Yesterday, Sound Transit’s Rider Experience and Operations Committee shared an update on the agency’s 2026 Service Plan. Along with aggregated community feedback on the previously shared ST Express service changes, the Sound Transit announced a pilot night-owl route between Sea-Tac Airport and downtown Seattle.
Last year, Sound Transit shared a proposal for three night-owl routes between Seattle and Everett, Lakewood, and Redmond. This proposal was well received with about 90% of survey respondents supporting the proposed routes. Negative feedback on the night-owl routes focused on travel times and the proposed routes not serving all stations.
Before the three night-owl routes are (likely) implemented in Fall 2026, Sound Transit will pilot a night-owl route between downtown Seattle and SeaTac Airport. The new route will provide service between SeaTac/Airport, Tukwila International Blvd, SODO, Stadium, International District / Chinatown, Pioneer Square, Symphony, and Westlake stations. Buses will run every 30 minutes between 12am and 4am.
This route will begin service on March 28 and connect late night/early morning travelers and airport employees with most of King County Metro’s night owl network.
This is an open thread.
I booked a super cheap flight 500a flight this summer. My plan was to take an assortment of routes between Ballard and the airport. But this night owl will certainly help. I just hope the connection between the D-line and the new route isn’t atrocious.
You should be able to make the connection downtown. I believe that the nightlink service will be serving the bus stops that ST Express normally serves during the day, sort of mirroring the 578/512/550 routing in that area. IDK if Metro will be timing them to the ST service, but night owl service runs frequently enough that it might not be an issue. The 44 to transfer at the freeway stop on 45th might also be another potential option for transfers, though maybe a tad longer (and sketchier imo) than downtown.
Sketchy? I’d expect there’d be nobody around near the freeway stops at night.
Sketchy as in the it’s dark, cars with drunk drivers like to crash into things, people coming off the freeway kind of way. The freeway stops are rather exposed, in my opinion. Also maybe not the most comfortable place to wait if you’re more vulnerable, but again where is at 2am? Plus u district isn’t the greatest at night in terms of people roaming around, again especially if you;re vulnerable.
This bus should be available all day, even if at lower frequencies. A trip to TIBS or SeaTac would easily be shortened by 10-15 mins from Stadium/SODO where you can transfer from Link.
Wouldn’t help; by the time you factor in local congestion around the airport, longer loading times for a bus (compared to a train), and longer wait times for a less frequent, less predictable service, all the time savings over Link would be gone.
TIBS is the answer. Riders could use use Link between TIBS and the airport. The bus approach gets mired in congestion. See former Route 194 and current routes 560 and 574.
my comment above related to the all-day point, not the night owl pilot. In the wee hours, the roadway is probably not congested.
I wonder is Stride 1 should be truncated at TIBS and not extend to/from Burien.
Right now, according to Google Maps, to go from Stadium to Seatac Stations:
Link: about 30 minutes
Driving: about 15 minutes
To guarantee that a rider saves at least 10 minutes, (versus staying on Link) you would need to run the express bus between Stadium and Seatac at a 5 minute headway- all day. I believe that is more frequent than any existing bus route, Metro or Sound Transit.
If this express bus terminated at TIBS, a Seatac-bound passenger would need to transfer a second time (with their luggage), making it nearly impossible to beat just staying on Link.
Would running a Stadium to TIBS/Seatac express bus at 5 minute headways all day be a good use of our transit funding?
Since most people don’t go to the airport very often, I wouldn’t put a high priority on making trips to the airport slightly faster, but if we’re going to do it, I’d improve Link service:
Short term: run the 1 Line at peak headways all day (every 8 minutes)
Medium term: expand the Link fleet and work with SDOT to get 5 minute headways on the 1 Line
Long term: Grade separate the Rainier Valley segment
Again, I’m not sure this is how I’d prioritize capital or operational spending, but it would benefit not only people going to/from the airport, but everyone using Link.
Tourists exist. They’d enjoy a lightning fast trip from the airport to Seattle over the sluggish Link. Traffic is the only thing that saves the Link, but that’s offset by the 10-15 walk to the Link station across the massive parking lot.
My guess is a high percentage of the people who take transit to the airport are workers. But it is the same idea. Most of these folks are shift workers. They have specific start and end times. The bus might be a bit faster but if it is running infrequently it doesn’t matter. I doubt you can get enough riders to justify a frequent bus. I think your other suggestions (for making Link more frequent) are good.
Trust me, outside of the average transit enthusiast like you all, nobody likes Link. Only people who can’t find parking, lives near a station, or relies on it to get into Seattle use it for the most part.
I’ve taken a few people on board and all of them complain it’s too slow and feels sketchy.
We got to add faster options. People want speed and reliability. Transit is supposed to be faster (or at least nearly as fast) than driving, and we’re probably one of the few countries that can’t even come close to the point it’s miserably slow.
No one wants to spend 40+ mins traveling 10 miles…. On a train…. when they already have to deal with transfers, wait time, walking time etc.
Tourists seem to me to be the least likely to use this- people who are mostly unfamiliar with the area and our transit systems seem unlikely to want to transfer (possibly twice) with their luggage, just to maybe save about 10 minutes under perfect conditions.
Ross, I take your point- the ideal user of this service would be someone who commutes to the airport daily from north of Stadium Station- but I’d want to see numbers for how any people are doing that. Whenever the topic of extending Link further south comes up, I’ve frequently seen the claim that airport workers tend to live in south King or Pierce counties, so I assumed that was true, but maybe there’s enough in central/north King to justify an express bus.
My general feeling is that most people (who don’t work at the airport) over value the airport as a destination in a transit network. Most people go to the airport at most a few times a year, but make trips to work/school, the grocery store, other shoping, visit family and friends, restaurants/bars/cafes, etc, just about every day. We should prioritize speeding up the trips that most people make frequently before we focus on speeding up trips that most people only make occasionally.
“My guess is a high percentage of the people who take transit to the airport are workers.”
I have seen workers use Link at SeaTac, but mostly I see travelers. It appears to be locals, especially ones seemingly headed to UW or I see visitors who bristle at the cost of Uber or Lyft. While Link takes longer, the wait is just a few minutes for a train any time before late night.
Link has a higher frequency to SeaTac than is provided at many other airports in the US or Canada with rail connections. The fare is also amazingly low compared to other places.
Consider that BART trains only leave SFO every 15 minutes or Pearson Express trains only leave every 15 minutes. Both charge a hefty premium fare too. We got it pretty good!
If ST started running buses for trips that took way longer than Link by driving on the freeway, there would be a massive fleet at a staggering operating cost and a huge loss of Link ridership. That’s especially true for many ST3 Link extensions. To me it only makes sense if there’s Link overcrowding.
Interestingly, 2026 will provide an inadvertent opportunity for ST to test for express bus vs Link trade offs. The decision to keep running express buses upon FWLE opening provides this. The-soon-to be-cancelled STX 515 has also been a test of sorts (although it’s officially provided to ease Link overcrowding).
I am not sure if the regulatory ramifications, but I do sometimes wonder if the Port should have remote terminals located around the region served by direct SeaTac buses. Those buses could have more room for luggage. It could have a different fare. It could be in a place like Colman Dock to serve Kitsap fliers or at King St. The cruise ships already fund something like this, I believe these have been tried elsewhere in decades past with mixed success.
Fliers are probably more concerned about reliability than travel time. A guarantee of 40 minutes is better than planning to ride a bus that may take anywhere from 25-50 minutes. Plus, there is a benefit to Link in that it does have a smooth ride and ample leg room when it’s not crammed. I seem more prone to motion sickness on a bus too. I’ve often taken the time on a Link train headed to SeaTac to text my neighbor that I’m leaving town or to check the weather at my destination or to confirm a needed car rental when I get there. I’ll also pull out a beanie or a hoodie or gloves while on the train back home when I’ve been somewhere warm. A little extra time on Link can be useful.
It is about time that ST did this. It will however likely take time to gain rider awareness/ confidence and then ridership boardings. In particular the service should offer schedules that coordinate reliable transfer meeting times since the 30-minute service will not be frequent. I hope that ST allows enough months of the service to get riders using it.
I will be curious how ST brands and promotes this service.
How about ST Express Ride Night Owl?
I like “NightLink” as D M used above.
“Take NightLink home from The Nightlife”
I like Nightlink, too. It’s clear what the service does.
To be honest, I don’t see why the bus can’t just take the freeway between International District station and SeaTac. The extra stops at Stadium, SODO, and TIBS station just add travel time but don’t help anyone. TIBS, in particular, the purpose of the station is to connect with other buses, but the only buses to connect with that run late at night are the 124 (which serves downtown and SeaTac anyway (*)) and the A line (which serves SeaTac anyway).
(*) King County Metro already extends late night trips on route 124 to serve SeaTac airport, even if the bus doesn’t do that during the daytime.
Good point. In the middle of the night, there’s little-to-no benefit of serving SODO or TIBS.
Well I see some potential in serving those stations. Idk why but it’s just that industrial vibe that seems like we’ll get some riders, I mean the only good thing to actually do in SODO is visit their McDonald’s, me and my friends went there once and was actually pleasant.
There are some bars within walking distance to SoDo, so it might help some people get home in the wee hours. Probably not many airport riders though.
Stadium and SODO don’t seem valuable to me but TIBS is reasonably busy late-night and the A/124 provide pretty poor service. I think it serves as a parking lot for airport workers.
The Greyhound Station is at Stadium Station. That will probably be a better ridership source than any of the Central Business District stations.
Given the options, where would transferring bus riders feel safest waiting for this bus southbound?
I can foresee this bus being more popular northbound, with a more comfortable wait at SeaTac Airport Station.
If ST decides to open a night owl route covering Beacon Hill, MLK, and the airport, leaving Stadium and SODO on the express night owl makes more sense to me.
If this route ends up staying basically as is, adding Capitol Hill would provide an even better place to wait for it, and a better place to lay over, with more food and restroom options.
Fair point about Greyhound being at Stadium, but looking at Greyhound’s schedules I see one regularly scheduled bus that arrives after end of service on Link (a 12:50 AM arrival from Portland). There are some buses that get in around 4 AM from Spokane, but those stop at King Street Station before the bus station which will always be a better transfer point that time of night.
Obviously, it’s Greyhound so you never know for sure what will happen and having service down there is important, but it’s something that could be accomplished by more night owl service on the 124 as well.
I’m pretty confident there’s zero Greyhound service after midnight. If there is, I don’t think a chunk of them are heading to SeaTac. But adding Capitol Hill is also a good idea. It would allow early flyers to catch their flight 5a and 6a flights.
The more obvious solution for Greyhound might be to move to a leased space at the airport, close to the station.
@Brent … moving Greyhound to outside of the city is a bad customer experience. It forces customers from all directions to take a train into the city. This especially doesn’t make sense for travelers coming from I-90. Plus, Greyhound/Flix added the U-Dist to their services. So the bus would still have to go through downtown regardless.
Uh… Anyone who has taken Greyhound knows that the schedules are not to be trusted. I used to regularly ride it between Olympia and Seattle late at night, and it was usually at least an hour late, often two hours late.
Christopher, Yeah, Greyhound sucks but it’s not like ST can just run buses to the Greyhound station at a time other than their scheduled arrival time. I think given that only one bus is reasonably going to be served by this service (the 10 PM from Portland arriving at 12:50 AM), it’s worth considering if other segments might be more critical to serve.
Brent, lots of Greyhound/Flix routes already stop at SeaTac. The better solution is for Greyhound to serve King Street Station.
“Uh… Anyone who has taken Greyhound knows that the schedules are not to be trusted.”
Yeah but does it make it sound less likely that people would take Greyhound to catch a late-night flight or the other way around?
I Certainly agree that having a choice of Amtrak trains, Amtrak buses, and Greyhound buses at King St would be a huge improvement for riders.
Something really needs to be done about the bus situation at Kung Street. It can take 10 minutes for the buses to exteact themselves from that mess.
Maybe they could put a lid over the BNSF and have that be the bus boarding area?
I question TIB too. The F doesn’t run those hours, and the A serves at SeaTac station. I struggle to see many park-n-riders at TIB using it. SODO and Stadium are of little value too. It may be for night-shift workers in SODO, since the 150 and 161 run late/early in north Kent for that reason (and I know two people who used them).
The last train to go through the DSTT leaves from SeaTac at 12:05 am currently. After that, the last train through the RV from SeaTac is at 12:57 am. ( It used to be a tad later.)
Once the full 2 Line is operating, ST may make sone changes to the last train from SeaTac. We don’t yet know what the new schedule looks like or if it will change the time of either last train.
ST also used to run the last train to SODO rather than just Beacon Hill. The addition of owl service could end up changing the 12-1 Link end point again if ST chooses to coordinate the last Link trains with the owl service.
Another consideration is that Greyhound is next to Stadium station. Intercity bus connectivity is important — even overnight.
So there are elements of connectivity that are offered by stopping at SODO and Stadium.
I also make this comment because owl service scheduling needs to be scheduled as a system rather than a stand-alone service. Today, the RV gets a last train almost an hour longer than Downtown and stations further north do. So when the owl service connects to Link at different times of night is important. For that matter, the transition between owl service and light rail service will need to be gradual rather than all at one exact time.
We have to hope that the train schedulers make travel as seamless as possible. But I think skipping more stations will make that harder to do.
There is only one Greyhound bus (a 12:50 am arrival from Portland) to serve that is scheduled to arrive or depart after end of service on Link. It would be better to serve the Greyhound Station via the 124 and send this night owl bus to Capitol Hill (and maybe U District if budget allows). But definitely agree that the night owl network needs better coordination.
The SODO/Stadium diversion is totally pointless. There are no buses to transfer to in the middle of the night! They should really run the bus all the way to U District via Capitol Hill, or least to Capitol Hill. Leaving out the two busiest late night destinations in the city in favor of serving bus transfer points with no buses is completely baffling to me.
I agree, that seems like the biggest miss to me.
Well, I don’t. Do you even know about the upcoming bus from Everett to Seattle at night or the 49? Yeah, I bet riding along Broadway at night when there’s no traffic versus running on I-5 which is getting a revive project which will cause delays is better. With that happening, maybe the night time route can be truncated at Northgate or U District? (if the I-5 work is still happening)
Sure, you can take the hourly (or worse) 49 that won’t be scheduled to interface with this route. Or ST could just run marginally more service hours to serve the central parts of Seattle’s nightlife (and the two densest neighborhoods). Feels like a better corridor to serve late at night than SODO-Stadium (even accounting for the Greyhound Station)
It will transfer to the 49 to Capitol Hill, but it’s unlikely to be time-coordinated.
@blumdrew… I just submitted your suggestion to ST.
Does anyone else think building Judkins Park station was a massive mistake? The walkshed it serves is pretty bad (the hint is in the name: ‘park’), plus it probably makes building a connection to Mt. Baker impossible due to the significantly worse curve geometry and grade change required to make the turn onto Rainier while avoiding I-90 infrastructure. That connection would have significantly improved the utility of the cross-lake line (rather than forcing an IDC transfer to get from Bellevue to Seatac) and would have allowed for a station in a much denser stretch of Rainier in between the I-90 tunnel portal and Mt. Baker.
I think it’s worth having. There is some housing there, plus a connection to bus routes 7, 8, and 48. The people who do use it, it will avoid a lot of backtracking.
In particular, reaching Bellevue from anywhere along route 7, it is faster to take route 7 to Judkins Park station vs. alternatives of taking route 7 to ID station or taking route 7 to Mt. Baker, followed by Link to ID station. Route 7 is a busy route, so this covers a lot of people. This in addition to routes 8 and 48.
Actually, I think the opposite. It should have been picked for the original Central Link routing rather than the at-grade alignment in SODO leading to the Beacon Hill tunnel (which caused someone to die). The I-90 alignment for Central Link would have been better because it would have been the final stop until the 1 Line and 2 Line part their ways, and it would have reduced Cross-Lake to only one station. Though it could have made a little bit of a delay.
Yeah, it is not a great station but it is definitely worth having. Ideally the train would run cut and cover starting there with another station (or two) between it and CID. But given the pathway ST chose — which saved a lot of money — it is about as good as you can get. As asdf2 mentioned, you have good connections with the buses and there are some apartments nearby. Rainier Valley/Central Area to the East Side (via the buses) seems like it is as important than East Side to TIBS, SeaTac, etc.
It is hard to see how the train could connect to Mount Baker and still go downtown. Were you thinking of a tunnel from the west end of the bridge to Mount Baker and then north again?
I’m not sure there’s much need for any additional stations between Judkins Park and C/ID. Most of the places to serve would be on Beacon Hill, and it would have been way more expensive to do a second Beacon Hill tunnel (though it would have gone almost directly under my apartment, and I’d probably have a station steps away from home). As it stands, I’ll have a 15 minute walk down the hill to catch a train to the east side, or I can take it to Chinatown and backtrack on the 36. Not really sure what will make more sense for me, though I imagine it will be “walk down the hill, take the bus on the way home”,
Going via Rainier/Jackson would have been too difficult to tie the line in at C/ID, though that’d be the best route for ridership purposes. Via Rainier/Dearborn, maybe there could have been a station at Rainier/Dearborn. That’d be a plus, but I’m not sure about cost/benefit. I guess it’d probably be a better investment than 90% of ST projects.
And I expect Judkins Park to be a fairly popular station. I am fairly sure that it will be faster for riders on the 7/106 to transfer here than it will be at Mount Baker (since the way via Beacon Hill and SODO has more stops and is physically longer). But unless there’s some significant changes to the freeway ramps, riders in the apartments north of I-90 will probably be better off on the 7 and transferring at C/ID if they are heading north of downtown.
Going via Rainier/Jackson would have been too difficult to tie the line in at C/ID, though that’d be the best route for ridership purposes.
Yeah, that is what I meant. What they chose wasn’t the best option from a rider standpoint but it was a lot cheaper. So much so that I get why they did it. I really have no qualms with the routing (or the station itself).
In contrast the lack of stations north of Westlake is a travesty. They skipped First Hill because of soil complications. OK, but at least put a station at 23rd & Madison. At least then riders could quickly transfer to what is now a bus running every six minutes (to go on Madison) or the 48 (which should run more often. The area around there isn’t as developed as First Hill but very little in the region is. Then they only had two stations in the U-District. A station on Campus Parkway would make a huge difference not only to people who would walk to that station but also the buses.
No, it’s a good station. C/ID is not very much backtracking for an eastside – SeaTac trip. It’s only three minutes longer to go from C/ID to Mount Baker via Link than it is to go from Judkins Park to Mount Baker via the 7/106, and an Eastside – SeaTac line that bypasses downtown doesn’t really seem to be a strong enough corridor for a dedicated line.
And the east portal of the Judkins Park station is probably one of the better walksheds in the system. There’s a lot of apartments on Rainer, some of whom are closer to the 23rd entrance than the Rainier one. Plus, there’s plenty of middle density stuff in the immediate vicinity. Plus the NW African American museum, an elementary school. And I think it’s good to have Link serve parks directly as well, though maybe that’s controversial. Judkins Park is really nice.
I think Judkins Park is a great station. It adds a good connection to the Eastside for much of RV/CD. Otherwise riders would have to head all the way to downtown to transfer to the 2 Line. The area has also exploded with midrise apartment buildings in the last few years in anticipation of the station.
Don’t you see? Catching a bus between Mt Baker and Judkins Park will be the future preferred path for transferring between the 1 and 2 Lines.
Mhm, Brent. Though it adds travel time if you’re coming from Stadium or SODO stations. Maybe also Beacon Hill Station, but it’s better to use this method if you’re coming from the SOUTH. Though you would need to race the 1 Line as it reaches IDC Station plus the 2 Line how long it takes from IDC to Judkins Park plus how long you need to cross to the other damn platform and wait for the 2 Line. What you said is actually kind of fire.
No, Judkins Park station is a really good idea and many of us are eagerly waiting for it. Initially it was just part of East Link and North King didn’t think it would get much benefit from it. But after Bellevue’s insistence on a downtown tunnel got it to beg North King to pay for the CID-Judkins Park segment and station, a couple years later more Seattle transit fans realized how good the station would actually be for North King.
It has entrances to transfer to both to the Rainier bus routes (7, 9, 106, future RapidRide R) and the 23rd routes (48, potentially 4). That will greatly improve Rainier Valley-Eastside trips It’s nonstop between Judkins Park and CID (the 1 Line has 3 intervening stations). That will attract some people on the 7 and 48 going to downtown or other 1 Line stations. It improves overall transit access in Rainier Valley. And the station area has been growing a lot with denser housing. It’s not just the park land.
“It should have been picked for the original Central Link routing rather than the at-grade alignment in SODO leading to the Beacon Hill tunnel ”
I don’t know what you’re talking about. It was never brought up in the 1990s when the 1 Line alignment was being decided. This is the first time I’ve heard of it.
The initial concept was to go around the northeast corner of Beacon Hill on the surface, via either Jackson-Rainier or Dearborn-Rainier (I’m not sure which). That was changed to SODO after Paul Allen asked for a Stadium station. So the closest alternative to yours was even more surface than what we have.
As far as massive Link stations go, I think it’s obviously not a mistake.
The east entrance will act like a new station. It will offer a new reasonable walkshed almost as far as Jackson and 23rd. It can be used to get to all of the CD and even UW if DSTT service is disrupted. Both Routes 48 and 8 will have a whole new market in 2 Line transfers since they don’t cross any Eastside ST Express route today.
It will also get more use than many other stations. Consider how the station ranked #5 of 12 in forecasted boardings on the East Link segment of the 2 Line in 2040 as forecasted a few years ago:
https://seattletransitblog.com/2020/01/27/sound-transits-station-ridership-in-2040/
(Note: The full East Link opening soon will make for an interesting check on the overall quality of ST ridership forecasts.)
That’s on top of serving the new residential buildings in various stages of opening within walking distance of the station. There has been something like over 3,000 new units added in the area. Even though the area has a history of frequent bus service, the Link station has played a big role in making building them more financially desirable for a developer to build.
All of this is before any sort of Metro bus restructuring done to offer direct Link service between First Hill medical offices and the station (especially Harborview and Swedish Cherry Hill) which some of us wish for (beyond Route 630).
In so many ways, this station offers so much more than Pinehurst ever will — from offering a bigger (having a two entrance or “barbell” walkshed as opposed to having one entrance or “circular” walkshed) and much denser walkshed to offering many more Metro buses per hour to transfer to. It may even end up having higher ridership than other RV stations after several months if not in the opening month.
I think the point is mainly the transfer.
I wouldn’t be surprised if Judkins park ends up being in the top 3 stations exclusive to the 2 line. Unlike the Eastside, the majority of residents around Judkins park are probably either car-free or car-light.
ST be like: I’m oblivious of the 124 serving SeaTac/Airport Station at night.
Seriously!? Another pilot project that will only bring money down the drain, they should be giving money to King County Metro to improve the 124 and be aware of what they currently have and make a change for once! Though I get that many people endorse this proposal, I definitely don’t. The 124 would serve Downtown Seattle and the airport better than this service ever could! Though I do support night owl service to the airport from Downtown. We seriously don’t have to do the hard way.
The 124 night owl headway is over an hour. That’s not much to work with.
I explained this in my proposal for new night service.
ST’s thinking is this is the first phase of a route to Lakewood, so it’s putting the same stops on it that that route would have. People coming from Lakewood would not want the travel time of going on East Marginal Way and Georgetown, and that’s not typical of an ST Express route.
I wonder if there’s an opportunity to sync all the night owl buses for a mass transfer somewhere in downtown Seattle. Nobody wants to wait 30-60 minutes to transfer in the middle of the night, and most nighttime transit security could be centralized at a single point.
Looking at the night owl bus map, 3rd/Madison would serve almost all routes. A secondary sync point at U District might be good as well.
https://seattletransitmap.com/app/
Metro used to do this back in the day. The night owl buses would all arrive and depart 3rd & Pike/Union within 15 minutes of each other.
Doesn’t work for crosstown night routes like the 67 or 44.
Blumdrew suggested starting the NightOwl in the U-Dist, which I support. If this were to happen, it would connect with these routes.
Did they sit for a few minutes to allow for an all-way transfer (“pulse”)? I’m suggesting the buses sit so that all the infrequent service through downtown can connect to each other:
https://humantransit.org/2010/11/basics-finding-your-pulse.html
Yes, they would all arrive around 155a, 200a. Then leave at 215a. Then again around 310-315 and leave at 330a. Today’s night owl routes are the 7, 49, 70, C, D, E (not sure if there are others). I do not believe they are coordinated.
As I recall, they did pulse. They all stopped for 5 minutes at Pike.
4th & Union.
If the shadow Link bus doesn’t serve all stations, it has less utility, fewer riders, and lower fare revenue. What ST has done is create an express service exclusively for last resort riders who can’t use it.
Metro and Sound Transit don’t even try with transfers anymore.
For low frequency routes, it’s literally impossible to find a good transfer from Link or an express bus to a local LOW FREQ route that is only a few minutes, except if you’re lucky at a random time.
Route 148 for example. What’s even the use? It runs every 30 mins only and almost always leaves a couple mins before Route 101/560/566 arrives at Renton TC. Most of the time it’s like 15-20 mins late but doesn’t matter when some of them arrive on time or early… there’s zero predictability. Then who is supposed to use it? Only people who go to Renton? The buses are always empty, I wonder why. There are riders, just not riders willing to wait 20-30 mins for it in the cold.
Same thing I can’t transfer between the 150/906 without forcing myself to wait 15 minutes. There’s very few trips with a transfer window that is normal despite the stops being the exact same!
I don’t think they care about transfers.. I’ve emailed them and they haven’t updated it in the recent schedule change.
For the most part the 148 turned into a homeless / drug use express..waste of money and might as well be deleted. Always only 2-3 people in the 35 foot buses that already have to continue through Route 107 as well.
Timing transfers is tricky. If a bus runs every 10 minutes it’s wasteful to make a bus pause for 5 just to allow for time to connect with another bus. However, it seems almost essential when buses run only every 30 minutes or especially less.
The tough choice is with the 15 or 20 minute frequency category. (Note buses are often less frequent after 7 pm.) Or when one bus route is frequent and the other one is not.
Maybe I’m too technologically unrealistic, but I’d like to be able to “register” a transit trip somehow while on Link to tell a bus driver that I’m coming — so that I don’t have to worry that I’ll miss my bus by getting there a few seconds late. That worry is often an issue compounded by the long crosswalk waits (sometimes two crosswalk waits) on MLK.
I like this starter line!
ST may have taken the feedback to heart to have the airport be the terminus for night owls from both the north and south.
I hope this becomes the south end of the proposed night owl Link shadow from Everett or Lynnwood. Having to transfer downtown in the middle of the night would kill ridership.
Here’s my proposal for a new overnight system in South King:
A Line:
Truncate the line north at SeaTac/Airport nights only (every 60 minutes).
124:
Increase overnight service (every 60 minutes). No changes to the routing.
126:
New overnight route from Downtown Seattle to SeaTac/Airport, it would duplicate the 124 from it’s northern terminus to SODO Station, where it would then turn east to Beacon Hill, and take S McClellan St to Mount Baker. The route would then turn south onto MLK to serve the at-grade stations along this corridor, and turn west onto S Boeing Access Road. At this point the rest of the route would follow the 124 all the way to SeaTac/Airport. The route would run every 60 minutes like the 124, doubling frequency to every 30 minutes on portions that overlap (which is what ST proposes). Most of the overlapping segments (and stations) will be covered by this “new overnight route” between Downtown Seattle and SeaTac/Airport.
594:
Add overnight service (every 60 minutes), and have it stop at Kent Des Moines, Star Lake, and Federal Way Downtown nights only.
I tried to be “as simple” as possible. No extra stuff, though the biggest change is probably adding the 126.
The 124 and 126 would combine service every 30 minutes. The A and 594 would also combine service every 30 minutes. This proposal also abandons the new ST overnight route from Downtown Seattle to SeaTac/Airport and the proposed night owl route from Seattle to Lakewood in the Annual Service Plan, see Sound Transit’s website for more information.
Sound Transit can’t order Metro to do things. You’d need to convince Metro to do a Metro restructure. If Metro were inclined to and had nothing else to do, it would take several months of planning and hearings and scheduling a county council vote on it. ST can offer Metro money for extra service, but Metro wouldn’t take kindly to ST telling it to reroute several routes and ST knows best what they should be. The council would say service changes should be on the twice-yearly days, March and September, which are now simultaneous with several agencies.
In any case it’s moot, because ST has already decided to do the pilot in March, and that’s just two months away. Too late for Metro to do anything in response to it. The next opportunity is September. And Metro might say, “It’s a pilot. We don’t change things around a pilot that may only be there for six months or a year.”
Then there’s the issue of where the A would lay over. It has space at TIB. It can’t even park or turn around on Intl Blvd.
It would lay over on the terminal, Mike. Also this is a restructure for the future, not for March. Also I’m not saying that Sound Transit is going to order Metro to do things. I’m just saying is that Metro could POSSIBLY run night service without being ordered (doing it by reaching out) like how they did back in the day, and Sound Transit only runs the 594 at night, stopping at Federal Way Downtown, Star Lake, and Kent Des Moines. Though like I said it could be replaced with the idea of having the 574 operate more at night, but since I’m kind of worried about this pilot idea, this is a future restructure that will depend on the outcome of the pilot.
“It would lay over on the terminal”
At the airport terminal? TIB is just a mile away.
Yes, though I didn’t have it end at TIB because of the 124.
We don’t need a new route complicating the network. We don’t need express service picking and choosing Link stations (and thus riders). And :30 frequency isn’t enough given the insane amount of revenue ST collects.
Just run the existing Route 97 shuttle overnight, every night, to all stops, at headways closer to :20. Run trains until 2:30am Friday and Saturday nights.
I agree almost with the whole thing, though the only thing I disagree with is 20 minute service. 20 minutes is too frequent for night time, and I think it’s just fine being average Community Transit local route standard frequency. I mean CT does a good job operating their buses half hourly in the day, then why doesn’t ST do this at night… You know, when ridership is quieter? I’m 50-50 with Friday/Saturday night service being until 2:30, and using the 97 number is kind of clever, nice job.
“I mean CT does a good job operating their buses half hourly in the day”
They should be every 15 minutes to make them more usable. Swift should be 10 minutes on weekends.
“then why doesn’t ST do this at night”
Because late-night is not daytime.
Using Route 97 should be obvious. Never add more complication than necessary.
We seem to be forgetting both the phenomenon of drunk driving and the disproportionately minority and low-income people who work late nights and early mornings. They’re stuck in cars now.
This isn’t Metro; it’s ST with $54B. On a temporary pilot. Serving a handful of stations on less than half of one line. They can afford :20 headways.
Yeah, but night service for ST (I mean SeaTac/Airport) should be like ridership on a CT 30 minute local route. I would not expect anyone to want higher frequencies such as 20 minutes which for me is way too frequent for night service. My proposal combines 30 minute service for connecting destinations. I agree, Mike. There should be 10 minute service on Swift routes on weekends, and I would also like to see 15 minute frequencies on local routes. Though CT just doesn’t have the audacity to do so.
Uh, Jon. Actually using the 97 is just or when there’s a disruption or something, not for late nights or anything like that. Route 97 was never intended to be operated on a full schedule. I now disagree with the fact of running the 97 late at night because of what I said, and its path is different from the overnight route(s). In your second paragraph, I have no idea what you really wanted to say. And what you said about Metro and ST, ST is facing a BUDGET SHORTFALL, so let’s have Metro run the routes (with the exception of night service on the 594). They CAN afford 20 minute frequencies, but what you’re missing is if anybody would really take it that often, so you might want to reconsider and learn from me.
Generically, I would rather have longer layovers for timed connections at 30 minutes than random arrivals at 20 minutes.
Random arrivals are fine for a 10 or maybe even a 15 minute frequency. But just missing a bus transfer and waiting another 20 minutes for the next bus is a really awful experience in the middle of the night.
And it will feel much safer sitting on an idling bus with a driver at a meeting point (waiting for connecting buses to arrive) than standing outdoors on the street in the middle of the night.
Nowadays, real-time bus arrival apps can let riders know when to start their journey. But bad timing at the transfer point isn’t something that a rider can adjust to at the onset of the trip.
The 97 should run every night, to all stations, funded by ST. An original intention doesn’t matter (it certainly doesn’t to ST) and definitely isn’t an excuse to complicate the transit network.
Metro has tight budgets and lacks extra money. ST is heavily overfunded but poorly run. That’s not the public’s fault, especially since ST consistently ignores us. Metro already operates the 97, and Link. ST only operates the T Line.
Yeah, if you serve 8 stations, you won’t get much ridership. It’s a very stupid idea of the kind we routinely see from ST. Another thoughtless token from Dow for the FIFA games.
You expect people to wait 30 minutes for a trip that takes less time than that? That would only apply to last resort riders, who aren’t likely to need trips between SeaTac and downtown.
Yes, again, to serve 8 stops, for 4 hours a day, for a couple months, with the extreme amounts of money ST already collects and wastes ($1B on parking in ST3, 1700 stall parking garage at Lynnwood, precluding driverless trains on all lines…), they absolutely can and should fund :20 headways for the pilot. Or is the point just to run a dumb service, so it’s lightly used, to argue there’s no need for it? If CT’s :30 routes need more frequency, as I’ve often said they do, so does this pilot.
If we believed in Vision Zero, we’d take meaningful steps to get drunk drivers out of cars. That includes later weekend Link service. That should’ve started at least with U Link in 2016. If we prioritized race and social justice, we’d provide useful transit service to people who work overnight instead of relegating them to cars. That means regular, predictable overnight service that skips no stops.
“If we believed in Vision Zero, we’d take meaningful steps to get drunk drivers out of cars.”
OMG! You mean street design isn’t the biggest reason for collisions! Tell that to SDOT! Speed cushions every block and no right turn on red signs won’t make injury collisions stop!
Relatedly, I sure see lots of distracted drivers with their cell phones and glaring vehicle touch screens too. (Allowing Teslas that require that drivers take their eyes off the road to see any info should be downright illegal.) It’s not all drunks.
> You mean street design isn’t the biggest reason for collisions!
That’s not what Jon said at all, and clearly not what they meant.
@ Nathan:
It’s just an aside — rooted in how putting in speed cushions and no right turn on red signs on arterial streets with no major collision history doesn’t reduce collisions. If it’s already 0 it can’t be reduced further.
His suggestion is that public money could be spent on keeping drunk drivers off the road by offering late night service. After all, Google tells me that 24 percent of fatalities in Seattle in 2024 were DUIs and 30 percent statewide. Offering better late night transit service would appear to me to get further in achieving the Vision Zero goal than introducing new street treatments on street segments and intersections where the Vision Zero goal has been met all along.
Jon is clearly missing the whole point of this project, it’s a PILOT project, not a permanent project. 20 minutes for a pilot project just seems like too much, you need 30 minute frequencies, if crowding occurs or if the ridership is close to booming then you can increase it to 15 or 20 minutes, that way you can actually test the waters rather than jumping to temptations. Did you even see my proposal? I’ll recap in case you didn’t see it:
– The A Line would be truncated north at SeaTac/Airport Station only at night.
– The 124 would have more night trips (that run to SeaTac/Airport).
– The 126 would be a new overnight route (fully operated by Metro) from Downtown Seattle to SeaTac/Airport every 60 minutes, it would combine service every 30 minutes along the portions that overlap with the 124. It follows the same path, but at Holgate/Airport, it doesn’t turn on Airport and continues on Holgate (which becomes Beacon Ave), after serving Beacon Hill Station it turns east on McClellan St to serve Mount Baker Station. After serving Mount Baker Station it heads south on MLK to serve the Rainier Valley stations, AFTER that it turns west onto Boeing Access Road and when it reaches Marginal Way it overlaps the 124 all the way to the southern terminus. This actually serves all light rail stations (with the exception of SODO) from Symphony to SeaTac/Airport.
– The 594 would have more night service (running hourly), and at night would serve Federal Way Downtown, Star Lake, and Kent Des Moines (like the 574). Though this idea could possibly be replaced by running the 574 more often at night (and making people at SeaTac/Airport transfer), there are many pros and cons of my options.
Feedback is always welcome. Feel free to express your opinion. I updated the proposal a little bit.
I’m curious just how many drunk drivers late night bus service would prevent. I think the answer is definitely more than zero, but exactly how much isn’t clear. You can use the late-night bus ridership as an upper bound, but not everybody on the bus is coming from drinking, and not everybody that is coming from from drinking would have driven had the bus not been running (many would have taken Uber, for example).
And, then, of course, there’s the route. I would imagine the night routes that would be the most cost-effective of getting drunk drivers off the roads would be short-distance frequent-stop routes in central Seattle, not the freeway expresses that Sound Transit likes to run. And, of course, if the person is getting into their car at a park and ride lot, they’re still driving the rest of the way home drunk anyway.
Here’s some data, asdf2:
https://wtsc.wa.gov/dashboards/fatal-crash-dashboard/
You’ll find that 39 percent of Seattle fatalities were from impaired driving in the last decade.
You’ll find that about 43 percent of all fatal crashes (133 of 308) in Seattle in the last decade were between 9 pm and 6 am. That’s very remarkable given how many fewer cars are on the road at that time. If you filter for that time period, you’ll find that 51 percent of fatalities were impaired drivers.
Sadly, you’ll see that fatalities are up since Covid — and the bigger roll out of the Vision Zero effort. It was 40 in 2024.
You’ll sadly also see that about half of fatalities were non-motorists.
It’s true that it’s hard to assume that more transit service could have prevented the fatalities. But one can’t deny that the stats show that late night drunk driving is a significant cause of Seattle traffic fatalities.
And you’ll find no mention of drunk or impaired driving in SDOT’s report here:
https://www.seattle.gov/documents/departments/sdot/about/documentlibrary/reports/2024_traffic_report.pdf
@Al S.
You’re speaking as though people don’t generally support better traffic enforcement. It’s not an either-or situation; we can have better traffic enforcement in addition to better road design.
SDOT doesn’t have jurisdiction on enforcement, except with automated traffic cameras. They’ve been a bit more aggressive with those recently but my impression is that they are politically unpopular so they’re hard to expand outside of school zones
That’s not it at all, jd. I’m saying that SDOT needs to think way way beyond street geometry and signage if they are sincere about reducing fatalities. Changing street geometry isn’t the only strategy available, and it seems to do little to address the more significant collision factors.
There’s an old adage that if you have a hammer, everything looks like a nail. Changing signs and pavement are things that SDOT staff can unilaterally do. So rather install one or two, suddenly an arterial may have four new speed cushions! Or installing no turn on red signs at intersections where almost no one turns and there’s never been a pedestrian collision.
I personally think the nighttime collision increase is somewhat due to the increasing popularity of LED lights for street lighting and for headlights. Even though LEDs can technically illuminate the street more cheaply and brightly, the difference in brightness between the source and the area lit creates temporary night blindness for a driver’s field of vision. These accidents are occurring in the dark when there are fewer cars for a reason!
But certainly impaired driving is another major factor in nighttime accidents. SDOT could do different things with its budget like subsidize additional transit at night or make transit free between 1 and 2 am — and get bars to promote a transit option, for example. Or the City could require bars to stay open at least an hour after the last drink is served so patrons could sober up before leaving.
Certainly school zones can feel dangerous. However school zone congestion is driven partly if not mostly by the manic driver behavior of parents going to and from that very school. Parents picking up kids are sone of the most inconsiderate drivers I’ve ever seen! I’m a big proponent for things like “walking school bus” programs to get kids active and to ease a parent’s (possibly misguided) fear of their kids not returning home safely unless the parent picks them up at the school’s front door.
WSP seems to often be on the lookout for drunk drivers at night. In contrast, I rarely see SPD out late at night observing or pulling over drivers.
Again, signage and pavement are tools — but are just a few of many tools to get to zero. Buying more of the same tool doesn’t help when other tools seem more needed but not ever used. Bringing four screwdrivers rather than one doesn’t address the problem when you need a wrench or a move evenly illuminated environment or tool use supervision and training.
@Al S.
I am saying that SDOT does not talk about enforcement because it is not within their jurisdiction. In this case they literally only have the hammer; they don’t have any other tools.
My understanding is that SPD knows Seattle wants more traffic enforcement but they do not have the capacity, though more recently they’ve been picking it up a bit:
https://spdblotter.seattle.gov/2025/12/03/seattle-police-hitting-streets-this-month-watching-for-impaired-drivers/
Thinking about perceived rider safety when waiting for a transfer, I think that a major transfer point may need overnight security assigned to it. Well-timed connections are most important — but I could see how having a bustling hub of people in one spot standing along a street at predictable times each night would be a magnet for attracting types of people that could make overnight transfers more uncomfortable for those waiting riders — especially if some look like visitors with luggage. It could make a big difference in how desirable the owl service will be. All it takes is a few unfortunate encounters in the news to scare away overnight riders.
I see 4th and Jackson as a good place for that. Overnight buses from all directions could converge there. It’s where owl service from both 1 and 2 Lines can meet. While not immediately adjacent to a freeway ramp, it’s near the I-90 and I-5 interchange, which is the regional freeway crossroads for Seattle. Stadium station could offer similar functionality but the added presence of having Union Station’s comfort stations and the proximity to King Street station would seem to favor 4th and Jackson.
If you want a single meeting point for all night owl service (other than the Westlake Station area), make it somewhere useful like Harborview. It’s not that far to the James Street ramps to I-5 and the Columbia Street pathway, though the buses would get a workout coming up the hill. The trolley lines can even get there.
I appreciate y’all keeping an eye on Pierce County.
Updating update:
https://www.thenewstribune.com/news/local/article314352823.html
“One of the pieces of rail needed for construction is not correct for the situation, Jackson said. A new piece needs to be prefabricated, which in Jackson’s experience can “take a long time.” There is no new date for when construction will begin.”
I rode the Federal Way Link extension this afternoon, for the first time (yeah, finally…) since it opened. Fun! Some thoughts –
* Coming into Federal Way Station, there were 20 people in my train car. If this’s at all representative, that’s 80 people every ~10 minutes on a Saturday afternoon, plus everyone on the still-running 578 bus. Federal Way needed good transit!
* Coming back north out of Federal Way, almost every seat was taken. But that’s unrepresentative because of the Seahawks game this evening.
* One of those sports fans asked me whether this train was the train to Seattle. (“Every train goes to Seattle here,” I answered.) Do we need better wayfinding?
* Hardly anyone got on or off at Star Lake. Hopefully there’ll be some more development there soon.
* The Commons at Federal Way looks like it’s in a bad state. Hopefully the new development around the station will reinvigorate it?
* Also, on the 560 bus south to the train, the run through Renton really does suck up a lot of time. I see why Renton wants to move its transit center… though unfortunately the lack of HOV ramps means the upcoming new one isn’t going to be much better.
* Speaking of the 560 – we need better connections. The (hourly!) 560 pulled out of Bellevue TC just as I could see an East Link train heading in to the station. (I’d been on the previous one.) Hourly buses should be given protected transfers to and from trains!
Yeah it’s kind of absurd how almost every bus does this. They pull out of the TC before the main connection comes by.
I’ve complained about this for Renton TC, where I can’t connect to any bus reliably with the 101 since every bus leaves right before it gets there. And this is the scheduled time. Extremely ridiculous. If it’s 5-10 mins after whatever, but why leave a minute before especially when the next bus is only 30+ mins later? Maybe have the buses wait at least and push out the schedule.
Basically forces us to take an earlier train /bus and wait at the TC for extra for no good reason. And if your connecting bus is late on top of that it’s even worse.
Write a complaint note to Metro? I’ll be writing one to Sound Transit, I think. I don’t see why the 148 would be timed with any other route, so maybe you’ll get some success there.
The biggest problem is frequency. If every Renton route ran every 15 minutes, the average to worst wait would be 7-15 minutes rather than 15-30 minutes.
Timing bus transfers is not realistic when bus routes are vulnerable to traffic congestion and missing stoplights that Metro has no control over. It can work in a rural area with little congestion, but not in a city like Renton with typical suburban congestion; e.g., in downtown Renton and west of it there’s an oversupply of cars queuing up to get on/off highways.
The second-biggest problem in Renton is too many routes terminate at Renton TC. The vast majority of Renton residents live east or south of downtown Renton but their major destinations are west and north of it (Seattle, Southcenter, Bellevue, airport). So more routes should continue through downtown Renton to one of these or something like that. A route like the 105 is very short: there’s no reason for it to be that short. The only non-neighborhood destinations on it are at downtown Renton, and there are few of those. So more people than necessary need to transfer for a lot of their needs. This is an artificial problem the transit network creates.
Metro plan to fix this long-term by combining it with part of the 106, a Renton Highlands-Rainier Beach route. That would have a transfer to 10-minute Link to all of Link’s destinations. It would also hopefully fill in its 30-minute periods, although Metro has been hedging on what the frequent span of “Frequent” routes would be. Some have also suggested extending the 160 (future RapidRide I) to Rainier Beach. But all this is a decade or more in the future even though the need is now. Metro and the governments just aren’t urgent enough about addressing these issues soon.
I’ve traditionally hated routes like the 102 that give a one-seat ride to one arbitrary neighborhood while other neighborhoods are left out. But something needs to be done to address Renton’s travel-time issues, so making the 102 all-day replacing the 101 might not be such a bad idea after all. If reform is done to both the 101/102, 105, and 160/I, that would give three parts of Renton better transit access to areas beyond just Renton.
The practical barrier is service hours. Metro needs more service hours to address congenstion unreliability and increase frequency. This has also been lagging. Metro defined a long-range plan in 2016, and the county has been talking for nine years about a countywide Metro levy to implement it but it still hasn’t happened. Seattle stepped up with its own levy to fill part of the gap, but other cities’ contributions haven’t been more than one route here and there (e.g., the Des Moines community shuttle). I really wish there would be more prioritization on getting all core routes up to 15 minutes full time. This isn’t the 1980s anymore or North Bend.
I’d like to see a 101-105 in particular. The 105 pulls very good ridership for a short, infrequent, suburban route (better productivity than the 101!) and I think combining the two would add a lot of value for both routes
Moving the TC to South Renton would unfortunately be a big detour for a theoretical 101-105 though.
I think the 105 could extend to Rainier Beach Station. The 101 can be changed to its old routing that goes from the old TC to the new TC, then back up to MLK way onto I-5. This would fix the transfer issue, and also riders from the old TC area can transfer to Link or the 101 pretty easily.
Also about Mike Orrs comment, I understand lining up transfers is tricky but at the very minimum the scheduled time could at least try. If there’s traffic and it gets messed up then that may be only occasional.
But I kid you not, the scheduled times for the 148 leaving Renton is always 1-2 mins before the 101 arriving in Renton. At least make it 2-3 mins, or 5-10 mins to be safe from late 101 buses especially on peak? The buses are mostly somewhat on time. The 148 can be pretty absurdly late sometimes though, but it’s fairly trackable by finding the matching bus running the 107.
About the 102 serving an arbitrary neighborhood, I completely agree. I think they could make a small P&R in Fairwood instead so at least people from various neighborhoods can catch the bus. But overall the routing actually gets most of the apartments in Fairwood (it has more of those than you would think), they chose it pretty specifically.
The 101 also needs a bus lane on the on ramp from I-5 into SR 900. It’s a simple change but will make the buses fare far better. That’s the main point of congestion, in addition to some of the traffic signals in the main city itself.