garfield. The 2 can also be truncated at MLK. Now you can put service back into the 27. Go ahead and run it every half hour and it becomes a classic trade-off. Folks along the eastern end of the city have less direct service to downtown, but it runs more often (every 15 minutes or so).

      7. RossB, the latter solution would cause me and presumably everyone else in Madrona to drive to work. Both the 2 and 3 have significant ridership east of 23rd, and one frequent route needs to at least reach 34th.

        The former solution makes it difficult to get to Harborview.

      8. Richard, whatever the route number becomes, wouldn’t it be a good idea to have a straight trolley route from the courthouse to Leschi via Yesler?

        And for thirty years, we’ve had wire hanging unused between the Colman Dock bridge wired up the hill to the Courthouse. Could we do an electric route- pretty well mandatory for those grades and likely loads- between the ferry landing and Lake Washington? Think about it.


      9. RossB, the latter solution would cause me and presumably everyone else in Madrona to drive to work. Both the 2 and 3 have significant ridership east of 23rd, and one frequent route needs to at least reach 34th.

        Yeah, I made a mistake when I wrote that. The 2 would be truncated at 34th (where the 3 currently truncates). It is only a few blocks (might as well). Even that seems generous, though. I don’t doubt that there is good ridership there, but I find it hard to see why Madrona is supposed to get a higher level of service than Yesler (east of 14th). I don’t want to Rob Peter to pay Paul, but if I do, at least Paul has higher density. But sending the 2 to 34th seems like it makes a lot of sense (the two lines would connect, as they do now).

        The former solution makes it difficult to get to Harborview.

        Good point. Overall, I like my original idea better (the one listed below — The main thing is, you send the 4 on Yesler, and combine the tails of the 27 and 14. In terms of how far the 3 goes, that is debatable and flexible (it has a truncated version now).

        The 4 as planned really doesn’t make much sense with this change. you send a bus on Yesler a few blocks, then up north, then back south again. As Mike said below, the 4 is debatable anyway. It made sense many years ago, but is now redundant.

        Personally, I’m not too excited about this change, as long as Metro insists on running buses on James. I get the idea — run a less important bus route there. The problem is, there are no unimportant bus routes in the C. D. Every single bus route is important, and if you end up sending a bus to James, that route doesn’t get more service, and ends up being slower. That makes for a very convoluted and squiggly set of bus routes in a fairly densely populated area. For example, the 27 — one of the more straight forward runs in the region — could suddenly be sent up to James, into the thick of traffic. Meanwhile, the 4 is even messier than it is today. It really gets down to my main point. I’m not saying I have the perfect solution, but I don’t think this should be done without a major restructure in the area, unless Metro is willing to not worry about service on James.

      10. As I indicated in a separate comment, my preference is to use James only in the eastbound direction as part of a live loop of a north-end route, and not to have James affect any CD route. I’d kill the 4S and replace it with more frequency on both the 3S and the 14. I think that punts the necessity of making any decisions about the 27, an outcome I’m OK with.

        “Yesler east of 14th” is served very well by frequent service on Jefferson and Jackson east of 14th.

      11. Mark,

        It wasn’t I who originally suggested rerouting the 27 to James between Third and the top of the hill. asdf2 mentioned that Metro was considering it and said it should jog at 23rd. I don’t agree with the jog, and so suggested the possibility of using the same route between Yesler and Jefferson that the 3/4 reroute would follow.

        I’m not stuck on the idea. If there is enough traffic from east of 23rd to make an “express” route along Yesler viable, sure, keep the bus on Yesler and find some other route to make the slog up James. Maybe the rump of the 13 could do it. Nineteenth to Madison to 9th to James and down the hill? Maybe it wouldn’t do to have the local trolleys on Madison, but it would be a good knitting route. But if it would work it would preserve between 19th Avenue and the hospital district while getting the 13 out of the way of the BRT on the most critical part of its route.

      12. Perhaps a route that heads down Yesler to Broadway, jogs over to Harborview, then heads up 9th or 8th to Madison or even under the convention center to Denny Triangle? It still doesn’t serve James and 5th, but that’s difficult to serve with a non-commuter route without getting stuck in I-5 traffic.

      13. “Yesler east of 14th” is served very well by frequent service on Jefferson and Jackson east of 14th.

        Say what? Someone is supposed to walk from Yesler all the way up to Jefferson just to get downtown?

        Service is better on Jackson, but I certainly wouldn’t say it is served “very well”. It is every 20 minutes. The 2 is every 15 minutes. So someone here ( is supposed to walk 5 minutes, and wait for a 20 minute bus, while someone here ( gets 15 minute service just by walking out their front door. Sorry, that just doesn’t seem right to me, but inertia is a powerful thing.

      14. If you read the posts above you’ll see that I’m consistently advocating for at least 15-minute frequency on the 14. You could almost get there today by deleting the Mt Baker Dr tail. In the long term I’d like to see a streetcar extension to 23rd/Jackson.

        I just don’t think it makes sense to split resources between two corridors three blocks apart trying to make them both frequent, especially when there is yet another frequent corridor six blocks further north. In most of the city, we’ve got those corridors half a mile to a mile apart. (Including that spot you put on the map — if you pulled the 2 out of there, the closest frequent corridor would be the 8, a steep half mile away.)

    2. Wouldn’t rerouting Route 27 require using diesel buses on James? That would seem to be a fatal flaw for that alternative.

      1. The 27 + 33 isn’t so long as to prevent it from being converted to one of the battery bus models.

      2. Good Point. I find it difficult in general to come up with a route that makes sense for James, if the 3/4 aren’t it. The problem is, Metro doesn’t want to send a frequent run up James. Fair enough, but what then? There are sections of bus routes that are low demand, but only sections (and mostly well to the east). The trolley wire situation just makes it more complicated. Even if you are willing to spend a little money to connect or reroute something, then what? I can think of some possibilities, but I don’t really like them.

        One is to move the 12, after the Madison BRT is here. From the north, follow the initial route down 19th and onto Madison. But this time, take a left on 14th or 12th. Then turn right on Jefferson, and follow the existing 3/4 route. That adds some service on 14th or 12th, and provides for a bit more of a grid (sort of). That would still require some new wire.

        Another possibility is to make a variation with the 9. Start at the same spot (Aloha) head south, and then follow the current 3/4 route at Jefferson. Since the 9 is very infrequent and doesn’t go downtown, it wouldn’t hurt service there. It might eat into streetcar ridership, but only if it ran often, and they don’t want to run on James often. That wouldn’t require any new wire.

        Better yet, combine these two ideas. Basically modify the 12 like so: Start at the same spot (heading south on 19th) but turn west on Thomas. Then head south on Broadway to Jefferson, and join the existing 3/4. That wouldn’t require any extra wire (except maybe some curves). It would provide both 19th and James with service, but not too much service. Of course that only makes sense after Madison BRT is in place.

      3. Pretty likely my ideas about the 27 don’t really fit into this discussion.Much longer time-frame. And because its main purpose is to create a straight trolley bus corridor between Downtown and Lake Washington on present 27 route exactly.

        And likewise turn south along the present route when it reaches the Lake. But proceed straight south along the shore to Mt. Baker Beach, and then angle west through the park to McClellan. And terminate at Mr. Baker Transit Center.

        My plan would take a lot of trolleywire, one, because of some of the hills, and two, because as I see Seattle developing, I can see heavier passenger loads, by size and weight. But three, because I think all the neighborhoods along the route would appreciate some quiet.

        Though in time-frame I’m considering, by opening day, batteries could push standing loads whole route. For the mean-time consideration here, my suggestion would be to run 3 and 4 with diesels or battery-assist, so no change will be “set in copper” ’til we find ideal route. For now, I still wouldn’t plan to use James for a “funicular”, like it is now, at all. I don’t see anything to lose following the Route 2 from Benaroya Hall to Virginia Mason, then 9th to Harborview, and then present routes. Using University Street Station instead of Pioneer Square, I don’t see any passenger inconvenience, and a lot better speed and reliability.


  2. Perhaps Metro should approach the 3/4 question as if it were a brand of toothpaste. First, survey the existing riders and other stakeholders about the service quality of routes 3/4 and see if the existing riders are happy with the existing service quality. If riders were first asked “are you happy with existing 3/4 service. If 100% of the riders are happy with the existing service, end of project. But, I would imagine only about 25% of the riders would respond positively. If Metro finds <100% customer satisfaction, try to find out what the dissatisfaction is caused by. Ask the riders to list the things that cause dissatisfaction. Could it be the slow slog on James? What other factors are disappointing riders? If it appears that a large percentage of the existing riders would benefit from a switch to Yesler, then Metro needs to get out and sell the change like Proctor & Gamble sells a tube of toothpaste.

    A 53/40 split really isn't "hazy", in politics that's almost a mandate. There may be some real concerns in the 40% group, but Metro needs to look at the bigger picture. How much of the opposition is Edith Bunker shrieking "gee that old LaSalle ran great" and how much of it legitimate? I know politics is a huge part of Metro changes–but if it's a better brand of toothpaste, it needs support.

  3. The area could use some changes. The slow speeds on James are a big issue, but I see another weakness in the system there. Both central James and Jackson have infrequent service. The 27 and 14 run only every half hour. This is adequate and appropriate for the eastern end, but not for service around 23rd and MLK. It is kind of crazy that Wedgewood gets 10 minute service to the U-District, but parts of the C. D. get half hour service to downtown.

    I support moving the buses to Yesler, but it makes the 4 fairly wonky. It will start out on Yesler, then head north to Jefferson, then back to Yesler before heading south on 23rd. That is less than ideal. I would address both of these problems like so:

    This is a significant change, but doesn’t require laying a huge amount of wire. Eventually you would send the new 4 straight down 23rd (to connect to Judkins Park station) and end at Mount Baker Station, but for now, you would only run wire on Yesler.

    The 27 becomes much more of a coverage route, which means running it every half hour is fine. The 3 runs frequently most of the time. The 27 is timed to it as best as possible, which means that service to Garfield remains very frequent (either bus will do). The new 4 runs a lot more frequently (roughly every 15 minutes, or as often as the combined 3/4 run).

    That would cost a little money, but there are savings to be had. You can truncate the 2 at MLK, since the new 27 serves that area just as often. The 27 is a slower route, but the new 3 and 4 are much faster. Since those buses run a lot more often, the overall savings are substantial.

    We would probably change the bus numbers after all of this.

      1. If you got rid of the 4S, then you are basically getting rid of the 4. On Queen Anne it is just one route, so one side benefit is that we simplify the numbering.

        I can see why you (and others) object to the 4, but it really isn’t the biggest problem in the area. The tail of the 4 may be redundant, but it probably is stronger than the tail of the 2, 3, 14 and 27 (the tail of the 14 is especially bad, but that is another story). All of those routes go through low density areas, while the 4 does not. It simply makes a curve, which while not essential, does provide a one stop connection from the Judkins Park Area to Cherry Hill. With my proposal, it goes away, and is replaced with a one seat ride from that same area to Yesler.

        The two issues are related. Getting rid of the old 4 (or my new one) sounds fine, except that you would need to greatly increase service on the east-west routes. Now the weak tails become the problem. In other words, running the 27 frequently would make a lot of sense if it stopped at MLK, but since it doesn’t, and spends a lot of time covering fairly low density areas, it is hard to justify.

        That is why I think it makes sense to combing the coverage buses that run on the eastern end. If you can combine them, then you might be able to justify better service. If you still can’t, it becomes less of an issue, since you simply have less of them. I’m not saying we should go to the extremes found with the 24, but we should head that direction.

        Here is one possibility:

        There are some interesting things about this. First of all, this provides service to James (which Metro has said is essential) along with coverage on the tails of all of the routes (2, 3, 27, 14). Not all of it is covered, but close to it. There is only one section of the 27, between Lakeside and 31st, which wouldn’t have service. The bus routes seem to be fairly well matched, which means you could run them each every half hour, which means the core gets service every 15 minutes.

        Then you start truncating. Truncate the 2 at 34th. Truncate the 27 at 31st. Then you have the 14 and the 3. If you want to keep the 3 (and run it on Yesler and Jefferson as proposed in this post) then truncate it at 21st/23rd (a truncation variation that exists now). Send the 27 to 23rd. Those changes, by themselves, should ensure some very frequent service on all of the core east-west lines (2, 3, 14, 27). It simplifies things, in that each bus can run as frequently as it needs to, without being tied down by its poorly performing tails.

        Of course we could also just not worry about Cherry Hill and whether it has very fast service to downtown. With this proposal, it would have service similar to what it has now. So, at this point, we could just get rid of the 3. Just have the 14 turn on 23rd, and end close to Garfield. That is a huge savings and simplification of the system.

        But there is flexibility here. If an “express” to Cherry Hill/Garfield makes sense, then I am fine with it. Even with a new 3, that heads that way, you have a lot of savings here. You have trimmed the trunk of most of the lines, which means that if you add more service, it doesn’t cost you much. That means the truncated 2, 27 and 14 can run a lot more often, and running a new 3 is fairly easy as well.

    1. The 14 has been upgraded to 15 minute peak headways, 20 minute midday and 30 minute at night. Saturday gets 20 minute headways.

  4. Is there a constituency for a coverage route that soaks up all the unproductive sections while making room for Metro LRP goals? For instance, if the 3 moves to Yesler and the 2 to Pine and Madison BRT causes an axe of the 12, you could have a half-hourly *trolley* route that does the following: Interlaken, 19th, Madison, Union, Seneca, 9th, James, 3rd, with a turnaround in Pioneer Square.

    1. Metro has been doing a lot of that recently, stringing coverage tails together into a single route, so it’s possible. On the other hand Metro may be counting on the 12′ s hours for Madison RapidRide.

    2. Your proposed route would require several additions to trolley overhead.

      The simplest way to cover James with trolleys using existing and already-planned overhead would be to run a terminator route that just goes up James and comes back down Yesler. My suggestion would be to do that as part of a restructure that eliminates route 4S and separates the north and south halves of the 3/4. You’d have the following routes after the restructure:

      – “Route 3” that is the current route 3S. The terminals would be 3rd/Virginia on the west end (as with today’s downtown turnback trips), and either 34th/Union or 21st/James on the east end. Frequencies along the core part of the route would be 5 minutes peak and 7.5 minutes off-peak, with half of trips continuing to Madrona and half terminating at 21/James. Late nights you would have 15-minute service with all trips continuing to Madrona.

      – “Route 4” that is the current route 3/4N to Seattle Pacific. On the downtown end, it would live-loop using James eastbound and Yesler westbound. Using James only eastbound would avoid the worst of the traffic issues while still serving both uphill and downhill trips reasonably directly. Frequency would be 8 minutes peak, 10-15 minutes at other times.

      – Late nights, when all critical services along James are closed, the two routes could be through-routed and use Yesler in both directions.

      The savings from eliminating the 4S tail and its half-hour recovery should cover, or almost cover, the extra time spent downtown under this solution

      1. Isn’t the worst traffic on James at the I-5 southbound on-ramp? How would using James eastbound avoid that?

      2. The worst traffic “on James” for purposes of these routes is really on both James westbound and 9th northbound. Both lanes of James back up as everyone tries to squeeze left to turn onto I-5. 9th backs up because no one can make the left turn onto James. Delays in the eastbound direction do happen approaching 6th (where the onramp is), but are minor compared with those in the westbound direction.

      3. Only switches, right? There’s already wire hung for all of that proposed route on 19th, Madison, Union, Seneca, 9th, and James, right?

    3. I proposed something similar (after you — I should have read ahead), as a way to move the 12, and get coverage on James: I think that would require only one switch (at Jefferson and James) and no new wire. I think your route would require a little bit of wire on Boren. Either way works for me.

  5. OK, we’ve got it!

    The Queen Anne Counterbalance is an excellent example, and I think some trolleybus routes in San Francisco too. Trolleybuses are the natural replacements for cable cars.

    However else service gets to Harborview- through DSTT construction, pretty sure 3 and 4 ran 9th- we’ll have a straight, strong electric route to Leschi. Where wire can be extended north to the Route 2 terminal.

    And even better, follow present 27 route under the I-90 bridge, and join the 14 at McClellan, where it’ll turn west over the hill to Mt. Baker transit center.

    Procedurally, would defintely speed up progress. Because I think there’d enough more agreement than on side-routes that we could be all the way to the lake while other routes are still being negotiated.

    And (sorry about this, but it’s YouTube’s fault, because Emmylou Harris has gone from cute to gorgeous in one lifetime)…

    Green and cream color. Myrtle Edwards to Yesler to IDS, take the cable at Fifth Avenue stop. Could get some flak between 23rd and the Lake, but free tickets for homeowners’ kids should turn our polite requests into screaming five year old demands.

    Already checked: Cable motor won’t fit into south mezzanine Pioneer Square. And besides, SDOT, the Waterfront Project, and Seattle Art Museum owe us BIGTIME! Call it Astro Hungarian Italian Maximinimal Surrealist Conceptual and we’ve got them!

    Besides, any grief about it, and Connector will go to Ballard via Queen Anne Avenue and Seattle Pacific.

    Mark Dublin

  6. What if the 3 continued its present routing and the 4 followed the new route on Yesler? There’s still frequent service to Harborview and along Jefferson, and this way, both areas between 3rd Ave and Harborview still have decent service (though not necessarily “frequent”).

    1. FHIA proposed this, but I think it’s a “worst of all worlds” scenario.

      – Yesler Terrace service will be bifurcated between two infrequent routes that cannot realistically have staggered scheduling.
      – All PM commuters from First Hill will try to pack themselves onto the few 4 buses to avoid James, leaving the much more numerous 3 buses empty.
      – People trying to get from Pioneer Square to Harborview won’t know where to wait.

  7. As much as it’s tempting to quickly shift routes, the response here suggests using an incremental approach.

    1. Begin by moving only Route 3 to Yesler. That will keep service on James and give Metro real-world data on travel times for Routes 3 and 4 to compare. It would also ease fears of the impact of potential loss of bus service on James.

    2. After a year, consider moving Route 4 Also to Yesler, depending on what data and opinions are about James service segment.

    3. Initiate community discussion on major route restructuring in light of Link now open at Capitol Hill and the profound opportunity of the 23rd Avenue entrance to Judkins Park Station. In particular, direct access without a transfer between Judkins Park and Cherry Hill/ Seattle U/ First Hill makes a great deal of sense to not only connect these areas outside of the out-of-direction congested Downtown core, but also to balance the loads to be more even along a route (rather than crushing everyone into a bus where it crosses I-5).

    Metro could apply some lessons about recent restructuring processes to a renewed effort. I think public perception has changed about using Link (now more desirable and less feared) as well as having to always transfer on Third Avenue (less desirable because of crowded and slow buses, elimination of the no-fare zone. and perceived safety and security issues). Surely the communities involved will be more interested in making changes than they were in 2015.

    1. I think the problem with the changes in 2015 was that they weren’t much better than what existed before. As good a station as CHS is, it is terrible from a bus transfer standpoint. It is just tough to get to. Imagine instead if they had a station at 23rd and Madison. No one would care about the 43, and folks in Madison Valley wouldn’t mind if they sent the 11 down Thomas. That gives Metro a lot of extra service to work with, which can really make a difference.

      I personally don’t think that Judkins Park is the game changer that a lot of people think it is. It is nice, to be clear, but I don’t see it as saving oodles of hours of service, or making everyone welcome a transfer. If I am at 23rd and Yesler (for some reason I’ve been at 23rd and Yesler a lot today) and want to just go downtown, what do I do? Take a bus south (?) and then walk up the stairs to the station? Yeah, maybe. Or maybe they just run the 27 more often (as they should) and you take that bus. It is only a bit more than a mile, and it is really crazy that they run that bus every half hour. Put it another way — if SDOT actually does what they want to do, and make the 7 very fast (with transit only lanes from downtown to Mount Baker) will anyone get off the bus if they are headed downtown? I doubt it.

      Again, I don’t want to downplay the station too much, but it really is nowhere near what the C. D. needs, or deserves. It is OK, and very nice for the people who work in Bellevue. More to the point, it won’t enable enormous changes in the way that buses work in the area.

      Neither will the Madison BRT, but I think it is bigger. Transfers aren’t fun. You have to add it up — inconvenience, extra walking, extra waiting. Mount Baker Station managed to hit the trifecta: inconvenient, lots of extra walking, and plenty of waiting to boot. Despite the fact that the train is fast, it just doesn’t add up — folks would rather just sit on the bus.

      But I don’t think that will be the case with Madison BRT. You still have the inconvenience, but it will be very frequent (more frequent than Link), with less walking (no escalators, or multiple crosswalks just to get to the platform). Even if it isn’t as fast, it will still be significantly faster, which means that it will actually be worth the transfer. Since it actually serves the norther crux of the area (23rd and Madison) it really is a game changer. We can finally kill the 43, for example, but that is just the start. Make the 11 turns on Thomas, and follow the old 43 route all the way downtown. That means folks in Madison Park have a bus that connects to Madison BRT, Link and still offers them a one seat ride to downtown (lucky them). Is the 12 even needed? It is four blocks to the 10 (which goes downtown, and connects with Link). It is four blocks to the 48, which connects with the new 11, as well as the Madison BRT. If you simply shifted service from the 12 to the 48, you would have a much faster route to downtown (even after the walk) and plenty of options for getting to Link. But maybe you don’t want to go that route, and do something different, like adding a north-south bus between 23rd and Broadway. That means something that actually resembles a real grid in the C. D. and enables a reasonably fast trip between Swedish Cherry Hill and K. P. (formally Group Health).

      Which is not to say that I dislike your idea. I agree that a bus that goes up 23rd, connects to Judkins Park Station, and then turns at Yesler sounds good. I think that is certainly on the table, and I think we should think big, and should think about major changes in the area. But I think the impetus for such changes will be Madison BRT. As it turns out, it comes on line at about the same time, so either way, it represents some major opportunities. Until then, it may be better to just nibble around the edges, as you suggest.

      1. Madison BRT should obviously be part of a restructure discussion.

        The halfway mark between Madison BRT and Judkins Park Station entrance is about Alder Street. University Street Station will be only about 8 minutes from Judkins Park while Madison BRT will be more like 15 minutes between 23rd Avenue and Third Avenue and riders will still be a block away from a Link entrance.

        Planning for the C.D. transit service is a complex task, with all sorts of issues. I think Judkins Park Station will however be more significant from Yesler South. Keep in mind that Link will also serve destinations in two directions while Madison BRT only will serve one, which is more inefficient.

        Regardless, I hope Metro will be innovative enough to roll out more than one to two alternatives at a time in a restructuring. A “like it or hate it” public process approach is a terrible way to present tradeoffs.

        Finally, I really think a final long-term restructuring needs to be decided to better inform where to put physical improvements. Wires, stops, shelters and (gasp!) turn-arounds take advance planning.

      2. Judkins Park station is more about connecting Rainier Valley to the Eastside. It will be more frequent and reliable and less overcrowded than the 550, and it will have entrances on both Rainier and 23rd, whereas the current flyer stop has only entrances on Rainier and is a derelict and depressing place to wait for a bus with nothing but concrete and cars around you telling you, “Get a car.”

        For going from east Seattle to downtown it may be less transfer walking than Mt Baker station, and it will bypass three intervening stations. Plus it will serve the 48. Although it’s unclear what the bus configuration of Rainier will be. Currently the 7 goes from Columbia City on Rainier to the station, but if the 7/48 route holds then it will turn on 23rd to the station, or about the same thing (and more pleasant because it’s next to a park rather than a freeway entrance). The 7 would be just north of Mt Baker, and may not be as popular to get to the station because it won’t go to Columbia City, Hillman City, and Rainier Beach on the other end.

      3. It really isn’t about how many people will use Judkins Park Station, it is about how the bus system changes in relation to it. I just don’t see anything changing, except maybe the 4 gets altered to serve 23rd. I don’t see any changes that would save service hours. We certainly aren’t going to replace any routes and it is highly unlikely that you would have truncations there. There are very few people there (it isn’t a real destination) and it sits between a major destination (downtown) and a major transit confluence (Mount Baker*). With both destinations fairly close, it makes sense to just keep going.

        In contrast, I’ve already outlined the changes that will likely happen with Madison BRT. From a savings standpoint, you have the 43 and 12. The 12 could just be eliminated, with service put elsewhere, or we create a brand new line, that helps strengthen the grid. Even if we keep it roughly the same, but tweak it a bit (e. g. run it over to James as proposed) it probably won’t run that often, which means we have some extra service hours to work with. The 11 changes as I said, which gives a different community (one not directly served by Madison BRT) a much better transit system. Even if we kept the system more or less the same, the extra savings could boost frequency along the 27 to provide needed service. I could easily see a truncated version of the 27 (just as there is a truncated version of the 3). But with all of the other changes (to the 11, 12 and 43) along with the extra service to work with, it would make sense to go bigger, and rethink the way transit functions in the area.

        * Not only does Mount Baker allow you to connect to the other Link line, but it connects you to major bus routes. Even if Link didn’t exist it would make sense as a termination point, just because of the way the street grid converges there. The 8, 48 and 7 all meet there.
        Judkins Park is the opposite — you can’t run buses east-west, and north-south ones should just keep going.

      4. It’s not so much about actual routes changing (although the 8 will likely be restructured and its replacement moved to serve JPS) as the relative usefulness of routes.

        Even with potential transfer penalties, I’m going to prefer Link if I’m anywhere east of 19th and south of about Marion, especially at peak. Let’s take a trip from Garfield High to Westlake at PM peak as an example.

        On route 3/4 directly, it goes like this:

        4 min expected wait
        25 min ride + any additional James St delay
        3 min walk from 3/Pike to Westlake
        total trip 32 minutes + high risk of additional delay

        On 48 + Link it goes like this:
        5 min expected wait for bus
        5 min ride
        1 min walk into Link station
        3 min expected wait for train
        10 min ride
        1 min to exit station
        total trip 25 minutes with minimal risk of delay

        The implication of this is that quite a few people now using slow service such as the 3/4, 7, or 14 will start using 8/48 + Link instead.

      5. @Mike — Yeah, if things go as planned:, then it will change transportation for folks who live close to Rainier Avenue. Right now, people take the 7, and basically stay on it, until they get downtown (or to other places along the way). People aren’t transferring to Mount Baker Station because it isn’t worth it. But with this new route, if they want to get downtown, they won’t have a choice. They will either have to transfer to the new, truncated 7 ( or they will transfer to Link. Compared to MBS, Judkins Park is a faster and more pleasant transfer (and the train ride to downtown is faster). Transferring to the new truncated 7 is an option, but it is hard to see how that would save you any time, even if they make it extremely fast and the transfer is easier. East Link operates as an express, with no stops between Judkins Park and I. D., and the transfer won’t be nearly as bad as Mount Baker or the UW. I see a lot of people using Judkins Park Station.

      6. A few comments:

        I’m glad to read that you guys understand how Judkins Park Station will be important for Metro+Link trips. It already is projected to get more riders by 2030 than Mount Baker has today and perhaps by 2030. That’s even without Metro restructuring.

        Mount Baker Transit Center is also horrible for east-west bus service. McClellan? Really? A transit center at Columbia City would have been better for east-west service.

        I expect a significant interest from Leschi, Mount Baker perhaps southern Madrona residents to use Judkins Park Station. Routes 4, 8, 27 and 14 are all candidates to be rerouted but each route has some issues about doing that. Perhaps the Metro vision to extend Route 27 to McClellan to Mount Baker could continue to Judkins Park. Perhaps then the Route 14 Hunter Blvd tail could be flipped to instead serve the areas south of I90 currently served by Route 4 ( a few blocks of new overhead wire), allowing Route 4 to also end at Judkins Park station.

        I also expect a latent demand to get to Judkins Park from Cherry Hill, First Hill and Yesler Terrace to develop. Route 7 will pick up some of those riders but reconfiguring Route 4 to run by the station would be serve so many more!

        Anyway, if any route ends at Judkins Park, turn-around capabilities will be needed. Frankly I think drop-off demand circulation issues warrant some sort of turn-around. A bus route terminating there would make the need more pronounced.

        I see two good options on how to do it. A simple one-way “crescent” across 23rd from the station entrance on the lid is one; it can be named Jimi Hendrix Plaza and be special. A second would be to straighten the loop ramp off of westbound I90 west of Rainier to intersect with Rainier at more of. 90-degree angle (making the west side of a Rainier crosswalk much safer and potentially signalized) and using the newly opened space (including the unutilized green space in the loop today) for a bus-only turn-around.

        As much fun as it is to draw reroutes on a map, it should be decided through a larger discussion with the many interests involved. That’s why I recommend that the restructuring discussion should start soon rather than wait until 2021 or later.

      7. >> It’s not so much about actual routes changing

        But that is the focus of this thread. Al and I both think Judkins Park will get lots of riders, it is just that I think there will be very few bus route changes because of it, while he thinks there will be.

        I think it will be similar to CHS. We all know that Capitol Hill Station is great. But it didn’t bring in dramatic changes in bus routes to the area. Once people started weighing the different options, nothing seemed clearly much better than the old set of lines. So basically, they nibbled around the edges (and it isn’t even clear if those changes were worth it).

        Judkins Park is like that. It will be very popular. But that doesn’t mean that folks will embrace significant routing changes as a result of its addition.

        But Madison BRT could create a cascade of welcome changes. The 11 no longer makes sense as currently run. That is a big change, and a huge change for the better. It is highly likely that the 43 will be axed, and it is quite possible the 12 will be too. Those are all significant changes in the system, and if they do things right, could result in a very nice network and grid in eastern Seattle.

        I just don’t see that with Judkins Park. Al’s ideas for ways to take advantage of the station are intriguing, but they actually run counter to David’s, which shows how controversial they will be. Axe the 4, or move it to 23rd? I’m actually with David. If we do anything with the 4, I think we should just kill it. If we moved it to 23rd, it would be a better run, but it still isn’t great. Cherry Hill is a decent destination, but you can get there from both ends. If you are coming from the north on Link (e. g. Northgate) you would get off the train sooner (at Pioneer Square) and take the 3 or 4 (more frequent than just the 4) headed the other direction. If you are approaching it from the new Corridor 48/7S ( then you just stay on that bus until Garfield. So the only real time savings are if you are coming from Bellevue or riding Link northbound (from, say, SeaTac). That is an improvement, but given the cost in extra wire and extra service, I just don’t see it as being worth it.

        The latter is really the problem I have with bus routes like this. Sure, it is great if we have more buses like this, but it really stretches us too thin. We end up with certain trips that work out great, but others that are terrible, because we don’t have the frequency to make the transfers painless. If this runs every half hour, how many people will actually take it? Even though it would be faster for trips from east or south on Link, a lot of people would just stay on the train, and make the transfer downtown, because there would be less waiting.

        There is a decent argument for buses like this simply because the combination of trains and bus routes prevent an effective grid. But in my opinion, this should be addressed head on, with as much of a grid as possible. So, for example, how about replacing the 4 and 12 with this:

        Now you can get to Cherry Hill from every direction on Link. If you come from the south, you transfer at MBS. From the north at CHS and the east at Judkins. Madison to Cherry Hill becomes a lot better. Local trips up 14th aren’t ideal (because it makes a detour to CHS) but still much better than what exists now. It isn’t a perfect grid by any means (because of the detour to serve CHS) but it is still providing something our system doesn’t provide now. For that reason, I see it as a 15 minute bus.

        Anyway, if any route ends at Judkins Park

        I could see a route bend to serve Judkins Park, but I can’t see any route ending there. For example, let’s say we run a different version of the 27, one without the eastern tail. You have three choices:

        1) Just end this short run at MLK.

        2) Turn at 23rd, and end at Judkins Park.

        3) Turn at 23rd, and end at Mount Baker Station.

        I just don’t see option number 2 making sense. The first option is short and sweet (no turns — you just find a turnaround spot in the neighborhood). It is an efficient grid, although it doesn’t connect well with Link. The third connects with both the southern and eastern end of Link. The second seems like a weird compromise — connecting with the eastern end of Link, but not the southern end. It is roughly the same distance from Jackson to Judkins Park as it is from Judkins Park to MBS, so it just doesn’t make sense to me. Either extend it all the way, or end at MLK.

  8. Tempting to keep trolley service on James- really shortest route to Harborview and best transit up a very steep hill. But pretty sure I remember driving Routes 3 and 4 up Seneca from Third to Virginia Mason, turning south on 9th to Harborview, and then east on present routes.

    Couldn’t we accomplish same thing using University Street Station- major trolley stop with easy DSTT access for same purpose as Pioneer Square Station serves now? If memory serves, there’s only one stop between I-5 and the hospital. What’s everybody think?


    1. You’re right. That’s the way the Harborview buses went until James was wired in the late-’70’s rebuild.

    2. Lots and lots of riders use 3/4 to get to Harborview and/or Cherry Hill from services in the Pioneer Square area. At many times of day you’ll find 20-30 people waiting at 3rd and James EB. The Yesler routing wouldn’t negatively affect those people, but using Spring for all service would put a hurt on them.

      1. Can’t these people ride the Tunnel to Benaroya Hall and get their 3 and 4 buses there, if routed south from Virginia Mason on 9th? And for Cherry Hill, egular routes from Harborview.

        But avoiding both blocking along James, and waiting in the cold. Benaroya stop very sheltered, can sometimes wait inside the concert hall. But of course we’d want to time the new route first.


  9. Not surprised that First Hill Streetcar hasn’t been mentioned today. But main obstacles (literally) are political problems, not civil engineering. Some lane-rearrangement and priority-reverse- streetcars over cars this time- might make it credible service from Pioneer Square and IDS through the Yesler Terrace neighborhood and Swedish.


  10. Route 27 lost its through route partner in fall 2016; Route 33 is hooked with Route 124.

    The First Hill improvement point about east-west route spacing is very weak, as routes 3 and 4 would return to Jefferson Street east of 9th Avenue.

    Note that Route 60 now has 15-minute headway service.

    Jefferson Terrace is well-served by the stops on 9th Avenue; it has elevators; it does not need the stops on James at 8th Avenue for access.

    Route 27 would not be helped by going through congestion. A slow unreliable service does not help the network or its riders.

    Note that the survey did not draw respondents well from QA and Belltown riders who are impacted by poor reliability and bunching.

    Yesler Terrace is growing.

    The proposal has majority support from the survey and is consistent with the service guidelines. It is in the Seattle TMP, Move Seattle, and the AWV RapidTrolley scenarios.

  11. Not making comment on moving 3/4 itself, but the statement of “Yesler Terrace has no connection to downtown”… Once the Central City Connector gets built, will that not count? Yesler has a Central streetcar stop, which gets you to Jackson and 1st. Once the new line is in, that’ll get you all the way around to Westlake.

  12. Jacob,
    Route 27 connects Yesler Terraace and downtown. The connection provided by the First Hill and CCC streetcars is very indirect. In general, riders do not like slow and indirect service. The revised routes 3 and 4 would provide a much better connection. Yesler Terrace is growing.

Comments are closed.