Planned transit hub connecting Stride BRT routes in Bothell (image: Sound Transit)

A transit hub northwest of a rebuilt SR 522 and I-405 interchange will serve as the eastern terminus for Sound Transit’s planned SR 522 BRT. It will connect Stride BRT on SR 522 to Stride BRT on I-405. Sound Transit is dropping plans for some BRT buses to serve Woodinville, and replacing those with a Sound Transit Express branded connection.

These, and other updates to SR 522 BRT plans, are shared in a Sound Transit online open house that runs through August 23.

With expanded express toll lanes on I-405 between SR 522 and SR 527, the Stride BRT on I-405 will be accessible at SR 522 from BRT platforms on new direct access ramps. The SR 522 BRT, meanwhile, will terminate at a bus loop and layover area just to the northwest. That makes for a straightforward connection between the two services and to other buses operating in the area.

The transit hub resolves several other tricky issues in Bothell that Martin explored last year.

SR 522 Stride will route through downtown Bothell with stops at UW Bothell/Cascadia College. This routing improves transit access in growing areas of downtown Bothell. I-405 Stride will still miss downtown Bothell, but the latest plans mean it is one frequent connection away via 10-minute service on SR 522 Stride. Alternately, sidewalks from the transit hub will connect to the North Creek trail and Campus Way, so walking access to the south end of campus will be more direct.

The revised SR 522 BRT route in Bothell serves downtown and UW Bothell while connecting to I-405 BRT every ten minutes (image: Sound Transit)

There are revisions to the BRT plan for Woodinville too. The ST3 representative project envisioned half of BRT service terminating to UW Bothell, and the other half continuing on to Woodinville. The last 3.5 miles nearest Woodinville were not to see any capital improvements and the BRT would operate in general traffic, potentially reducing reliability along the entire route. Ridership estimates were very low, with recent estimates as low as 100 daily boardings in Woodinville in 2042.

The solution for Woodinville (as briefed to the Woodinville Council in April) is an ST Express branded service that appears better matched to riders’ needs. At peak, it would operate between Woodinville and the Bellevue transit center, making several stops in Woodinville and connecting to the BRT lines at the Bothell hub. At off-peak and weekends, it would operate only between Woodinville and the Bothell transit hub. At comparable operational expense to the original BRT, this would be a rather speedy direct connection to Bellevue and to East Link. The connector would operate at the same 20 minute headways as the representative ST3 project, peak and off-peak. It would be 10 minutes faster to Seattle (50 minutes + 1 transfer, vs 60 minutes + 1 transfer).

This WSDOT graphic shows the highway configuration with the added direct access to express toll lanes on I-405. The Bothell transit hub, not shown here, would be to the upper left from the intersection & within the loop of the exit ramp (image: WSDOT)

The open house describes several more revisions elsewhere. Most notable are added stations at 61st Avenue NE in Kenmore and NE 165th Street in Lake Forest Park, and the deletion of a station at 25th Avenue NE in Shoreline/Seattle. There are street changes including new roundabouts in Shoreline. There are other other design changes of more local significance.

The SR 522 BRT and north I-405 BRT are currently scheduled to open in 2025. That may be delayed further as Sound Transit ‘realigns’ the ST3 plan after COVID reduced future revenue forecasts.

81 Replies to “Bothell transit hub will connect SR 522 & I-405 BRT”

  1. Overall, these changes look positive, except for those who want to get to Woodinville during the off-peak hours.

    I didn’t see how frequent the Woodinville shuttle is expected to run, but if it gets the ridership I’m expecting, I’m having a hard time seeing it running more often than every 30 minutes. That’s a lot of waiting and, quite likely, worse, than today’s ST 522, which also runs every 30 minutes, but at least, gets you all the way downtown, without multiple transfers. With three routes to connect to, even a timed connection isn’t possible.

    In practice, I’m expecting the Woodinville shuttle to only get really used during the weekday midday periods, after the parking further along the line is filled up. On weekends, people won’t bother with the transfer – they’ll just drive the entire 522 BRT route in their car, and park at the Link Station directly.

    At least there’s a trail connection, though. A bike ride from downtown Woodinville to the planned transit hub via the Sammamish River Trail/North Creek Trail could be very fast. Hopefully, there will be bike parking and lockers when you get there.

    1. I didn’t see how frequent the Woodinville shuttle is expected to run, but if it gets the ridership I’m expecting, I’m having a hard time seeing it running more often than every 30 minutes.

      I don’t see that at all. I see the shuttle as an implicit trade between a one-seat connection and frequency. In effect, ST is promising to give Woodinville frequent service to the hub indefinitely, even though it means a transfer. I would expect off-peak service somewhere in the 10 to 15 minute range. (Inbound it would be not be times, although an outbound bus might wait a minute or two if there was an approaching bus).

      Of course ST could always cut the frequency in the future, but that is true of any bus route, or the trains for that matter.

      1. I wouldn’t trust that implicit trade. They are going to blow a ton of money to have a new bus to Bellevue during peak. I wouldn’t think they would spend even more money to keep the shorter connector bus frequent off-peak unless I was told. I have to assume that it’s going to be infrequent because 1. it was always going to be infrequent, and 2. they will have even less money to work with off-peak than they did before.

      2. The commitment at this time is the Woodinville bus runs every 15 20 minutes at peak (with continuing service to Bellevue), and every 20 minutes off peak. So that’s the same off-peak frequency they would have gotten with the 50% BRT experience, though with an added transfer perhaps depending on how one got to Bothell. See the deck shared with the Woodinville Council for details.

        update, 20 minutes at peak too.

      3. Thanks Dan. Yeah, that’s not very good, and I don’t think it will be very popular. I think ending at the transit hub is a bad idea. The hub is nowhere. It saves money, but it would be much better if it continued to the UW Bothell campus. The campus is not very far away, but for a rider, it avoids a half mile walk to one of the bigger destinations in the area. UW Bothell will continue to be a hub, unless the plan is to send all of those buses to the other hub. Doing that would be a lot more expensive, and in some cases gain you very little. The 239, for example, goes to Brickyard, so there is nothing to be gained by looping around to serve the hub, other than making this connection. That connection is better made via this Woodinville bus, since it would enable a one seat-ride to UW Bothell.

        I have no general objection to terminating the 522 Stride at this transit hub. There is nothing much to the east. But whenever possible, a transit hub should be an actual destination, not in the middle of the freeway. Buses serving this hub from the east should continue and at the very least serve the campus.

      4. “nothing to the east” – what about all of those jobs immediately east of 405? Not dense enough to serve easily, but it’s one of the larger secondary job centers on the east side.

      5. The Woodinville travel patterns on page 24 is interesting. Commutes from Woodinville on all transportation modes are: 28% Redmond, 22% Seattle, 13% Bellevue, 11% Everett, 7% Kirkland, 6% Bothell, 3% Woodinville (people who both live and work in Woodinville), 3% Renton, 8% Other.

        It’s surprising Woodinville is so Seattle-oriented. You’d think Seattle workers would live in closer cities like Kenmore, Bothell, Bellevue, or down around Factoria; and Woodinville would be almost completely geared toward the Eastside.

        I’d also like to know what the off-peak travel patterns from Woodinville are. That would help inform what kinds of bus routes it should have.

      6. RossB

        I think ending at the transit hub is a bad idea. The hub is nowhere. It saves money, but it would be much better if it continued to the UW Bothell campus.

        Yes, UW Bothell is the biggest transit likely population. Not just students commuting to classes (which are all day and continue into the evening for working students and continuing ed) but currently in the design stage is a large expansion of student housing and other amenities to turn this into a real college campus.

        Moving the “transit hub” to Beardslee has a number of advantages. For starters it eliminates the at grade traffic light crossing of 522. There’s so much wrong with this I should write a page 2 post. Second it serves the business park; not that big a deal since everyone working there has free parking and it’s not likely to be a big ridership gain (there is a hotel or two close by but patrons probably rented a car). Perhaps most important is it is better suited to include a loop between DT Bothell and DT Woodinville. Woodinville P&R is surrounded by apartments. UW Bothell Campus has housing and more is in the pipeline. DT Bothell is becoming a dense “urban village” with all the development on the old HS/bus barn site .

        Do it right and nix the Rose Hill stupidity.

      7. @RossB – As per the article, 522 BRT plans to serve UW Bothell on the way to the transit hub. We can consider it as the layover space with the added benefit of a better transfer to 405 BRT

      8. It’s surprising Woodinville is so Seattle-oriented.

        I’m not suprised. Seattle is where most of the jobs in the region are. Besides, according to your study, Seattle jobs are still a minority for Woodinville (only 22%). It is pretty easy to see how someone would end up working in Seattle, and living in Woodinville. Maybe they bought a house there, then switched jobs to Seattle. Maybe their spouse works in Woodinville, and they work in Seattle.

        If anything, I’m a bit surprised the numbers for Bellevue are so low. Redmond leads the way, with over twice as many riders. Maybe instead of running those express buses to Bellevue in the morning, they should run express buses to Redmond. People could always transfer if they are going to downtown Bellevue (and do the opposite if they are going from Lynnwood to Redmond). I’m not sure the best way to do that, though. I’m not sure which streets have the HOV lanes. It may be that there is no fast way, which means that the express bus doesn’t save you much time over going to downtown Bellevue and taking the train. It may also be the case the jobs are too spread out, and when they aren’t, Microsoft runs a shuttle for many of them. Unless you find a fast express that can beat the train, you also wouldn’t get any to transfer. Folks going from Lynnwood to Redmond would just take the train in Bellevue. If that was the case, this would do little to deal with crowding on the 405 Stride.

      9. @KB — Oops, sorry, I wasn’t very clear. Ending the 522 BRT at the transit hub is fine.

        The issue I have is with the shuttle. A shuttle from Woodinville to the transit hub (by the freeway) is less than ideal. It should go a little bit further, and end inside the UW Bothell Campus, where there is plenty of layover space, and lots more connecting buses.

        That is a bit of an overlap, but very minor. It means Woodinville riders get a one seat ride to the campus, and two seat rides to more places.

  2. Kudos to ST for thinking about drop-off and pick-up with a turnaround in the Stride layouts. I’ve seen lots of other station layouts for not just Stride BRT but also sometimes Link that ignores or diminishes this way to get to or from transit at a number of key stations.

    I do hope it won’t be too cumbersome for drop offs. Without easy access and well-signed directions, some transit riders may dangerously hop out of cars at some of these intersections instead. There are lots of mindless people out there that do dangerous things.

  3. Interesting. This looks great.

    I’m still a little confused as to how this will work though. From what I can gather, it will work like this:

    1) The 405 bus will have a stop in the middle of the freeway, similar to the Totem Lake stop. The bus may have to stop at a traffic light, but it won’t have to deal with traffic. Riders wanting to transfer will walk a relatively short distance over to the 522 bus stop, which is fairly close to 405.

    2) Outbound, the 522 bus will go through Bothell on 185th/Beardslee, then get on the southbound on-ramp at Beardslee/195th. But rather than get on 405, it will immediately exit, on the special transit-only ramp. It will stop at the stop, and lay over.

    3) But how does the 522 bus get back? The only way out is via SR 522. It has to serve the stops on Beardslee, so it has to get back up north. The only reasonable way I see of doing that is to get on SR 522 eastbound, then get on 405 north and exit at the next stop (Beardslee/195th). It can’t take the HOT lane north, because there is no exit from the HOT lanes to Beardslee/195th. So it will be in the general purpose lanes for that segment.

    That isn’t the end of the world, but it is less than ideal, especially for the beginning of the line. For the end of the line (the bus getting to the transit hub) it doesn’t matter that much. As long as ST allows enough layover time, the bus can be delayed a few minutes getting to the terminus. But it isn’t good if a bus is a couple minutes late (because of freeway traffic on 405) before its second stop.

    I don’t see a cheap alternative though. Ideally you would build a two lane road from Beardslee to the transit hub. It would start west of the ramp, and then cross over (or under) the exit ramp of 405. That would allow buses to get between Beardslee and the transit hub without ever experiencing freeway traffic. That would cost a lot though (mostly for the overpass, but also because the existing ramps have bridges to go over the creek).

    A cheap alternative would be to do a loop. That would mean the Beardlee/195th bus stop would only be outbound (eastbound). The bus would exit the transit hub and turn right onto 522 westbound. Then it would take the next right and head through campus, like so: Of course this might be worse than the alternative. It means having stops only going one direction, which can be confusing. It also means dealing with 522 traffic (although for a about half the distance than the 405 traffic). It also means going through campus, which may be slow.

    Regardless, I think this looks like a very good solution, even if there are small concerns. At worst the bus could also take the loop in the future, since it wouldn’t require much work (just getting rid of a bus stop, and adding a new one in campus).

    1. You make a good point about how 522 Stride (S3) leave the transit center and ultimately get to Beardslee.

      One other non-structure alternative would be to add a short-distance bus lane running just north of 522 from the transit center to the main 522 northbound non-HOV ramp east of 405. That however would have to plow through or next to the proposed transit plaza under the 405 mainline and the 405 Stride (S2) boarding areas so it’s probably not a safe idea.

      Of course, if 522 was changed to a diverging diamond Interchange it would be less of an issue. That however would be a whole other interchange design though so it probably has other fatal flaws.

    2. I like the loop through campus to serve campus directly.

      Having Beardslee/195th bus stop would only be outbound would be a bummer, but that could be mitigated by having the new STX Woodinville route do the same loop (clockwise) but using it in the opposite direction (goes through UWB eastbound instead of westbound). That way, if you are at the transfer hub and want to get to 195th* you can just take the STX route . It’s less than ideal to have one-way stops, but probably necessary given the geography. For Stride 522 eastbound to Woodinville, you can use the Beardslee/195th as your eastbound transfer, while you’d use the transit center for a west bound transfer.

      *Or more likely, a job in one of those office parks east of 405, which are the heart of Bothel’s bioscience hub. By having a few stops on 195th st and 120th ave before getting to Woodinville’s town center, the STX route does good work providing last mile connections for both 522 Stride and 405 Stride. This route could get decent ‘reverse commute’ ridership to supplement the Woodinville to Bellevue/Seattle ‘primary’ commute.

      1. Ah, realized the Woodinville route will serve Bellevue directly during peak. Well then, I’d argue there needs to be a local route that goes from Woodinville to the transit center via 120th & 195th during the peak. Or, I could see a shuttle service to those office facilities using the transit center, either through a public-private partnership or by chipping in for the local bus service.

        Either way, ST is really erroring by focusing on serving only Woodinville with this STX route. I hope Bothell realizes they should be using this route to serve their city east of 405.

      2. I wouldn’t be too concerned about that stop at 195th/Beardslee. There is some development there, but I don’t see that stop as being that big — it will be nothing like UW Bothell. It also isn’t that far from the stop that serves the campus. The riders heading to the school — which will greatly outnumber those heading to 195th/Beardslee — will have a longer walk.

        They can also be served by the 239, if it was extended as far as the 230 (or they could swap tails, at no cost). The 239 would then connect that neighborhood you mentioned (with the office parks) to the main part of Bothell as well as both of the Stride lines (at Brickyard for 405, and along Beardlee for 522). Someone who commutes to Bellevue from Beardslee/195th would take the 522 Stride, then the 405 Stride in the morning. In the evening they would either do the reverse (and walk an extra 5 minutes), or get off at Brickyard and take the 239.

        You would want the 239 to run more often, especially during peak.

        As far as Woodinville to the Bothell office parks, I think it would make more sense to run the 931 more often, especially during rush hour. Realistically though, very few people will take the bus from Woodinville to those office parks — they will simply drive.

        All that being said, I’m just throwing out a concern that may amount to nothing. It may be that getting to that stop via 405 isn’t that bad. If so, then I wouldn’t bother with the loop.

        Either way, though, I would probably swap the tails of the 230 and 239, to serve the office parks you mentioned. It is quite reasonable for someone from, say, Renton to ride the Swift 405 all the way to Kingsgate, then transfer and get there via an extended 239. The 230 doesn’t make that connection, and while the Swift 522 does, it doesn’t get that close to those office parks. The opposite is true as well — someone who lives over there would benefit a lot from a connection to the Swift 405, while still retaining the connection to downtown Kirkland. That means you could bump up frequency during rush hour both directions, and have increased ridership.

      3. All true. I was thinking the new STX need to be leveraged, but taking a look at our favorite greater eastside frequent transit map, I see those routes you mention can also do the job. I’d also imagine this transit center becomes a new node for KCM, as other routes bend slightly to go through the TC to provide transfers to 405 stride, plus the layover space makes this a logical terminus for a route like the 931 and 239, rather than UWB itself.

        This may end up looking a lot like the Eastgate TC, where you trade an inconvenient (distance & incline) but doable walk to the college campus for a high quality transfer to a freeway station, and a bunch of local routes pass through not because there is any “there” there but because it has the bus facilities for a great transfer nexus (layover space, charging stations, etc.) and a freeway station with super frequent buses at peak. Trade Bellevue College for UWB/Cascadia and Seattle for Bellevue.

        I guess the best use of this STX route really comes down to the context of the Eastside restructure once Stride comes online (hopefully near the same time as East Link).

      4. I’d also imagine this transit center becomes a new node for KCM

        Yes, but I don’t think things will change that much. The transit center offers connections to 522 Stride, 405 Stride, and other express buses on 405. Candidates for extensions or detours to the transit hub already connect to the 522 Stride line. It is really the 405 connection that matters. I’ll put aside express buses for now and just focus on the 405 Stride line.

        Looking at the map you mentioned ( there are a number of buses that serve UW Bothell. I think the 522 will go away, and the 372 will be truncated at Kenmore (if not sooner). The 312 will either be truncated or killed. That leaves several buses that could be extended to the transit hub:

        105 and 106 — Maybe, but these buses already cross Canyon Park, which means that most riders trying to access the 405 Swift already have a better option.

        239 — Similarly, this already connects to the 405 BRT at Brickyard.

        230 — If the 239 takes over the tail of the 230 (as I proposed elsewhere) then this could be extended to the transit hub.

        931 — I suppose. It is a slow way to get there, but might as well (since there is no other service along much of that line).

        OK, back to the express buses. Several of them come from Woodinville. Theoretically, there is value in extending the 105 and 106 for this purpose. During rush hour, someone would ride a bus to Woodinville to the transit hub, then transfer to the 105 or 106. Sorry, but I just don’t see it. People will drive there. That is a two seat ride instead of a short drive. The only place with significant 9-5 employment is an especially short drive, and an especially long bus ride. Office workers tend to have money, and office workers in Woodinville tend to have cars. I think additional ridership would be in the single digits (as in a handful of riders for the entire day). I just don’t see it being worth it.

        So, in terms of changes, I would do the following:

        1) Extend the Woodinville shuttle to UW Bothell. Mainly because it is a destination by itself. The prospect of a fairly quick bus with 20 minute frequency to the campus is likely to be popular.

        2) Have the 239 take over the tail of the 230 (the section from UW Bothell to 240th and 35th).

        3) Have the 230 go through campus and end at the transit hub, instead of at UW Bothell. Alternatively, it could follow the same path as the Stride 522, which would add the stop at Beardslee and 195th.

        So that’s it. One additional bus extended to the transit hub, while the shuttle is extended to the campus. Oh, and if both buses ran through campus, and stopped in campus, then it would help those that have trouble getting around. You would avoid a ten minute walk like this:, or something similar from the transit hub. That is a pretty long trip for someone who uses a wheelchair or cane.

  4. I’m admittedly concerned about the capacity and operation of the proposed 145th St/ 5th Ave NE roundabout. There are lots of ways this design could go badly. One way is that 145th drivers may drop off people to get to the Link station on 145th instead of the designated drop-off area.

    I once lived in Boston when there were several rotaries (their term) that were over capacity. When they overflow with too much traffic or stopping for pedestrians, the situation can result in backups for miles. That would be bad for 522 Stride operations.

    1. I feel the opposite. I think a roundabout is a great idea — with one caveat. First off, it makes getting to the station better than ever. Westbound on 145th, the far right lane is bus-only. Not BAT, but bus-only. It is also right turn only. Northbound on 5th, just north of 145th is also bus-only. That means that a bus turns without ever encountering traffic. At some point it needs to merge left, but that can happen north of the turn for the freeway. This was the plan before, but might have to wait for traffic.

      Going the other direction means avoiding the left turn lane. It may take a while to get into the roundabout, but that is usually better than waiting for a light cycle. After the turn, the bus moves into the far right lane (another bus lane). Overall, it should be much faster.

      The other thing that is being added is a ramp from eastbound 145th to northbound I-5. This will allow cars to take a right, and then loop around to head north. This means fewer cars in the roundabout, since again, these cars will just take a right.

      The only concern I have is how pedestrians will get around. They are adding a pedestrian bridge at 148th over I-5. But there will still be people that want to cross close to the roundabouts. Heading north-south should be fine. There is a crosswalk on First Avenue (with beg buttons) west of the roundabout. They may also add one on Third. You could do the same on the other side, at 6th Avenue or further east (there is nothing to the south of that crossing, other than the golf course). But there will still be some people that want to go across the freeway on 145th.

      I’m not sure if there is an easy way to avoid the danger. Roundabouts are dangerous for pedestrians, especially the type discussed here (two lanes turning). These are roundabouts close to the freeway as well. People will be focused on the other cars, and then start accelerating on the on-ramp. Crosswalks without a signal (which is typical around roundabouts) are inherently dangerous. If you add a signal, then you would want it a ways away from the roundabout, but that means a detour for pedestrians.

      I think the only safe solution is to simply ban pedestrian access on 145th. This is harsh, but it is worth noting that crossing there is no picnic right now, and there are various areas where it is challenging. My only concern there is that people often break the law and risk their lives to avoid a detour. The problem is that the 148th pedestrian bridge (as good as it is) is on 148th, not 145th. It isn’t a substitute for 145th if you are coming from the south. The only long term solution is to build another pedestrian bridge over the freeway a bit to the south, at roughly 140th (where the freeway narrows).

      (Of course all of this could have been avoided if they had simply put the station at 155th. ST would have saved a fortune in widening 145th, and the bus would be nowhere near freeway traffic. Drivers on the 522 Stride would spend an extra minute on the bus (if that) to get to Link. The bus would have kept going, connecting to Aurora and terminating at high ridership Shoreline College. No one on this blog would care about these interchanges, since they wouldn’t involve the buses.)

      1. Roundabouts should be safer for pedestrians, in theory, because you only deal with traffic in one direction at a time. The addition of a refuge island will make this an easier crossing for those less able, compared to the current 7 lane wide signalized intersection. Converting a signalized intersection to a roundabout is a move to safety, not throughput.

        Also, I’ve almost always seen pedestrians get immediate priority through a yield sign or flashing yellow, because traffic can continue through flow through the central roundabout as a pedestrian crosses one half-street at a time.

      2. There is no evidence that roundabouts are safer, even though people often think they are: All the literature points to increased safety for drivers, not for pedestrians, nor bikers. In fact, they recommend against the use of roundabouts for bikes. You can see various roundabouts in the above document — e. g. These aren’t in the actual roundabout, but just a bit outside it. You can imagine the hazard. You are driving, having just looped around, focusing on what the other cars are doing (merging, changing lanes, etc.) and just when you finally come out it and start focusing straight ahead — Aach@! There is a pedestrian! You slam on your breaks and hope no one rear ends you.

        Keep in mind, Seattle got rid of all of the four lane crosswalks that lack signals. Either it is one lane each direction, or there is a signal. If the crosswalks next to the roundabouts lack beg buttons, then this has the same basic problem as crossing a four lane road with a walk signal. A truck (or bus) stops in the outside lane. The pedestrian walks out from behind the first car, and Bam! — another car hits them (from the inside lane). They never saw them, and never thought the other car was slowing down for a pedestrian.

        To be clear, if the plan is to add signals, then that is great. I’m all for it. But this really isn’t about pedestrian safety.

        Roundabouts are put in to increase throughput. Undoubtedly.

      3. Roundabouts just plain suck when you’re on a bike. Drivers are looking only for other cars when making their merge/weave. A cyclist is pretty much just another bug on the bumper.

      4. Roundabouts just plain suck when you’re on a bike. Drivers are looking only for other cars when making their merge/weave. A cyclist is pretty much just another bug on the bumper.

        Yep, that is pretty much what all the literature says. If you have roundabouts, you need to provide a different way for bikes to get around (a completely separate bike path). In this case, that seems plausible, as bikes would essentially detour around the roundabouts (perhaps using the crosswalks and sidewalks), and cross over the freeway at 148th. I still think the key is to have traffic signals with the crosswalks (with a beg button).

      5. Roundabouts can be much safer for pedestrians and bikes, if they’re designed that way – look at the Netherlands for example. The difference is how cars enter and exit the roundabout. There are whole YouTube videos I’ve watched that explain the differences, but essentially when the Dutch want to give pedestrians/bikes priority, they force cars to make almost 90 degree turns into and out of the roundabout. That slows them down to 10 mph.

        Contrast that to US roundabouts where cars barely need to slow down because of the angle of the lanes. The Dutch have similar ones, but they’re all in the countryside where they don’t expect bikes.

      6. Not all roundabouts are alike. These have multiple lanes and will carry lots of traffic. That means they will be much more dangerous than a typical one lane roundabout that we are used to seeing. It’s also at a freeway so some drivers will be adjusting from being at higher speeds.

    2. Oh as a generic design, I like it! It will make Stride buses operate faster (if the capacity is there)!

      I’m mainly concerned about if it will overflow.

      As for pedestrians, Shoreline’s 148th Street pedestrian and bicycle bridge project over I-5 will help lots! There is already pedestrian access under 145th too.

      Given how proactive Shoreline has been about 148th, I would think they would be considering this challenge too. Getting pedestrians away at least a few hundred feet from each of these multi-lane roundabout approaches should be the goal. It may ultimately take some barriers on the sides of the roundabouts to discourage people from hopping out of cars at the roundabout though. Human nature often misuses good designs.

      1. Given how proactive Shoreline has been about 148th, I would think they would be considering this challenge too.

        Absolutely. This is the first I’ve heard about the roundabouts, and I have no doubt that everyone on the various pedestrian lists in Seattle and Shoreline will be all over this, and asking a lot of questions (staring with whether the crosswalks will have signals). If the answer is “no”, then that will be a big issue. As I wrote up above, crossing two lanes of traffic going the same direction without a signal is just not safe. (Oh, and there will be three lanes if you count the bus).

  5. Should we be using the new designations for these Stride lines? 405N Stride is S2 and 522 Stride is S3. If the naming is going ahead as planned, it’s not too early to start using these line designations.

    1. Only if people say which one it is next to it. It’s hard to keep track of which one is S1 or S2, or L1/2/3/4, or RapidRide G/H/J, or Metro’s 4-digit routes, before they start running.

  6. So, in trying to solve the reliability issue with Woodinville, they have stumbled into almost the most expensive option imaginable which will still require 3-seat rides off-peak. Seems to me, the best option is to extend the BRT route to Woodinville, with a 5-minute wait at the transfer hub. Since they are building a new transfer station with layover spaces, it seems trivial to add a couple bays for 522 Stride where the bus can wait 5 minutes (as a reliability buffer), while picking up passengers. Then, at this point, just extend every BRT bus to Woodinville (which, at least at peak, will be much cheaper than a new bus to Bellevue. And off-peak may be about break-even if the connector bus is going to be frequent anyway).

    Not to mention that off-peak, no matter what, it’s a 3-seat ride to Seattle. This is the problem with always further breaking up bus routes. Breaking up a bus route to truncate it at Link is almost always a good idea. Breaking up one-seat rides into two-seat rides with a better network is likewise almost always good. But breaking up a bus route twice when there is no apparent advantage for the second transfer just seems silly. 3-seat rides are *very* hard to convince riders to take, and reduce the usability of the system.

    At minimum, they should ditch the special off-peak bus and extend every other BRT bus to Woodinville off-peak only. This solves the transfer problem off-peak, and applies only to when traffic is least an issue. It should be cheaper too. That way, they can keep the expensive Bellevue bus during peak without needlessly hindering the usability of the system for Woodinville during off-peak hours.

    1. That said, assuming the willingness both to spend the money and make it frequent full-time, I am not at all opposed to a good frequent bus to connecting service that adds value by dramatically expanding the reach and speed of two-seat rides. But this would be better served by a bus from Woodinville to UW, via I-405 (stopping at Totem Lake and 85th St when that opens), and 520 (stopping at the 520 freeway stations, with connections to Redmond and S. Kirkland. These stations are always undervalued for their use as transfer points by planners, and should be more heavily utilized). This is basically the same as the “super 311” idea that has been previously discussed in the comments.

      This would speed up trips to downtown Seattle by a bit (going over 520 and a shorter U-Link train ride), speed trips to UW by a lot (vs having to choose between a J-shaped trip or a 3-seat ride), and in north Kirkland, would be a fast alternative to the 255. Bellevue becomes a two-seat ride, but it would be a very easy 2-seat ride.

    2. If the Woodinville-Bellevue bus makes all the intermediate stops Stride does, then it will supplement Stride’s peak capacity as well as being a one-seat ride to Woodinville. Not unlike the 522 and 312.

      1. That’s a fine use for such a route if it’s needed, but I’m not sure it is. Stride will provide frequent service, and ST still plans on operating the 532 at peak (with a stop added at UWB to replace the 535), which runs every 10 minutes. I don’t think this section necessarily needs another frequent route for capacity.

      2. I don’t think this section necessarily needs another frequent route for capacity.

        I would bet that it does. The 532 is full during rush hour. An express from Lynnwood to Bellevue running every ten minutes (i. e. 405 Stride) will be full as well. Either you run the buses more often, or run different buses.

        Lucky Woodinville, you get an express bus. My guess is, most of the riders actually come from Totem Lake. No matter, it still works. Woodinville benefits because they were promised something else, while Totem Lake riders schlep to the freeway station or have a two seat ride to Bellevue. It isn’t necessarily the best option, but it isn’t that bad, and given the political realities, likely the best we can do.

      3. “My guess is, most of the riders actually come from Totem Lake. No matter, it still works.” For sure. This reminds me of riding the I90 buses, where the 554 and 216 have tails that are useful for a number of riders, but most people still board at the Eastgate TC. You could get rid of those routes and just run the 212 at super high frequency and still get like 80% of the same riders, but we break them into multiple routes to get that extra 20%.

        My bet this is not the only route that partially overlaps with Stride. If you are Totem Lake trying to get to Bellevue, you might end up with 4 or 5 routes to choose from at peak.

      4. “The 532 is full during rush hour. An express from Lynnwood to Bellevue running every ten minutes (i. e. 405 Stride) will be full as well.”

        Actually, I took the bus from Totem Lake freeway station to Bellevue twice a week for several months, and in my experience the 532 was full. But there was always seats available on the 535, despite running 1/3 as frequently. I often passed up a 532 if a 535 was coming in a couple minutes. The best was a 237. It was always almost empty and (now) cheaper than ST (but also now, non-existent until the economy improves).

        So it’s a fair point that the 532 won’t improve capacity much since it’s already packed with Everett commuters, and the Bellevue line would improve matters the most because it’s basically the 237. But it’s also apparent to me that the portion between Bellevue and 522 is well served by bus service today, and since Stride will essentially triple the frequency of the 535 (which was one of the empty buses), I think it’s no longer the priority for dumping more expensive part-time peak-hour service.

      5. Yeah, your personal experience matches the data (

        The 532 is crowded, the 535 is not. But keep in mind that the 532 is not really packed with Everett commuters. Everett commuter (for all three stops together) make up less than a third of the riders. Canyon Park makes up almost a third of the ridership of the 532. They ride it because it is is faster than the 535 and frequent enough that riders will let a 535 go by.

        But things will change. The Stride line will be just as fast for those riders as the 535. There will be no reason for those riders to take the 535 over the 405 Stride.

        The improved speed will likely dramatically increase ridership from Lynnwood. The detour to serve the campus increases all-day ridership, but greatly suppresses ridership north of there. It isn’t clear how much ridership will increase, but we can look at other routes. If you look at the 511 (an express to downtown Seattle) Lynnwood has about twice the ridership of Ash Way. I see no reason why that same ratio won’t exist with the Stride route. Right now, about a third of the ridership on the (crowded) 532 is from Ash Way. That means way more riders from Lynnwood.

        I also think it is quite likely that the 532 gets a cut in service. Canyon Park and Totem Lake riders make up about 40% of the ridership. If they shift to Stride, then the 532 can get by with 15 minutes service. I could also see them just ditching the 532 altogether.

        I’m not saying that is the best option, but it is reasonable to run a bus like this to avoid crowding. To be clear, the particular choice is all political. This is a relatively arbitrary choice, and likely not the best (there are a number of things I would do first). But Woodinville was promised something by ST, and now it will get something. What they were promised was not a very good value — I think this will be better.

        The problem is that Woodinville — like most suburbs — has very peak oriented demand. In the middle of the day, there just aren’t that many riders. Those that exist are OK with a transfer or two. Ridership won’t be hurt much by the shuttle, but it could be helped quite a bit with this route. The 237 (that makes this existing connection) only runs three times a day each direction. This will make it much easier for Woodinville riders to get to Bellevue (and in turn, Redmond). Those that are trying to go to downtown Seattle during rush hour will continue to take the fastest option — the 311.

      6. “Those that are trying to go to downtown Seattle during rush hour will continue to take the fastest option — the 311.”

        The 311 is in Metro’s 2025 plan under the guise 2998, so that’s a reasonable assumption.

    3. I think the basic problem is that Woodinville has very low ridership, especially along the 522 corridor. It just isn’t worth watering down the line for a handful of riders. The shuttle is simply a low cost way to provide something for riders outside of rush hour. If you missed your 237 express from Bellevue to Woodinville, at least you can get home. Likewise with the 311 from downtown Seattle. You can also get to Kenmore in a straightforward, two seat ride (unlike today).

      I agree that a three seat ride is going to turn a lot of people off — but what do you expect? You live in Woodinville. It is a far flung suburb, and even if it had elite transit — even if it was served with a subway train right into downtown Seattle and downtown Bellevue — ridership would be low. That’s just life in Woodinville.

      Meanwhile, there are a ton of people who will have a three seat ride to their destination. That is what happens when you build a very linear, very limited subway. You are worried about poor little Woodinville — how about Lake City, which has orders of magnitude more riders. Lake City riders who want to get to say, Lower Queen Anne, will take a bus to Roosevelt, take a train, then another bus. The same is true for trips to First Hill. These are trips that are far more popular, far more urban, and far less automobile-centric than any trip from Woodinville. And yet, once they start doing it, they will embrace it, and never look back.

      Because frequency is most important, followed closely by speed. So what if you make a few transfers — if they are short, it doesn’t matter. Link will be fast and frequent. So will these BRT lines (although not as fast, and not as frequent). The big issue is the line through Woodinville. Twenty minute frequency is not frequent — some would not even call it transit — but it is all that was expected for Woodinville, and frankly, it is all that should be expected, given its nature.

      If you really wanted to improve things, then you could truncate the 311 at the UW, and then run it all day long (with a tiny detour to this station). The problem is, again, ridership would be terrible to Woodinville outside of rush hour. It would be good to Totem Lake. It would be good to this station, but bad to Woodinville. It makes more sense to run an express from the main UW campus to UW Bothell, since the vast majority of riders north of Totem Lake would be heading there. That would give the Woodinville riders another very good two seat ride (to the UW) and a much faster three seat ride to downtown Seattle.

      Which means that the best bet for this shuttle is to simply extend it to UW Bothell. That is not much longer than the proposed shuttle, but with a solid terminus (at the campus) and a lot more connections (to other neighborhood destinations). No, ridership won’t be high, but it will be decent, which is about as high as you can expect for the area.

      1. For the bus being almost empty, I have Jarrett Walker to thank for a good reason to run it at all:
        But part of the point is that it can be normal for a route to be almost empty near a terminus but still be worth it. That is the case here. What doesn’t make sense is to separate the least productive section, forcing a transfer for the people who really need it, and making it less attractive (will *needlessly* forcing a second transfer make more people take the bus or less?).

        As an aside, often what this looks like (to me anyway, at least cynically) is that there is a section of a bus route that an agency doesn’t really want to run but would face a lot of resistance to outright remove. So they water it down, while still “serving” the area. Then when ridership inevitably falls, they go “oh well, I guess no one rides the bus” and they delete that section with less resistance. You can’t know the motives for sure, but the effect is often the same.

        The solution to the traffic problem is to build in a 5 minute or so wait buffer at the 522 hub. Not only does that mitigate traffic (and let people wait inside the bus rather than outside if the bus is on time), but it hides the inconvenience to the few riders in Woodinville. They probably won’t even notice that it takes 5 minutes longer than it “should” to get where they’re going, but they will definitely notice if they have to get on a third bus (which will cause them to stress about making a transfer, or give up and drive). Notice, that can be worth it if it makes a better grid network to transfer (this is what U-Link is, and the north Eastside mobility project as well), but in this case, the long bus ENDS where the short bus BEGINS. It has no where else to go. It’s just going to turn around after a layover buffer anyway. It might as well continue after another brief wait, and then you can optimize the schedule better than if it was a separate bus anyway.

        You could do this with the 311 as well. I think it makes sense. As does cutting the frequency if it really doesn’t warrant it, but doing so in a way that preserves the connection to the train. I don’t think people will mind a 20 minute headway if they can take the bus all the way to the train. I think they would mind that less than a 15 minute bus that takes them to “the middle of the freeway,” where they have to take another bus and then a train. At that point, it’s just, why not drive on the freeway.

      2. They probably won’t even notice that it takes 5 minutes longer than it “should” to get where they’re going, but they will definitely notice if they have to get on a third bus

        Bingo, the reason I rarely used the route when I lived there is that anytime out of the peak it took multiple transfers. And even if one end of the trip worked the other didn’t.

        I repeat, this is not your grandfathers Woodinville & Bothell. Looking at past ridership failure doesn’t take into account the huge changes that have already happened in Bothell and are occurring in Woodinville. There’s a large senior living center near Woodiville City Hall and another by the P&R. There’s not a lot to do in DT Woodinville (the library is out on Avondale!) but a good connection with DT Bothell changes that.

      3. there is a section of a bus route that an agency doesn’t really want to run but would face a lot of resistance to outright remove

        YES! Now you are getting it. If this agency was really focused on ridership per dollar spent, they would just ignore Woodinville. But Woodinville was promised something, so they are getting something. It is not the same thing, but it is probably a better value. It solves the biggest weakness — commute trips to downtown Bellevue — while at least providing something similar for the handful of people that would consider riding the Stride 522 from Woodinville.

        You are missing the whole point of Walker’s post. He was writing about the value of seemingly empty routes, not making the ridership case for bad tails. In fact, he has made the opposite. It is especially important to have strong “anchors” at the end of the line, as explained here (

        The greater point is that extending the 522 line to Woodinville is a terrible value. Even at rush hour, very few people will ride it. It makes no sense as a way to downtown Seattle — those riders would just take the 311. Yet there are very few people riding it from say, Woodinville to Kenmore. Those riders, of course, can live with a two seat ride.

        Check out the numbers for Woodinville commuting. Redmond is by far the most popular destination, making up 28% of the trips. Yet Redmond only makes up 3% of the transit commutes. Why is that? Because everyone has a car. It isn’t that hard to drive to Redmond, even if traffic is terrible. Almost all commute trips (74%) are to Seattle. This makes sense, because Seattle commute trips are actually faster with a bus (and frequent enough to be convenient). People who own cars just leave them in their garage.

        That means that ridership in the middle of the day will be low. If someone is going to Bothell or Kenmore, they just drive. Even with the faster bus, it will take a long time to get to Seattle via 522, which means that those headed to the UW or downtown Seattle will probably pay for parking, just like they paid to cross the 520 bridge.

        Of course you would get more riders from extending this. But not many more. The whole idea that a same stop transfer — with no time penalty — is going to cripple otherwise strong ridership seems absurd. Even if you extend this, you still aren’t going to get that many riders, and it would cost a lot more. There are much better values elsewhere.

      4. By the way, the transit weakness with Woodinville and this corridor is more about its geography, not its general makeup. Consider the existing 522. Trips in Seattle (along Lake City Way) dominate ridership. In that case, it is all about urban versus suburban ridership.

        But Kenmore has way more riders than Bothell, and Bothell has way more riders than Woodinville. They are all similar in their suburban nature. The difference is the geography.

        Kenmore is at the top of the lake. The fastest way to drive to downtown Seattle or the UW is via 522. As you get to Bothell, it gets to be about 50/50. Sometimes it is faster to go on 522, sometimes it is faster to go via 405. As you get to Woodinville, it isn’t even close. It is much faster to go 405.

        This means that if you are trying to get downtown from Kenmore, the bus is going the way you would drive. But from Woodinville, it isn’t. Furthermore, Woodinville already has an express bus to downtown, the 311. Thus there aren’t going to be that many riders from Woodinville to downtown going via a 522 bus, even when that bus is BRT and connected to Link.

        Of course there is no express bus from Woodinville to downtown most of the day. There isn’t even an express to the UW. But with high automobile ownership in these suburbs, it is even more likely that riders will take their car. The buses don’t get much faster, while they become less frequent. At the same time, driving becomes a lot faster. There just aren’t that many people who will take transit along that corridor from Woodinville.

      5. ST publicly estimated that 522 BRT will take 22 minutes to go between Bothell (undisclosed location) and Link at Shoreline South. Then it will be about another 20 minutes to get to Downtown Seattle on Link plus the time it takes to walk and wait between the two services.

        ST reports that Downtown Bellevue to Westlake is 24 minutes. I can’t find the time it will take between UWB and the Bellevue Transit Center but it appears to be about 20 minutes. That means that getting between this 495/522 transit center and Downtown Seattle takes about the same amount of time regardless of if it’s 522 or 405.

        I know it’s awkward to talk about Stride overcrowding during this pandemic, but 405 Stride buses may be so full of riders by the time they leave Snohomish County that may keep those in North Kirkland off a bus or two. After all, it’s more dangerous to have riders standing on 405 than on 522. The “left behind” bus and train overcrowding issues are often missing from the line extension discussions.

        The beauty of BRT and the new naming (S#) is that ST can quickly introduce S4 a a separate line that serves Woodinville and can make it an S4 “shuttle”, or an overlay service on S2 or S3. They could also subsidize a Metro route. I’m cool letting ST serve Woodinville based on a real-world systems need that is in place on opening year.

  7. The Woodinville shuttle seems like a really weird route … it’s like ~6-7 minutes to drive from 405 to Woodinville P&R via the route there, even accounting for a couple bus stops along the way. Trading off against system complexity, it seems easier to just extend BRT.

    1. I don’t see this shuttle being practical In the long-term either. It looks to me to be a mere political placeholder for either eventually dropping the Woodinville connection entirely from Stride, or creating a new day-long Woodinville-Bellevue line. The good thing about operating buses is that it’s often not too difficult to change operations.

    2. If each way is 6-7 minutes of drive time, this means one bus can’t quite do 15 minute headways without sacrificing layover time, the result being that Sound Transit would need to either run the shuttle with two buses, or run the shuttle only every 30 minutes.

      The former is more expensive than just extending the BRT (due to all that extra layover time), so Sound Transit is probably envisioning the latter.

      I could be wrong, but I’m not reading this at all as a trade-off between an extra transfer and better frequency, but a statement that hardly anybody in Woodinville rides the bus, except at rush hour, and Sound Transit wants to spend as little money serving it in the off-peak as they can get away with.

      The good news is that at least the Woodinville shuttle is actually serving Woodinville with a local stop pattern, rather than just the park and ride. This is what the 522 should have been doing from the beginning.

      Long-term, I think the Woodinville shuttle is going to have to be extended and combined with some other route in order to survive, and I can easily see this route being the domain of King County Metro, rather than Sound Transit. For example, perhaps route 230 could do it, provided that some other route can be found to take over route 230’s current northern tail. Or, if Totem Lake develops enough, maybe an all day version of the 311 (truncated to UW Station)?

      1. Maybe the right way to serve Woodinville, then, is to extend the 372, or at least some trips on it.

        It would do everything the ST shuttle would do, but would also get you to Bothell and Kenmore with a one seat ride, with more stop options.

        For the cost of running this shuttle, ST could just pay KCM for adding a couple extra buses to the 372.

        As an added bonus, those traveling from 405 BRT to DT Bothell would now be able to take either the STRIDE or the 372, whichever comes first, making the wait time very minimal.

      2. A bus going on to Monroe would be useful. Mike, referring to your earlier comment about how many east side people work in Seattle, back beginning in 1975 I lived in Gold Bar for many, many years. Lots of people had daily commutes to Seattle from Sky Valley areas. Hardly anyone worked in Everett. People had interesting jobs with work hours that allowed them to leave super early and drive home in the afternoon before rush hour. This was before Microsoft, etc., on the Eastside. Seattle was where the action was, and still is.

      3. I think it would probably be up to Community Transit to serve Monroe. They only have one bus a day (the 424) and if it was truncated, maybe they could have two, or three. One of the problems is that the highway doesn’t have HOV lanes, so a bus at rush hour spends a huge amount of time on the highway. This cuts into ridership, and makes it very expensive.

        A shuttle could be extended out to Monroe and it would be much faster. But it is still along ways, and ridership per service hour would likely be abysmal.

      4. Maybe the right way to serve Woodinville, then, is to extend the 372, or at least some trips on it.

        Sorry, no. Metro knows there isn’t enough ridership to justify an extension, and so does Sound Transit. There just aren’t enough riders. Why should we subsidize the heck out of the handful of riders from Woodinville? There are much greater needs for the East Side (and the north side, and the south side …).

        The 372 north of 145th is redundant after 522 Stride gets built. At the same time, we need another bus route that mimics the current 522, and follows Lake City Way all the way to Roosevelt, and either ends there or continue to the UW. There are more riders at the 20th/85th stop on Lake City Way (served by the 522, but not the 372) then anything north of Seattle. Similar stops (at 95th, 90th, 15th) should be added to that route. Stops closer to Lake City (105th, 110th, etc.) should be served by both buses, so that riders along that part of Lake City have a choice between the slow backside route to the UW, or a fast connection to Link.

        That means two routes on Lake City Way — one that veers off to go on 25th, and the other that stays on the main corridor. Neither should go past Kenmore. The shorter one (that stays on Lake City Way) should end where the current 41 ends (close to Fred Meyer, at around 130th). The other one should end somewhere between 145th and Kenmore. If they can find layover space, then 145th is ideal. If not, then just go to Kenmore.

        Meanwhile, there should be an express from the UW to UW Bothell. This will provide those Woodinville shuttle riders with a fast connection to the UW (and in turn, a faster three seat ride to downtown Seattle). Ideally the shuttle continues to UW Bothell, but if not, then this express would do a little loop around the transfer hub stop before ending at UW Bothell.

      5. Monroe is outside the taxing district, so that’s definitely a Community Transit problem, unless the city signs a contract with ST like Gig Harbor does. That said, CT could use this transit center (particularly for an off-peak truncation) for a Monroe route, as the transit agencies use each other’s facilities for no cost.

      6. Meanwhile, there should be an express from the UW to UW Bothell. This will provide those Woodinville shuttle riders with a fast connection to the UW (and in turn, a faster three seat ride to downtown Seattle).

        Fast connection from Woodinville to UW Bothell; faster (maybe) to Montlake. Fast in any form can’t be part of a 3 seat ride on the eastside. Woodinville P&R ridership sucks for the same reason any route in Seattle would given the same frequency and stop/distance ratio. I lived in Hooterville for decades and used the P&R less than a dozen times because even if I had to go DT it was just about useless. And I hate to drive to DT Seattle. From Bellevue I’ll beg a ride and go to S. Kirkland or Bellevue TC.

        One thing that should be considered is that recent development and projects already in the planning stage change the equation. Woodinville’s main draw used to be Molbacks; Bothell was Hillcrest Bakery? The times they are a changing.

      7. “Metro knows there isn’t enough ridership to justify an extension, and so does Sound Transit. ”

        If there’s enough ridership to run a shuttle to Bothell every 20 minutes, there’s enough ridership to extend a bus that already goes to Bothell every 20 minutes. Of course, running the extension with every trip (every 6 minutes peak) would be overkill and a waste of money. I’m not suggesting that. I’m just suggesting extending select trips so that the Bothell->Woodinville bus runs every 20 minutes. Just like the Sound Transit Shuttle, but more efficient in that you’re not paying for as much layover as you would with a dedicated route.

        It’s also worth asking why the ridership in Woodinville is so bad and whether we should really be giving up on the area, rather than asking what can be done to make the transit there more attractive. The downtown area of Woodinville is not acres of sprawly houses. There’s a big commercial area with a bunch of shops and a movie theater that a UW Bothell student living next to campus might actually have reason to take a bus to. The housing around downtown Woodinville also includes many more apartments than houses.

        All in all, the land use seems similar to the area around Crossroads Mall. Not skyscrapers, but certainly enough there that we shouldn’t be just giving up transit in the area and telling everyone who wants to visit there to go get a car.

        Having the 231 connect to Kirkland in a straight line without an hour of zig-zagging was a good place to start. Having the 522 serve the whole town center, rather than just Woodinville P&R would also help.

        I’m not asking for a bus every 5 minutes, I’m just saying that the bus every 20 minutes needs to go somewhere useful, not just end at a transit center in the middle of nowhere, and extending the 372 is one potential option for making it go somewhere useful.

        Relying on the 931 as the sole Bothell->Woodinville route, I don’t think is going to cut it. It’s slow. It’s a lot of meandering. And it doesn’t connect with the I-405 bus without yet another transfer (unless you want to add 10 minutes to everybody’s trip on the I-405 bus to exit the freeway and serve the UW Bothell loop). I think the route Sound Transit chose for Woodinville->Bothell is fine. It just needs to continue on and do something past Bothell.

        “The 372 north of 145th is redundant after 522 Stride gets built.”

        Is it? The stop spacing on the STRIDE looks pretty wide, like over a mile. Are you proposing to just leave the intermediate areas without any bus service? Or are you saying ST should slow down the STRIDE by making it stop every 1/4 mile or 1/2 mile, to avoid the expense of running a shadow route?

      8. Stride is supposed to be a limited-stop route, a poor man’s light rail like Swift.

      9. If there’s enough ridership to run a shuttle to Bothell every 20 minutes, there’s enough ridership to extend a bus that already goes to Bothell every 20 minutes.

        There isn’t enough ridership. That’s the point. This is being done for political purposes. This is being done *only* because ST promised Woodinville this service, not because it is a good value.

        The only reason this is a shuttle instead of an extension is because it is cheaper. Why else would they do this? Do you think this costs the same, or more? That makes no sense. This is being done to save money. It is a bad value either way — at least this doesn’t cost as much.

        In exchange for the transfer in the middle of the day, they give riders something they might actually use — more rush hour trips to Bellevue. That might be a bad value, too. Or it might not. Time will tell, but it has a much better chance of being a good value than an extension (or for that matter, the shuttle).

      10. It’s also worth asking why the ridership in Woodinville is so bad and whether we should really be giving up on the area, rather than asking what can be done to make the transit there more attractive.

        It is not horrible — it is just not good *for this corridor*. It gets back to what I wrote elsewhere. If you live in Kenmore, the Stride 522 changes your world. Stride is your transit ticket to every major trip. If you are headed to Seattle, it is faster than driving during rush hour, and fairly competitive the rest of the time. If you are going to Bellevue, then you take that bus the other direction and again, at rush hour, that is faster than driving.

        This just isn’t the case with Woodinville. The fastest way downtown is the 311. If they truncate the 311 at the UW (which I hope they do) then it is still faster. That would also mean a one seat ride to the UW, which again is faster than taking the 522 bus and transferring to Link. Meanwhile, the 522 bus is irrelevant if you are headed to Bellevue, Redmond or Lynnwood — the shuttle is just as good. The two biggest destinations that would benefit from extending the 522 bus are Northgate and UW Bothell. The former isn’t worth it, and the latter can easily be served by just extending this five minute right into campus (as I’ve written earlier).

      11. “The 372 north of 145th is redundant after 522 Stride gets built.”

        Is it? The stop spacing on the STRIDE looks pretty wide, like over a mile. Are you proposing to just leave the intermediate areas without any bus service? Or are you saying ST should slow down the STRIDE by making it stop every 1/4 mile or 1/2 mile, to avoid the expense of running a shadow route?

        The stop spacing is wide, but it really doesn’t miss many stops. The wide stop spacing on 145th is OK, simply because other buses will pick up the slack. That is towards the busy end of the line (where everyone is headed) so it makes sense to spread out there (and save a lot of people time). Most of the rest of the line has wide stop spacing, but are very similar to the existing service for the 372 (and 522). There just isn’t much in between a lot of these stops.

        The only problem area I see is between Kenmore Park and Ride and Bothell. There is no stop at 80th or 83rd. Maybe this is just a mistake on the map, but there should be. The 522 stops at 80th, with decent ridership. This is also towards the far end of the line, which means you don’t cost that many riders that much time. The same is true for a couple other stops in the same area. It is quite possible that ST just didn’t have the time or money to build “stations”.

        But as it turns out, that doesn’t matter that much. If ST leaves a service hole that Metro feels must be served, then they can do that by simply extending the 331. That would be less than ideal, but not as cumbersome as sending the 372 there. You also provide more value, in that it gives Ballinger Way folks a one seat ride to Bothell (and the college) whereas the 372 is more redundant.

        The bigger challenge is finding a layover and turnaround spot. This gets complicated (I may write a Page 2 post) but basically there are two bus routes that are likely to run on Lake City Way north of 130th. One is the 372 (to the UW) and the other follows the current 522 path, but is truncated at Roosevelt (or maybe the U-District). I’ll call that bus the “new 522”.

        You want at least one of them to go as far as 145th — otherwise you have a big service hole (although the hole could be filled by moving the 65 or simply asking people to walk a couple blocks). It also makes sense to have the buses go as far as 145th because of the density and to enable a connection with the Stride bus. It is bad enough that someone traveling along the same corridor (SR 522) needs to transfer, but they should transfer only once. If layover space at 145th is found, then both buses go there. But if not, I would send the new 522 to Kenmore, and truncate the 372 at the Fred Meyer.

        Keep in mind, any bus stop that Stride skips is a low ridership stop (by definition). It would be nice to provide service to those stops, but there other places, with more potential riders, that are in a similar predicament ( If you do provide service to stops along SR 522, then it should be done with a coverage bus, not something as frequent as the 372. That is why something like the 331 could work.

      12. Fast in any form can’t be part of a 3 seat ride on the eastside. Woodinville P&R ridership sucks for the same reason any route in Seattle would given the same frequency and stop/distance ratio.

        Do you see the contradiction there? You are saying that you can’t possibly have a three seat ride from the East Side to anywhere. Because on the East Side that would kill ridership.

        At the same time, you are saying that any place in Seattle with the same service as Woodinville would be doomed to the same low ridership.

        That is an obvious contradiction. It is also absurd. Ridership varies in part because of service levels, but also because of the inherit nature of the neighborhood. Woodinville is a distant suburb, with high automobile use. There is no way they are going to have the same sort of ridership as say, Fremont. Density matters, but so does proximity. What is true for light rail (as recently reported here: is true for buses.

        If anything, the farther out you go, the less one-seat rides matter. In Woodinville, speed is important — so important that transfers (and frequency) are less important. Put it this way: Imagine a bus that delivered a one-seat ride to Woodinville, via surface streets. It avoids the nearby expressway, and averages about 10 miles an hour in the middle of the day, when it isn’t stopped by the frequently opening drawbridge. It is much slower during rush hour, but at least there is no drawbridge opening to worry about. I’ve just described the 40, which has over 12,000 riders a day.

        At that rate, a trip from Woodinville would take about two hours. Oh, but wait! It runs every ten minutes! It is a one-seat ride!

        Do you really think that Woodinville would suddenly have ridership approaching Fremont if it only had what Fremont had? Give me a break.

        You just aren’t going to get the ridership in these areas that you can get in the core of the city. Maybe at rush hour you could have that — oh wait — you already do! The 311 isn’t going away, even though it is an extremely poor value compared to slow motion buses like the 40. Running express buses to Woodinville in the middle of the day are extremely expensive, and a terrible value. The ridership per dollar would be horrible, and the value added would be minimal. Asking an agency that struggles with basic coverage — i. e. has no service any time of day for various apartment complexes — is ridiculous.

        For now, the best Woodinville can expect in the middle of the day is a shuttle to an express bus which goes to a major destination and connects to Link. Turns out that will have that.

    3. Is there a way the Woodinville shuttle can also serve Bothell east of 405? It’s very a suburban layout, but those office parks are a big bioscience cluster and could be developed into a much denser job node in the near future.

      1. You mean like Google Bothell, and the places around there (

        If so, see my comment up above. The problem is that the direction is all wrong. It is basically a loop — a bad idea. Someone from Woodinville would go way out of their way before finally getting there. If your goal is to connect Woodinville with that area, then just run the 931 more often.

        Otherwise, it makes way more sense to serve it from the south or the west. That is what the 230 does. The only weakness with the 230 is that it doesn’t connect with a Stride 405 stop, which means downtown Bellevue to that area would be a three seat ride. That is why I would simply give the tail of the 230 to the 239. Riders from that area would transfer at Brickyard to get to downtown Bellevue. During rush hour all you need to do is add a few trips (given the overall strength of that route, I think those trips would pay for themselves).

    4. There are two basic problems:

      1) There is very little ridership in Woodinville.
      2) There are traffic issues in Woodinville.

      The first means you start thinking about truncating half the lines. But the second causes a problem with the first. Say the truncated bus is suppose to leave Bothell at 9:00. The Woodinville bus is supposed to pass by the stop at 9:10. But because of traffic, it goes by at 9:15, only a few minutes before the next Bothell bus, leaving at 9:20. That is a prescription for bus bunching, even if there is off-board payment.

      As Alex suggested, you could have a long layover in Bothell. I suppose, but that cuts into ridership again. It means that trip to Seattle takes even longer. Is it fundamentally different than a transfer? Not really.

      You could also spend a bunch of money and make that trip to Woodinville faster. but that means spending a bunch of money on very few riders.

      Having a long and a short route also complicates logistics. There are ways in which this can be carried out, but it is a lot messier. Basically, it gets solved by having the bus spend a lot in Woodinville, which means spending a lot of money on only a handful or riders. Sorry, it just isn’t worth it.

      This shuttle isn’t ideal. It is too short for maximum ridership. I think it should be extended into UW Bothell, and/or extended to cover more of Woodinville. But it is also quite possible that ST and Metro have it figured out, and this is just enough time to do a live loop to the station, come back, and layover in Woodinville. If you ran it farther, you would run it less often. If you had more money, then you would. But of course, there just isn’t enough money for good bus service. We are spending it all on other stuff (light rail to South Kirkland, BRT stations on NE 85th, widening the road at 145th so that the bus can get to 148th instead of 155th…).

      1. The problem with very short routes is that you end with a very high layover-time-to-service-time ratio, which results in achieving relatively poor frequency for the amount of money spent. Each minute of layover costs nearly as much as a minute of motion because the driver still has to get paid and the bus is still tied up, unable for use by another route.

        In this particular route, the length is 2 miles and Google Maps is estimating a non-stop drive time of 6 minutes. Throw in a couple minutes for passenger loading and the round trip becomes 16 minutes. This means that to run the route at 15-minute frequency, you need two buses even with a live loop. Even at 20-minute frequency, you’d still likely need two buses, otherwise, the layover at the end of each round trip becomes too tight, plus violates union rules about giving the bus driver enough time to go to the bathroom. So, if all Sound Transit can afford is to run the route on just one bus, what you end up getting is a bus that comes every 30 minutes, with only 50% of that bus’s service time actually carrying passengers, the rest just spent sitting empty at the endpoints.

        In general, scanning through the schedules, it seems as though Metro and Sound Transit have found the optimal route length to be around 45-60 minutes. Anything longer makes for service that’s too unreliable. Anything shorter means spending too much taxpayer money paying drivers to sit with an empty, stationary bus.

        For this case, I tend to favor the approach of extending the 372, whose reliability is less critical than the 522 BRT bus. You get the one-seat ride to Bothell, plus solve the inefficient-short-shuttle problem.

      2. I think the point about the short route makes the case for extending the shuttle to UW Bothell. Doing so might not cost much at all.

        But extending the main route is clearly more expensive than running the shuttle. Otherwise, ST wouldn’t propose this change. The shuttle will be unpopular — people would prefer the extended line. But the shuttle saves the agency money. It also makes the main line more reliable. Those are the two reasons this is being done — to save money and to make the main line more reliable. It is a trade-off, but likely worth it.

        I commented on the 372 up above. It should be truncated, not extended, once Stride 522 starts.

      3. The only reason the shuttle is cheaper than just extending the route is that it’s running at a lower frequency.

        If you compare the shuttle to simply extending every other trip, the shuttle actually becomes more expensive – you’ve got the same number of bus-hours driving around Woodinville, but more bus-hours sitting in layover. The reason why Sound Transit is choosing the shuttle over this is to ensure that every bus on the main route leaves Bothell exactly on time, regardless of traffic in Woodinville. In other words, they’re spending more money on operations (to pay for the additional layover time) to improve reliability on the main route.

        Extending the shuttle to UW Bothell would help somewhat, and could possibly be nearly free if it doesn’t impact the number of buses required to run the route at a given frequency. But, once you do that, you may as well make it an extension of some other route that ends there. Assuming that the reliability of that other route is not critical, making it an extension both saves money and reduces transfers. DT Woodinville->DT Bothell is still the type of very short trip in a straight line that makes a forced transfer particularly ugly.

        Assuming the 372 still needs to run to Bothell, it becomes a logical extension point, simply because it ends there. Another option is to switch the tail of route 230 to continue to Woodinville after UW Bothell, instead of going a mile to the northeast. (The current northern tail of the 230 is already covered by Community Transit route 106, so maybe that’s good enough). The point is, pick some route that already ends in Bothell (I don’t have too strong opinions as to which one) and extend it, rather than run a separate route.

        As to whether the 372 is still necessary to go to Bothell, that’s another topic. But, unless the new 522 is going to be stopping more often than the current 522 (the plans show it isn’t), you have to have some local shadow to avoid places being left without service. This could involve some restructure, rather than a literal running of today’s route 372. For instance, maybe the 372 only goes to Lake City, then you have some other (less frequent) route running Bitter Lake->130th St. Station->Lake City->Kenmore->Bothell, serving all the stops on SR-522 that the Sound Transit bus skips.

        Or, are you saying that the Sound Transit bus needs to just stop every 1/4-mile throughout SR-522 and people will simply have to put up with the extra travel time?

      4. The only reason the shuttle is cheaper than just extending the route is that it’s running at a lower frequency.

        Yep. But that doesn’t change the fundamental problem. There will be very few riders from Woodinville. Not enough to justify running all of the buses out there. That has always been the case. That leaves two options:

        1) Run half the buses out to Woodinville (the original plan).

        2) Run a shuttle, but with frequency in Woodinville equal to original plan.

        Obviously option number two is cheaper. Otherwise, ST wouldn’t consider it. This seems very counter intuitive. After all, Woodinville will have the same frequency. I believe the problem is reliability. To keep the core of the line (from Bothell to Shoreline) as reliable as they want, you would have to have a big layover at the transit hub. This complicates things, and adds cost.

        But that is just speculation about the particulars. What is clear is that a shuttle saves money over the original plan. Otherwise, they wouldn’t propose it.

        Extending the shuttle to UW Bothell would help somewhat, and could possibly be nearly free if it doesn’t impact the number of buses required to run the route at a given frequency. But, once you do that, you may as well make it an extension of some other route that ends there.

        Good point. That came to me as well (a bit later). As I wrote here ( I think it makes sense to have the 239 take over the northern tail of the 230, and have the 230 run through the UW Bothell campus to the transit hub. Having it continue to Woodinville would make a lot of sense. This would give folks in Woodinville two ways to get to a 405 BRT station, and two ways to get to Juanita and downtown Kirkland.

        The problem is timing. The 230 runs every half hour. The shuttle runs every 20 minutes. Lots of folks are complaining about the shuttle, but its frequency exceeds the frequency of most bus service in the region. There is no ridership or financial case for this added frequency. Woodinville is not “on the way”; in fact, it is “out of the way”. It is expensive to serve with anything but a shuttle. The 20 minute frequency is based on a promise. The adjustment is based on the fact that very few riders from Woodinville will head to Seattle via SR 522, no matter what you do.

        But it does beg the question: Which is better, and extended 230, which runs to Woodinville every half hour, or a twenty minute shuttle to the transit hub. If you are headed to Bellevue, the shuttle is better. If you are headed to Kenmore, the shuttle is better. If you are headed to Bothell, it is a tossup, but personally, I prefer the more frequent bus. My guess is that the vast majority of folks are heading to Bellevue, which is why the shuttle isn’t that bad of an idea. Extending a little bit (to an actual destination) might be a cheap (or even free) way to add value, but it probably won’t add a lot of riders.

  8. Will this configuration take advantage of the Sammamish River trail that connects D/T Bothell and UW/Cascadia campus, enabling foot traffic to the new transit hub?

    1. It’s on the opposite side of 522 from the trail. There is a 12′ wide sidewalk leading towards the North Creek Trail so I assume that’s where it’s headed.

    2. Yes, the North Creek trail goes under 522 and connects to the Sammamish River trail right where North Creek itself connects to the Sammamish River.


    Fast, pretty, and ORCA-card amenable. Stops at the new transit center, the campus, and downtown Bothell. Would be equally popular to have another line connecting Ballard Link terminal with Shilshole via Nordic Heritage.

    Since I think present freight-rail conflict owes more to trail-resistance than to railroading, should be possible to negotiate elevated structure so there’s plenty of room for bikers and hikers underneath it.

    Same for south end of the same Link line, where it might even be economical to put it in a tunnel if nobody wants to look at either it or the business it’ll attract. Though recalling Oakland, think the views would add both votes and ridership. Wills and Way, Wills and Way.

    Mark Dublin

  10. ST3 is a mess; I am not sure this cleans it up. It promised Bothell I-405 BRT and was infeasible without two service patterns. The 522 BRT reliability issue was one created by their goofy turn back trips; the obvious service solution was to extend all trips to and from Woodinville. ST tends to go for bells and whistles and not for service. Today, Route 522 provides less evening service to the Northshore than Route 307 did in 2002. I doubt the ST Woodinville ridership forecast. The new 522 alignment misses the campus, stops short of Woodinville, and ends in the middle of two freeways. If the replacement Bellevue Route was to be acceptable, it would be full time. Frequency baby.

    1. Hey, Eddiew it’s Saturday. Meaning, like with every single other thing in the world right now, whatever was wrong with ST-3 yesterday, something completely different will be wrong with it Monday, which is the next workday. Since the COVIDS landed, pick up any phone and the recording will be telling you something that used to be wrong, but is now out of date.

      Failing a Force 9 or a meteor hit, Bothell and its whole street and road grid will still be there long after masks are a memory. And the antique world shows what kind of condition a bus can be in and still start. My favorite little people mover which, maybe in tribute to Andrew Hallidie’s historic system in SF, are actually cable cars!

      So when California collapses, Governor Newsom will not only sell us the Oakland Airport people-mover but personally help dismantle it and put on the truck for its trip to its new home in Bothell. Where does it say he won’t? See you in today’s comments.

      Mark Dublin

    2. ST3 is a mess; I am not sure this cleans it up.

      Maybe not, but I think this is about as good as they can do at this point. A big part of the problem is cost. The 522 BRT project is really expensive. A lot of that money is going into making 145th wider. This is where they blew it. They weren’t looking at the system from a holistic standpoint. A station at 155th with connecting bus service would have been a much better value. Folks along the north shore didn’t want that, because it would be a bit slower. Getting to 155th would involve an extra couple turns (a right and a left inbound). Ironically, though, the bus will have to do that anyway! It will need to take a right onto 5th, and then a left into the station at 148th. Thus they are spending a fortune for a route that won’t be much faster than if they simply put the station a few blocks to the north. Once they realized they couldn’t put the Link station at 145th proper — i. e. that a bus couldn’t serve it via 145th itself — they should have moved it all the way to 155th, and used the savings to improve the rest of the BRT line.

      Spending all that money on 145th is a huge waste, but necessary at this point if this “BRT” project is going to come close to living up to its name. But it also means that money for other parts of the project is limited. Bothell doesn’t have an underpass like Shoreline has at 155th. The 522 BRT has to connect to the 405 BRT at either 195th or the 405/522 interchange. Ideally 195th would be rebuilt like the Totem Lake area. 405 buses would stay in the HOV lanes, while making a stop. Other buses could not only drop people off there, but continue on. But that would be extremely expensive (way outside of the remaining budget for this project). As it turns out, WSDOT is rebuilding the 405/522 interchange, and will build just such a station. This is great, really, and about as good as you can hope for. The 405 bus will move much faster than today, and yet you still have that key connection. The buses go frequent enough that transferring is not a huge issue. Meanwhile, someone taking the 405 bus still doesn’t have to walk that far to get to campus.

      My only quibble is the tail of 522 line. This asks UW Bothell riders to walk a ways, so that Beardslee and 195th riders don’t. I’m not convinced that was the right choice. I would have gone through campus (with a stop right in the middle) then got on 522 and ended at the transit hub. I think that could be more reliable than the current plan, with less effort. You would need to add bus lanes on SR 522 and maybe do something through campus (I have no idea if there are traffic issues there). This could help other (express) buses as well. A bus that goes from the Bothell Park and Ride to downtown Bellevue would take advantage of the bus/HOV lanes. (It is possible that the idea of going through campus died because the UW administration didn’t want to make it harder for students to park. The sad fact is that universities are run like businesses, and they make a ton of money off of parking).

      If you ran through campus, that would leave Bearsdlee and 195th a long walk from the 522 BRT. In exchange, they could have something similar to Woodinville. Run a few express buses to Bellevue during rush hour, and extend the Woodinville shuttle to follow the tail of the 230 (as AJ suggested). That would be a bit squiggly, but it would get the job done.

      Which brings up the issue with Woodinville. The fundamental problem with Woodinville and the 522 BRT project is that it doesn’t have enough riders to justify the expense necessary to provide reliability. That is all about geography. The fastest way to get from Kenmore to anywhere in Seattle is via SR 522. The fastest way to get from Woodinville to Seattle is via 405 (and then over the water). During rush hour, the ST 522 bus sometimes skips Woodinville for this reason. Overall, Woodinville ridership is low. Outside of rush hour, the 522 is the only bus along the SR 522 corridor to serve Woodinville. Thus someone in Kenmore might take the 372, while someone in Woodinville has only one choice: take the 522. Yet Woodinville ridership on the 522 is poor — much worse than Kenmore and Bothell (and much, much worse than Lake City). It just isn’t worth the cost to serve the handful of Woodinville riders while trying to maintain reliability.

      This shuttle will be unreliable. So will the express buses to Bellevue. It doesn’t matter. What matters is that the reliability of the core of the line, which is Bothell to Shoreline.

      Oh, and as far as my quibbles/suggestions, the good news is, this is a bus! It wouldn’t be that hard to change things around. You would basically have to add one “BRT Station” bus stop in the middle of campus (better than any bus stop right now) and then remove the stop at Beardslee and 195th (that stop would be served via other buses). All of that could be done later. That is one of the big benefits of bus systems like this — you can fix the weak points later at very little cost. In contrast, fixing a rail system is usually extremely expensive (otherwise we would move the Mount Baker Station to the transit center triangle in between Rainier and MLK).

      1. RossB is correct as usual. ST and the small cities provided themselves lemons (e.g., I-5 alignment, and NE 145th Street station in an interchange with traffic). They are now searching for a lemonade recipe. Northshore riders could have met Link at the Roosevelt or NE 130th Street stations or a NE 155th Street station as suggested by RossB. The first two options would have also served Lake City and its density and transit connections. ST3 is spending lots of capital; it will be delayed.

  11. How would this be different than what the 237 (Woodinville–>Bellevue weekday peak route)?

Comments are closed.