King County Metro 28 in Broadview
King County Metro 28 in Broadview

Perched on a ridge north of beautiful Carkeek Park and west of Bitter Lake, Broadview enjoys a broad view of Puget Sound — sort of like living on Sounder North — and shares the quiet, leafy feel and chip sealed, sidewalk-less roads that characterize much of north Seattle past 85th St. Homes here are exclusively single-family, and tend to be on lots a little larger and more widely set-back than streetcar suburbs, but this neighborhood feels similar to walk through, possessing a mostly-intact street grid and houses in a jumble of styles and ages and states of repair.

Broadview is served by two Metro routes: 5 and 28, and the express variants 355 and 28X. Route 5 currently operates on Aurora and Phinney/Greenwood before splitting, with half of the trips continuing straight up Greenwood Ave to serve Broadview, Bitter Lake and Shoreline, terminating at Shoreline Community College; the other half continue to Northgate via Holman Rd . Route 28 currently serves Dexter and Fremont, then jogs west to 8th Ave NW before heading up to Whittier Heights, Crown Hill and Broadview, terminating on 145th St near Aurora.

Notably, due to the absence of suitable roads, Route 28 operates three blocks west of Greenwood Ave, on 3rd Ave NW, for most of its alignment north of Crown Hill, even though no intervening terrain feature acts as a barrier. I’m not sure when this alignment was created, although it goes back at least to the 80s and streetcars existed south of 85th St on both 8th Ave NW and 3rd Ave NW.

Long-time readers can probably guess where I’m heading with this, after the jump.

Let’s look now at the ridership charts for Route 5:

Ridership Chart for King County Metro 5
Ridership Chart for King County Metro 5

This chart requires some caution to interpret, because of the distortive effects of the much-less-used Northgate tail, which joins the Shoreline tail at Holman. I have omitted data entirely for the Northgate tail, but it is not possible to separate out Northgate trip data out from Shoreline tail data south of that point. This means, for instance, that the big drop in midday loads at Holman is not “real”, i.e. does not indicate that half the people on the bus got off there. Similarly, even though there are more boardings per trip in Bitter Lake/Broadview than in Greenwood, there are actually less riders served there over the course of a day than in Greenwood due to the half-frequency.

That said, here’s what I see in the data:

  • Route 5 (Shoreline) performs very well at all times, with only evening and night service to downtown carrying relatively light loads.
  • Route 5 (Northgate) not so much, probably due to superior competing service from Route 75. I have omitted the data so you’ll have to take my word for it, but it’s pretty weak.
  • Most stops show a steady churn of riders, particularly those at transfer points on 46th, 85th, and Holman; this precisely parallels Route 358. The only stops that are little used are those between 145th and Shoreline CC, plus those on Aurora south of the ship canal.
  • Shoreline CC is a major ridership driver, particularly in the midday, surprisingly strong for an institution at a distance from major urban population centers. Without Shoreline CC, ridership in Shoreline would almost completely vanish outside the peaks, whereas with the college, only night ridership flatlines.
  • Route 5 is in dire need of stop reduction, with stops often two blocks apart in Greenwood, and not much further apart to the north. Between weaving around in Greenwood and constant stops, riding the 5 any significant distance can feel interminable. With stops that close, eliminating some will lose no riders and will probably gain a significant number due to improved speed and reliability.

And for Route 28:

Ridership Chart for King County Metro 28
Ridership Chart for King County Metro 28

There is a more subtle distortion at work on this chart. Some of the morning and evening trips for Route 28 begin or end at 85th St causing the apparent plunge in morning peak ridership there. The same artifact isn’t observed in the evening, as riders presumably read the schedule or headsign before boarding, rather than engaging in a surprised mass exodus at 85th St. Note that this chart has a different scale to that of Route 5.

  • Great performance to Fremont. Stops on Dexter are well used, although riders seem to be almost entirely headed downtown. The stops on 34th are extremely busy, and show significant turnover, which makes sense giving the entertainment, residential and employment land uses in that area.
  • Good performance to Crown Hill. The bus is mostly unloading, but some riders are using the 28 to travel to the commercial district at 100th St.
  • Poor performance in Broadview. Past 100th St, stop usage is minuscule and the bus is almost empty outside of the peaks, in stark contrast to Route 5, which carries good loads all the way to to Shoreline except late at night. This illustrates the importance of terminating routes at ridership centers if at all possible.

The initial proposal for the Fall 2012 restructure suggested deleting Route 28 north of 100th St off-peak, and switching those Route 5 trips currently going to Northgate to Shoreline. Based on the closeness of the routes, and the ridership data I’ve presented here, it should be fairly obvious where this idea comes from. Deleting the 28 off-peak is unlikely to seriously impact many riders, due to the modest walking distance to Route 5, and extending all 5 trips to Shoreline will focus service where it is most cost effective and likely to garner additional riders, while making schedules simpler for riders and more efficient for Metro.

25 Replies to “Streamlining Service in Broadview”

  1. This is the neighborhood of my youth, having lived there while in High School. I can say from personal experience that the 28 got poor ridership because it was a north south route and it didn’t take people where they most wanted to go which was to points east e.g. to Aurora, Lake City, Northgate. Not to mention infrequent and unreliable headways. Combine that with significant grade hills and the bus route was really competing with the car, which Broadview residents (west of 3rd Ave) generally have no trouble affording. While I’m a bit nostalgic, I can say Broadview residents can generally readily make other choices.

      1. My tenure in that neighborhood was “a few decades ago” but, anything that would allow people to get directly east to Greenwood, Aurora, and Northgate. If connections were frequent e.g. 20 minutes and could reliably meet up with north/south lines on either Greenwood or Aurora, then getting to stores, schools and restaurants by transit could be come viable. The logical east/west streets would be 145th or 125th with a jog over to 115th by the cemetery to get to Northgate. It would make sense to connect people with 358 or the 5 to go downtown instead of a 1 seat ride on a 28 to go south.

        The reality is that the area isn’t dense enough to warrant frequent service and it is affluent enough to have have access to personal vehicles. It is an area that will not see any meaningful densificaton in our life times (remoteness, and land covenants). It is also an area whose terrain is steep enough that the average person is simply not going to bother. The stop consolidation that was done on this route earlier this year probably contributed to waning ridership because at some point when you’re already making a trek of several streets up a steep hill, having to go an additional 1000 feet to get to a bus stop for 30+ minute headways isn’t worth it.

        Of the nearly 9,000 households in the 98177 zipcode, the portion that lies west of 3rd avenue presents a problem for attracting riders to bus service because of terrain and density and the orientation of bus service being North South versus demand for east/west service. BTW, this is a similar problem I’m experiencing in my present living situation in Seward Park not having frequent east/west connections to MLK.

        This could also be a case to be made that favorable parking solutions near Greenwood or Aurora that then connected with frequent service to Northgate could be the solution to at least cutting miles traveled for Broadview residents even if it didn’t eliminate the need for them.

  2. The other part of the Fall 2012 restructure for route 28 also calls for eliminating the Local variant entirely in favor of an all-day bidirectional 28 Express that would skip the heart of Fremont and Dexter Ave, instead cutting across N 39th St and taking Aurora to downtown, with that service replaced by putting the 5 onto Dexter instead. It would be interesting to see what Metro’s ideas are for stop reduction along the express leg of the route (presumably retain most of the existing stops north of the ship canal and mirror the existing 28X stop pattern on Aurora?) Although it will likely be slower than the existing 28X routing, by skipping the major slow point in downtown Fremont that could be an interesting change since it should presumably speed up the trip for everyone north of Fremont who currently uses the 28 local to get to/from downtown, but conversely it is going to probably slow down the 5 quite a bit from Fremont south, so there might be some migration of riders between the two routes, or from the 5 to the 358/RR E.

  3. Why isn’t it considered to have the 28 route go all the way to Shorline CC to increase north end ridership, and not terminate around the 145th and Greenwood Ave area?

    The 28 is a bit faster than the 5 and the 28X is much faster than both. As a resident who lives at the bottom of the hill on 12th Ave NW, losing the 28 stops on 8th Ave NW would increase commute time immensely by making riders walk farther for the 5, adding an extra 10-15 min, and then adding another 10-20 minutes of bus ride time. Not to mention, the 5 never seems to be as reliably on time as the 28.

    1. I think the problem with the first suggestion is that there isn’t sufficient additional pent up demand at Shoreline CC to justify another route. You would probably see some increased use of the 28 as existing 5 users might switch depending on which bus came first, but I doubt you’d see enough increased usage to equal the added platform hours.

      The reason the 28 is probably much more reliably on time than the 5 in Broadview is that it starts its route much closer to where you get on, and with the exception of two stops (Linden/143 and Greenwood/145) there are virtually no riders getting on and no signals or traffic to wait through, so if the bus leaves on time it’s all-but guaranteed to be at each stop on time until Holman.

      Unfortunately you seem to be the exception rather than the rule for your neighborhood; based on the graphs there is almost no one using the route past Holman who couldn’t be easily served by another nearby route, with the exception of the (three?) boardings taking place on the upper segment of 8th NW. The Linden/143 boardings are one block from Aurora and the 358, while the Greenwood/145 boardings are at the same intersection as the 5. If people are going cross-town there are good E/W connections there for nearly everyone. It sucks to be the odd duck out, but living on that close to the waterfront without any attraction to drive demand for travel to the location is basically asking to be isolated from convenient transportation connections.

      It would be great if Metro could find an E/W connecting route with less frequency that could cut over to 8th Ave NW in that area to provide a link for the few bus commuters in the neighborhood, but the only current route that comes remotely nearby is the 345, and that already has a number of borderline ridiculous diversions.

      1. That could be interesting if it terminated on 8th NW by doing a loop via 3rd NW/NW 132nd/8th NW/NW 125th, thereby serving that little cut-over. I’d worry that you lose a lot of ridership by skipping Northgate, but it would provide one consistent E/W corridor instead of keeping it disjointed like now with the 41/345 and the big gap around I-5.

        In an ideal world you’d serve Northgate Link station in both directions with a route like this, but the topography would make that difficult (maybe via Meridian/Corliss/1st Ave NE to connect between Northgate and 130th?) Of course, that’s far enough away that other solutions might arise between now and then, such as finalizing the North Link extension, or just making the transfer to RR E on Aurora if that service works well.

      2. Putting the 75 on that route cuts everyone on Sand Point Way off from service to Northgate. This used to be the tail of the 41 and gave a one-seat ride to downtown, which was supplanted by the 75 when the 41 was truncated at Northgate/Lake City. The route did not need a one-seat ride to downtown, but the service seems to get decent ridership to Northgate where the 41 (and, in a few years, Link) provides extremely frequent service downtown–frequent enough where the arrival time of the 75 is not relevant as another 41 will be along shortly.

        If the 522 was as frequent as the 41 (and placed in the tunnel), then perhaps a transfer at Lake City would be more palatable as it is quicker should you get lucky with the transfer…but that seems to be rare and most riders seem to just stay on to Northgate and hop the 41, and this behavior would only increase once Link is available there (unless there is a station at N 130th, which there really should be). The transfers in general at LCW/125th are poor and the shelters the same you’d see at any minor stop that warrants one, rather than what you’d expect at a location with several crossing routes including a major one. Improving the transfer experience at this hub with bigger, well-lit structures, signage and–hopefully, someday–Orca vending machines would possibly draw enough people off the 75 to make this change possible. You still have the issue of getting people from NE Seattle to their major shopping area at Northgate.

        NE Seattle has horrible bus service south of 125th and north of 65th. Both Sand Point and 35th NE used to have direct downtown service all day, whereas neither do now (except for a few “expresses” on 35th during peak hours). The frequency of the 75 with its service to a major center ameliorates that a bit, but until Lake City gets upzoned and more frequent 522 service–or, better yet, a high-capacity, high-frequency route–routing the 75 away from Northgate is a bad idea.

      3. I’m going to have to agree with ScottS on this one. Not only is the 75 important for Sand Point->Northgate, but it’s also an important East->West route on Northgate Way, both east and west of Northgate.

        The 345 route, however, does seem a bit questionable. Geographically, I can’t think of a trip you could take on it longer than 1/2 mile for which there isn’t another route that can do the job faster. What if we killed the 345, replacing it with an East->West route on 145th St. This could provide important connections between Shoreline, the 510/511/512 buses to go downtown, and Lake City. The route I’m envisioning is this:,-122.326241&spn=0.043408,0.111494&sll=47.729972,-122.31637&sspn=0.043414,0.111494&geocode=FRCI2AIdKgC1-A%3BFUki2AIdiOu1-A&vpsrc=6&mra=dme&mrsp=1&sz=14&t=h&z=14. Google estimates 13 minutes nonstop drive time end to end, so assuming a 1 hour round trip cycle, including layover time, we could get 30 minute headways with 2 buses, or 15 minute headways with 4 buses.

      4. What I was thinking was sending the 28 to Northgate and maybe Lake City on the 75’s route once it hits Crown Hill, and send the 522 and/or 372 to Northgate once Link opens there.

        Not a big fan of your 145th route – two golf courses kill its walkshed – were it not for the freeway station at 145th the current route of the 330 would work at least as well. Even discounting the problems with the I-5 alignment itself, I’m not a fan of putting a station on 145th on that alignment that would lose a quarter of its walkshed unless the Jackson Park golf course gets redeveloped. I’m not sure why it’s so important that the 510/511 stop there outside peak hours, other than that it’s there – there’s not even a park-and-ride there, and it’s not like it has much of any connecting service right now.

      5. There are also riders wanting to get from Sand Point to Aurora in less than the 45 minutes it currently takes.

        I tried to explain to a colleague how to get from Sand Point to 160th & Aurora or Shoreline CC on a Saturday afternoon (for two different events that happen to be near each other), and it would take an hour or 1:15 to take the 75 and transfer to the 358 or 5. So then I looked at going the other way, transferring to the 44 to the 358, or going all the way to Fremont and walking the span between the 30 and the 5, and none of those helped at all. I told him the time for taking the 75 and 358 and he just looked at me and said, “I’ll probably rent a car.” Not just because of the time, but because it sounded complicated and he’s new to the area.

      6. ScottS,

        Imagine consolidating the 522, 372, and 72 into a BRT line serving 25th Ave NE between Lake City Way-ish and UW Station starting in 2016, except keeping some peak-hour service directly to downtown, then back. Maybe the 306, 307, 308, and 312 could have increased runs in order to provide the peak service, including at least some in the reverse-peak direction.

      7. @Morgan:

        While the golf course does cut some of the walkshed of a 145th route, there are still a fair number of homes it does serve. It would also provide access to Lakeside High School, which currently has virtually no east/west transit access.

        The I-5 transfer point is also important because someone from shoreline who wants to get to downtown can get there much faster along I-5 than along Aurora with stops every 1/4 mile. Even with the transfer, 510 and 511 have a frequent enough combined frequency so the wait wouldn’t be too bad. Of course, to keep the wait time reasonable for the outbound trip, the 145th St. bus would also have to run frequently because, with traffic, that connection would be almost impossible to time.

        This is even more so for traveling between Shoreline and the U-district. With 510/511, it’s a straight shot down I-5. Without 510/511, you get to choose between 358->44, 358->48, or 346->66, all of which really suck. The 373 is a little better, except its single direction peak only. I used to ride it a bit and it was always overcrowded.

        It always seemed stupid that we have express buses going up and down I-5, but the local service is oriented to get you to take milk runs all the way downtown, or transfer to the 41 at Northgate, rather than take a much shorter milk run, transferring to a fast 510/511 at 145th St.

        The most logical reason I can think of for this is institutional inertia – legacy routes leftover from before the 510 and 511 existed, or Metro thinking only of the KC metro system, without thinking of Metro and Sound Transit as one regional system.

    2. Considering the restructuring proposal for the 28 and the 5, truncating the 5 at Northgate Way and extending the 28 to cover the 5’s Northgate Way to Shoreline CC routing. The 28, travelling on the Aurora bridge, will be a much faster and more reliable route for riders north of Northgate Way. Riders heading to Greenwood or Fremont could transfer from 28 to 5 at Greenwood/Northgate Way.

  4. I wonder how much the zone boundary is an impediment to ridership.

    Right now, pay-after-you-shove-to-the-exit enables riders to avoid the zone-fumbling problem, as long as they get off by 145th. When the Ride Free Area is gone, a whole lot of e-purse users will either be getting charged two zones during peak, even though most of the riders get off before Shoreline, or Metro will simply give up on trying to charge any two-zone fares on this route.

    I suspect the additional revenue from charging two-zone fares to those who actually cross the boundary is miniscule compared to the cost to the system to have to change the ORCA reader setting for each rider who lets the operator know they are only riding one zone. Moreover, charging the right fare is logistically impossible anyway, without a tap-on-tap-off system.

    Want a cheap way to save service hours and improve ridership? Make the fare on the 5 “local” (i.e. one zone) 24/7.

    1. Two-zone routes seem to be breaking down anyway in with ORCA. I rode the 271 last week from UW to Bellevue and the reader said “2 zones” but when I tapped it just said “PASS”” rather than “PASS + 25c”. Conversely I went UW to Lynnwood on a CT UW-Marysville bus and I was thinking of ST zones instead of CT zones, so I ended up paying twice as much as I should have. And on other Metro routes the driver hasn’t bothered setting the reader to 2 zones even when most riders were going to pass the zone boundary. So there’s both overcharging and undercharging all over the place.


  5. I’m one of those riding the 5 north to 130th at night. Depending on when I get off work, I’m catching the 5 at Macys either at 11pm or 11:30pm. Either one, there’s a good number–15-25–that go north and are still on by the time we get to Holman Road. I hope that Metro doesn’t screw with the timing of the route, or get rid of these later runs or I would just have to drive to work each day. When my schedule was more ‘day-ish’, yes, going along Greenwood seemed to take forever with all those stops, but the nighttime run just over 25 minutes, giving me time to relax after my day of work. The 358 is not a pleasant bus to ride and I wouldn’t feel comfortable walking from Aurora to Greenwood at midnight. The 26/28 route is a better walk, but the 5 drops me off really close to my home.

    1. yes, in the zeal to maximize efficiency of bus routes, the significant percentage of people who work other than M/F 9-5 and in deed, in places other than the Central Business District is often overlooked. The bias towards north south routing in this town is maddening.

      1. I don’t think people who want to maximize efficiency are the people responsible for today’s route structure. Quite on the contrary, in West Seattle in particular, this restructure switches from an almost entirely radial network to a network with much better east-west connectivity. Getting people where they want to go today rather than where they wanted to go in 1890 is actually kind of the point.

        That said, downtown is a massive ridership center, and there will always be a crapload of people wanting to go there, and, therefore, lots of bus service to take them.

Comments are closed.