Seattle’s Transit Lanes in 2025

65 Replies to “Seattle’s Transit Lanes in 2025”

1. Thanks for sharing.
   I’ve been working on a similar dataset as personal project in the past few months and will cross reference yours!
   Transit Network by ROW type

   1. Thanks for sharing your map. I was just starting to create my own map from scratch when I found SDOT’s dataset. A regional map would be great to have as a reference, let us know if you would like any help from the community.

      1. Yeah I found the SDOT dataset too as well. I was trying to start something with existing data, but all the PSRC transit network/Trolley network data don’t have the precision I wanted which drove me nuts, so I ended up started the entire thing from drawing board.

         This dataset can be accessed through Rest API
         but happy to send you the file in certain format familiar to you.

      2. I also want to note that it is still a working progress as I recently discovered that Snohomish County has Q-jump for SWIFT all over the place.

         I don’t want to keep splitting my link just to distinguish a small section of Q-jump lane or bus bay that has no passing capacity if the length is shorter than a block, so I am starting a separate layer to document that kind transit priority as well as transit signal. It is less obvious to spot those feature, so I haven’t published that content to my Seattle region transit map.

2. Please detail the J line miles. BAT lanes on Fairview Avenue north? What else?

   1. And a short section of it on Eastlake Ave between Allison and Harvard (under I-5 bridge).

   2. https://seattletransitblog.com/2024/09/23/rapidride-j-line-final-design-and-recap/

      We have an article on it detailing it block by block if you want to know more

   3. and 3rd Avenue with transit priority….

3. It’s pretty exciting with the transit plus 40 and rapidride j almost every major bus route will now have at least some BAT/bus lanes

   I do hope in the future we can potentially get more bus routes with center running bus lanes though

   1. Why do you consider center running bus lanes superior? They require always crossing the street instead of half the time, and require extra island platforms and therefore more impermeable surface.

      It works on Madison because all left turns are banned. But most roads can’t reasonably ban left turns because there aren’t alternatives. So you’ll have left turns conflicting with the bus lanes, which is probably more disruptive than right turns conflicting/sharing the lane.

      1. “Why do you consider center running bus lanes superior?”

         Because side lanes have to allow cars to enter or cross them to access driveways on the right or turn right. Center lanes are faster for buses because they can travel full speed without any cars.

      2. > Why do you consider center running bus lanes superior? They require always crossing the street instead of half the time, and require extra island platforms and therefore more impermeable surface.

         It works in many american cities. I’m a bit confused why so many talk here about how it won’t work. it’s effectively a much cheaper but almost as effective for of a streetcar/tram to build.

         > It works on Madison because all left turns are banned. But most roads can’t reasonably ban left turns because there aren’t alternatives.

         On eastlake SDOT originally investigated adding center running buses there. rainier avenue was the original alignment for the light rail (well 2-car streetcar variant). 15th avenue for rapidride D is more than wide enough for center running brt etc…

      3. But most roads can’t reasonably ban left turns because there aren’t alternatives.

         I think it is the other way around. Most roads can ban left turns. Making three rights is quite reasonable just in terms of safety and better traffic flow. The other alternative is to just ban unprotected left turns. Again, this is often just for safety. If you have a pair of unprotected left turns then running buses in the center doesn’t hurt anything. You already have a special cycle for those left turns (both directions). Then when those buses are done turning, the bus (and cars) can go straight. Where it is less than ideal is if you have an imbalance. For example a left turn for northbound cars but not a left turn for southbound. In those cases it is common for the northbound cars to go straight while the cars are turning left (while the southbound cars wait). Unfortunately a bus would have to wait as well. But those situations are relatively rare.

         In short, center running is usually significantly faster than curbside running. This is especially true in urban environments where you have lots of cars turning right along with lots of pedestrians crossing the street. These sorts of urban environments tend to have the most potential for transit (making center-running buses a good option).

         Meanwhile, if you have one-way streets you can often benefit from contraflow buses. These have their own set of advantages and disadvantages. If you have only one lane for buses then buses can’t pass each other. It can change the nature of the traffic light cycle (making it longer as they now have to deal with traffic coming from a direction it didn’t before). But again, most of the time it can save the bus quite a bit of time. I think the lack of contraflow downtown is the biggest weakness with the RapidRide G while the center running is it’s biggest strength.

      4. Oh, and it is worth noting that Link is center-running down Rainier Valley. It would be a lot slower if it ran curbside.

      5. “I’m a bit confused why so many talk here about how it won’t work.”

         It’s not that it won’t work. It’s that there is a safety concern — especially in the middle of higher speed arterials.

         The design concept is similar to Link on MLK but with typically smaller and more frequently spaced stations. Even though MLK is technically designed to be fully safe even with curbs and barriers at stations, it’s known to have had pedestrians get hit crossing the street or the tracks.

         Note too that Madison isn’t a higher speed arterial.

         And while streets can get lower speed limits, that doesn’t automatically slow down drivers if they’re feeling comfortable driving at a higher speed.

      6. @al

         We’ve been over this many times already. It is very normal to have center running buses throughout the world. I don’t see why Seattle roads are magically incapable of accommodating them

      7. “It is very normal to have center running buses throughout the world. I don’t see why Seattle roads are magically incapable of accommodating them.”

         Yes and people all around the world get hit by buses too.

         And my comment about MLK is more an indictment about dangerous driving habits among a few people in Seattle. As I said, it’s not that Seattle is incapable; it’s instead a matter of accepting risk and designing for safer bus stop crossings.

         I might even suggest that median bus lane stops are better if they’re have crosswalk signals that are only for reaching the stop or crossing the street, rather than have them combined with turning auto traffic.

      8. @Al

         > Yes and people all around the world get hit by buses too…. And my comment about MLK is more an indictment about dangerous driving habits among a few people in Seattle. As I said, it’s not that Seattle is incapable; it’s instead a matter of accepting risk and designing for safer bus stop crossings.

         We’ve talked about this point before as well. trains have a harder time to stop because they are much heavier. a bus can stop much faster. Or if you are comparing it against vehicle fatalities/collisions I really fail to see how replacing a vehicle lane with a bus lane somehow has an increase in collisions.

         Typically injury collisions decrease after center running brt is installed since there are less cars anyways or combined with the restricted movements

         Even for the mlk light rail it is of course unfortunate that there were light rail collisions and deaths, but it is also completely false to say that there would be less deaths from retaining the third car lane. Mlk way had a very high number of auto collisions before the light rail was added.

         > I might even suggest that median bus lane stops are better if they’re have crosswalk signals that are only for reaching the stop or crossing the street, rather than have them combined with turning auto traffic.

         That is very rarely done. Also again as we have previously mentioned it is not as if a pedestrian doesn’t need to cross the road in the returning direction.

      9. “> I might even suggest that median bus lane stops are better if they’re have crosswalk signals that are only for reaching the stop or crossing the street, rather than have them combined with turning auto traffic.

         That is very rarely done.”

         Curitiba is full of them.

      10. @Al

          Most of them are still at the intersection for curibita’s brt. they have like 300+ stations the vast majority have street level intersections.

          I know you like the elevated mid block station, but first of all even for curitiba most of them are at intersections not midblock. they have it more to infill when the intersection spacing is large. secondly you’ve said you hated adding more stops that would slow the rapidride E/swift blue now are you saying you prefer more stops? lastly, come on didn’t you already dislike the property takings required for center bus stations. now we’d then need to build a 3 sets of elevators an elevated bridge.

          honestly i don’t even know why you keep bringing up curibita brt for elevated pedestrian overpasses. have you actually used it? they really don’t. like 99% percent of them all have normal intersections to cross.

          https://maps.app.goo.gl/BmxDT9B5NMSj5YJu9
          https://maps.app.goo.gl/ToVJHYGMXdVc6Z5C8

      11. @ Wesley:

          I did not mention grade separated or “elevated” crosswalks here. I’ve mentioned it elsewhere admittedly, but here I’m referring to signalized crosswalks specifically. I think your opinions of me are causing you to misread my post.

      12. I think for a wider arterial like Elliot Ave in Interbay, center running is actually a pretty good idea.

      13. @Al

          They build it at mid block for enough space to build passing bus lanes for the express service. and many/most of the bus stations are still at the intersection as well.

          > I think your opinions of me are causing you to misread my post.
          I mean you did explicitly talk about how you loved the elevated bus stations at mid block in the past so you can’t say I didn’t read your comments in the past lol

   2. I personally dislike center bus lanes because I often walk along the street, checking behind me for the bus, and if I spot it I wait at the nearest stop. This is impossible with center bus lanes.

      Also, the first time I tried crossing the new Madison st by car (South on 15th) I was blocked, forced to turn right on Madison, and then blocked from turning left at light after light until Broadway, where the terribly inefficient light cycle kept me waiting for over 10 minutes to turn left. My journey was lengthened by 20 minutes for the terrible detour. Now I know better routes, but my point is that it is a terrible headache for uninitiated drivers.

      1. That really isn’t any different from denny way which bans left turns and doesn’t have a bus lane west of fairview. or honestly in other suburbs with their 1 mile block grids you can’t even turn left until you drive a mile sometimes.

         Yes it can be a bit annoying but plenty of cities have left turn restrictions and also in the suburbs with their cul de sacs even without the bus lane

      2. It is certainly a compromise for stop accessibility, but if there is really a need to preserve continuous curb-side area for pickup/drop off/delivery, wouldn’t center-running be a better solution than no bus lane but still keep business happy?

      3. The lack of left turns on Madison are a bigger problem than on Denny, because between Broadway and 17th there is only a single road that can cross Madison (12th). On Denny almost all cross streets allow you to continue, makeing the 3-right turn strategy more viable.

      4. I think the big problem is just the change. People are used to driving a particular way. Then they change it and things get crazy. I’m not sure if they’ve made the change yet, but at some point you won’t be able to make a *right* from northbound Westlake to eastbound Denny. For someone obeying the law but not being aware of the change it will be a big pain in the ass. They will be forced to go across the street and then make a bunch of rights followed by a left. But once they get used to it — and once the navigation systems are aware of it — people just use alternatives. There is nothing special or unique about Madison. There are alternative streets you can use to get there. The problem is adjusting to the new way to get there.

         There are some unusual streets that would cause big detours if they ban left turns. For example consider this trip from Fremont to a Marina off of Westlake (https://maps.app.goo.gl/8kh8scCHkaWCxGp26). If they banned left turns into the Marina you would turn around at Highland Drive (https://maps.app.goo.gl/DidoFYruzRuPSKxz9) or Aloha (https://maps.app.goo.gl/ud2fAARqkDMn8RYM8). This isn’t the end of the world but it is a significant detour. But once you get used to it, it becomes second nature. This type of thing is actually pretty common with divided highways. You can’t always take a direct route but it isn’t that hard to figure out (especially with a good navigation system).

4. Using the ridership per service hour (from Sep 2025) and a subjective assessment of transit priority (partially based on this map), I created the chart below. The “Transit Priority Level” is subjective but is based on the criteria below.

   – Transit Priority Level 4: Excellent transit priority. As good as it gets.
   – Transit Priority Level 3: Good transit priority. Could be improved, but enough to provide a significant benefit during most of the day.
   – Transit Priority Level 2: Some transit priority. Enough to provide a small benefit, but can choke up with moderate traffic.
   – Transit Priority Level 1: Basically no transit priority.

   https://imgur.com/ylLwhJy

   *FHSC ridership/hour is based on 2024 data; it isn’t included in the System Evaluation
   https://www.seattle.gov/documents/departments/sdot/streetcar/2024_streetcar_operations_report.pdf

   Notes:
   – This includes all routes at or above 25 riders/hour as of Sept 2025
   – Transit priority under construction (including RapidRide I) is included; 3rd Ave is intentionally excluded.

   1. Notes part 2:
      – The trendline is meaningless here, I shouldn’t have included it; it’s drawn against a non-quantitative value

   2. Based on my own subjective analysis (hah), the routes most in need of additional transit priority are the 8 and the FHSC

      Other routes that could use additional investment are the 4, 5, 36, 44, and 62

      1. The slowest part of the 5 is how many stops it makes. There’s rarely enough traffic on Greenwood Ave to slow it down. The only big improvements I can think of are queue-jump lanes and transit priority signals at 80th and 85th to let the bus go ahead of cars. But even then the bus will almost immediately stop within the next 3 blocks and end up behind the cars again.

      2. I wouldn’t put the First Hill Streetcar in the top 10 or 20 routes that need investment in.

   3. I’m not sure there is a great way to determine transit priority. Sure, you can look at how much of the roadway has bus or BAT lanes but it is hard to determine whether they are focusing on the right area or not. Worse case scenario they add a bus lanes where they aren’t needed and neglect the most important part. One measure is on-time performance. That sounds promising but I can see some potential flaws. If a route is consistently late then they change the schedule. It will then be “on time” but really slow. I’m not sure if there is a great measure of how much congestion a bus encounters. Maybe a comparison between peak and light night travel times (although time spent picking up riders influences that).

      1. Speed limit vs bus speed. If the bus isn’t traveling full speed between stops and stoplights, there’s congestion.

      2. That would be very difficult to measure as well. Buses are constantly speeding up and slowing down to pick up riders (and stop at stoplights or stop signs).

         Turns out Metro has already done a lot of this work (see this comment: https://seattletransitblog.com/2025/12/26/seattles-transit-lanes-in-2025/#comment-976453). I’m just not sure where the report is.

      3. The bus speed is the same as the car speed. SDOT knows which streets are congested. It has vehicle counts. Naturally we don’t count when it slows down for stops, because that is the purpose of a bus. That makes it all the more important to go full speed between stops, because that’s how to make buses competitive with driving.

         So we take the most important bus corridors, which Metro has already identified by having at least 15 minute frequency midday weekdays. Intersect that with the most congested streets, and those are the ones that most need transit-priority lanes.

         But you also have go further than Metro or SDOT do, because they say there’s no congestion issue but the bus still slows down behind cars sometimes. This was the issue with Aurora and 15th Ave NW. So you go down the congestion street list further than they do, and then you get all of them. Then we’ll have a transit network where buses have priority everywhere they need it.

      4. “The bus speed is the same as the car speed.”

         I would argue that’s too simplistic, and that buses are often more prone to traffic-induced delay than cars. First, there’s the bus bunching effect, where even a minor delay cause the bus to pick up more people at upcoming stops, creating further delay. Then, there’s those situations where serving bus stops drops the bus’s average speed below what the traffic signals are optimized for, causing the bus to wait at far more red lights than cars do. And, of course, those pull-out bus stops in the parking lane, where every bus stop requires the bus to wait for an opening in the traffic to start moving again. So, letting one person off the bus involves 5 seconds to open the doors, followed by 20 seconds of merging back into traffic. Which doesn’t sound like much, but when happening over and over again, over and entire route, the delay adds up.

         Another factor is that, to make sure that buses to at least leave their terminals on time, the layover times in the schedules must be padded to account for the worst-case traffic delays, which effectively means that the transit agency must pay (in terms of how many buses and drivers to allocate to the route) for worst-case traffic conditions every single day, whether traffic is actually bad that day or not. This results in lower frequency, with the brunt of the frequency cuts happening on nights and weekends. So, how congested a bus is on Friday actually indirectly impacts how long a person has to wait for the bus on a Sunday.

         And then, of course, there’s transfers. When bus travel times are unpredictable, waits at transfer points become long and unpredictable, even between routes that are supposed to be running frequently. This is why bus->Link transfers are so much less annoying than Link->bus transfers, because Link can be depended on to run like clockwork, while a bus cannot. The transfer effect, amplifying minor traffic delays, of course, has no impact on people driving cars, as they do not make connections (except for a few special trips, like riding a ferry).

      5. Agreed that this might not be a great way to determine transit priority need. Not saying this analysis is useless, but more like demand cannot be the only factor to be looked at.

         Usually a very congested existing traffic condition (like Ship Canal crossings) will create broader benefit for a transit lane. Even if the area is consider congested, you still need to make sure the transit lane will actually help transit. If all the bus are turning left, a curb side BAT lane for the block probably won’t do much because bus need to switch to the left when approaching the intersection. Something like this make it hard to generalize a project identification simply based on demand or another factor or two. More in-depth analysis is needed.
         I think the approach today tend to focus on a specific high-demand route with terrible on-time performance (like 8), then start digging into the issue along the route.
         A speed comparison analysis won’t do much in a urban core network setup because of dense intersection spacing which slow down everything and bus making stops drags down the average speed. Analyzing on-time performance is probably a better way to identify the problem.

5. Another place that needs transit priority is along 85th and Northgate Way, cross Aurora, for routes 45 and 40, as it routinely takes the bus multiple cycles to get through these intersections, even outside of rush hour. Route 40 crossing Greenwood Ave. has a similar issue.

   The catch, of course, is the impact that such a bus lane would have on car traffic.

   1. I think Metro has already done a lot of the analysis. The hard part is finding it. This article on The Urbanist mentions it and shows a great map. But I can’t find the actual document. The Urbanist doesn’t link to it. I did find the Spot Improvement documents but they don’t seem to have that data. The article also includes the MASS proposal. This is great but obviously insufficient now.

   2. There’s the delay map for the route 40 here: https://i0.wp.com/seattletransitblog.com/wp-content/uploads/2024/07/Screenshot-2024-07-24-at-5.02.01%E2%80%AFPM.png?ssl=1

      as you noted the bottlenecks are crossing 15th avenue and aurora avenue. Unfortunately bat lanes weren’t proposed for there though potentially it could be added in the future.

      Both that delay map and the rapidride variant for route 40 were discussed in
      https://seattletransitblog.com/2024/07/29/rapidride-route-40-corridor-1993/

      1. Yeah that area is very congested, but ROW is tight. There is no room to add a additional Q-jump lane, thus signal priority is pointless.
         If it was not for the transfer to D, 40 should totally use NW 46 to bypass the intersection. I’d argue walking distance from existing D Line stops to 15th NW at NW 46 is also not that bad but definitely can use some sidewalk improvement.

      2. It is the same type of measurement but the map shows a lot more than the 40. The screen-shot has over a dozen small corridors and only one includes the 40. There is another excerpt that is focused on corridors — in this case Jackson. The latter is similar to what they show with the “Spot Improvements” document (like this one) but nowhere near as broad. This makes sense — the spot improvements are focused on areas they are actually hoping to fix in the near term, not all the areas that are congested. I’m not sure what document(s) The Urbanist referenced, but it appears to be the type of thing that lots of people (including me) have requested in the comments: A map showing how congested all the corridors are in the city and how many people it impacts*. This doesn’t tell the whole story (some areas are easier to fix than others) but it would still be useful.

         *OK, it might now show all of the corridors but it appears like it does. It is certainly a lot more comprehensive than anything else I’ve seen.

6. In my opinion, overnight traffic on 3rd is definitely an issue. I’ve seen it clogged at night with cars for mariners games an and other stuff as well as daytime for things like the pride parade, some of the most important times that it’s clear.

   1. I think the issue is deliveries. I could be wrong but it wouldn’t surprise me if there are unique delivery entrances from Third. But that doesn’t mean the hours can’t be extended. 6:00 am to 7:00 pm seems too small. Maybe 6:00 am to midnight. Or 6:00 am to 2:00 am. I’m actually kind of surprised it starts at 6:00 am given how many buses run at that hour. Maybe 5:00 am to 1:00 am. In other words, if you have a delivery you make it from 1:00 am to 5:00 am.

7. One of the structural challenges that we face in Seattle is that Metro runs most of the buses, ST runs light rail and a few buses, and SDOT maintains the streets in which they run. That’s having several cooks in different kitchens preparing one meal — each seemingly cooking the cuisine they prefer.

   On top of that, Seattle decision-makers seem nonchalant when it comes to looking at benefit or performance metrics in designing programs. They seem to want to sketch out ideas more out of imagination rather than on analysis. I commend you for trying to introduce quantitative logic into the process!!

   When looking at bus lanes, I think the most important benefit is bus speeds or travel time improvements. After that it’s probably reliability.

   This lack of using travel speed to define transit investments leads to things like the ultra-slow FHSC. Part of that is that the path of the FHSC actually crosses several other routes, many of which have more riders than the FHSC itself. When routes cross each other, transit signal priority becomes more of a zero-sum challenge. Even if the FHSC had signal priority, its overall operation probably can’t be sped by much because so much of it is delayed by both traffic on cross streets with frequent buses and crosswalk/ bicycle track signal cycles.

   I think too that the SDOT Madison curb height design mistake also needs to be called out as a case study in agency coordination failures. It’s embarrassing to have spent so much money on it only to have metal plates added at the stops to raise the bus door heights.

   Finally, the recent SDOT decision to push no right turns on red slows down buses. So does adding early walk cycles at signalized intersections. Most curb bus lanes (even 24/7 ones) double as traffic right turn pockets at intersections too so buses get delayed behind cars in that short shared right turn pocket more often. I think most of us support making intersections safer — but it can add travel time to bus routes, even when a route has mostly an exclusive curb bus lanes for the route. And it of course adds more time when buses must operate in mixed traffic.

   Would a different structural solution help, like a joint agency transit rider design review advisory committee? Should SDOT require transit priority engineers learn to drive a bus, or at least more do on-board travel time research (as opposed to a short field visit as a spectator)? Should SDOT at least survey Metro drivers to identify what and where transit travel time improvements are most needed, rather than merely originate the ideas by themselves?

   I feel like making further street changes for transit would yield more benefit for riders if the actual process itself changes to become more objective-driven and better coordinated among all the relevant agencies.

   1. “Would a different structural solution help, like a joint agency transit rider design review advisory committee?”

      I don’t think SDOT or any transportation agencies would plan bus lane without consulting transit operating partners.
      The question is when transit agencies provide feedback to the city, did they just run it by their planning team or they actually communicate with the drivers.

   2. “I think too that the SDOT Madison curb height design mistake also needs to be called out as a case study in agency coordination failures.”

      That was completely a construction mistake as all the engineering design calls out the right height.

      1. I may be wrong, but that seems to point to a process flaw in construction management oversight. Would a Metro field inspector caught the problem earlier? Is SDOT training inspectors on transit design thoroughly enough? When the first stop was installed, shouldn’t someone from some agency have tested it with an actual bus before signing off on what was installed?

         I’m sure SDOT won’t make the same mistake again — at least for a few years. But the mistake has a procedural component to it that needs to be identified and corrected.

      2. I agree. That’s the whole reason inspection exists in the process. It is disappointing that this thing was not caught before the opening.
         It was their first time to build such thing so they probably didn’t know where to inspect.

   3. “When looking at bus lanes, I think the most important benefit is bus speeds or travel time improvements. After that it’s probably reliability.”

      Speed is not very useful in this context. How do you identify the area with low speed that is actually fixable by a bus lane, especially when the task is to prioritizes bus lane implementation among several different projects?
      Low speed can be result of traffic, longer dwell time, or simply higher stop density. The latter two are not traffic issue that can be fixed by bus lane.

      Travel time improvement in transit operation essentially is the reliability improvement or travel time reliability improvement. When you separate these two things, it made it sound like bus lane warrant a higher speed limit, thus bus can travel faster even in the early morning and late night, which is not true because bus lane doesn’t really cut travel time when there is no traffic.

      Bus lane help reduce variance of travel time between peak and off-peak because bus lane bypass traffic during peak hour. This is basically a mitigation on travel time reliability issue. To actually reduce travel time, you will need to reduce stops.

      1. “How do you identify the area with low speed that is actually fixable by a bus lane, especially when the task is to prioritizes bus lane implementation among several different projects?”

         It’s possible to use the APC data to see travel imes. It can be parsed by weather or time of day. That is a pretty good way to get macro data.

         The time the doors are open versus closed is also hopefully recorded. That help separate the travel time spent with doors opened (like boarding riders) versus doors closed (like traffic congestion and signal timing).

         I would suggest sending interns or junior planners out with a running stop watch and a way to record and better understand the causes in detail to augment the APC and GPS records. I would look to seeing how many times that a bus is stopped at a particular signal and for how long, for example.

         Once the biggest time delays are identified, then the best ways to respond can be examined. Maybe it’s a full-on bus lane. Maybe it’s an intersection queue jump. Maybe it’s basic transit signal priority strategy like shifting left turn cycles to run after through movement buses clear if a bus is approaching. .lits depends on what the data show are causes.

         I actually think there is value for an occasional travel time “physical” like this for every route over time.

         But bus travel time is what I think is important. It’s part of the riders per service hour measure is calculated — as faster buses mean shorter round trip times and thus fewer buses and/or bus hours needed to provide the same frequency of service. On top of that, faster buses either less waiting can then attract more riders.

         Even if someone isn’t a transit rider, their taxes are still subsidizing transit. Researching and improving transit productivity benefits them too.

      2. I’ve been told by a transit advocate in Europe that one of the criteria they use there is transfers.

         Eg: suppose there was a half hourly train from Seattle to Bellingham. You also have a bus that runs every 2 hours from Leavenworth and connects to the train in Everett. So, you need to have the bus arrive there 5 minutes before the train arrives, and depart 5 minutes afterward.

         Does the existing infrastructure allow this? If not, then some part or other of the infrastructure needs to be improved to increase speed and reliability so the timed transfer works.

         I’ve been told the Swiss are particularly good at this, as they regularly look at transit networks on a national scale and see what isn’t working, and what needs to increase in speed to make connections better at as many points as possible. This include regional and local services by various providers.

      3. What Metro calculates (apparently) is “Passenger Delay (passenger hour) per mile per day”. Here is an example of a handful of corridors in central Seattle: https://www.theurbanist.org/wp-content/uploads/2022/06/Screenshot-2022-06-28-8.23.51-AM.png.

         It seems like this metric is sufficient. It takes into account how much delay there is as well as how many riders are hurt by it. I’m not sure if it only counts riders on the bus or those waiting but even then, it is probably similar either way. The point being that they have already calculated how much the bus is being delayed.

         What remains is actually doing something about it. It should be noted that BAT lanes aren’t perfect. There are delays from cars turning right. Thus you might have a corridor that goes from really bad to just pretty bad. That is still an improvement but it doesn’t mean the problem is solved. But quite often the challenge is just getting the money or political will to add the BAT lanes.

      4. Glenn in Portland,

         Planning works in the Europe might just be different.
         I used to work for a Dutch AEC company. Some Dutch guy told me that their travel demand model back home is calibrated/validated in extreme accuracy and precision that they are comfortable to use the data directly to run more microscope level analysis without much of post-processing.
         This quality just doesn’t exist in the US. Here we do things in similar complexity as NL but with super crappy regional demand data.

   4. I agree that communication between agencies can be an issue but I don’t think it is a big problem in this case. Look at that article by The Urbanist again: https://www.theurbanist.org/2022/06/28/rainier-avenue-bus-lane-proposal-leaves-out-metros-highest-priority-segments/. Now look at the screen-shot midway down: https://www.theurbanist.org/wp-content/uploads/2022/06/Screenshot-2022-06-28-8.23.51-AM.png. That is exactly the type of thing that you want as the first step. You’ve documented the various corridors; the amount of delay as well as the number of people effected. The next step is up to the city. In other words, Metro did the work, now it is up to the city (and specifically SDOT) to figure out what to do about it. It would be no different if the city ran the buses. Unless it is a tiny city (with one bus a day) you are bound to have two different agencies.

      Sound Transit is a different beast. It is has been established — after spending a bunch of money on consultants — that Sound Transit is *internally* dysfunctional. They have trouble communicating within themselves. Of course they also have trouble communicating with Metro, SDOT, the city of Tacoma and everyone else. The sad part is that they really should be the one bridging the gap between different *transit* agencies. Pierce, King and Snohomish County all have their own concerns. Of course their buses overlap. An agency with broad jurisdiction would be well suited to figure out solutions that involve cross-border service. But not when that agency is dysfunctional. Someone who can’t tie their own shoes is not the person you consult for help untangling a tricky knot.

8. One related topic is that WSDOT has built a fair number of HOV facilities over the years that no or almost no use by transit. For example, the ramps I-405 to the I-90 HOV ramps seem to have been built with the expectation that the 550 would take the freeway all the way to Bellevue Transit Center, except it doesn’t. The only buses I have ever seen use these ramps are the 255 and 545 on their detour route when the 520 bridge is closed.

   Another example is the ramp connecting the HOV lanes on I-405 and SR-167. The only way a bus can use it is if it connects Kent to Bellevue in a way that skips Renton completely. ST 567 used to do this, but it only ran a few trips per day, and was discontinued at the start of the pandemic. Today, the ramps are usable only by carpools and vanpools, and have no transit on them.

   Is there any communication between WSDOT and King County Metro or Sound Transit when HOV facilities are built? Both of the above are pretty expensive projects, and if the justification was helping transit, they are doing little to nothing to actually help transit, given actual bus routes.

   I hope Sound Transit is serious about funding continuing operations for the I-405 STRIDE routes, or we’ll just end up with yet more HOV facilities that have no transit on them.

   1. I think WSDOT builds infrastructure like that for carpools as much as buses. At least in the short term. In the long term they could be used for buses.

      Of course they are overbuilding in the suburbs, but that is largely political. The swing districts are in the suburbs. You make improvements in those areas to keep control of the legislature (or even the governor’s mansion). You could ignore HOV but that would be stupid. You would upset the greens and throw away future capability. If we continue to sprawl than it is the only real solution (run more buses on those corridors). In contrast if we just “add a lane” it won’t help in the least. So at worst you’ve built something that not that many people actually use but costs a fortune. Welcome to American politics.

   2. > Another example is the ramp connecting the HOV lanes on I-405 and SR-167.

      That direct connector ramp was built focused on car commuters not really about transit. sound transit move didn’t fund it.

      > One related topic is that WSDOT has built a fair number of HOV facilities over the years that no or almost no use by transit. For example, the ramps I-405 to the I-90 HOV ramps seem to have been built with the expectation that the 550 would take the freeway all the way to Bellevue Transit Center, except it doesn’t.
      > Is there any communication between WSDOT and King County Metro or Sound Transit when HOV facilities are built?

      There was a bit of a different vision back then in the 1990s. WSDOT and others wasn’t really sure about light rail on the eastside would ever get funded. The idea was for a lot more freeway express buses with bellevue transit center as the lynchpin. You can actually see part of the original express bus route on sound move 1 where a express bus goes from Seattle to mercer island to bellevue and then on to redmond.

      The problem of course was how to get buses over from the i405 southbound over to i90. it was exceedingly expensive to build a i405 to i90 center direct connector ramp. Secondly back then there was a larger vision of using right side ramps and collector distributor ramp plus hov lanes to provide the bus lanes.

      Stride 1 and 2 freeway express buses were actually originally going to be built using right-side shoulder lanes as well. but of course it was kind of mess with half of using center bus lanes (kirkland) and the other half (woodinville to lynnwood) using right side lanes. with the new etl ramps they decided to just fully convert them over to purely using center lanes.

      https://www.soundtransit.org/sites/default/files/documents/199605-sound-move-ten-year-regional-transit-system-plan.pdf

      there’s probably more detail behind their thinking but unfortunately it is not readily available online for me to find as it is all in the 1990s and not uploaded.

   3. I feel like WSDOT wants to build things like connector ramps nowadays mainly to enhance express lane toll revenue. The express lane expansions are grandiose (especially wider) when compared to a modest HOV lane installations. 405 between 90 and 167 is a massive reshaping of the freeway segment. (Hmmm maybe highways with express toll lanes should no longer be called “freeways”.)

      And it’s time that WSDOT dedicates of proportion of that express lane revenue to transit service. Right now they’re keeping it all.

      1. I feel like WSDOT wants to build things like connector ramps nowadays mainly to enhance express lane toll revenue.

         Again, that is all politics. WSDOT builds what the state legislature (and the governor) want to build. But there is an obvious conflict within the Democratic Party. On the one hand you have those on the left (especially in Seattle) who want to focus on transit. They don’t want any more freeways. On the other hand you have those in suburban swing districts who want more and more lanes. The folks at WSDOT know that won’t actually solve the problem. As a result you have a “compromise” where they build a bunch of infrastructure designed to let buses and carpools move faster. It includes toll roads because those are seen as a way to raise money without slowing down the buses/carpools. It is all considered a “win-win”.

         But that doesn’t mean it is the best use of money. But that is the nature of politics. For a little background, consider the 167/509 “Gateway Project”. It is clearly a huge waste of money. I think most people understand that. But the representatives in those areas (Democrats in swing districts) wanted it, so it is being built.

         Now consider the Columbia crossing. The politics are similar. You have Democrats representing Vancouver and they want to make it easier for their constituents to drive to Portland. But there are also those that think it is a bad idea to spend so much money just for cars. So they will likely work out some sort of compromise like a massive bridge costing a huge amount of money with bus lanes or an extension of MAX. Of course this would still be a terrible waste of money but as long as put a little “transit dressing” on it, people are OK with it. (Or at least those in Vancouver are OK with it.)

         The point is it is not being driving by the transit agencies. Of course they are consulted but rarely do they build “bus lanes” — they are almost always HOV lanes. The transit agencies aren’t about to complain (maybe they will run a few buses there, eventually) but they have their own set of projects that are bound to be different.

      2. Actually, I think the 405->167 HOV ramps are literally HOV-only, and not open to single-occupancy vehicles paying a toll. The tolled section of 167 begins about a mile south of 405.

         For what it’s worth, as the regular 167->405 ramp backs up, even on weekends, I have found the ramp useful when traveling in a carpool to places like Mt. Rainier. Still, from a transit perspective, it is maddening to see such large investments in HOV infrastructure be inaccessible to transit riders without a transit-specific replacement. For instance, to commute from Kent to Bellevue, you’re just supposed to sit through a grand tour of Renton on a bus that’s stuck in traffic. And, off-peak, you don’t even get to use the freeway to go to Renton, instead, you’re supposed to slog it out on surface streets with stoplights. The time advantages of driving over transit in the South King area are huge, and even for people that don’t want to drive, the time calculations strongly favor forming a carpool over riding the bus. The root of the problem, of course, is that the land use down there makes the area essentially impossible to serve with transit without either absurdly long travel times (except for a handful of trips along a few specific transit-heavy corridors) or an absurdly large taxpayer subsidy per passenger boarding (e.g. bringing back buses like ST 567).

      3. For instance, to commute from Kent to Bellevue, you’re just supposed to sit through a grand tour of Renton on a bus that’s stuck in traffic.

         They used to have an express bus (the 567) but hardly anyone used it. That is one of the challenges. If you skip Renton you don’t get many riders. If you include Renton you delay the riders from Kent. It is quite possible that the best option is carpools. Not everyone is headed to Downtown Bellevue. Then again, the 567 was cancelled before East Link. So it is possible it might get more riders now.

   4. Yes, and we could use more HOV + bus infrastructure.

      I-5 needs to restrict the HOV lane to bus only or 3+ during high traffic hours. This could be dynamic with a goal of maintaining 30-45+ mph speed. SR 900 needs an extra bus lane eastbound to bypass the traffic signal that gets backed up every afternoon. Route 101/102/150 should have reliable travel times throughout the day.

      I-405 needs HOV ramps from most directions into I-90. The only exception is from Bellevue towards Issaquah. This ramp pretty much never sees traffic worth building HOV for. The Issaquah-Renton and Mercer Island-Renton interchanges need them the most.

      SR 520 could use a special entrance into the toll lanes as well.

      Finally, buses should take advantage of this. More express buses that skip stops and take you between two popular points.

      For example, instead of running a route at higher frequency during peak, just run express service in addition to those routes. People who live near a bus route can time their ride and don’t need higher frequency at the expense of trip time. People who are transferring/park&riding need a faster trip.

      Route 270 in Bellevue should run 15-20 minutes as planned, but an express bus that uses 405 and 520 with only freeway stops should complement the route during peak travel time.

      Similarly Route 566 should skip Renton altogether and utilize the new HOV flyover.

      And Route 554 should be kept but revised to go to Bellevue TC via S Bellevue station. But retain peak hour bidirectional 556 service that uses I-405 to directly reach Bellevue TC, then continue to Totem Lake TC.

Comments are closed.