We’re about a month into the Fall shakeup, where a substantial number of service changes were implemented to improve connections to Link. This is a recurring playbook that Metro has used ever since light rail service began: identify duplicative services and restructure existing bus routes to connect to Link stations rather than running all the way into downtown Seattle.
When Northgate Link opened, Sound Transit truncated many of its 510 series routes to connect at Link instead. After Lynnwood Link, all of Community Transit’s direct services to downtown Seattle were restructured. And with the August changes, both Metro and Sound Transit no longer offer any one-seat rides from Bellevue or Issaquah to Seattle, save for the 2 Line.
Critics of this approach generally lament the loss of the one-seat ride, that trip times generally increase when factoring in wait times, and that Link is not necessarily faster point-to-point than express routes. The Sound Transit subreddit and other various community forums feature countless examples of riders testifying to that end. There is even now a “Restore 554” petition circulating among Issaquah residents, urging the restoration of a direct Issaquah-Seattle route.
The planning justifications for this type of restructuring are quite straightforward: Link provides a frequent high-capacity serve to major regional centers; instead of running costly peak-directional bus service, reinvest the hours into boosting local connecting service. This implementation aligns with Jarrett Walker’s “frequency is freedom” mantra and is generally a position this blog has supported.
The plight of congestion
In practice, however, real-world conditions can muddy the picture. Take, for example, the 554’s successor: the 556, which now provides frequent all-day service between Issaquah and Bellevue. Those previously using the 554 (or the 550 local stops along Bellevue Way) to get to Seattle must now connect at South Bellevue Station for the same trip.
These transfer conditions can greatly deteriorate due to local northbound congestion on Bellevue Way just in the short span between the I-90 off-ramp and the station bus stops. Riders agonizingly wait for traffic to crawl up Bellevue Way only to see Seattle-bound trains glide past over their heads. Even in the southbound direction, buses have to yield to northbound traffic in order to turn into the station bays, which adds travel time variability.
Constraints of connecting infrastructure
The physical infrastructure in between the connecting services is important too. Jarrett Walker touches on this point in one of his foundational articles on transferring:
… the unpleasantness of the connection experience depends on many details of how the connection works. If two buses or trains arrive on opposite sides of a platform, facing one another five meters apart, with their doors open at the same time, walking out of one and into the other is a pretty low level of inconvenience for most passengers. If the connection involves getting off a bus, crossing a busy street, and waiting for another bus not knowing when it will arrive, the inconvenience is much greater.
As it were, the connecting infrastructure at South Bellevue and Mercer Island stations is less than ideal. Both settings require a moderate walk (across a signalized intersection in the latter case, largely thanks to Mercer Island’s opposition to an 80th Ave transfer point) and ascending or descending to the train platform. This can add a lot of variability, particularly for those with limited mobility, on top of the wait for the connecting service.
Tradeoffs across the system
Negative feedback is often amplified over positive feedback, as human beings are far more likely to complain about things that hurt them rather than praise the things that help them. But in many ways, what is often seen as a degradation in one part of the network typically means an improvement in another part.
554 riders, for example, might glowingly reminisce about the quick travel times across I-90, but forget about the frequent outbound reliability issues inflicted by 2nd Avenue congestion (as I so often personally experienced). And although not quite a zero sum game, the changes have meant a substantially better commute for Issaquah-Bellevue riders, among other parts of the system.
Big restructures in the regional network will invariably result in system-wide tradeoffs, but that doesn’t mean agencies’ hands are tied. Barring a miraculous way of sending buses congestion-free right up (or down) to the Link platforms, transfer penalties can be offset by improving frequency and reliability across all segments.
Making the 556 a frequent all-day route that connects to frequent 2 Line service is certainly an excellent start, but continued work on reliability improvements are always in order. This might mean more attentive maintenance to ensure fewer breakdowns of Link trains. Or it could mean stronger partnerships with WSDOT and local jurisdictions to give signal or lane priority to buses.
Whatever the solution, transit agencies cannot go it alone here. Route 8’s dramatic reliability leaps were only possible by some repainting (and other minor infrastructural improvements) from SDOT. In the same way, local jurisdictions owe it to the taxpaying public to squeeze out the full return on our massive regional investment in transit. Helping create reliable well-connected services is a small, but important step to get there.
And then there’s this weekend, when if you want to go from Westlake to Issaquah you are expected to take Light Rail to ID, transfer to a shuttle bus to S Bellevue, and transfer again to the 556
Route 255 riders know this scenario all too well
In normal operation the forced transfers to Link are OK. But when portions of Link are replaced by shuttles, the expected routing becomes a farce. ST/Metro could come up with a more sensible rerouting in those scenarios.
Unfortunately, this is kind of an unsolvable problem. There are several options that come to mind, all of which have drawbacks:
1) design the bus network to ignore Link. The cost: less frequency on the buses all the time, even when Link isn’t disrupted, and/or extra transfers and/or backtracking required to reach non-downtown destinations.
2) special bus network when Link is disrupted. The cost: riders would be confused, people would wait at the wrong stops.
3) thru-route Link shuttles with connecting routes to minimize transfers. The cost: would cause Link shuttles to run with irregular spacing, leading to crowding and/or longer waits for people who aren’t making connections. Would also not work when the connecting route continues onward in a different direction.
Ultimately, this is a problem with no good answer.
The answer is to improve Link resilience so that it doesn’t need to close as often.
The 255 was victim of several things right after each other: the covid Link reduction, 520 construction, and the bus driver shortage.
It’s a consequence of how Link has not planned for disruptions. What I see emerging is a Capitol Hill — Judkins Park — Mt Baker shuttle when Link is disrupted. It’s a darned shame that BLE doesn’t cross for transfers at Capitol Hill and continue as an automated Metro 8 subway to Mt Baker in the long term. Instead we get DSTT2.
Except there isn’t a perfect solution for Kirkland because there isn’t a Link station outside Downtown Seattle more accessible from SR 520, but there is Mercer Island for 556.
556 would serve those going to work in Seattle better if it terminated at Mercer Island, but it would be much more difficult for those who go to Bellevue to use it. It is a dilemma and I think WSDOT and transit agency should do something to improve transit access to South Bellevue from east of I-405.
It might be true that bigger share of Issaquah residents work in eastside, but I don’t think that proportionally translate to more transit ridership. Issaquah residents working in Seattle is far more likely to take public transportation than those going to work in Bellevue. It occurs to me that 554 before this fall is more popular than 556 is today. Not sure how long it will take to change that.
Per LEHD, commuting to Issaquah is mostly from Seattle (8.6%, 2.2k) and points south (Renton and Kent specifically). Bellevue is 4th (behind Seattle, Issaquah, and Sammamish) at 5.2%. Coming from Issaquah, Seattle (31.7%, 5.4k) outpaces Bellevue (14.6%, 2.5k) and Redmond (12.4%, 2.1k) combined. It’s not surprising to me that Seattle-oriented commuting is more important to people in this context, and I think the 556 is a sort of compromise solution that doesn’t serve Issaquah-Seattle commuters all that well. Granted, LEHD may have some methodological issues for this specific corridor, as I am not certain how they handle firms with multiple major locations (i.e., Amazon in Seattle and Bellevue).
Of course, it serves everyone else better by sheer virtue of running more frequently, but I don’t see why ST couldn’t have kept the 554 as commute hours only (maybe truncated to Issaquah TC), or why Metro couldn’t run the 218 from Issaquah TC to downtown for the fairly large share of Issaquah-Seattle workers. Outside peak hours, the existing set up with the 556 is quite good, but I am sympathetic to folks who feel like they got the short end of the stick here.
Yes. I think they should have kept 554 at least during peak.
Saying 556 is an improvement to Issaquah is mostly because it runs ever 15 minutes, not because it is a better route than 554. If 545 can stay, I don’t see why 554 cannot.
Besides 554, 212 was also a popular route during peak period. I imagine the end of one-seat ride from Eastgate P&R to Seattle might push some people to drive to South Bellevue to catch light rail directly. It doesn’t seem very efficient to drive to Eastgate for a 2-seat ride to Seattle.
Does it affect 218 riders much to drive to Issaquah Highlands P&R instead of Issaquah TC? They can’t be walk-ups because Issaquah TC is in the middle of nowhere.
Yes, I think Eastgate and Issaquah TC are the general losers in this service change. Maybe that’s fine, but I think we could have gone to peak-hour only service first before a total cut.
And Mike Orr, the Highlands and the rest of Issaquah are very poorly connected. If you live in Talus or downtown, trekking up the hill to the Issaquah Highlands P&R is probably just as much driving time as heading to Eastgate. This maybe an exaggeration, but they aren’t very interchangeable.
Route 554 between Issaquah and Mercer Island should replace Route 556. We should not ask for Route 554 between Issaquah and downtown Seattle.
The difference between the 554 and the 556 is coverage on Bellevue Way. If you want to switch the Issaquah to Link connection over to MI Station, fine, but then how will you backfill service along Bellevue Way south of downtown Bellevue?
Yes, I think Eastgate and Issaquah TC are the general losers in this service change.
I’ve been writing about the Issaquah situation since this was first proposed. The Issaquah Highlands come out way ahead. Central Issaquah does not. From the Highlands you have a fairly frequent set of buses that will quickly get you to Mercer Island, which is the fastest way to get to Link if you are headed to Seattle. You can also get to Eastgate/Bellevue College. If you are headed to Downtown Bellevue then you can transfer at Eastgate or take the bus through Central Issaquah directly there. This is a big improvement for the Highlands as the old all-day option was the 554. While this was a direct bus to downtown it is quite possible that riding to Mercer Island and then taking Link is substantially faster (since the 215 skips the rest of Issaquah).
It is a different story for Central Issaquah, especially Issaquah Transit Center. On the old 554, that was the last stop in Issaquah for those headed to downtown. Now they have a bus that instead goes to Downtown Bellevue. This adds value but for those headed to Seattle it is a significant downgrade compared to Mercer Island. I can see why they are disappointed in this change.
Eastgate is basically a wash. The 554 ran about every twenty minutes midday. The 215/269 combination runs every fifteen minutes. Riders also have the option of taking the 556 to South Bellevue (if they miss the other bus). This helps make up for the transfer. It is also easier to get from Eastgate to the south end of Bellevue Way. Overall it seems about the same.
how will you backfill service along Bellevue Way south of downtown Bellevue?
I would probably just shift the 240 over there. I don’t see the need for service along 112th between Bellevue Way and Main (where the 220 comes in). The stops in the area that were once served by the 240 had very few riders (about 30, total). (See this map where you can click on the individual stops). It is worth noting that you would likely save service. It takes about ten minutes longer to get from Eastgate to Bellevue TC than from Eastgate to Mercer Island.
I’m curious to see ridership numbers for the 556. I would consider anything around 20 boardings per hour reasonably successful. For reference, in 2024 the 550 had 26.6; the 554 had 22.3; and the 566 had 14.0.
One date to check for this connection: Sound Transit currently lists the Int’l District/Chinatown–South Bellevue 2 Line shuttle replacement for October 17–18, with shuttles every 10–15 minutes. If you mean a different weekend, that particular notice may not apply. The live alert page is here: https://www.soundtransit.org/ride-with-us/service-alerts . This is an AI-assisted check of the current official notice, not a firsthand trip report.
As someone who is an infrequent visitor (every other month?) to Issaquah (and a former resident), I think the 556 is a massive improvement over the 554. The 554 had a pretty miserable weekend and off peak schedule relative to the 556. I used to have to plan trips from my Beacon Hill apartment pretty carefully to get the right bus to transfer to the 554. Now I can just walk down to Judkins Park and get the first train without much consideration. It makes me more inclined to go to Issaquah in general.
That said, I think a peak-hour express route serving Seattle should have stayed. My memory from riding the 554 in the mornings when I did (I preferred the 218 as I was staying in the Highlands, but I took the 554 too) is that it was very well used from Issaquah TC. I think the 218 should have had a stop at the TC added and retained its downtown leg, as this would have also retained downtown service to Eastgate. I imagine the Eastgate garage is one of the least well-used in the system now. Who parks and rides for a bus-to-Link trip when there’s garages for Link? This would be a fairy small investment of service hours, but it would have improved most Issaquah and Eastgate commuters view of the Eastlake restructure.
Keep in mind that the 556 is not the only Issaquah-Seattle route. There’s also the 215/269, which goes to Mercer Island, avoiding the bottleneck at the Bellevue Way ramp (which is really only an issue during the morning peak).
I think that’s good enough. In practice, a bus that went all the way downtown would spend about as much time waiting at stoplights (if going all the way to Westlake) than would be saved by avoiding the transfer. Such a bus would also get stuck in traffic downtown for the evening commute, especially when there’s a Mariners game. I don’t think it’s worth it.
Neither the 215 nor 269 stop at Issaquah TC. Service from the Highlands is quite good – and much better than it was pre service change, particularly off-peak as skipping downtown Issaquah is a big time savings. Overall, I personally prefer the Link connection, but I do think that between Eastgate and Issaquah TC, there’s enough there for an express bus even if it’s a duplication with Link west of Mercer Island. These are large, existing garages that may as well see at least one express to Seattle. I would take say that people petitioning for an express bus to return is evidence that it would be utilized, and something like 4 or 5 daily peak-hour round trips from Issaquah TC to downtown via Eastgate only is worth considering in that context.
It’s not like there’s perfect fidelity region-wide to non-duplication. The 510 still exists (and seemingly always will), as does the 106. The 594 and 574 overlap in a lot of ways too. Most other parts of the region have some form of an express bus to Seattle, even if its imperfect or even wasteful. Why should Issaquah be different?
The problem is, service hours is a zero sum game, and peak hour service is especially expensive, since every single trip you add requires buying a new bus and hiring a new bus driver. Not to mention half the mileage on this new bus being deadhead miles.
What service would you cut to pay for this? Would you end up the 556 earlier in the evening? Reduce it to every 30 minutes on Sundays? Something would have to give.
Sure, but something like moving the already peak-hour only 218 (Highlands – Eastgate – Mercer Island) to run (maybe from the Highlands -) Issaquah TC – Eastgate – Seattle isn’t a massive additional investment. Even if it’s 50% longer, a 50% cut in runs (from the current 8 down to 4) would be worthwhile, especially since the 215/269/556 already cover the route. I’d rather have the 218 serve a psychologically important route (Issaquah – Downtown) than a peak-hour duplicate route.
@asdf2
The 218 already provides duplicative express transit service. Reduce its span/frequency and run it to Seattle. Riders outside of peak-of-peak can take the 215/269 and transfer at Mercer Island.
You raised a good point about the 218, as I forgot that that route still exists.
Nonetheless, there is still the issue of opportunity costs. Given that the 215 and 269 already provide frequent service on the corridor, the purpose of the 218 is strictly about peak hour capacity. If the number of trips can be reduced to offer a one seat ride into downtown, then the number of trips can also be reduced to redirect buses to some other corridor, that needs the service more.
For example, one could also make the argument that some peak-hour 556 trips should extend to the U district, as it would save time over the 270, and the 270’s peak frequency is only every 15 minutes.
So, to justify a one seat Issaquah to downtown ride, you have to make the case as to why it is more important than, not just this, but all other underserved corridors in the system. Without that, it’s essentially just a shouting match where everyone believes that their individual commute is the most important, and most deserving of resources, and that other people should endure a longer commute to make theirs shorter. That’s not a good way to run a transit system.
I have a hard time believing that peak hour capacity in the Highlands is that much of an issue. In the last service change, when the 218 was still the sole Highlands express route, it got 343 daily riders with about 3/4ths of those being Highlands – Downtown and the other 1/4th being Eastgate – Downtown (also served by the 212 and 554 at the time). This was on (I think) 4 round trips per day. Comparing that with the 554, I see marginally more ridership in the AM peak at Issaquah TC, though on more buses.
This (plus my comment in a different thread about commute patterns) leads me to two conclusions: that Issaquah Highlands is a stronger transit market for a downtown Seattle connection. I think that this is reflected in the service change. The 215 and 269 provide high-quality, all-day service with a relatively better connection point at Mercer Island. But this is still an over correction. The 554 got pretty strong peak hour ridership from both Issaquah TC — which I think indicates, along with prevailing commute patterns, that at least some Seattle-oriented service should have remained.
ST hasn’t ever cut a bus like the 554 before and it’s not like the 2 Line is the first rail expansion project. I think if Redmond keeps the 545, if Everett keeps the 510, if Federal Way keeps the 577/8, and if Tacoma keeps the 594, then Issaquah really should also keep the 554. It just feels arbitrary to cut this one and leave the rest. Buses run alongside trains on every corridor that trains run except for I-90.
Neither the 215 nor 269 stop at Issaquah TC. Service from the Highlands is quite good … Overall, I personally prefer the Link connection, but I do think that between Eastgate and Issaquah TC, there’s enough there for an express bus even if it’s a duplication with Link west of Mercer Island.
I think it makes way more sense to just run a peak express from central Issaquah (including Issaquah TC) to Mercer Island. That way those in central Issaquah can avoid the issues with serving South Bellevue (they get to live like those in the Highlands — at least during peak). There are issues with that though. A rider going the other direction would have to know the schedule. Otherwise they won’t know whether to use the station at Mercer Island or the one at South Bellevue. The other issue is that it might exceed the limit Mercer Island has on buses there.
Of course the other big argument is that it just isn’t worth the money. Sure, there are similar buses, but that doesn’t mean those are a good value either. One of the main reasons the 510 exists is that ST has had trouble coming up with alternative routes in the north subarea. Ridership on some of the routes has been terrible. If they were more creative then I could see them coming up with better routes. That might mean working with both Community and Everett Transit. Given they haven’t worked well with Metro (their main partner) that is a big ask. In contrast, the ST routes on the East Side have a bunch of riders and clearly have major needs (like running the 542 every fifteen minutes).
As for south end buses, the problem there is that there isn’t a great terminus, other than Federal Way. Federal Way has HOV access from the freeway making it a good terminus for buses coming from the south. But it is a really long trip to Seattle from there. Thus an express saves a huge amount of time. A similar issue exists at Rainier Beach. You could terminate there but you have to spend a lot of time getting there and then when you do, it will take a while before you get to downtown. In contrast, Mercer Island is easy to get to (the bus never leaves the HOV lanes) and a quick, straight shot into downtown (with only one stop in between). Boeing Access Road was originally designed to solve this problem. Not only was it meant to connect Link to the freeway but also to Sounder. Thus a bus from say, Renton, could terminate there and riders could get to downtown relatively quickly. Yet the current board has forgotten the whole point of the station and now it will be nothing more than a suburban station with a parking garage in a low-density area. Without a multi-modal station it makes sense to continue to have express buses from the south but not the north and east.
As for the Eastgate garage, it isn’t that much different than Ash Way. It is a parking garage with very frequent connecting service to Link. Some people park there because it is easier than driving all the way to South Bellevue (and they don’t worry about finding a parking space). If this is not the case and people just don’t use it then it should be sold off.
” One of the main reasons the 510 exists is that ST has had trouble coming up with alternative routes in the north subarea… In contrast, the ST routes on the East Side have a bunch of riders and clearly have major needs (like running the 542 every fifteen minutes).”
Should East King have been two subareas? Then one would have one set of needs, and another would have another set of needs, so Issaquah and Renton wouldn’t be competing against Kirkland and Redmond for resources.
But the Eastside was far different in the 1990s when the subareas were created. A larger proportion commuted to Seattle, reverse-commuting was unusual, inter-suburb trips were fewer, and Issaquah was a small town.
Should East King have been two subareas?
The subarea concept was a really bad idea. So too was having Sound Transit build the mass transit system instead of Metro. Either Sound Transit should have incorporated Metro (and the other transit agencies) or Metro should have built it. Having one agency handle most of the buses in the region (and thus most of the ridership) while a different agency builds a multi-billion dollar mass transit system is a bad idea. That doesn’t mean that Link couldn’t have eventually gone across the county border (to serve Lynnwood) but that would not be the priority (let along getting to Everett).
One of the oddities of the recent restructures is how riders from Stanwood, Marysville, Lake Stephens, and Snohomish now have express routes to Lynnwood Station, and Everett does not.
Sure, Everett has a bus route to Lynnwood Station, but it is a hard stretch to call it “express”. Gobs of service hours are being wasted all day, every day, to make the Everett-Lynnwood-Station connection intolerably slow.
Let ST Express 512 become a fast express option, while CT 201/202 does a better job serving Ash Way (not just the P&R) and Mariner P&R area.
A restructured 512 that bypasses Ash Way could run more frequently, saving ca. 25 minutes every day in the lives of Everett-Link commuters. Compare that to Ash Way P&R commuters, who would have fewer than 8 minutes (counting both directions) added to their daily commute. Given the random timing between those two routes and the availability of ST Express 513, it will probably cost Ash Way commuters much less than 8 minutes a day.
Also compare the service levels between Ash Way and Lynnwood Station (8 express trips per hour each way off-peak) with that between Tacoma Dome Station and Federal Way Station (4 trips per hour per direction off-peak).
Brent: this is due to ST choosing to provide a one-seat ride on Route 510 all the way to/from downtown Seattle. I expect they may store the buses in a SODO yard during the day. Otherwise, they would have long ugly deadhead trips on the I-5 general-purpose lanes; the operators may go back; CT used to do that. ST could instead provide shorter waits by running Route 510 to/from Lynnwood. It seems odd that ST is the more conservative agency. CT and Metro have truncated more.
I recommend the Golden Rule of Link integration; ST should integrate with Link as it would have its partner agencies integrate. The choices seem hard sometimes: transfers v. waiting. It all breaks down to minutes. In ridership forecasting, waiting is what riders like least, so improved frequency is best. Hence the Jarrett Walker motto, frequency is liberty.
ST, with its 2027 SIP, will continue being conservative. Our hope for shorter waits is in fall 2028! Opportunities are to consolidate routes 542 and 545 on SR-520, truncate Route 510, and consolidate the Pierce County routes to meet Link at Federal Way. A key to good bus structure is to use the Sound Move center access ramps (e.g., NE 128th Street, NE 6th Street, and 317th Avenue South) and the WSDOT HOT lanes and center HOV lanes.
I left the 510 out of my comment since bringing it up seems to make Everett politicians close their ears. Even leaving the 510 as is, removing Ash Way from the 512 is a huge improvement that should induce more ridership, and a big cost-saver as well. The savings might even prolong the life of the 510.
Yes, this. It is really two different issues: The 512 Should not stop at Ash Way. Whether we should keep the 510 is a completely different matter.
The 510 is more related to North Sounder. In my opinion we should keep the 510 but get rid of the North Sounder. Sell the space back to BNSF or sell it to Amtrak/Washington State. Then use the savings to run the 510. Also run the bus to Mukilteo more often (if they need to). The big thing to do is run a new (all-day) route from Edmonds to 185th Station. This should be an all-day route running every fifteen minutes. The savings from abandoning North Sounder should (hopefully) pay for all of this.
Baby north Sounder had only four trips per peak( is it relevant?
It is relevant because it is (still) very expensive. They could do a lot with that money. Not that many people ride it but it is still a train. They might very well have to add more trips for the 510 and 512 (during peak) just to deal with crowding if they get rid of it. I think they would still come out way ahead. Running the trains is very expensive and the time slots (they leased from BNSF) cost a fortune. Before the pandemic I remember costs of somewhere around $100 per rider — it may be worse now.
If ST is going to consolidate 542/545, it should probably put it all into the 545, and let those riders destined to UW transfer to 255 or 270 at one of the SR-520 freeway stations or backtrack on Link. There is already lots of SR-520 service to UW and downtown Seattle is also an important destination.
The 545 serves the Amazon/Denny area much better than Link at Westlake, and providing one 7-day route on SR-520 to downtown Seattle creates important redundancy when Montlake/UW are gummed up which happens often enough with sporting events, graduations, Opening Day etc.
If you want to consolidate the hours to 542, then restore the 255 to downtown Seattle. One or the other route should provide service to downtown imo. That also leverages the investment made in the SR-520 freeway stations.
Feed Link.
@Carl — If Link didn’t exist then that approach would make a lot of sense. But it does. The problem with sending buses to downtown is that it is expensive. It means that other buses don’t run as often or at all. This is justified when the time savings for a direct bus to downtown is huge and/or the transfer point offers little else. That isn’t true with the UW Station. It is a big destination in itself (the second biggest in the region). It connects to Link which goes to many other places (not just downtown). Most importantly the transfer doesn’t cost riders that much time. It takes a while to get from the bus to the train platform but the buses are fast and frequent. They also travel through downtown quite quickly. Spending money giving some riders a slightly faster trip to downtown just isn’t worth it.
In my opinion the biggest problem with our transit system is that the buses don’t run frequently enough. Running more express buses to downtown will just make the situation worse.
What are some of the best and worst bus to Link truncations?
I don’t know that it is the worst but I think the 255 was handled fairly poorly. It should not have been truncated before work was complete on the 520/Montlake exit. It was stuck in traffic due to no center exit and it was constantly disrupted due to construction
Yes, Montlake was disruptive. But that should be compared with the congestion faced on the I-5 general purpose lanes and in downtown Seattle.
Route 255 does use the SR-520 center access ramp on 108th/112th avenues NE at SR-520. It is new Route 270 that does not use.
Initially, the outbound Route 255 was assigned to southbound Montlake Boulevard (UW Link station bay 3). This was a mistake, as Route 255 buses had to wait for a break in traffic to reach the lanes going south to Montlake; when waiting, they delayed Route 44 trips that used the outside lane to go east on NE Pacific Street. Metro changed the pathway last year. In 2020, planning mistakes could be hidden by low Covid ridership.
Note that the March 2020 restructure was too narrow; routes 545, 237, 249, 252, 257, and 268 were not included.
The center exit lanes were not open when the 255 was routed to UW. For many years it was very, very consistently stuck in westbound Montlake exit traffic. On top of that every few weekends (and many weekday evenings!) would have a Montlake closure that rendered the 255 unusable.
The bus should have been truncated but the timing was terrible.
A lot of local service in areas served by Link has improved considerably now that buses don’t have to travel all the way downtown. North Seattle used to have almost no frequent service before ULink or Northgate Link, and now has a pretty extensive grid of frequent service. I think that benefit gets forgotten when we talk about the loss of one seat rides downtown. Also, in more congested parts of the city (like anywhere north of Seattle), dropping bus riders off at Link often leads to faster travel times downtown, not slower travel times.
That being said, the reduce in express service downtown from NW Seattle has surprised me (seeing as there’s no Link service to replace it).
I feel that there is a high correlation between transfer difficulty (as well as reliability) and public support for the feeder concept. Bad transfers can feel onerous when replacing a direct trip.
In the case of Downtown Bellevue, a rider leaving a bus has to walk down stairs or take the elevator to get to the Link platform after crossing 110th. And it’s something like 50 or so steps or three floors down if it was in an office building. That is because ST believes in forcing people to walk downward at most stations, and removed down escalators through last phase budgeting cutbacks at many stations. And they didn’t design things to enable easily adding them back later.
I don’t recall a Bellevue TC diagram prominent at the Link station entrance the shows where different bus stops are either.
And if you think that’s bad, look at what ST wants 1 Line riders to do in Downtown Seattle in DSTT2 and at SODO to merely transfer to another light rail train. Those pretty ST diagrams look streamlined, but the ruder experience will be really, really awful.
Ironically, transferring is easier at South Bellevue. Southbound buses have an exclusive left turn picket that lets them turn to then stop under the platform.
So any discussion about losing direct service needs to also discuss transfer difficulty. They should be inseparable issues for a transit rider.
Imo the 556 is better than the 554. I nearly hated the 554 because it just felt boring and stale and infrequent, I love the change!
This is a good example of why cities should not have veto power over transit infrastructure. The city of Mercer Island imposed an inferior bus connection at their Link station, and Bellevue won’t put bus lanes on Bellevue Way, and this primarily disadvantages people who don’t live in either city.
“This is a good example of why cities should not have veto power over transit infrastructure.”
What? That’s the whole point!
Sherwin: nice piece. The ST and Metro planners needed this a few years ago. I tried.
In general, east-west routes should meet the 2 Line at Mercer Island station to use the WSDOT center HOV lanes and the Eastgate freeway station and north-south routes could meet the 2 Line at downtown Bellevue, Main Street, or South Bellevue. The ST and Metro planners got focused on South Bellevue station, probably out of habit. Up until the 2 Line crossed the lake, that was where the radial service was, Route 550. Route 550 has been there due to the ramps and the large P&R. There is no village; there is no density; the pedestrian infrastructure is poor and hostile. ST has provided bus layover. South Bellevue is actually an awkward point to meet the 2 Line; riders have to go out-of-direction. Bellevue Way SE has congestion; the ramps to eastbound I-90 have congestion; the buses have to weave to reach the center HOV lane to/from the east; the east to south ramps used by Route 111 are especially congested. (See Google Maps and ask for congestion feature; it shows different times of the typical day).
The Issaquah petitions could ask for a better answer. It is not Route 554 to/from Seattle, but Route 554 to/from Mercer Island. (They should want Link). That would save most riders at least five minutes per trip over Route 556 of today. There are two reasons for the time difference: connecting at South Bellevue involves out-of-direction travel and the eastbound trips face congestion on Bellevue Way SE and on the ramps. There is probably more variation in running times from congestion. The AVL data on new Route 556 should show the issue to ST and Metro. ST could ask the public about this Route 554 for the fall 2028 SIP; their outreach to date has not asked about it. The fall 2026 Route 556 is a service consolidation but a weak one due to the congestion and out of direction travel. ST Route 556 covers the arterial segment of former Route 550, but there is no reason that market has to have a blue bus; it did not before Route 550; it had former Route 226 after fall 1997. It could be served by Route 240 (or routes 532-535).
Since the teens, when ST began constructing the 2 Line, Route 554 riders have had slower travel times: they lost the center roadway on the bridge; they lost the D-2 roadway; and, they were sacrificed to serve the Rainier Avenue South stops. Together, those meant several added minutes. In about 2010, the Route 554 Sunday service was reduced due to the recession and it was never restored after recovery. When the I-90 served was slowed, it took more buses; ST should not have been surprised, but seemed to be unprepared and cut trips to fit into the fleet size.
I expect new Route 111 will also have issues. There is not a good southern terminal and layover. They seem to be attempting a live loop. So, the variable lateness imposed by congestion on the east-to-south ramps between I-90 and I-405 will be passed on to the northbound trips. IMO, this stems from Metro not setting up the project scope well enough; Route 105 should have been included; then the route extended to the 2 Line would have had the Renton TC as a terminal and anchor; the route would have had markets at both ends.
The Mercer Island passage is incomplete. Both 77th and 80th avenues SE were relevant. The bilateral agreement between ST and MI prevents off-island buses from going south of I-90. First, the MI rationales were specious. Second, ST should not have bilateral agreements; the transit operator should be at the table at zero percent design. This mistake is on the elected leaders and managements of both transit agencies. As Link expands, ST should have the transit operators at the table. But even with the MI agreement, a better network could have been provided by hooking intra island local routes to off-island routes. Two concepts I suggested: Metro routes 245-212-204 or Route 269 could have extended to Mercer Island high. They could have served the south station entrance on 77th Avenue SE. Bus stops would have to be added to 77th Avenue SE. It would reduce the transfer walk to/from Link. Note that MI complained about riders having to cross Island Crest Way to transfer; but the restrictions of their own bilateral agreement caused it! Route 630 is backward; it ignores Link and does not integrate.
In rail integration, some planners get focused on inbound trips, but outbound trips need attention as well. The outbound trip can be more difficult as the riders may be going to a local bus route with a longer headway. A useful diction would be to say “connect with Link” rather than “connect to Link”. Think of both directions. The local bus headway get longer at off-peak times. The outbound trips are relevant to the South Bellevue v. Mercer Island choice as well. With the fall 2026 network, outbound 2 Line riders will face a choice of alighting at Mercer Island or South Bellevue. The waits for bus routes will differ by time of day. The waits could have been shorter with more service oriented to Mercer Island. An Eastgate rider could use MI routes 215, 218, or 269 or South Bellevue Route 556; but the rider must make a choice. An Issaquah TC rider will probably choose Route 556 at South Bellevue.
The integration network used for East Link modeling in the aughts had Route 554 oriented to Mercer Island. The Route 556 concept arose in more recent years. The planners must not have noted the congestion or used the “pony race” test of measuring the travel.
“Route 630 is backward; it ignores Link and does not integrate.”
Can’t have all those pesky First Hill to Bellevue commuters taking advantage of their dedicated and funded bus route…
Some of what MI has done hurts MI residents.
It would be nice for MI residents to be able to hop on a bus in the populated village area and have a one-seat ride to Issaquah and other east destinations. Doing this does nothing to attract those dirty Seattleites.
But MI does have a bus that can bring those unwanted Seattleites to MI in the evening: the 630!
All that said, it will be interesting to see the data on how many east I-90 riders are trying to get to downtown Bellevue vs Seattle. It may sober the eastside up on whether the Link 4 Line serves much of a purpose.
One way to look at this is, there are people with cars and without cars that have two different needs. The people with cars only really care about transit to downtown because they expect to drive everywhere else. These people will mostly drive directly to Link, not drive to a bus to Link, so the restructure is not even really an extra transfer. It’s just a tradeoff between needing to drive a few minutes further up I-90 to get to the transit, in exchange for the transit being more frequent and reliable, and offering a one-seat ride to the UW, in addition to downtown. The only way a person with a car even considers taking a bus to the train is if they’re worried about the parking at South Bellevue being full.
For people without cars, quicker bus access to Bellevue and more frequent bus service throughout the day, in both directions, is more important than a difference of a few minutes travel time to downtown.
If there were a case for instituting a peak-hour version of the 554, it would require conditions like South Bellevue parking garage filling up nearly every weekday, traffic being chronically bad on I-90 in SOV lanes around Eastgate, and the 2-line itself being at capacity.
So far, I haven’t seen signs of any of these conditions. Although, I have observed a full south Bellevue garage one Saturday evening, when both the mariners and Sounders were playing at the same time.
“it would require conditions like South Bellevue parking garage filling up nearly every weekday“
Bellevue Way off-ramp from I-90 WB will be packed before the parking garage is filling up.
What I don’t understand is why didn’t they send one route between 215 and 269 to South Bellevue. The two all-day routes are identical west of Issaquah Highlands.
As a weekend 554 rider the 556 is way better. 554 was prone to getting stuck in event traffic and had limited morning service. I’ll take the few minutes of transfer time over getting stuck in traffic. Also the routing thru the ID was a waste of time. It would look really lame to have a bus running along next to the train on I-90. Adding a bus lane to Bellevue way is the way to go.
Problem is the bus is faster than the train on I-90 about 90% of the time. These changes generally have led me to feel frustrated. Why are the trains so slow? Why are we truncating the buses to go into traffic when we’re finally getting to the part of the freeway without traffic? I feel like there’s so much potential but my commutes have become worse. I’ve traveled more miles in my car than before.
Problem is the bus is faster than the train on I-90 about 90% of the time.
That is true of I-5 as well. Link can’t compete with express buses. Riders of the old 41 dearly miss it. They avoided a transfer and their bus was faster.
The thing is, the bus isn’t that much faster. It just isn’t worth the extra money to run downtown when Link does a very good job of that. It is just not a good use of limited funds to spend money on express service of this nature. This includes express on more popular routes like Aurora as well. The old 355 skipped much of Aurora and got riders from Aurora Village to downtown quite quickly. It was popular but is no longer worth it. We are much better off just running the buses more often.
Complicating matters in Issaquah is that the buses run express to and from the Highlands (connecting to Mercer Island Station) but Central Issaquah has to deal with the slower connection at South Bellevue. That is just ST favoring the connection from Issaquah to Bellevue over the trip from Issaquah to Seattle. It is worth noting that Issaquah Link does the same thing (and it will cost a bundle to actually make it happen). But there is nothing inherently wrong with truncating before the bus reaches Seattle — how the bus does this is the bigger issue.
Route 301? Route 355 served north Greenwood.
Yeah, the 301. I couldn’t remember the number.
It is hard to watch while traveling on the floating bridge, but at least the overall travel time between Downtown Seattle and Downtown Bellevue doesn’t get longer if not shorter sometimes, so I made piece of it.
The fact Link is so slow is a shame. Some people keep saying it’s because light rail is incompatible with commute travel. And that’s true. But if Link is going to cross the bridge or follow the freeway, they should be using faster trains at least for those segments. Trains should reliably travel over 75-80 mph.
Personally I think we need a proper Seattle subway and replace Link with fast commuter service. 2 Line focuses on Bellevue downtown, South Bellevue, Mercer Island, Chinatown, Westlake, Cap Hill, and UW. Interline high frequency automated local trains with a transfer.
Same for the 1 Line. Add a bypass line. That is technically the Sounder but the Sounder serves a different area and it doesn’t go north of Chinatown into high traveled places like Westlake and UW. And it misses the airport too.
… And it doesn’t run all day.
“As a weekend 554 rider the 556 is way better.”
Yeah, especially that the trade-off is that there’s a new express connection between Bellevue and Issaquah that runs all-day (since they’re both Eastside communities and makes more sense to connect them two rather than Seattle).
Now getting from Bellevue to Issaquah is easier! If I were to want to go to Issaquah before from Lynnwood I would have used Link and the 554 but since I don’t like transferring in Downtown Seattle now I can transfer in Bellevue!
You could also take the 535, but it takes almost an hour (57 minutes) just to get to Bellevue on weekdays. Link to South Bellevue takes about 50.
“You could also take the 535, but it takes almost an hour (57 minutes) just to get to Bellevue on weekdays.”
I did take the 535 to the 556 one time and it was better than the 2 Line to South Bellevue Station in my opinion. I just can’t help it with those awesome double deckers and how safe I feel in Downtown Bellevue compared to Downtown Seattle.
Long live the new 556 y’all, these guys shouldn’t grieve over losing the 554 and make a petition to restore it when the new 556 is better. At least they now have a connection to Bellevue instead of Seattle which makes more sense for an Eastside community like Issaquah. And if you want to continue to Seattle you get to ride a light rail on a floating bridge which no other countries have accomplished yet so these guys should feel lucky and be grateful on what they have.
Why did MI oppose the 80th st transfer point? Seems like half the bridges already taken up by up and landscaped sidewalk anyway.
“Why did MI oppose the 80th st transfer point?”
Because they wanted to avoid Sound Transit making it a regional hub (which they said they weren’t going to do but broke the promise). I guess that’s another reason why the 554 wasn’t truncated at Mercer Island Station. Of course some may take the 215/218/269 as a loophole but that’s a whole new level, Metro is less regional and more of a local agency over Sound Transit. Sound Transit was the one who made the promise anyway.
The city sued Sound Transit twice over the Crosslake Connection. The first was the residents of Mercer Islands expressing concern over losing the express lanes and the second was the regional hub problem.
I guess they want to be an isolated community (which they really should be). If Mercer Island were it’s own Link extension, Sound Transit would immediately forget about it. Since a stop in Mercer Island is necessary for a light rail line from Seattle to the Eastside (and it wouldn’t really make sense to bypass it), they’re just taking advantage of it and upzoning it with multi-family and mixed use developments. If it weren’t for that, I can assure the most comparable community to Mercer Island would be Medina or Clyde Hill.
“Why did MI oppose the 80th st transfer point?”
Because they wanted to avoid Sound Transit making it a regional hub
Yes, that is what happened. The leaders treat it as a provincial little island much like a gated community in the city. They want it both ways. They want the convenience of a light rail line (to Seattle, Bellevue and Redmond) but they don’t want buses from the east laying over on the island. (I guess no one cares about going from Mercer Island to Issaquah, Eastgate or Bellevue College.) To be clear, I’m sure some people don’t mind but lots of people (and lots of the representatives) pushed back.
Small nitpick: I think if you factor in waiting time, the new 556 is a huge win. Yea you gotta wait around at South Bellevue for a bit but now I’ll never have to wait in Issaquah or Seattle for an hour!
Sice link launched I’ve always found the new prevailing attitude of ” let then take link” to be annoying as it typically accompanies a large upheaval in bus service. the price of getting rail service has too often the decimation of bus services that may or may not have been very popular with ridership. The issues for 255 began long before covid when it was deemed link had to have the tunnel all to itself. .then it was banished completely from downtown. The added frequencies to compensate never materialized. Cutting rhe brickyard terminal probably made more sense. As things have developed now, a lot of places I need to go have become:
Bus to link
Transfer to train
Transfer back to a different bus.
Arrive destination at last
Much of this might be unique to my individual needs but it’s not a great user experience.
“The added frequencies to compensate never materialized.”
Evening service doubled, and weekend service may have increased too. But then then it lost some of it in the covid recession and driver shortage. The restructure didn’t cause that. If the 255 had still gone downtown, the reductions would have been even more drastic because it would be using more service hours.
To my surprise, the changes proposed for Renton after Stride keeps service largely the same. That means preserving commuter routes including the 102. But that also means nothing new. No connections to Link still (besides SODO), and no new Bellevue-oriented service besides Stride and Route 111. The East Link connections was by far the best update they did. South continues to not get much for commuting purposes. I think Metro views ridership down here as a poor people shuttle and prioritizes local service only.
The transfer penalty is very real, especially when connecting bus frequencies don’t line up cleanly with Link schedules. Forcing transfers before the full 2 Line actually connects across Lake Washington just adds unnecessary friction to an already long commute.
I’m in the having a plan B option, especially given the maintenance, accident, and other issues associated with Link. Put another way, “don’t put all of one’s eggs in the same basket,” which is what the region has done here, in part to bolster ST’s ridership numbers while allowing the individual transit agencies to redirect those hours elsewhere, too bad if the riders are inconvenienced. The latter seems to be a short-sighted response. I’d much rather have an alternative, even if it’s as infrequent to an hourly service in some cases, half-hourly in the case of Sea-Tac bound buses (“bring back the 194!”). While I’m also wishful thinking, I’d like to see Sound Transit run a bus line that follows the southern part of the Sounder route on weekends, when the Sounder does not operate, for there is, anecdotally, a fair amount of demand for Lakewood to Tacoma to Sumner to Auburn to Federal Way service on weekends.